Fresenius Environmental Bulletin Founded Jointly by F. Korte and F

FEB – Fresenius Environmental Bulletin founded jointly by F. Korte and F. Coulston

Production by PSP – Parlar Scientific Publications, Angerstr. 12, 85354 Freising, Germany in cooperation with Lehrstuhl für Chemisch-Technische Analyse und Lebensmitteltechnologie, Technische Universität München, 85350 Freising - Weihenstephan, Germany Copyright © by PSP – Parlar Scientific Publications, Angerstr. 12, 85354 Freising, Germany. All rights are reserved, especially the right to translate into foreign language. No part of the journal may be reproduced in any form- through photocopying, microfilming or other processes- or converted to a machine language, especially for data processing equipment- without the written permission of the publisher. The rights of reproduction by lecture, radio and television transmission, magnetic sound recording or similar means are also reserved.

Printed in GERMANY – ISSN 1018-4619

FEB - EDITORIAL BOARD Environmental Toxicology: Prof. Dr. H. Greim Chief Editor: Senatskomm. d. DFG z. Prüfung gesundheitsschädl. Arbeitsstoffe TU München, 85350 Freising-Weihenstephan, Germany Prof. Dr. H. Parlar Institut für Lebensmitteltechnologie und Analytische Chemie Prof. Dr. A. Kettrup TU München - 85350 Freising-Weihenstephan, Germany Institut für Lebensmitteltechnologie und Analytische Chemie e-mail: [email protected] TU München - 85350 Freising-Weihenstephan, Germany

FEB - ADVISORY BOARD Co-Editors: Environmental Analytical Chemistry: Environmental Analytical Chemistry: K. Ballschmitter, D - K. Bester, D - K. Fischer, D - R. Kallenborn, N Dr. D. Kotzias D.C.G. Muir, CAN - R. Niessner, D - W. Vetter, D – R. Spaccini, I Via Germania 29 21027 Barza (Va) ITALY Environmental Proteomic and Biology:

D. Adelung, D - G.I. Kvesitadze, GEOR A. Reichlmayr-Lais, D - C. Steinberg, D Environmental Proteomic and Biology:

Environmental Chemistry: Prof. Dr. A. Görg Fachgebiet Proteomik J.P. Lay, D - J. Burhenne, D - S. Nitz, D - R. Kreuzig, D TU München - 85350 Freising-Weihenstephan, Germany D. L. Swackhammer, U.S.A. - R. Zepp, U.S.A. – T. Alpay, TR V. Librando; I

Prof. Dr. A. Piccolo Università di Napoli “Frederico II”, Environmental Management: Dipto. Di Scienze Chimico-Agrarie Via Università 100, 80055 Portici (Napoli), Italy L.O. Ruzo, U.S.A - U. Schlottmann, D

Prof. Dr. G. Schüürmann UFZ-Umweltforschungszentrum, Environmental Toxicology: Sektion Chemische Ökotoxikologie Leipzig-Halle GmbH, K.-W. Schramm, D - H. Frank, D - D. Schulz-Jander, U.S.A. - Permoserstr.15, 04318 Leipzig, Germany H.U. Wolf, D – M. McLachlan, S

Environmental Chemistry: Prof. Dr. M. Bahadir Managing Editor: Institut für Ökologische Chemie und Abfallanalytik TU Braunschweig Dr. G. Leupold Hagenring 30, 38106 Braunschweig, Germany Editorial Chief-Officer:

Prof. Dr. M. Spiteller Selma Parlar Institut für Umweltforschung Universität Dortmund PSP- Parlar Scientific Publications Otto-Hahn-Str. 6, 44221 Dortmund, Germany Angerstr.12, 85354 Freising, Germany

e-mail: [email protected] - www.psp-parlar.de Prof. Dr. Ivan Holoubek RECETOX_TOCOEN Marketing Chief Manager: Kamenice 126/3, 62500 Brno, Czech Republic Max-Josef Kirchmaier MASELL-Agency for Marketing & Communication, Public-Rela- tions Environmental Management: Angerstr.12, 85354 Freising, Germany Dr. H. Schlesing e-mail: [email protected] - www.masell.com

Secretary General, EARTO, Rue de Luxembourg,3, 1000 Brussels, Belgium

Prof. Dr. F. Vosniakos T.E.I. of Thessaloniki, Applied Physics Lab. Abstracted/ indexed in: Biology & Environmental Sciences, P.O. Box 14561, 54101 Thessaloniki, Greece BIOSIS, C.A.B. International, Cambridge Scientific Abstracts, Chemical Abstracts, Current Awareness, Current Contents/ Agricul- Dr. K.I. Nikolaou ture, CSA Civil Engineering Abstracts, CSA Mechanical & Trans- Organization of the Master Plan & portation Engineering, IBIDS database, Information Ventures, NISC, Environmental Protection of Thessaloniki (OMPEPT) Research Alert, Science Citation Index Expanded (SCI Expanded), 54636 Thessaloniki, Greece SciSearch, Selected Water Resources Abstracts

0 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

CONTENTS

ORIGINAL PAPERS

CHARACTERISTICS OF TOTAL DISSOLVED IRON OUTPUT AND 1169 ITS SPECIES DURING DIFFERENT FLOOD EVENTS IN THE SONGHUA RIVER Jiunian Guan, Baixing Yan, Yingying Xu, Lixia Wang and Hui Zhu

IRRIGATION WATER MANAGEMENT IN THE JORDAN VALLEY UNDER WATER SCARCITY 1176 Abbas S. Al-Omari, Emad K. Al- Karablieh, Zain M. Al-Houri, Amer Z. Salman and Radwan A. Al-Weshah

ANTHROPOGENIC CHARACTERISTICS AND CONSERVATION STATUS OF THE 1189 VASCULAR FLORA OF KOZAN CASTLE AND ITS SURROUNDING AREA (TURKEY) Necattin Türkmen, Atabay Düzenli, Havva Karakuş and Medine M. Uma

RESISTANCE TO THE RECLAMATION OF ENVIRONMENTALLY SENSITIVE 1195 AREAS THROUGH THE ESTABLISHMENT OF A NEW FOREST ECOSYSTEM Turgay Dindaroğlu

DETERMINING THE TOTAL ANTIOXIDANT STATUS AND OXIDATIVE STRESS INDEXES OF 1204 HONEY SAMPLES OBTAINED FROM DIFFERENT PHYTOGEOGRAPHICAL REGIONS IN TURKEY Ethem Akyol, Zeliha Selamoglu, Hamide Dogan, Hasan Akgul and Adnan Unalan

TRANSFORMATION OF NITROGEN IN BIOFILTERS USED IN DRINKING WATER TREATMENT 1209 Huining Zhang, Huihui Gan, Huixa Jin and Kefeng Zhang

LIFE CYCLE ASSESSMENT OF WASTE TIRE PYROLYSIS 1215 Müfide Banar

PHYSICAL AND WATER PROPERTIES OF SOILS DEVELOPING 1227 FROM POST-MINING MATERIALS OF KONIN BROWN COAL MINE Krzysztof Otremba, Mirosława Gilewska, Andrzej Mocek, Wojciech Owczarzak, Piotr Gajewski and Zbigniew Kaczmarek

PHYSICAL PROPERTIES AND EXCHANGE SYSTEM OF SEAWATER 232 IN AYLA LAGOONS IN THE NORTHERN GULF OF AQABA, RED SEA Riyad Manasrah

ELEMENTAL DISTRIBUTION AND CHEMICAL COMPOSITION 1250 OF AMADUMBE (COLOCASIA ESCULENTA) LEAF AND IMPACT OF SOIL QUALITY Sihle Mngadi, Roshila Moodley and Sreekanth B. Jonnalagadda

EFFECTS OF AZOXYSTROBIN AND FLUSILAZOLE ON 1258 GROWTH AND PROTEIN AMOUNT OF Scenedesmus acutus Burcin Bedil, Gokce Kendirlioglu, Nur Agirman and A.Kadri Cetin

IN VITRO PLANTLET REGENERATION FROM NODAL SEGMENTS OF 1263 CREEPING JENNY (LYSIMACHIA NUMMULARIA L.) - A MEDICINAL AQUATIC PLANT Mehmet Karataş and Muhammad Aasim

1167 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FLUORIDE REMOVAL FROM AQUEOUS SOLUTIONS USING MORINGA 1269 OLEIFERA SEED ASH AS AN ENVIRONMENTAL FRIENDLY AND CHEAP BIOSORBENT Sina Dobaradaran, Maryam Kakuee, Iraj Nabipour, Abdolrahim Pazira, Mohammad Ali Zazouli, Mozhgan Keshtkar and Maryam Khorsand

THE TOXIC EFFECT OF CELL MEMBRANE OF 1275 E. COLI CAUSED BY CdTe/MPA QUANTUM DOTS Jinhui Zhou, Ling Ding, Zhenyu He, Ling Jin, Maolan Liu, Xiujin Han, Qingzhu Zhao and Xusheng Jia

THE MECHANISM OF ALUMINIUM TOXICITY AND RESISTANCE ON SOME 1282 BIOCHEMICAL AND HORMONAL CONTENTS OF HORDEUM VULGARE SEEDLINGS Mona M. Abdalla

EFFECTS OF Mn-LOADING ON ACTIVATED CARBON TREATED BY HNO3 FOR SO2 REMOVAL 1291 Xuejiao Wang, Mengdan Gong, Jiaxiu Guo, Yifan Qu, Huaqiang Yin, Yongjun Liu and Jianjun Li

HEAVY METAL AND MINERAL LEVELS OF SOME FRUIT 1302 SPECIES GROWN AT THE ROADSIDE IN THE EAST PART OF TURKEY Mücahit Pehluvan, Metin Turan, Tuncay Kaya and Uğur Şimsek

EFFECTS OF ROADS AND RAILWAYS ON LARGE GAME IN 1310 THE BELGRADE AREA: A CASE-STUDY OF NINE MUNICIPALITIES Dragan P. Gacic, Milorad Danilovic, Jasmina Gacic and Dusan Stojnic

PCCD/Fs AND BTEXs IN THE VICINITY OF AN 1318 INDUSTRIAL WASTE INCINERATOR IN NORTHERN ALGERIA Fetta Ait Ahsene-Aissat, Yacine Moussaoui, Yacine Kerchich, Feriel Guenane and Messaoud Hachemi

MEASUREMENTS OF MERCURY ASSOCIATED WITH 1326 AIRBORNE PARTICLE MATTER: USE OF TWO SAMPLING DEVICES Zia Mahmood Siddiqi and Julia Lu

INDEX 1333

1168 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

CHARACTERISTICS OF TOTAL DISSOLVED IRON OUTPUT AND ITS SPECIES DURING DIFFERENT FLOOD EVENTS IN THE SONGHUA RIVER

Jiunian Guan1,2, Baixing Yan1, Yingying Xu3, Lixia Wang1 and Hui Zhu1, *

1Key Laboratory of Wetland Ecology and Environment, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Shengbei Road 4888, Changchun 130102, China 2University of Chinese Academy of Sciences, Beijing 100039, China 3Key Laboratory of Songliao Aquatic Environment, Ministry of Education, Jilin Jianzhu University, Changchun 130118, China

ABSTRACT 1. INTRODUCTION

As a result of climate changes, the occurrence of flood As a result of climate changes, the occurrence of ex- events has been increasing. In order to reveal the charac- treme precipitation events and associated floods has been teristics of total dissolved iron (TD Fe) output and its spe- increasing worldwide, especially for countries in the tem- cies during different type of flood events, the water was perate zone at high latitudes [1-4]. Besides, the snowfall has sampled from the Songhua River at Harbin City, during the also been increasing gradually and snowstorms occurred flood event caused by extreme precipitation associated more frequently, which leads to an enhancement of runoff to Typhoon Bolaven struck northeastern China during flow during spring flood which is an important hydrological 28th-30th August, 2012 and spring flood period in 2013, which event in the cold region [5-8]. Consequently, the flood was the severest spring flood in decades. Filtration and cross- events will significantly change the regional hydrology [9, flow ultrafiltration were employed to analyze the chemical 10], directly affect the ecological processes and services of speciation of TD Fe. The results showed that the TD Fe con- terrestrial ecosystem [11, 12], and further alter the biogeo- centrations varied between 0.183 and 0.318 mg/L (mean chemical cycle of nutrient elements [13, 14]. However, its 0.251 mg/L) during the flood event caused by Typhoon and impacts on the output and species characteristics of nutri- fluctuated from 0.199 to 0.326 mg/L (mean 0.256 mg/L) ent metal elements such as iron, were rarely reported, es- during the spring flood period, with the output of 32.53 ton/ pecially in the Songhua River. day and 44.24 ton/day, respectively. The proportion of TD Iron is one of the most abundant metals of the Earth's Fe species is in the following order: complexed iron> ionic crust. It is not only a key metal controlling biogeochemical iron> colloidal iron. Further, the increase of TD Fe output cycle of other elements in aquatic environment, but also an during the flood event can be attributed to iron-rich terres- essential element for physiological functioning of terrestrial runoff and DOM transported into the river due to the trial and oceanic organisms, in particular of phytoplankton, significant positive correlation between TD Fe and DOC. which is responsible for the primary productivity in the Besides, a significantly negative correlation was observed ocean [15, 16]. The North Pacific is known as the high- - - between TD Fe and NO3 -N, indicating that NO3 inhibits nitrate and low-chlorophyll (HNLC) ocean area, where dis- Fe(III) reduction cycle. Therefore, flood events can exert solved macronutrients such as N, P, in the surface water significant influence on the out and species of TD Fe in the cannot be sufficiently utilized by phytoplankton due to low Songhua River, further affect the transport and cycling of concentration of dissolved iron. However, it is not the case nutrients in the river. for the Sea of Okhotsk, which may be attributed to the sufficient transport of dissolved iron from the Amur River [17- 20]. As the largest tributary of Amur River, the Songhua River can actually play a critical role in iron export [21]. KEYWORDS: total dissolved iron; iron species; DOM; flood event; Songhua River Rivers receive large iron loads from groundwater supply and direct terrestrial runoff, wherein, iron occurs in forms as

ionic iron including ferrous (Fe(II)) and ferric (Fe(III)) iron,

organically and inorganically complexed iron and colloidal

iron [22]. In most cases, the total dissolved iron (TD Fe)

concentration is relatively low in surface water, since in an

oxic region, Fe(II) could be rapidly oxidized by oxygen * Corresponding author into the species of Fe(III), which speedily hydrolyze and

1169 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

then precipitate as iron oxyhydroxides [23]. Nevertheless, 2.2.2 Spring flood in 2013 dissolved organic matters (DOM) can stabilize Fe(II) and During the winter from November 2012 to April 2013, retard the oxidation and precipitation [24]. Hence, the mo- Northeastern China suffered from intensive and frequent bilization and transport of TD Fe largely depends on DOM snowfall, the average amount of snow reached 78 mm in in rivers, and the iron dynamics is controlled by the factors Jilin Province and 70 mm in Heilongjiang Province, which that cause changes of DOM transported into the river, such was 119% and 109% more than annual average snowfall, as flood events [25]. respectively, resulting in the severest spring flood in the The basin of Songhua River is an important path trans- Songhua River since 1961. The ice cover of the river began ferring iron into Okhotsk Sea. However, the rapid develop- to thaw on 19th April, 2013 in Harbin City. The average ment of agriculture, industry and urbanization during the discharge of Songhua River was 2,000 m3/s during the past decades changed land use in the river basin, which spring flood period (April to May), which almost equaled could increase leaching of chemicals to surface waters and to the average discharge of summer flood period. consequently change the aquatic chemicals. Much attention had been given to the cycle of organic compounds and heavy 2.3 Sample collection and analysis metals in the Songhua River [26,27], however, the charac- One liter of river water was sampled at a depth of 50cm teristics of TD Fe output and its species during the flood of the Songhua River at the Road Bridge in Harbin City by events were limited documented, though there were studies polythene water sampler in each sampling event. The water on the iron contents and species [28,29]. Therefore, the pre- samples were collected from September 1st to 10th, in 2012 sent study is aimed to reveal the characteristics on TD Fe out- and from April to May in 2013. Three samples were ob- put during different flood events and investigate the critical tained at the middle (thread of stream) and nearby the left factors that influencing the iron concentration and species. and right bank of the Songhua River, respectively. The same volume of these three samples were mixed by siphon method to be a representative sample. Three replicate sam- 2. MATERIALS AND METHODS ples were collected during each sampling. Water samples were immediately stored in a portable refrigerator (0-4ºC) 2.1 Study site after collection till further treatment and analysis. The Songhua River Basin (41º42’-51º38’N, 119º52’- 132º31’E) mainly located in Jilin and Heilongjiang Prov- Chemical speciation of iron was analyzed by filtration inces, Northeast China, which is a typical cold region. It is and ultrafiltration methods according to the method estab- the largest tributary of the Amur River, with an area of lished in previous research by our research team [21, 32, 5.57×105 km2. The river flows 1, 434 km from Changbai 33]. The samples were firstly filtered by acid-cleaned Mountains joining the Amur River at Tongjiang City, with Whatman GF/F membrane (Whatman International Ltd., an annual discharge of 1, 266 m3/s at Harbin City [30]. The England) to analyze TD Fe concentration. Then the TD Fe river freezes from November until April. was divided by cross-flow ultrafiltration into low-molecular-weight (LMW Fe), medium-molecular-weight (MMW The Harbin City is located at the middle reaches of the Fe) and high-molecular-weight (HMW Fe) with the size of Songhua River, with the water catchment area of 3.9×105 0.01 μm (10 kDa MWCO PES), 0.01-0.05 μm (50 kDa km2, taking 70% of total basin area. It controls the main- MWCO PES), and 0.05-0.7 μm (Whatman GF/F), respec- stream of middle Songhua River after converging the Nen tively [32,33]. In the present study, the colloidal iron frac- River and Second Songhua River. The city belongs to mon- tion is defined as the sum of HMW and MMW Fe. The soon area of the north temperate zone, with average annual LMW Fe mainly contains ionic and complexed iron. The temperature of 4ºC, the precipitation is around 500 mm per ionic iron includes Fe2+ and Fe3+; complexed iron species year and over 70% of annual precipitation occurred from mainly represents Fe(III) binding with organic matters. June to September [31]. The recovery of cross-flow ultrafiltration method was 93.4-103.5% and the detection limit is 0.002 mg/L. 2.2 Flood events 2.2.1 Flood associated to Typhoon Bolaven The ionic iron concentration measurements were con- Accompanied with Typhoon Bolaven, which was re- ducted by ET7406 Fe Concentration Tester (Lovibond, garded as the most powerful storm to strike the Northeast Germany). The concentrations of different species of iron Asia in a decade, the extreme precipitation occurred in and dissolved manganese were determined by GBC 906 northeastern China leading to localized flooding in the AAS, Australia. The pH and electrical conductivity (EC) Songhua River. The precipitation lasted from 28th August values were measured using pH and EC electrode (Rex to 30th August, 2012 in Jilin and Heilongjiang Province. Shanghai, China). Dissolved organic carbon (DOC) was The average precipitation amount of 51.2 mm in Jilin Prov- analyzed using a non-dispersive IR detector to quantita- ince, with the maximum recorded value of 171.4 mm, and tively measure CO2 levels (Shimadzu TOC-VCPH, Japan). + - 3- the precipitation reached 150.5 mm in Harbin City. The av- The concentrations of NH4 -N, NO3 -N, and PO4 -P in erage discharge of Songhua River was 1500 m3/s at the sec- each sample were measured using a discrete auto analyzer tion of Harbin City during this flood period. (SmartChem, Westco Inc., Italy).

1170 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.4 Data analysis 3.2 Characteristics of iron concentrations and species during the flood events All statistical analyses were conducted using the SPSS 22.0 statistical software (SPSS Inc., Chicago, IL, USA). To The statistical results for the iron concentrations dur- test the normal distribution of data, Q-Q probability plots ing the two flood events were presented in Table 2. The TD were employed. Normal distribution is the precondition to Fe concentrations varied between 0.183 and 0.318 mg/L, compare the mean of the data. By Q-Q probability plot an- with a mean concentration of 0.251 mg/L during the flood alyzing, majority of the data showed normal distribution event associated to Typhoon Bolaven and fluctuated from pattern, so the correlation between the analyzed variables 0.199 to 0.326 mg/L (mean 0.256 mg/L) during the spring was measured using the Spearman correlation coefficient. flood period. The TD Fe concentrations were significantly ANOVA, the abbreviation of analysis of variance, is used higher (p<0.01) than that monitored previously during the to test the significant differences among the mean of sam- flood season by our research team, nevertheless, the ionic ples. The difference were considered significant if p <0.05. Fe concentrations were significantly lower (p<0.01) [21]. Similarly, significant increase of TD Fe concentration was also observed during the spring and summer flood season 3. RESULTS AND DISCUSSION as well as flood events in different rivers [38,42-44]. As it is shown in Figure 1, dissolved iron in the river 3.1 The physicochemical characters during the flood events most existed in LMW form taking account for 72.9% and The statics of physicochemical parameters of the water 73.4% of TD Fe concentration, respectively, in which com- samples are listed in Table 1. Most of the parameters were plexed iron took 49.0% and 53.9% of TD Fe, respectively, fit for the Grade-III surface water criterion according to En- during these two flood events. The relative proportion of vironmental Quality Standard for Surface Water (GB3838- iron species is in the following order: complexed iron> 2002) in PR China during the flood events, except N concen- ionic iron> colloidal iron. Comparing to the previous re- + trations (Both NH4 -N and TN are required to be ≤ 1.0 mg/L). search, the proportion of complexed iron raised signifi- - The mean concentrations of NO3 -N were 1.73 mg/L and cantly and became the main contributor to TD Fe concen- 1.39 mg/L, respectively. The Songhua River is affected by tration during these two flood events. anthropogenic activities intensively and the excessive ferti- The increase of TD Fe concentration in the Songhua lizer application and domestic sewage have become the River during the flood events can be attributed to the high principal sources of nitrogen and phosphorus in this area background. The rocks (basalts, andesite-basalts, and ande- [34, 35]. The runoff can transport the nitrogen into the re- sites) presenting widely in this area are rich in ferruginous ceiving river, especially during the flood events, inducing minerals, besides, high contents of iron also occurs in differ- eutrophication and loss of biodiversity in the aquatic eco- ent accumulative formations in the river basin [45,46]. Dur- system [36]. The mean DOC values during the two flood ing the flood event, the runoff can transport a large amount events were 7.66 mg/L and 10.34 mg/L, respectively. The of iron-rich sediment and soil into the river, resulting high values were significantly higher than that reported previ- concentration of TD Fe [47]. In such conditions, a strong ously (p<0.01) [37, 38], which is due to the effect of per- positive relationship between iron concentration and dis- colating organic-rich soil horizons by runoff and raising of charge in the rivers was observed during rainstorms and floods shallow groundwater [39]. Furthermore, it is reported that [44]. Atmospheric input is an important source of iron, how- majority of the annual DOC export occurs during the in- ever, the iron input via precipitation is indiscernible in this tensive precipitation events and following flood in differ- area due to its low concentration [33]. The groundwater is an- ent catchments [40, 41]. other contributor of iron, however, during the flood events, it is lack of possibility for groundwater to supply surface water with iron due to the high surface water level.

TABLE 1 - The physicochemical parameters of the Songhua River during the flood events

pH EC DOC NH +-N NO --N PO 3--P Mn Statistics 4 3 4 (μS/cm) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) Mean 7.12 162 7.66 0.21 1.73 0.09 0.05 Median 7.13 161 8.04 0.20 1.71 0.08 0.05 Typhoon Bolaven Max 7.23 168 9.24 0.32 2.06 0.15 0.07 (N=30) Min 6.89 159 5.82 0.12 1.39 0.05 0.03 Std. D 0.14 3.91 1.45 0.10 0.27 0.05 0.02 Mean 7.08 140 10.34 0.12 1.39 0.06 0.03 Spring Median 7.08 141 10.03 0.11 1.41 0.07 0.03 Flood Max 7.29 146 12.84 0.21 1.91 0.10 0.05 (N=45) Min 6.98 137 9.31 0.08 0.85 0.05 0.02 Std. D 0.14 3.64 1.56 0.05 0.42 0.03 0.01

1171 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 1 - The TD Fe species during the flood events in the Songhua River

TABLE 2 - The concentrations of different species Fe in the Songhua River during the flood events

Ionic Fe TD Fe HMW Fe MMW Fe LMW Fe Statistics (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) Mean 0.06 0.251 0.041 0.024 0.183 Median 0.06 0.253 0.039 0.022 0.185 Typhoon Bolaven Max 0.08 0.318 0.057 0.035 0.251 (N=30) Min 0.05 0.183 0.029 0.014 0.145 Std. D 0.01 0.05 0.01 0.01 0.03 Mean 0.05 0.256 0.045 0.029 0.188 Spring Median 0.06 0.254 0.041 0.028 0.189 Flood Max 0.10 0.326 0.067 0.039 0.267 (N=45) Min 0.02 0.199 0.031 0.016 0.143 Std. D 0.03 0.05 0.02 0.01 0.04

3.3 The estimated output of TD Fe during the flood events 3.4 Factors controlling TD Fe concentration and species during the flood events Based on the average discharge data during the flood events (1500 m3/s and 2000 m3/s, respectively), the esti- The Spearman correlation coefficient was selected to mated TD Fe output was 32.53 ton/day during the flood analyze the correlation between the typical variables pre- event associated to the Typhoon Bolaven and 44.24 ton/day sent in the river water and to identify the critical factors during the spring flood period (April-May) in 2013. Since during the flood events (Table 3). A significantly positive both the TD Fe concentration and river discharge were sig- correlation (p<0.01) was observed between the pairs of TD nificantly increased during the flood events, it can be de- Fe-Mn and TD Fe-DOC. A significantly negative correla- duced that the output of the TD Fe would increase signifi- tion (p<0.01) was observed between ionic Fe-DOC, TD - cantly and the flood events could change the pattern of TD Fe-NO3 -N. TD Fe showed a positive correlation with EC Fe transportation in the Songhua River. (p<0.05), implying that high ionic strength could increase TD Fe concentration via ion exchange.

1172 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 3 - Correlation of physicochemical parameters of the Songhua River (Spearman, n=75)

+ - 3- Ion-Fe TD Fe pH EC DOC NH4 N NO3 -N PO4 -P Mn 1 -0.519a 0.354 0.005 -0.725a 0.539a 0.698a 0.116 -0.136 - 1 -0.111 0.436b 0.708a -0.325 -0.572a -0.079 0.578a - - 1 0.197 -0.331 0.239 0.279 0.195 0.188 - - - 1 0.257 0.310 0.251 0.065 0.824a - - - - 1 -0.365 -0.629a -0.212 0.403b - - - - - 1 0.569a 0.224 0.164 ------1 0.251 -0.017 ------1 0.070 ------1 a. Correlation is significant at the 0.01 level (2-tailed). b. Correlation is significant at the 0.05 level (2-tailed).

In most cases, iron is predominantly in the particulate 4. CONCLUSION fraction in the river, and the oxidized iron particles are set- tled on the riverbed. During the flood event, they can be The concentrations of TD Fe in the Songhua River resuspended. Furthermore, DOM can not only release iron ranged from 0.183 to 0.318 mg/L (mean 0.251 mg/L) and from the suspended solid and sediment, but also prevent 0.199 to 0.326 mg/L (mean 0.256 mg/L) during the two and retard precipitation of iron oxyhydroxides to increase flood events, respectively. The output of TD Fe increased the solubility and transport capability of iron. And the sig- significantly during the flood events in the Songhua River. nificant increase of DOC concentration during the flood The proportion of TD Fe species exited followed this order: events would result in the increase of TD Fe solubility and complexed iron>ionic iron> colloidial iron. The correlation output of iron [25,39,48]. The significant positive correla- analysis suggested DOM can be the main factors affecting tion between TD Fe and DOC can also explain the signifi- the output and species of TD Fe during the flood events. The cant increase of complexed iron proportion (p<0.01) during increase of TD Fe output could exert significant influence on the flood events in current study. In addition, the reactions the nutrient cycle in the river and estuary ecosystems. between iron and DOM can control energy transfer [49], and further alter biological availability of iron [23,50]. ACKNOWLEDGMENTS In recent years, the interactions between iron and nutrient elements (e.g. carbon, nitrogen, and phosphorus) at- Funding support is gratefully acknowledged from Na- tracts increasing attention [51-53]. It has been proved that tional Nature Science Foundation of China, grant after iron fertilization in HNLC regions, phytoplankton bi- 41271499 and Special S&T Project on Treatment and Con- omass increased with a decrease of the main nutrient ele- trol of Water Pollution, grant 2012ZX07201004. We are ments contents in the surface water of ocean [16], which indebted to the seniors in our team for previous work in this indicates that the increase of TD Fe output during the flood research and their critical reading, kind remarks and rele- event could on one hand reduce the concentration of nutri- vant suggestions. ent elements via reinforcing the biological availability of nutrients, on the other hand exacerbate phytoplankton The authors have declared no conflict of interest. bloom, especially in the estuarial, offshore and coastal areas [54]. In addition, the nitrogen cycle and iron cycle are potentially coupled via biological reduction of Fe(III) and REFERENCES chemical or biological reduction of nitrate, nitrite and NO accompanied with Fe(II) oxidation [55] and microorgan- [1] Easterling D.R., Meehl G.A., Parmesan C., Changnon S.A., isms take Fe(III) as an electron acceptor while oxidizing Karl T.R., and Mearns L.O. (2000). Climate extremes: Obser- + - vations, modeling, and impacts. Science, 289(5487): 2068- NH4 to NO2 for energy production in sediments [56]. 2074. Consequently, the iron redox cycle can influence the nutrient exchange between sediments and water in riparian eco- [2] Grossmann I., and Morgan M.G. (2011). Tropical cyclones, climate change, and scientific uncertainty: what do we know, systems [57]. The negative correlation between TD Fe and what does it mean, and what should be done? Climatic Change, - - NO3 in the this study proves that NO3 inhibits Fe(III) re- 108(3): 543-579. duction cycle [58]. The study area is a key grain producing [3] IPCC, 2012. Summary for policy makers. In: Field, C.B., Bar- area and a large amount of N would be leached into the ros, V., Stocker, T.F., Qin, D., Dokken, D.J., Ebi, K.L., river, especially during the flood event [34,59], thus, the Mastrandrea, M.D., Mach, K.J., Plattner, G.-K., Allen, S.K., application of N fertilizer would reduce the TD Fe output Tignor, M., Midgley, P.M. (Eds.), Managing the Risks of Ex- in the rivers. treme Events and Disasters to Advance Climate Change Ad- aptation. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change. Cambridge Uni- versity Press, Cambridge, UK/New York, NY, USA, 1-19.

1173 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[4] Bates B.C., Kundzewicz Z.W., Wu S., Palutikof J.P., Bates [19] Suzuki K., Hattori-Saito A., Sekiguchi Y., Nishioka J., Shi- B.C., Kundzewicz Z.W., et al., (2008). Climate change and gemitsu M., Isada T., et. al. (2014). Spatial variability in iron water. Technical Paper of the Intergovernmental Panel on Cli- nutritional status of large diatoms in the Sea of Okhotsk with mate Change, Geneva, Switzerland. 210. special reference to the Amur River discharge. Biogeosci- ences, 11(1): 373-415. [5] Barnett T.P., Adam J.C., and Lettenmaier D.P. (2005). Poten- tial impacts of a warming climate on water availability in [20] Nishioka J., Mitsudera H., Yasuda I., Liu H., Nakatsuka T., snow-dominated regions. Nature, 438(7066): 303-309. and Volkov Y. N. (2014). Biogeochemical and physical processes in the Sea of Okhotsk and the linkage to the Pacific [6] Lemke P., Ren J., Alley R.B., Allison I., Carrasco J., Flato, G., Ocean. Progress in Oceanography, 126(8): 1-7. and Zhang T. (2007). Observations: Changes in snow, ice and frozen ground. Titel: Climate change 2007: the physical sci- [21] Pan X., Yan B., and Yoh M. (2011). Effects of land use and ence basis; summary for policymakers, technical summary and changes in cover on the transformation and transportation of frequently asked questions. Part of the Working Group I con- iron: A case study of the Sanjiang Plain, Northeast China. Sci- tribution to the Fourth Assessment Report of the Intergovern- ence China Earth Sciences, 54(5): 686-693. mental Panel on Climate Change, 337-383. [22] Stumm W., and Morgan J.J. (2012). Aquatic chemistry. John [7] Minville M., Brissette F., and Leconte R. (2008). Uncertainty Wiley & Sons., New York, 614-671. of the impact of climate change on the hydrology of a nordic watershed. Journal of Hydrology, 358(1): 70-83. [23] Rose A.L., Salmon T.P., Lukondeh T., Neilan B.A., and Waite T.D., (2005). Use of superoxide as an electron shuttle for iron [8] Zhu H., Xu Y., Yan B., and Guan J. (2012). Snowmelt Runoff: acquisition by the marine cyanobacterium Lyngbya majuscula. A New Focus of Urban Nonpoint Source Pollution. Interna- Environmental Science & Technology, 39(10): 3708-3715. tional Journal of Environmental Research and Public Health, 9(12): 4333-4345. [24] Dillon P.J., and Molot L.A. (1997). Effect of landscape form on export of dissolved organic carbon, iron, and phosphorus [9] Sokal M.A., Hall R.I., and Wolfe B.B. (2010). The role of from forested stream catchments. Water Resources Research, flooding on inter annual and seasonal variability of lake wa- 33(11): 2591-2600. ter chemistry, phytoplankton diatom communities and macro- phyte biomass in the Slave River Delta (Northwest Territories, [25] Vuori K.M. (1995). Direct and indirect effects of iron on river Canada). Ecohydrology, 3(1): 41-54. ecosystems. In Ann ales Zoologici Fennici, 32(3): 317-329. [10] Beyene T., Lettenmaier D.P., and Kabat P. (2010). Hydrologic [26] Jiang G.B., Shi J.B., and Feng X.B. (2006). Mercury pollution impacts of climate change on the Nile River Basin: implica- in China. Environmental Science & Technology, 40(12): tions of the 2007 IPCC scenarios. Climate Change, 100: 433- 3672-3678. 461. [27] Wang C., Feng Y., Zhao S., and Li B. L. (2012). A dynamic [11] Knapp A.K., Beier C., Briske D.D., Classen, A.T., Luo Y., contaminant fate model of organic compound: A case study of Reichstein M., et al. (2008). Consequences of more extreme Nitrobenzene pollution in Songhua River, China. Chemo- precipitation regimes for terrestrial ecosystems. Bioscience, sphere, 88(1): 69-76. 58(9): 811-821. [28] Levshina S.I. (2012). Iron distribution in surface waters in the [12] Kardol P., Campany C.E., Souza L., Norby R.J., Weltzin J.F., Middle and Lower Amur basin. Water Resources, 39(4): 375- and Classen A.T., (2010). Climate change effects on plant bi- 383. omass alter dominance patterns and community evenness in an experimental old-field ecosystem. Global Change Biology, [29] Chudaeva V.A., Shesterkin V.P. and Chudaev O.V. (2011). 16(10): 2676-2687. Trace elements in surface water in Amur River basin. Water Resources, 38(5): 650-661. [13] Knapp A.K., Fay P.A., Blair J.M., Collins S.L., Smith M.D., Carlisle J.D. et al., (2002). Rainfall variability, carbon cycling, [30] Lu Z.H., Xia Z.Q., Yu L.L., and Wang Z.J. Changes of annual and plant species diversity in a mesic grassland. Science, runoff distribution in middle reaches of Songhua River. Jour- 298(5601): 2202-2205. nal of Hehai University (Natural Sciences), 40(1): 63-69. (in Chinese) [14] Gao Y., Zhu B., Yu G., Chen W., He N., Wang T., et al. 2014. Coupled effects of biogeochemical and hydrological processes [31] Miao, C., Yang, L., Liu, B., Gao, Y., and Li, S. (2011). Stream- on C, N, and P export during extreme rainfall events in a purple flow changes and its influencing factors in the mainstream of soil watershed in southwestern China. Journal of Hydrology, the Songhua River basin, Northeast China over the past 50 511: 692-702. years. Environmental Earth Sciences, 63(3): 489-499. [15] Chen J.B., Busigny V., Gaillardet J., Louvat P., and Wang [32] Pan X., Yan B., Yoh M., Wang L., and Liu X. (2010). Tem- Y.N. (2014). Iron isotopes in the Seine River (France): Natural poral variability of iron concentrations and fractions in wet- versus anthropogenic sources. Geochimica et Cosmochimica land waters in Sanjiang Plain, Northeast China. Journal of En- Acta, 128: 128-143. vironmental Sciences, 22(7): 968-974. [16] Boyd P.W., Jickells T., Law C.S., Blain S., Boyle E.A., Bues- [33] Wang L.X., Yan B.X., Pan X.F., and Zhu H. (2012). The Spa- seler K.O. et al. (2007). Mesoscale iron enrichment experi- tial Variation and Factors Controlling the Concentration of To- ments 1993-2005: Synthesis and future directions. Science, tal Dissolved Iron in Rivers, Sanjiang Plain. CLEAN–Soil, 315(5812): 612-617. Air, Water, 40(7): 712-717. [17] Sugie K., Nishioka J., Kuma K., Volkov Y.N., and Nakatsuka [34] Ongley E. D., Xiaolan Z., and Tao Y. (2010). Current status of T. (2013). Availability of particulate Fe to phytoplankton in agricultural and rural non-point source pollution assessment in the Sea of Okhotsk. Marine Chemistry, 152: 20-31. China. Environmental Pollution, 158(5): 1159-1168. [18] Yoshimura T., Nishioka J., Nakatsuka T. (2010). Iron nutri- [35] Yang Y., Yan, B., and Shen W. (2010). Assessment of point tional status of the phytoplankton assemblage in the Okhotsk and nonpoint sources pollution in Songhua River Basin, Sea during summer. Deep Sea Research Part I: Oceanographic Northeast China by using revised water quality model. Chi- Research Papers, 57(11): 1454-1464. nese Geographical Science, 20(1): 30-36.

1174 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[36] Wu L., Long T.Y., Liu X., and Guo J.S. (2012). Impacts of [52] Achtnich C., Bak F., and Conrad R. (1995). Competition for climate and land-use changes on the migration of non-point electron donors among nitrate reducers, ferric iron reducers, source nitrogen and phosphorus during rainfall-runoff in the sulfate reducers, and methanogens in anoxic paddy soil. Biol- Jialing River Watershed, China. Journal of Hydrology, 475: ogy and Fertility of Soils, 19: 65-72. 26-41. [53] Jäckel U., Russo S., and Schnell S. (2005). Enhanced iron re- [37] Levshina S.I. (2008). Dissolved and suspended organic mater duction by iron supplement: a strategy to reduce methane in the Amur and Songhua River water. Water Resources, emission from paddies. Soil Biology and Biochemistry, 37: 35(6): 716-724. 2150-2154. [38] Wang D., Zhao Y., Xie J., Chow C.W., and Van Leeuwen J. [54] Mills M.M., Ridame C., Davey M., La Roche J., and Geider (2013). Characterizing DOM and removal by enhanced coag- R.J. (2004). Iron and phosphorus co-limit nitrogen fixation in ulation: A survey with typical Chinese source waters. Separa- the eastern tropical North Atlantic. Nature, 429: 292-294. tion and Purification Technology, 110: 188-195. [55] Kampschreur M. J., Kleerebezem R., deVet W. W., and Van [39] Andersson K., Dahlqvist R., Turner D., Stolpe B., Larsson T., Loosdrecht M. (2011). Reduced iron induced nitric oxide and Ingri J., and Andersson P. (2006). Colloidal rare earth ele- nitrous oxide emission. Water Research, 45(18): 5945-5952. ments in a boreal river: changing sources and distributions during the spring flood. Geochimica et Cosmochimica Acta, [56] Clément J.C., Shrestha J., Ehrenfeld J.G., and Jaffé P.R. 70(13): 3261-3274. (2005). Ammonium oxidation coupled to dissimilatory reduction of iron under anaerobic conditions in wetland soils. Soil [40] Hope D., Billett M.F., and Cresser M.S. (1994). A review of Biology and Biochemistry, 37(12): 2323-2328. the export of carbon in river water: fluxes and processes. En- vironmental Pollution, 84(3): 301-324. [57] Nealson K.H., Belz A., and McKee B. (2002). Breathing metals as a way of life: geobiology in action. Antonie Antonie van [41] Inamdar S.P., O'Leary N., Mitchell M.J., and Riley J.T. Leeuwenhoek, . 81: 215-222. (2006). The impact of storm events on solute exports from a glaciated forested watershed in western New York, USA. Hy- [58] Matocha C.J., and Coyne M.S. (2007). Short-term response of drological Processes, 20(16): 3423-3439. soil iron to nitrate addition. Soil Science Society of America Journal, 71(1): 108-117. [42] Zhou G.H., Sun B.B., Liu Z.Y., Wei H.L., Zeng D.M., and Zhang B.M. (2012). Geochemical feature of rare earth ele- [59] Zhang Y., Zhu H., Yan B., Ou Y., and Wang L. (2013). Nutri- ments in major rivers of eastern China. Geoscience, 26: 1028- ent removal in different overlying water layers and their vari- 1042. (in Chinese) ation in pore water of drainage ditches in Sanjiang Plain, Northeast China. Desalination and Water Treatment,51(28- [43] Kulakov V.V., Kondratyeva L.M., and Golubeva Y.M. (2010). 30): 5599-5607. Geological and biogeochemical prerequisites for high Fe and Mn contents in the Amur River water. Russian Journal of Pa- cific Geology, 4(6): 510-519.

[44] Johnson C.A., and Thornton I. (1987). Hydrological and chemical factors controlling the concentrations of Fe, Cu, Zn and As in a river system contaminated by acid mine drainage. Water Research, 21(3): 359-365. [45] Chen J, Hong S, Deng B, et al. (1999). Geographical tenden- cies of trace element contents in soils derived from granite, basalt and limestone of Eastern China. Soil and Environmental Sciences, 3: 161-167. (in Chinese) [46] Meng X, She Z, Liu G, et al. (1985). Background values of certain metal elements in main agriculture soil of middle northeastern in China. Acta Ecologica Sinica,5(2): 114-125. Received: April 14, 2014 (in Chinese) Revised: July 04, 2014 [47] Heikkinen K. (1990). Seasonal changes in iron transport and Accepted: July 23, 2014 nature of dissolved organic matter in a humic river in Northern Finland. Earth Surf Process Landf, 15(7): 583-596. [48] Laudon H., Köhler S., and Buffam I. (2004). Seasonal TOC CORRESPONDING AUTHOR export from seven boreal catchments in northern Sweden. Aquatic Sciences, 66(2): 223-230. Zhu Hui [49] McKnight D.M., Scott D.T., Hrncir D.C., and Lovley D.R. Key Laboratory of Wetland Ecology and Environment (2001). Photochemical and microbial processes influencing Northeast Institute of Geography and Agroecology iron humic interactions in stream and lake sediments. In: Hu- Chinese Academy of Sciences mic Substances and Chemical Contaminants. Soil Science So- ciety of America, 351-369. Shengbei Road 4888 Changchun 130102 [50] Kustka A.B., Shaked Y., Milligan A.J., King D.W., and Morel F.M. (2005). Extracellular production of superoxide by marine P.R. CHINA diatoms: Contrasting effects on iron redox chemistry and bioavailability. Limnology and Oceanography, 50(4): 1172-1180. Phone: +86-(0)431-85542215 [51] Smolders A., and Roelofs M. (1995). Internal eutrophication, Tax: +86-(0)431-85542298 iron limitation and sulphide accumulation due to the inlet of E-mail: [email protected] river Rhine water in peaty shallow waters in the Netherlands. Hydrobiology, 133: 9-65. FEB/ Vol 24/ No 4/ 2015 – pages 1169 - 1175

1175 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

IRRIGATION WATER MANAGEMENT IN THE JORDAN VALLEY UNDER WATER SCARCITY

Abbas S. Al-Omari1,*, Emad K. Al- Karablieh2, Zain M. Al-Houri3, Amer Z. Salman2 and Radwan A. Al-Weshah4

1 Water, Energy and Environment Center, the University of Jordan, Amman-Jordan 2 Agricultural Economics and Agribusiness, the University of Jordan, Amman-Jordan 3 Civil Engineering Department, Applied Sciences University, Amman-Jordan 4 Civil Engineering Department, the University of Jordan, Amman-Jordan

ABSTRACT development and on the well being of the people in the absence of proper and efficient adaptation measures. In addition This paper investigates the deficit in the irrigation de- to the projected negative impacts of climate change, frequent mand in the Jordan Valley for several scenarios of water droughts, high population growth rate both natural and invol- scarcity and looks into possible adaptation options. For this untary due to the political instability in the region, inefficient purpose, the Water Evaluation and Planning (WEAP) sys- use of the available resources in all the sectors, the non uni- tem is implemented to the Jordan Valley basin. Three sce- form spatial distribution of the population, and the lack of narios were analyzed by the developed WEAP model which funds to develop new resources are among the challenges are the Business As Usual (BAU) scenario, the climate that add to the complexity of the water crisis in Jordan. For change scenario, and the Red Dead Canal scenario. The re- Jordan to sustain its economic and social development un- sults showed that for the BAU scenario, the deficit in the ir- der these severe circumstances, it has to implement all pos- rigation demand will grow until the end of the planning pe- sible adaptation measures such as improve water use effi- riod despite the measures implemented which are: improv- ciency in the irrigation sector both conveyance and appli- ing irrigation efficiency, reducing non revenue water and cation efficiencies, implement demand management prac- implementing the Disi project. Under the climate change sce- tices in all the sectors, reduce Non Revenue Water (NRW), nario, the deficit in the Irrigation demand is projected to be and develop new resources. the most severe due to demand increase and resources re- In Jordan, the agricultural sector is the largest water duction. The results showed that by implementing the Red consumer. For the year 2007, water use in the agricultural Dead Canal project, the deficit in the irrigation demand in sector accounted for 64% of the available resources, half of the Jordan Valley for the year 2050 will drop from about which occurred in the Jordan Valley [2]. The deficit be- 177 MCM for the climate change scenario to zero for the tween supply and demand for that year was estimated at Red Dead Canal scenario. 638 Million Cubic Meter (MCM) for all uses, which is projected to drop to 503 MCM for the year 2022 [2] with the

KEYWORDS: Amman Zarqa basin . Climate change . Jordan Val- agricultural sector accounting for the largest proportion of ley . Water resources management. WEAP the deficit. It is worthy to note that the drop in the deficit

till the year 2022 is mainly due to implementing the Disi project by the year 2013. The DISI project which started in 1. INTRODUCTION mid. 2013, is projected to supply Jordan by about 100 MCM of fresh water from the DISI aquifer for domestic use for Projected climate change impacts on water resources the next 50 years. Knowing that approximately 60% of the in arid and semi arid regions such as Jordan are two folds, water used in the domestic sector becomes wastewater for they are expected to reduce the available resources as a re- the case of Jordan, the Disi project will also provide an ad- sult of reduced rainfall and increased evaporation, and in- ditional 60 MCM per year of treated wastewater to the ag- crease the demands as a result of temperature increase ricultural sector. It is important to note that the aforemen- which increases domestic as well as agricultural demand tioned projected deficit in the year 2022 does not take cli- due to increasing evapotranspiration of the planted areas mate change into consideration which means that the actual [1]. Reduced resources and increased demands in a country deficit can be even more severe when climate change im- struggling to bridge the gap between the limited resources pacts are considered. and the rapidly increasing demands such as Jordan are expected to have severe consequences on the socio economic 1.1 Background Due to the limited water resources in the Jordan Valley * Corresponding author and to the fact that the Jordan Valley is the food basket for

1176 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

Jordan, the management of water resources and demands lows for year around cultivation especially vegetables in in the Jordan Valley and their optimization have received winter. About 70% of Jordan’s production of fruit and veg- considerable and steady attention from several researchers etables is from the Jordan Valley. Irrigated area in the Jor- and scientists over the last three decades to sustain irrigated dan Valley is about 40,000 ha. Crops planted in the differ- agriculture and its role in Jordan’s socio-economic devel- ent agro climatic zones in the Jordan Valley are field crops, opment ([3-19]). Climate change impacts on water re- vegetables, fruit trees, banana, citrus, green house vegeta- sources and demands in the Jordan Valley were investi- bles, and dates. Irrigation technologies used in the different gated by Menzel et al. [1] in the context of the GLOWA agro-climatic zones in the Jordan Valley are drip irrigation, Jordan River project. The GLOWA JR project is an inter- surface irrigation and sprinkler irrigation ordered from disciplinary, international research project which aimed at high to low in terms of percentage use. However, by the providing scientific support for sustainable water manage- year 2020, drip irrigation and sprinkler irrigation percent- ment in the Jordan River basin taking into consideration ages will be increased at the expenses of surface irrigation climate and global changes. The GLOWA JR project was which will reflect positively on the application efficiency. financed by the German Federal Ministry of Education and Water resources in the Jordan Valley consist of ground Research (BMBF) as part of the GLOWA research initia- water, surface water, and treated wastewater from Amman tive: Global Change and the Hydrological Cycle. The pro- Zarqa Basin. Groundwater basins in the Jordan Valley are ject was launched in the year 2000 and lasted for ten years the Jordan Valley basin and the Jordan Valley side wadis [20]. It was found by Menzel et al. [1] that a projected de- basin, the safe yield of which are estimated at 20 MCM and crease in rainfall by 11% and a projected increase in evap- 31 MCM respectively [22]. Other water resources in the otranspiration by 2% will result in 25% decrease in water Jordan Valley are Yarmouk River, Taiberia Lake, and availability and 22% increase in irrigation water demand in Mukheba wells. The Yarmouk River flow to King Abdulla the Jordan Valley. Climate change modeling was based on Canal, which is the backbone of the transfer system in the the International Panel on Climate Change (IPCC) B2 valley, varies significantly from year to year which de- emission scenario [21]. pends on rainfall and on the upstream use by Syria. For ex- Despite the considerable number of studies that inves- ample Yarmouk River flow at Adasiya near the inlet to tigated the different management options for irrigation wa- King Abdulla Canal for the year 2004 was about 69 MCM ter to sustain irrigated agriculture in the Jordan Valley and which dropped to about 15 MCM for the years 2006, 2007 its role in the socio economic development in Jordan [3- and 2008. Release from Lake Taiberia to King Abdulla Ca- 19], no study to date has looked into the possible adaptation nal is governed by the Peace treaty which is about 50 MCM options to sustain irrigated agriculture in the Jordan Valley per year. Abstraction from Mukheiba wells is another wa- under climate change conditions in an integrated manner ter resource in the Jordan Valley, which oscillated between considering the available resources, the demands and the 18 and 35 MCM between 2000 and 2008 which mainly de- evolution of new resources. This paper investigates the def- pends on rainfall. In addition, several side wadis distributed icit in the irrigation demand in the Jordan Valley for the along the valley flow from east to west, the base flow of planning period, 2009-2050, taking into consideration cli- which is estimated at about 62 MCM [23]. These side mate change impacts and looks into the possible adaptation wadis flow to the Jordan River, however, some water from options to bridge the gap between supply and demand in these side wadis is used in the upstream eastern of King the context of the research subgroup “WEAP analysis” Abdulla Canal. In addition, four small dams are built on within the multinational joint research project network these wadis, the storage capacity of which is about 30 MCM. GLOWA JR (An integrated approach to sustainable man- Al-Karameh dam with a storage capacity of about 52 MCM agement of water resources under global change). is the largest dam in the Jordan Valley. Unfortunately its water is saline and can’t be used for irrigation without further treatment. A desalination plant was recently con- 2. MATERIALS AND METHODS structed to desalinate about 12 MCM per year to be used for domestic purposes. Furthermore, brackish springs exist 2.1 The Study Area in the valley some of which are desalinated and used for irrigation privately. The volume of As Samra WWTP ef- The Jordan Valley is a low-lying strip that extends fluent discharged to the Zarqa River increased from about along Jordan’s west border from northern Jordan near Lake 55 MCM for the year 2000 to about 61 MCM for the year Taiberia at an elevation of about 212 m b.s.l. to southern 2007. Water behind King Talal Dam is released to King Jordan near Aqaba. The part of the Jordan Valley covered Abdulla Canal where it gets mixed with the fresh water by this study extends from northern Jordan to near the Dead there and used for unrestricted irrigation in the Middle and Sea where elevation drops to about 424 m b.s.l., the lowest the southern Jordan Valley. point on the earth. The study area experiences a sharp gradient in rainfall from north to south. Average annual rain- Irrigation demand, which is the main demand in the fall in the Northern Jordan Valley is about 377 mm and Jordan Valley, is estimated at 320 MCM per year distrib- 77 mm in the southern Jordan Valley [7]. The prevailing uted from north to south among five main demand zones subtropical climate in the Jordan Valley and fertile soil al- namely: the Northern Jordan Valley, the middle Jordan

1177 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

Valley, the southern Jordan Valley, the northeastern Jordan ment nodes, and others. Demands in WEAP are calculated Valley and the Hesban Kafrein project in the southern Jor- by multiplying the annual activity level by the water use dan Valley. Domestic demand in the Jordan Valley is minor per unit activity level. The activity level for domestic de- compared to the Irrigation demand which is satisfied from mand sites is the population whereas for agricultural de- groundwater sources in the valley in addition to the desali- mand sites, the annual activity level is the area. In addition, nated water from Al Karameh dam mentioned earlier. the user has the option to enter the demands directly when demands are known. Agricultural demand can also be esti- It is important to note that fresh water upstream of the mated by built in software called MABIA which estimates confluence of As Samra WWTP with King Abdulla Canal crop water requirement based on evapotranspiration. In is pumped to Zai Water Treatment Plant which provides WEAP, the user defines the time step which can be annual, drinking water to west Amman. Pumping from King Ab- monthly or daily based on the user's specific need. WEAP dulla Canal to Zai WTP for the years between 2000 and has also the capacity to simulate scenarios as defined by 2008 oscillated between about 37 MCM for the year 2002 the user. For more details about the WEAP model and its to about 54 MCM for the year 2008. Figure 1 shows the allocation algorithm, the reader is referred to the WEAP study area. manual [24-26].

2.2 The Water Evaluation and Planning model 2.3 Model calibration The Water Evaluation and Planning (WEAP) system Non Revenue Water is the water pumped by the water developed by the Stockholm Environmental Institute (SEI) [24-26] is implemented to develop a network of resources supplier to the consumers but not billed which means that and demands for Amman Zarqa Basin and the Jordan Val- the water supplier does not generate revenue from this water. NRW can be estimated by subtracting the volume of the wa- ley connected by transfer lines. WEAP is water balancing and allocation software which seeks an optimum solution ter billed from the volume of the water pumped. According to the International Water Association (IWA), NRW is di- to the water allocation problem under water scarcity by employing a linear alogorithm that solves constrained optimi- vided into unbilled authorized consumption, and water zation problems. The constrained optimization problem losses. Water losses are further divided into apparent losses and real losses. Real losses are also referred to in literature consists of an objective function that maximizes coverage subject to a set of linear constraints. Coverage is defined in as physical losses. Real losses are divided into leakage from WEAP as the water delivered to a demand site divided by the transmission lines and the distribution mains, leaks and overflow from service reservoirs, leaks from service connec- the supply requirement for that demand site where supply requirement for a demand site is defined as demand plus tions up to the customer meter and losses due to pipes bursts. losses. The set of linear constraints consist of the physical Apparent losses are divided into unauthorized consumption due to illegal connections and customer meter inaccuracies. characteristics of the system and user defined criteria. The physical characteristics of the system include water quan- Unbilled authorized consumption refers to the water used for tity, water quality, and the capacity of the transfer system. public services such as irrigating public parks, water used for pipes’ maintenance and flushing, and water used for fire- The user defined criteria are demand priority and supply preference. When solving the linear optimization problem fighting [27, 28]. For the purpose of this paper, administra- under water scarcity, WEAP assumes equal coverage for tive losses are defined as commercial losses plus unbilled authorized consumption. demand sites of the same priority. Where several demands compete for the same resource, they are satisfied based on their demand priority levels assigned by the user. For ex- The breakdown of NRW into its two main components physical and administrative losses is important as it affects ample, domestic demand is usually given priority over other demands such as industrial and agricultural so it is the wastewater volume generated at a demand site. The fact that administrative losses are used within the demand site usually given demand priority one. In addition when a demand site receives water from more than one resource, wa- and returned to the wastewater collection system while ter is released from these resources based on their supply physical losses are lost from the system was used to breakdown NRW into its two main components physical and ad- preference assigned by the user. Furthermore, water balance is kept for each WEAP element while the optimum ministrative [28]. The method is based on iteratively ad- solution is sought. It is important to note that, the optimi- justing the breakdown of NRW into physical and administrative losses so that the difference between measured and zation problem is solved for each demand priority level independently starting at demand priority one. In addition, WEAP calculated inflow to As Samra WWTP is mini- WEAP simulates hydrological processes, i.e. rainfall-run- mized. Figure 2 shows the WEAP calculated versus measured wastewater volume inflow to As Samra WWTP for off and infiltration to groundwater. the calibrated model. This figure shows good agreement A basin in WEAP is expressed as a network of demand between calibrated and measured inflow to As Samra and supply nodes connected by transmission links. Other WWTP. However, in some instances considerable differ- elements of WEAP are wastewater treatment plants, return ence exists between calibrated and measured inflow to As flow lines which return wastewater from a demand site to Samra WWTP especially in the wet season which is at- a wastewater treatment plant, rivers, channels, dams, catch- tributed to infiltration/inflow to the sewer lines.

1178 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 1 - Study area and irrigation system in the Jordan Valley (not to scale)

1179 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

1180 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 2 - WEAP calculated versus measured inflow to As Samra WWTP

1181 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 1 - Irrigation water application efficiency in the Jordan Valley for the planning period, %, [29]

Year Northern & North Middle Southern Eastern Jordan Valley Jordan Valley Jordan Valley 2000 68 77 77 2005 68 77 77 2010 72 79 79 2015 75 81 81 2020 80 83 83 2050 80 83 83

TABLE 2 - Main social indicators used in for the BAU scenario

Year Population growth rate, % Per capita net water demand l/c/d Non-Revenue Water, % Source 2000 2.8 55 WIS1 2004 2.8 52 WIS 2008 2.6 48 WIS 2010 2.2 80 45 [2] 2015 2.2 100 37 [2] 2020 2.0 110 28 [2] 2022 1.9 120 25 [2] 2050 1.53 160 15 Proposed by the Ministry of Water and Irrigation WEAP team 1 Water Information System at the Ministry of Water and Irrigation

2.4 The Scenarios nated surface water form side wadis south of the Dead The developed WEAP model was run for three scenar- Sea, Zai water treatment plant, groundwater sources ios which are Business As Usaul (BAU) scenario, Climate from outside the Jordan Valley. Change scenario and the Red Dead Canal scenario. All scenarios were run for the planning period. Following is a 2.4.2 Climate Change Scenario brief description of the main features of these scenarios. The climate change scenario is based on the BAU scenario which means that all the inputs to the BAU are inher- 2.4.1 Business As Usual scenario ited by the climate change scenario except those changed or The main features of the BAU scenario are: updated by the user. The inputs to the climate change scenario are the outputs of the simulations made within the 1) No expansion in the irrigated areas in the Jordan Val- GLOWA JR project. Menzel et al. [1] used two models to ley is projected [2], simulate climate change and land use change impacts on wa- 2) Application efficiency for the different irrigation ter resources in the Jordan River basin. The Fifth-Generation zones in the valley is projected to improve as given in Penn State/NCAR Mesoscale (MM5) climate model which Table 1, and conveyance efficiency is projected to re- is maintained by Penn State University and the National main at 95% in all the Jordan Valley zones for the Center for Atmospheric Research (NCAR) was used to in- planning period [29], vestigate climate change impact on rainfall in the study area. 3) Population growth rates used are given in Table 2 The MM5 climate model was run for the Intergovernmental which shows that it is projected to drop from 2.2% for Panel on Climate Change (IPCC) B2 emission scenario. The the year 2010 to 1.5% for the year 2050. TRAIN model which is a physically based, spatially distrib- 4) The Per capita water demand is given in Table 2 uted hydrological model was then used to determine the im- which shows that the net per capita water demand pact of rainfall reduction on water availability in the study is projected to double between 2010 and 2050 from area. The TRAIN model simulates processes at the soil-veg- 80 l/c/d to 160 l/c/d. The main reason behind this in- etation-atmosphere interface with evapotranspiration as one crease is the socio economic development [2]. of the principal mechanisms. Evapotranspiration simulation 5) NRW is given in Table 2. The division of NRW into is based on Penman Monteith equation (Monteith 1965) administrative and physical is made by model calibra- [31]. Inputs to the TRAIN model are precipitation, air tem- tion as described in Al-Omari and Huber [30]. perature, humidity, wind speed, solar radiation which are 6) The DISI project started providing 100 MCM to Am- adopted from the MM5 climate model runs, in addition to man by the year 2013. soil information and land use/land cover information [1]. 7) Supply preference to Amman city is assigned in The main findings of the GLOWA JR which were input to the following order: Disi, Zara Ma’in which is desali- the climate change scenario in this study are:

1182 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

1) A gradual decrease in rainfall up to 11% by the year jected [29]. Conveyance efficiency depends on the age of 2050, the irrigation network and periodic maintenance. In the Jor- dan Valley, conveyance efficiency is satisfactory at the dif- 2) A gradual decrease in infiltration to groundwater and surface runoff to about 25% by the year 2050, and ferent agro-climatic zones, which is about 95%. It is projected to remain so for the planning period [29]. 3) A gradual increase in the irrigation demand up to 22% by the year 2050, Figure 4 shows sharp reduction in the deficit in the irrigation demand in the Jordan Valley by the year 2013 for

2.4.3 Red Dead Canal scenario the BAU scenario which is attributed to the additional treated wastewater delivered to the valley due to the imple- The Red Dead Canal scenario is based on the climate mentation of the Disi project. In addition, Figure 4 shows change scenario which means that the Red Dead Canal sce- that the reduction in the deficit in the irrigation demand for nario inherited all the inputs to the climate change scenario. the BAU continues as more of the Disi water is pumped to The main additional feature is the implementation of the Amman due to the increasing demand until about the year Red Dead Canal project by the year 2022 which will pro- 2022. The reduction in the deficit in the irrigation demand vide about 850 MCM of desalinated water, 550 MCM of is also attributed to the reduction in the demand due to the which are to Jordan and the rest are to Palestine and Israel. projected improvement in the irrigation water use effi- It is important to note that the project is proposed to save ciency, i.e. application efficiency. Furthermore, this figure the Dead Sea by disposing of the brine form the desalina- shows that for the BAU scenario there will be considerable tion plant to the Dead Sea. Environmentalists argue that the deficit in the irrigation demand in the valley despite the disposal of the brine to the Dead Sea will alter its chemistry measures taken which are the improvement in the applica- and may result in algal blooms, and color change from tur- tion efficiency, the implementation of the Disi project and quoise to brown which will impact the tourist industry at the NRW reduction. This means that these measures can both sides. In addition the projected negative environmen- help reduce the deficit in the irrigation demand in the valley tal impacts, the high cost of the project which is estimated but are not enough by themselves to overcome it. It is in- at 5.0 billion us dollar or more, is another serious barrier teresting to note that NRW reduction reflects positively on that may impede or at least delay its implementation [32]. the water availability for agriculture through two mechanisms; the first of which is that NRW reduction through leak reduction means an increase in the generated 3. RESULTS AND DISCUSSION wastewater volume which is reused for irrigation after treatment, and the second of which is that NRW reduction Figure 3 shows that there has been rapid growth in the simply means additional water resource which means more irrigation demand between 2000 and 2005 which is at- water becomes available to the agricultural sector. Further, tributed mainly to the expansion in the irrigated area. Irri- Figure 4 shows that the deficit in the irrigation demand is gated area in the Jordan Valley for the year 2000 was 19.9 projected to increase considerably for the climate change ha which expanded to 35.9 ha for the year 2005 and re- scenario which will reach about 177 MCM by the end of mained steady afterwards [29]. However beyond the year the planning period. Remembering that the climate change 2005 there has been a gradual decrease in the irrigation de- scenario is based on the BAU scenario, assures that the mand due to the projected improvement in the irrigation measures taken in the BAU are not sufficient by them- water use efficiency till about the year 2020 beyond which selves to effectively mitigate the projected negative im- the irrigation demand is projected to remain steady as no pacts of climate change. However, the implementation of further improvement in the irrigation efficiency is pro- the Red Dead Canal brings the deficit in the irrigation de- jected. Irrigation demand is a strong function in crop water mand to zero by the end of the planning period with a little requirement which in turn is direct function in evapotran- excess water of about 4 MCM. It is important to note that spiration. Difficulties and uncertainties in estimating crop the Red Dead Canal water will not be used for irrigation, water requirement will strongly be reflected on the esti- however, its impact comes through the additional waste- mated irrigation demand. Other inputs that impact the esti- water volume generated in Amman and Zarqa cities as a mated irrigation demand are water use efficiency, both result of the increasing fresh water supplied to these two conveyance and application in addition to the irrigated cities from the Red Dead Canal project. area. However, uncertainties in these two inputs are low as they can be determined to an acceptable level of accuracy Figure 4 shows that the deficit in the domestic demand especially the irrigated area. Furthermore, application effi- is expected to grow for the BAU scenario which will be ciency is a strong function of the irrigation technology even more under the climate change scenario. However, used. Generally, high losses are expected when using sur- the Red Dead Canal scenario will bring the deficit in the face irrigation and low losses are expected when using drip domestic demand for Amman and Zarqa cities from about irrigation while application efficiency for sprinkler irriga- 284 MCM for the climate change scenario to zero for the tion is in between. Application efficiency is projected to year 2050. Figure 4 shows that the implementation of the improve from about 71% for the year 2000 to about 81% Disi project in the year 2013, significantly reduced the def- for the year 2020, beyond which no improvement is pro- icit in the domestic demand for both Amman and Zarqa.

1183 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

350

300

250

BAU Climate Change Trend RDC 200 MCM

150 Demand, 100

0 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050

Year

FIGURE 3 - Projected irrigation demand in the Jordan Valley for the planning period for the three scenarios

FIGURE 4 - Projected deficit in the Irrigation demand in the Jordan Valley for the three scenarios

1184 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

300,00

250,00 BAU Climate Change Trend RDC Deficiency in MCM Deficiency 200,00

150,00

100,00

50,00

0,00 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050

FIGURE 5 - Projected deficit in the domestic demand in Amman and Zarqa cities for the three scenarios

FIGURE 6 - Projected Zarqa River flow downstream of the confluence with As Samra effluent

1185 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 7 - Projected Red Dead Canal project water balance

Furthermore, Figure 5 demonstrates that NRW reduc- change scenarios will prevail. This means that huge deficit tion which is part of the BAU scenario is not enough by in the irrigation demand in the Jordan Valley is expected itself to bridge the gap between supply and demand for especially under climate change scenario which will have Amman and Zarqa as this figure shows a continuous in- negative impacts on the socio economic development in crease in the deficit in the domestic demand under the BAU Jordan if no other actions are taken. scenario until the end of the planning period. Needless to talk about these measures as effective adaptation measures to the climate change scenario. 4. CONCLUSIONS Figure 6 shows that climate change will result in reducing the Zarqa River flow as compared to the BAU sce- The Water Evaluation And Planning system was im- nario significantly. However, the implementation of the plemented to investigate different management scenarios Red Dead Canal project will help increase the River flow for irrigation water in the Jordan Valley under water scar- significantly due to the additional treated wastewater dis- city. The developed WEAP model was run for three sce- charged to the river. As was shown earlier, this increase in narios which are the BAU scenario, the main features of the river flow helped bring the deficit in the irrigation de- which are improving irrigation efficiency, reducing NRW mand in the Jordan Valley to zero by the end of the plan- and implementing the Disi project. The second scenario is ning period. the climate change scenario the main features of which are Figure 7 shows that for the climate change scenario, reduced resources and increased demands due to climate about 225 MCM from the Red Dead Canal are needed by change. The third scenario is the Red Dead Canal, the main the year 2025 to satisfy the increasing domestic demand in feature of which is the implementation of the Red Dead Amman and Zarqa which will grow to about 500 MCM by Canal project. The results showed that under the climate the end of the planning period. change scenario, the deficit in the irrigation demand will grow to a maximum by the end of the planning period. Fur- Taking into consideration that serious barriers and ther, despite the measures taken under the BAU scenario challenges can impede the implementation of the Red Dead such as improving irrigation efficiency, NRW reduction Canal project such as the political situation in the region as and the implementation of the Disi project, the deficit in the benefits of the project are shared between Jordan, Pal- the irrigation demand will still be considerable. However, estine and Israel, the high investment needed, and the pro- by implementing the Red Dead Canal project, the deficit in jected negative environmental impacts of the project which the irrigation demand in the Jordan Valley will be reduced means that the projections of the BAU and the climate to zero by the end of the planning period due to the addi-

1186 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

tional treated wastewater that will flow to the Jordan Val- [12] Oroud I (2008) The impacts of climate change on water re- ley via the Zarqa River as a result of the additional fresh sources in Jordan. In: F. Zereini and H. Hötzl (eds). Climatic changes and water resources in the Middle East and North Af- water available for domestic use in Amman and Zarqa. rica. environmental science and engineering, Springer-Verlag Berlin Heidelberg. pp109-123.

[13] Salman A, Al-Karablieh E, Regner H-J, Wolff H-P, and Had- ACKNOWLEDGEMENT dadin M (2008) Participatory irrigation water management in the Jordan Valley. Water Policy 10 (4):305-322. The authors would like to express their gratitude to the [14] Toll M (2008) Investigating unconsolidated aquifers in an arid German Ministry of Science and Education (BMBF) who environment – A Case study from the lower Jordan Valley- funded the GLOWA Jordan River project under the grant Jordan'. In: Zereini, F., Hötzl, H. (Eds): Climatic changes and 01LW0501A1. The authors would further like to thank Ali water resources in the Middle East and in North Africa. Subah from Ministry of Water and Irrigation, Nayef Seder Springer Verlag, Berlin. pp 289-324. from the Jordan Valley Authority, H. Hoff ,K. Tielbeger [15] Toll M, Heinrichs T, Sauter M, Salameh E, Dietrich P (2008) and C. Bonzi, from the University of Tübingen, Germany. An integrated approach for the hydrogeological investigation of the unconsolidated aquifers in lower Jordan Valley'. In: Wa- The authors have declared no conflict of interest. ter resources of the Jordan and Dead Sea rift Valley, Springer Verlag, Berlin, pp 447-464.

[16] Wolff H-P, Salman A, Al-Karablieh E, Shechter M, Hijawi T, REFERENCES Fleischer A Kan, I (2008) Change in natural resources vs. socio-economic development – identification of bottlenecks for exploiting future agricultural potentials in the Jordan Valley. [1] Menzel L, Teichert E, and Weiß M (2007) Climate change im- Tropical day: Competition for Resources in a Changing pact on the water resources of the semi-arid Jordan region. In World. New Drive for Rural Development. University of Ho- Proceedings of the 3rd International Conference on Climate henheim, Oct. 7-9. and Water, Helsinki: 320-325. [17] Menzel L, Koch J, Onigkeit J, Schaldach R (2009) Modelling [2] Ministry of Water and Irrigation (2008) Water for life: Jor- the effects of land-use and land-cover change on water availa- dan’s water strategy, 2008-2022. Ministry of Water and Irri- bility in the Jordan River region. Advances in Geosciences gation, Amman, Jordan. 21:73–80. [3] Al-Weshah R (2000) Management of irrigation water in the Jordan Valley. Water Resources Management 14 (5): 327-338. [18] Al-farra A, Bennedict E, Hotzl H, Sader N, and Sonneveld, B (2011) Modeling water supply and demand for effective water [4] Salman A, Al-Karablieh E, and Fisher F (2001) An inter-sea- management allocation in the Jordan Valley. In proceedings of sonal agricultural water allocation system (SAWAS). Agricul- the 6th IWA Specialist Conference on Efficient Use and Man- tural Systems 68 (3):233–252. agement of Water, Water Demand Management Challenges & Opportunities, 29th, March - 2nd, April, Amman-Jordan. [5] Abu-Sharar T, and Battikhi A (2002) Water resources management under competitive sectoral demand: A Case Study [19] Gunkel A and Lange J (2012) New insights into the natural from Jordan. Water International 27(3): 354-378. variability of water resources in the lower Jordan River basin. [6] Doppler W, Salman A, Al-Karablieh E, and Wolff H (2002) Water Resource Management 26 (4):963-980. The impact of water price strategies on the allocation of irri- [20] (Project web site http://www.glowa-jordan- gation water: the case of the Jordan Valley. Agriculture Water river.de/Main/HomePag). Management 55 (3):171–182. [7] Al-Assaf A, Salman A , Fisher F, Al-Karablieh E (2007) A [21] IPCC (Intergovernmental Panel on Climate Change) (2007) trade –off analysis for the use of different water sources for Summary for policymakers. In: Solomon, S. Qin, D. Manning, irrigation (The case of Southern Shounah in the Jordan Val- M. Chen, Z. Marquis, M. Averyt, K.B. Tignor, M. and Miller, ley).Water International 32 (2): 224-253. H.B. (Eds.) Climate change 2007: The physical science basis. Contribution of working group I to the fourth assessment re- [8] Tabieh M (2007) An optimal irrigation water allocation port of the Intergovernmental Panel on Climate Change. Cam- model: management and pricing policy implications for the bridge: Cambridge University Press. Jordan Valley. Dissertation, University of Sciences, Penang- Malaysia. [22] Haddadin M, Salman A, and Karablieh E (2006) The role of trade in alleviating water shortage in: Haddadin M (ed.): Water [9] Wolff H-P, Shechter M, Fleischer A, Salman A, Hijawi T, Kan resources in Jordan: Evolving polices for development, the en- I (2007) Forecasting social and economic impacts from cli- vironment and conflict resolution, resources for the Future. mate change on farming systems in riparian countries of the Washington, DC, pp 150-183. Jordan River –a combined model-based approach. In: Tielkes, E. (ed.) Tropical Day: Utilization of diversity in land use sys- [23] Haddadin M, Sunna’ S, Al-Rashid H (2006) Development of tems: Sustainable and organic approaches to meet human water resources and irrigation. In: Haddadin M (ed,) Water re- needs, Universities of Kassel and Goettingen, Oct. 9-11. sources in Jordan: Evolving policies for development, the en- [10] Hötzl H, Guttman J, Salameh E, Tamimi A (2008) State of vironment, and conflict resolution, resources for the future, water strategy and policy. In: Hötzl, H, Möller P. and Rosen- Washington, DC, pp 66-87. thal E, (Eds.): The water of the Jordan Valley; SpringerVerlag, [24] Stockholm Environment Institute (SEI) (1999) WEAP: Water Berlin, pp 481-503. Evaluation and Planning system, Tellus Institute, Boston, USA. [11] Matouq M (2008) Predicting the impact of global warming on the Middle East region: Case Study on Hashemite Kingdom of [25] Stockholm Environment Institute (SEI) (2001) WEAP: Jordan using the application of geographical information sys- WEAP: Water Evaluation and Planning system, Tellus Insti- tem. Journal of Applied Sciences 8 (3):462-470. tute, Boston, USA.

1187 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[26] Stockholm Environment Institute (SEI) (2008) WEAP: User Guide for WEAP21. Tellus Institute, Boston, USA. [27] Kingdom B, Liemberger R, Marin P (2006) The challenge of reducing Non-Revenue Water (NRW) in developing countries how the private sector can help: A look at performance-based service contracting, water supply and sanitation sector board discussion paper series paper no.8. The world bank group

[28] Al-Omari, A. (2013) A methodology for the breakdown of NRW into physical and administrative losses. Water Re- sources Management, 27: 1913-1930, DOI 10.1007/s11269- 013-0262-y. [29] MWI (Ministry of Water and Irrigation) (2004) National Wa- ter Master Plan. Ministry of Water and Irrigation, Amman, Jor- dan. [30] Al-Omari A, and Huber M (2010) The Red Sea Dead Sea project: a solution to the water crisis in Jordan. In Proceedings of the International Sustainable Water and Wastewater Manage- ment Symposium (USAYS), 26-28 October, Konya, Turkey. [31] Monteith, J.L. (1965) Evaporation and environment. In G. Fogg (Ed.), the state and movement of water in living organisms, 205-234. Symposium. Soc. Exper. Biol. 19.

[32] Sharp, J.M. (2008). The “Red-Dead” Canal: Israeli-Arab ef- forts to restore the Dead Sea, Congressional Research Service, prepared for members and committees of congress. The library of congress.

Received: April 16, 2014 Revised: October 01, 2014 Accepted: November 27, 2014

CORRESPONDING AUTHOR

Abbas S. Al-Omari

Water, Energy and Environment Center

University of Jordan

Amman

JORDAN

E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1176 - 1188

1188 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

ANTHROPOGENIC CHARACTERISTICS AND CONSERVATION STATUS OF THE VASCULAR FLORA OF KOZAN CASTLE AND ITS SURROUNDING AREA (TURKEY)

Necattin Türkmen*, Atabay Düzenli, Havva Karakuş and Medine M. Uma

Department of Biology, Faculty of Sciences and Letters, University of Cukurova, 01330 Adana, Turkey

ABSTRACT transformations of natural plant cover on a global, regional, or smaller scale [5]. The alteration of ecosystems by hu- This study was carried out to determine the floristic mans is most apparent in and around urbanizing landscapes characteristics and the conservation status of the plants in [6]. There is no doubt that human civilization has had a the Kozan Castle and the surrounding area in the eastern negative impact on biodiversity, especially since the indus- Mediterranean, Turkey during the years 2008-2011. Total trial revolution. Exposure to pollution, disconnection from flora of the study area consists of 328 vascular plant taxa nature and wildlife, and surroundings of building and other belonging to 247 genera and 70 families. Life-form distri- structures can be elements of an urban lifestyle [7]. butions of the taxa are as follows: therophytes 150 species Urban habitats are often compared with rural or wil- (45.7%), hemicryptophytes 92 species (28.1%), phanerophytes 35 species (10.7%), geophytes 35 species (10.7%) derness areas and are typically characterized by lower spe- and chamaephytes 16 species (4.8%). The phytogeo- cies diversity, replacement of native wildlife with non-native species, simplified species composition, increased graphic elements are represented in the study as follows; Mediterranean 108 species (32.9%), Irano-Turanian 26 patchiness of habitat, low stability, and increased land- species (7.9%) and Euro-Siberian 10 species (3.1%). The scape conversion towards the urban centre [8-11]. phytogeographic region of taxa 184 species (56.1%) are The destruction of habitats through agriculture and ur- unknown or multiregional. Percentage of endemic species ban sprawl, excessive use of pesticides and artificial ferti- in the study area was 5% (16 species). According to the lizers, overfishing and hunting, and the release of other IUCN threat categories Sideritis cilicica Boiss. & Bal. en- toxic compounds into the environment are always harmful dangered, Bupleurum polyactis Post ex Snogerup, Feru- to the native biodiversity [12]. lago pachyloba (Fenzl) Boiss., and Verbascum antitauricum Hub.-Mor. vulnerable and the other 12 endemic spe- Castles and their surroundings are ideal places to study cies were classified as near threatened and least concerned. plant invasions and link them to historical and current hu- The flora of the study area has a moderate synanthropi- man activities [13]. zation index (30.2%). The main objective of this study was to determine the

degree of naturalness, floristic characteristics and conservation status of the flora in Kozan Castle and the surround- KEYWORDS: Anthropophytes, apophytes, castle flora, synan- ing area. There is no other specific study on plants of the thropization, threatened species, Kozan. study area.

1. INTRODUCTION 2. MATERIALS AND METHODS

In the past years, anthropogenic changes in the plant The research area (37o26’E – 35o48’ N), has elevations world have received greater attention from researchers [1-4]. ranging from 110 – 400 m, covers the area of 400 hectares These changes are manifested in a variety of forms, such as and is located in the middle of the city of Kozan which has total destruction of vegetation in certain areas or replacement 30.875 inhabitants and area of 1.690 square kilometers. of natural plant communities by cultivated communities (in Kozan castle (formerly known as Sis castle), lies in on a farmlands, windbreak forest strips, parks, gardens, lawns, high rocky ridge close to the city of Kozan in the province etc.), but the most significant aspect concerns profound of Adana in Turkey (Fig. 1). The castle was built as two separate groups in north-south direction; these sections are * Corresponding author connected by a wall because it is a long distance between

1189 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 1 - Location of the Kozan castle and the surrounding areas (Photo by H. Karakuş).

them on a mountain ridge. Environmental length of it is from 14.4 oC in January to 35.6 oC in August and the mean 6 km long and towers of the castle have 44 towers. The minimum temperatures from 6.2 oC in January to 23.3 oC castle and the surrounding area have been protected since in August. The highest recorded temperature was in July 1987 [14]. with 45.4 °C. The lowest recorded temperature was in Feb- The most ostentatious work remain from ancient Sis ruary −5.0 °C. According to the average climatic data for city where was established in the place of the Kozan district 30 years (1980-2010) obtained from the Meteorological is Kozan castle. The castle had a great strategic importance Station of Kozan, the dry period for the study area is from in terms of military, economic and cultural in ancient times. June to October. Kozan castle, in 3000 BC, was built by the Hittites, whose The main vegetation types in the study area are as fol- dominance lasted until the first century AC. In the castle, lows: from first to 8th centuries the Romans and Byzantines, in the Macchia vegetation, on hilly slopes, consists of ever- 9th century the Abbasid Arabs, in the 11th century the Seljuk green species. The common species include Olea europaea Turks, in the 13th century the Armenians, in the 14th century L. var. europaea, Phllyrea latifolia L., Fontanesia philli- the Egyptian Mamluks, in the 16th century the Ottoman raeoides Labill. ssp. philliraeoides, Ceratonia siliqua L., Turks began to sovereignty respectively [14]. Myrtus communis L. ssp. communis, Jasminum fruticans The hilly territory of the study area, the second era L., and Daphne sericea Vahl. (Mesozoic) Cretaceous epoch limestone (limestone) rocks, Rocky vegetation growing on the vertical surfaces of the and the third era (Cenozoic) is represented by conglomer- walls. The main species are as follows: Hyoscyamus aurea L., ates of the lower Miocene age. Whereas the plains territory, Galium canum Req.ex DC. ssp. ovatum, Onosma albo- Fourth time (Quaternary) is represented by the Holocene roseum Fish.& Mey. ssp. albo-roseum, Dianthus leptopetalus alluvial and colluvial [15]. Willd., Stachys rupestris Montbret & Aucher ex Bentham, The Mediterranean climate in the study area is charac- Euphorbia dendroides L., Micromeria fruticosa (L.) Druce terized by long summer droughts and mild and rainy win- ssp. brachycalyx, and Phagnalon graecum Boiss.. ters. The mean annual precipitation is about 818.4 mm, Synanthropic vegetation reflects the different vegeta- while the monthly precipitation approximates 25.3 mm in tion types that thrive in man-made habitats, can be gener- July and 104.2 mm in December. The mean annual temper- ally divided into two broad types: weed and ruderal vege- ature is 19.4 oC. The mean maximum temperatures range tation. Weed vegetation is found on arable land and ruderal

1190 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

vegetation is found in settlements, waste deposits, along an = number of anthropophytes (i.e., alien plant species) transportation routes, but also in semi-natural areas com- sp = number of non-synanthropic spontaneophytes prising disturbed river shores and woodland fringes [16- (i.e., non-synanthropic native plant species) 18]. The common species include Galium aparine L., Am- aranthus deflexus L., Amaranthus retroflexus L., Daucus broteri Ten., Scandix pecten-veneris L., Anthemis pseudo- 3. RESULTS AND DISCUSSION cotula Boiss., Calendula arvensis L., tenuis (Boiss. & Blanche) Bornm. ssp. tenuis, Carduus pycnocephalus L. ssp. During the floristic studies, total 328 vascular plant albidus, Centaurea calcitrapa L. ssp. calcitrapa, Conyza taxa belonging to 247 genera and 70 families were identi- canadensis (L.) Cronquıst, Lactuca serriola L., Senecio ver- fied. The distribution of taxa according to phytogeographic nalis Waldst. et Kit., Silybum marianum (L.) Gaertn., regions is as follows: Mediterranean elements 108 (32.9%), Sonchus asper (L.) Hill ssp. glaucescens, Xanthium stru- Irano-Turanian elements 26 (7.9%) and Euro-Siberian ele- marium L. ssp. strumarium, Stellaria media (L.) Vil. ssp. ments 10 (3.1%), pluriregional 184 taxa (56.1%) (Table 1). media, Polycarpon tetraphyllum (L.) L., Heliotrophium Most of the phytogeographical elements are naturally of europaeum L., Raphanus raphanistrum L., Ochthodium Mediterranean origin as a result of the area being situated aegyptiacum (L.) DC., Chenopodium album L. ssp. album, in the Mediterranean region [29]. Euphorbia peplus L. var. peplus, Mercurialis annua L., So- The great majority of taxa refer to the biological type lanum nigrum L. ssp. nigrum, Lamium amplexicaule L., of herbs (277 species), followed by shrubs (33 species) and Cynodon dactylon (L.) Pers. var. villosus, Erodium cicu- trees (18 species). The percentage of endemic species in tarium (L.) L'Herit. ssp. cicutarium, Malva nicaeensis All, the study area (4.9%) was found lower than the rate of en- Oxalis pes-caprea L., Pariateria judaica L., and Tribulus demism of the country (30%). terrestris L..

Today, Kozan castle is surrounded by vineyards, gar- TABLE 1 - A comparison of the phytogeographical elements in the dens, residential houses and macchia vegetation. In the last study area. decades Kozan castle has been a touristic highlight, with Phytogeographic Number Percent of many daily visitors. region of species sample total Mediterranean 108 32.9 Materials of this study, which are 328 vascular plant Irano-Turanian 26 7.9 samples, Kozan Castle and its surrounding area, in 2008- Euro-Siberian 10 3.1 2011, were collected in monthly periods. Multi-regional 184 56.1

The all specimens collected were dried according to Bidens bipinnata L. (Asteraceae) an invasive alien spe- known herbarium techniques and processes. These samples cies, has been reported as new records from Turkey for the were mainly determined with the help of Flora of Turkey first time in this study. It is widely distributed weedy spe- [19-21] and other related publications [22-24]. Authors of cies in South Africa, America, East Asia, Europe and Pa- plant names are abbreviated according to Brummitt and cific Islands [30]. Its wide spread in the world leads to us Powell [25]. Herbarium specimens are stored at the depart- to consider it as a species with a very high invasive poten- ment of Biology, Science and Letters Faculty, University tial into anthropogenic habitats in Turkey. of Çukurova. The basic life form spectrum is represented in the study Threatened categories were proposed for endemic and area as follows: Therophytes 45.7%, Hemicryptophytes rare taxa in the study area according to IUCN risk categories 28.1%, Phanerophytes 10.7%, Geophytes 10.7% and [26, 27]. Chameophytes 4.8% (Table 2). Turkey's natural species were divided into two groups: apophytes (synanthropic native plant species adapted to TABLE 2 - Life form spectrum of the species in the study area. conditions created or modified by human activity.) and spontaneophytes (non-synanthropic native plant species). Life-Type Distribution % Therophytes 150 45.7 Anthropophytes are synanthropic plant species of foreign Hemicryptophytes 92 28.1 origin, whether introduced voluntarily or involuntarily. Phanerophytes 35 10.7 Geophytes 35 10.7 The degree of anthropogenic transformation of plant Chamaephytes 16 4.8 species in the study area, calculated by the following formula [5, 28] The study area is located in the Mediterranean phyto- Wsc (%) = (ap+an) / (sp+ap+an) x100 geographic region and has a Mediterranean-type climate Wherein: which has long arid summers. Therefore, Mediterranean elements and Therophytes are represented by the highest per- WSc = synanthropization index indicates the level of centage in the flora. Euro-Siberian and Irano-Turanian el- floristic transformation ements are the least common in the area because it is very ap = number of apophytes (i.e., synanthropic native far from both phytogeographical regions. Therophytes plant species) are more resistant to summer drought than the hemicrypto-

1191 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

phytes, phanerophytes and geophytes, since the former The data obtained from the study area (apophytes: 86 spend the summer in the form of seeds and the latter in the species, spontaneophytes: 229 species, and anthropophytes: form of vegetative organs [31, 32]. 13 species), by applying synanthropization index formula The first five families with the highest number of spe- [Wsc (%) = (ap+an) / (sp+ap+an) x100)], synanthropization cies are Fabaceae 10.7%, Asteraceae l0.4%, Poaceae 8.8%, index of the flora was found to be 30.2% (Fig. 2). Apiaceae 5.2% and Liliaceae 5.2% (Table 3). The genera Anthropogenic species (i.e., non-native species) in the with the highest number of species (ranging from 7 to 4 study area ratio (4%) compared to the city of Berlin (41%) species) were, high to low, Trifolium, Silene and Galium is quite low [1]. This also shows us that the lower rate of These results are suitable with Flora of Turkey [19]. urbanization and native flora transformation in the research area. Synantropization index value height (30.2%) indi- TABLE 3 - The families represented with 9 and more species in the cates that human activities are quite effective in the re- study area. search site. Family Distribution % Of the 3504 endemic plants in Turkey, 12 have already Fabaceae 35 10.7 been extinct and 3492 are threatened [26]. Of 16 thretened Asteraceae 34 10.4 Poaceae 29 8.8 endemic species in the study area, were evaluated accord- Apiaceae 17 5.2 ing to IUCN [26, 27], 1 is in endangered category, 3 are in Liliaceae 17 5.2 vulnerable category, 2 are in near threatened category, and Lamiaceae 13 4.0 10 are in least concern category (Table 4). The factors Scrophulariaceae 12 3.6 Caryophyllaceae 11 3.4 threatening the endemics in the study area can be grouped Brassicaceae 10 3.1 as; the urbanization and industrialization, construction, Euphorbiaceae 9 2.7 over-grazing, domestic use and exportation, agricultural Ranunculaceae 9 2.7 fight and pollution, and anthropogenic fires. Others 132 40.2

Anthropophytes; Apophytes; 86; 26% 13; 4%

Spontaneophytes; 229; 70%

FIGURE 2 - Synanthropic and non-synanthropic species distribution in the study area.

TABLE 4 - The endangered species of the study area and its IUCN Red Data List categories.

Endemic species Conservation status Bupleurum cappadocicum Boiss. VU (B1 a,b and B2 a,b Ferulago pachyloba (Fenzl) Boiss. VU (B1 a,b and B2 a,b Kundmannia syriaca Boiss. LR (lc) Hieracium lasiochaetum (Bornm. et Zahn) Sell et West LR (lc) Alkanna kotschyana DC. LR (lc) Anchusa leptophylla Roemer et Schultes subsp. incana LR (lc) Alyssum huetii Boiss. LR (lc) Ricotia sinuata Boiss. & Heldr. LR (lc) Rosularia chrysantha (Boiss.) Tahkt. LR (lc) Hypericum aviculariifolium Jaub. & Spach subsp. aviculariifolium LR (lc) Crocus cancellatus Herbert subsp. cancellatus LR (lc) Sideritis cilicica Boiss. & Bal. EN (B1 a,b and B2 a,b) Stachys rupestris Montbret & Aucher ex Bentham LR (lc) Hyacinthella micratha (Boiss.) Chouard LR (nt) Thesium cilicicum Bornm. LR (nt) Verbascum antitauricum Hub.-Mor. VU (B1 a,b and B2 a,b Legend of abbreviations: CR (B1 a,b and B2 a,b) – [Critically Endangered: Extent of occurrence less than 5000 km2; area of occupancy less than 500 km2; known no more than five locations; inferred decline in the area, extent and/or quality of habitat], EN (B1 a,b and B2 a,b) – [Endangered: Extent of occurrence less than 100 km2; area of occupancy less than 10 km2; known to exist at only a single location; inferred decline in the area, extent and/or quality of habitat], VU (B1 a,b and B2 a,b) – [Vulnerable: Extent of occurrence less than 20,000 km2; area of occupancy less than 2000 km2; known no more than 10 locations; inferred decline in the area, extent and/or quality of habitat], NT – Near threatened, LC – Least concern.

1192 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

4. CONCLUSION [5] Gorchakovskii, P.L. and Kharitonova, O.V. (2007) Plant cover synanthropization in the Pechora–Ilych Biosphere Reserve along an altitudinal gradient. Russian Journal of Ecology 38, The flora of the study area comprises a total of 428 vas- 375-380. cular plant. As follows from Appendix, 99 out of 428 species [6] Lundholm, J.T. and Marlin A. (2006) Habitat origins and mi- in this flora are synanthropic (apophytic, anthropophytic) crohabitat preferences of urban plant species. Urban Ecosys- and, therefore, its synanthropization index is 30.2%. tems 9, 139-159. Synanthropization is a consequence of anthropogenic [7] Knapp, S., Kuhn, I. and Schweiger, S. (2008) Challenging urban impact on natural vegetation with consequent disturbances species diversity: contrasting phylogenetic patterns across plant of its dynamic balance and stability. The synanthropic functional groups in Germany. Ecology Letters 11, 1054-1064. component is represented by both indigenous species [8] Mckinney, M.L. (2002) Urbanization, biodiversity and conser- (apophytes) and adventive species (anthropophytes) intro- vation. Bioscience 52, 883-890. duced in the course of human activities. [9] Alberti, M., Marzluff, J.M., Shulenberger, E., Bradley, G., Ryan, C and Zumbrunnen, C. (2003) Integrating humans into The high rate of synanthropic species in the study area ecology): opportunities and challenges for studying urban eco- is explained by several factors: systems. Bioscience 53, 1169–1179. (a) The main routes of anthropophyte and apophyte [10] Turner, W.R., Nakamura, T. and Dinetti, M. (2004) Global ur- invasion from neighboring areas and subsequent expansion banization and the separation of humans from nature. BioSci- ence 54: 585. in the study area include roads, paths, vineyards, gardens, residential houses, ranger stations, grazing grounds, tourist [11] Ugulu, I, Dogan, Y. and Kesercioğlu, T. (2012) The vascular plants of Buca Faculty of Education campus (Izmir): contribu- campsites, and frequently visited areas near conspicuous tion to educational practices. Eurasian Journal of Biosciences natural landmarks. 6, 11-23. (b) The main factors that threaten native species in the [12] Hunter, P. (2007) The human impact on biological diver- study area are human impacts such as road construction, sity. EMBO Reports 8 (4), 316-318, retrieved: March 29, 2013, from:http://www.nature.com/embor/journal/v8/n4/ creation of recreational areas, clearing of the natural vege- pdf/ 7400951.pdf. tation for cultivation, overgrazing, illegal logging, and settlements. Loss of habitat due to anthropogenic activities [13] Iatrou, G., Trıgas, P. and Pettas, N. (2007) The vascular flora of Akrokorinthos Castle and its surrounding area (NE Pelo- makes these species more vulnerable to extirpation. ponnese, Greece). Phytologia Balcanica 13 (1), 83-89. (c) Disturbance also raises the possibility of encroach- [14] Mert, Y. and Ateş, F. (2010) The architect of Kozan. Istanbul ment and competition from invasive and non-native spe- (in Turkish). cies, including terrestrial and aquatic environments. [15] TOPRAKSU (1973) Ceyhan Basin Territories. Publication No: 285, Ankara (in Turkish). [16] Mucina, L. (1989) A coenocline of the high-ranked syntaxa of ruderal vegetation. Vegetatio 81, 117-125. ACKNOWLEDGEMENTS [17] Lososova, Z., Chytri, M., Kuhn, I., Hajek, O., Horakova, V., Pysek, P. and Tichy, L. (2006) Patterns of plant traits in annual The authors would like to thank the Çukurova Univer- vegetation of man-made habitats in Central Europe), perspec- sity Research Fund (Project No: BAP 2009-06) for finan- tives in plant ecology. Evolution and Systematics 8, 69-81. cial support. [18] Silc, U. (2010) Synanthropic vegetation pattern of various disturbances on life history traits. Acta Botanica Croatica 69, The authors have declared no conflict of interest. 215-227. [19] Davis, P.H. (ed.) (1965–1985) Flora of Turkey and the east Aegean Islands. Vols 1–9, Edinburgh University Press, Edin- burgh.

REFERENCES [20] Davis, P.H., Mill, R.R. and Tan, K. (eds.) (1988) Flora of Tur- key and the east Aegean Islands. Vol 10 (Supplement 1), Ed- inburgh University Press, Edinburgh. [1] Sukopp, H. (2003) Flora and vegetation reflecting the urban history of Berlin. Die Erde 134 (3), 295-316. [21] Güner, A., Özhatay, N., Ekim, T. and Başer, K.H.C. (eds.) (2000) Flora of Turkey and the east Aegean Islands. Vol 11 [2] Galera, H. and Sudnik-Vojcikowska, B. (2004) The structure (Supplement 2), Edinburgh University Press, Edinburgh. and differentiation of the synanthropic flora of the botanical gardens in Poland. Acta Societatis Botaniqorum Poloniae 73, [22] Petrides, G.A. and Wehr, J. (1988) A field guide to eastern 121–128. trees. Eastern United States and Canada, including the Mid- west, Houghton Mifflin Company, Boston, New York. [3] Bomanowska, A. and Witoslawski, P. (2008) Selected aspects of diversity of synanthropic flora in the chosen cities of central [23] Barwick, B. and Schans, A. (2004) Tropical and subtropical Poland. Biodiversity Research and Conservation 9, 35-42. trees), an encyclopedia. Timber Press, Portland, Oregon. [4] Tokarska-Guzik, B., Węgrzynek, B., Urbisz,.,Urbisz, A., [24] Chen, Y.S. and Hind, D.J.N. (2011) Heliantheae. In: Flora of Nowak, T. and Bzdęga, K. (2010) Alien vascular plants in the China (eds. Z.Y. Wu, P.H. Raven, D.Y. Hong), Vols 20–21 Silesian Upland of Poland: distribution, patterns, impacts and (Asteraceae), Science Press (Beijing) & Missouri Botanical threats. Biodiversity Research and Conservation 19, 33–54. Garden Press (St. Louis) 852–878.

1193 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[25] Brummitt, R.K. and Powell, C.E. (eds.) (1992) Authors of plant names. Royal Botanic Gardens, Kew, London. [26] Ekim, T., Koyuncu, M., Vural, M., Duman, H., Aytaç, Z. and Adigüzel, N. (eds.) (2000) Red Data Book of Turkish plants (Pteridophyta and Spermatophyta). Turkish Association for the Conservation of Nature-Van Centennial University, An- kara.

[27] IUCN (2011) Guidelines for using the IUCN Red List Catego- ries and Criteria). Version 9.0. Standards and Petitions Sub- committee of the IUCN Species Survival Commission.

[28] Kaminski, D. (2006) Floristic diversity on the medieval earth- works of Chelmo Land (Ziemia Chelminska) in north-west Po- land. Biodiversity Research and Conservation 3, 344-347. [29] Türkmen, N. and Düzenli, A. (1997) The Flora of Dörtyol and Erzin Districts of Hatay Province in Turkey. Turkish Journal of Botany 22, 121-141. [30] Sirbu, C. and Oprea, A. (2008) New alien species for the flora of Romania). Bidens bipinnata L. (Asteraceae). Turkish Jour- nal of Botany 32, 255-258. [31] Abdelghani, M.M. and Abdelkhalik, K.N. (2006) Floristic diversity and phytogeography of the Gebel Elba National Park, south-east Egypt. Turkish Journal of Botany 30, 121-136. [32] Floret, C., Galan, M.J., Lefloch’h, E., Orshan, G. and Romane, F. (1990) Growth forms and phenomorphology traits along an environmental gradient): tools for studying vegetation. Journal of Vegetation Science 1, 71-80.

Received: May 21, 2014 Revised: July 11, 2014, December 01, 2014 Accepted: February 05, 2015

CORRESPONDING AUTHOR

Necattin Türkmen

Department of Biology

Faculty of Sciences and Letters

University of Cukurova

01330 Adana

TURKEY

E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1189 - 1194

1194 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

RESISTANCE TO THE RECLAMATION OF ENVIRONMENTALLY SENSITIVE AREAS THROUGH THE ESTABLISHMENT OF A NEW FOREST ECOSYSTEM

Turgay Dindaroğlu

Kahramanmaraş Sutcu Imam University, Forest Faculty, Department of Forest Engineering, 46100, Kahramanmaras, Turkey

ABSTRACT 1. INTRODUCTION

The negative effects of desertification firstly emerge in Desertification affects approximately one-third of our ecologically sensitive areas. Thus, prioritizing these areas world [1, 2]. The southwestern United States, southern Eu- during rehabilitation and reforestation activities is essential rope, the Mediterranean, and the Middle East will experi- for improving them. The purpose of this study, conducted ence the harmful effects of drought in the future [3]. It in several areas outside the forest ecosystem, was to estab- means that there will be climate change, food poverty, and lish a forest ecosystem, to map environmentally sensitive a variety of other factors that will result in land degradation areas (ESAs), and to enable the monitoring of desertifica- in arid, semi-arid, and dry sub-humid areas. Desertification tion in the future. ESAs were evaluated using the Mediter- problems are often caused by human activities. Desertifi- ranean desertification and land-use (MEDALUS) method- cation creates non-sanitary environments and, at the end, ology, which includes the quality parameters of soil, vege- has the power to trigger human migration movements [4- tation, climate, and management. The research area, lo- 40]. cated in the town of Askale in Erzurum Province, was previously used as an armory. Those environmental areas of Desertification includes a complex set of variables as- the research field sensitive to desertification were catego- sociated with it, such as dynamic soil characteristics, mi- rized as the Critical (C2) area (5.4 ha; 3.72% of the total crometeorology, and human activities [5-8]. Both eliminat- area), the Critical (C1) area (19.6 ha; 13.52%), the Fragile ing or slowing desertification processes depend on human- (F3) area (48.5 ha; 33.45%), the Fragile (F2) area (65.6 ha; kind. Fragile and sensitive ecosystems constitute desertifi- 45.24%), the Fragile (F1) area (4.8 ha; 3.31%), and the Po- cation; therefore, decision-makers should be very careful tential (P) area (1.1 ha; 0.76%). In terms of desertification with fragile ecosystems. Vegetation and water quality rates in the research field, critical areas accounted for management are significantly under the influence of deser- 17.24% of the total area, fragile areas accounted for 82%, tification processes [9-41]. To achieve the sustainability of and potential area accounted for 0.76%. The main goals soil fertility, soil degradation and its associated risks were to determine the resistance to the reclamation pro- should be taken into account when related to the regional cesses: identification of environmental areas sensitive to dominant desertification processes, and should be resolved desertification in the study area, and the assessment of the using appropriate methods [10]. GIS technologies and successful planting ratios in the ESAs between 2009 and trends in the long-term monitoring of land degradation and 2012. Ultimately, 280,302 saplings were planted in the re- land cover maps at different times of the spatial compari- search field. Resistance to the reclamation processes of the son of multiple spatial changes is ideal, and is required to sensitivity areas in various subtypes of ESAs increased em- identify and assess the severity of land degradation [11, pirically from 2009 to 2012 (Critical ESAs>Fragile 12]. Mediterranean desertification and land-use (MED- ESAs>Potential ESAs). According to the research results, ALUS) [13] focus on the definition of environmentally the identification of ESAs is very important in the determi- sensitive areas with a multidimensional approach. Environ- nation of which seedling species can be used for successful mentally sensitive areas (ESAs) are determined by soil, cli- reclamation in arid and semi-arid areas. The decisive fac- mate, vegetation, and management quality indices. The tors in terms of desertification were the soil quality and MEDALUS project has been successfully used by many vegetation quality indices in the research area. In this re- researchers for identification of areas sensitive to desertifi- gion, reforestation projects should be implemented to pre- cation [14-17]. In Turkey, the problems of desertification vent desertification and to reduce ESAs. that could be fixed include the following ones: erosion, de- forestation, pasture degradation, loss of soil organic matter, KEYWORDS: Desertification, ecosystem, ecology, environmen- the physical degradation of land, wrong land use, decreased tally sensitive areas, MEDALUS water potential, contamination by toxic elements, soil sa-

1195 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

linity, pesticides, soil pollution, and degradation of the nat- The soil texture was analyzed according to Bouyoucos ural vegetation and biodiversity [18]. Desertification pro- [27]. The soil depth, rock fragment, and the drainage were cesses have been studied on a local scale by many research- identified [28]. The soil pH was measured by a pH-meter ers [19-22]. In this study, we aimed to identify and map with a "Glass Electrode" in a soil-water suspension with environmentally sensitive areas that had the potential of be- 1:2.5 scales potentiometrically [29]. The lime contents of ing at risk of desertification using the MEDALUS ap- the soils were determined by a Scheibler Calcimeter volu- proach. The methodology was applied in a case study that metrically [30]. The organic matter content of the soil was was proposed to create a new forest ecosystem, the purpose determined by the Smith-Weldon wet combustion proce- of which was to determine the responses of the reclamation dure [31]. processes applied during the years 2009-2012, and to eval- The parent material was identified [32] in the research uate the results. area. The vegetation and management quality index was

determined in accordance with the Forest Management

Plan [33]. The precipitation was determined by MGM [23]. 2. MATERIALS AND METHODS The Bagnouls-Gaussen drought index was used as shown in formula (2): This research was carried out around the town of Askale, which is located in the western province of Erzu- ∑ 2 . (2) rum, Turkey. The bounding geographical coordinates of the study area were 39°55’48’’ to 39°56’33’’ north lati- where, ti is the average air temperature for month i o tudes and 40°40’12’’ to 40°39’27’’ east longitudes. The re- ( C); Pi is the total precipitation for month i (mm); and ki search area is 145 ha, and its average altitude is 1700 m. is the proportion of the month (2ti - Pi >0). The climate in the research area is characterized by hot, dry Types of environmentally sensitive areas (ESAs) were summers but cold, snowy winters. The mean daily maxi- classified using Table 1. mum temperature ranges from 18 ºC in summer to approximately -13 ºC in winter, and the maximum and minimum TABLE 1 - Types of ESAs and corresponding ranges of indices. daily temperatures range from 36 ºC in summer to about - 36 ºC in winter. The rainfall is powerful in autumn and Type Subtype Range of ESAI spring, and the annual rainfall amount is 406 mm. Evapo- Critical C3 >1.53 Critical C2 1.42-1.53 ration is effective in the summer season. The average an- Critical C1 1.38-1.41 nual relative humidity is 64% [23]. The soils in the study Fragile F3 1.33-1.37 area have a "Ustic" moisture regime according to the U.S. Fragile F2 1.27-1.32 soil classification system [24]. Fragile F1 1.23-1.26 Potential P 1.17-1.22 Non affected N <1. 2.1 Assessment of environmentally sensitive areas The soil, climate, vegetation, and management param- TABLE 2 - Mapping symbols for ESAs to desertification. eters are assigned to each class based on the behavior of desertification and weight factors, and are grouped in vari- F 1 - c 2 s 1 v 2 m 1 ous uniforms to class. Then, the qualities of the soil, climate, vegetation, and management were evaluated using the formula (1) from the MEDALUS project [13]:

¼ ESAI = (SQI x CQI x VQI x MQI) (1) Where, ESAI is the environmentally sensitive areas index, SQI is the soil quality indicator (soil texture, soil depth, ESA type ESA subtype ESA climate degree of limitation soil degree of limitation vegetation degree of limitation management degree of limitation parent material, rock fragment, drainage, and slope gradient); CQI is the climate quality indicator (rainfall, aridity The soils were sampled systematically in the research and aspects); VQI represents the vegetation quality indica- area. In this context, 10 soil profiles and 23 surface soil tors (plant cover, erosion protection, drought resistance and samples were collected [28]. fire risk); and MQI is the management and policy quality indicator (land-use, land-use intensity and policy). Organic 2.2 Geostatistical analysis matter content (Poor; 0-2%, Good; 2-5%, Best; > 5%) was An Inverse Distance Weighted (IDW) interpolation also added to SQI. technique – the technique in which values at unknown The soil, climate, vegetation and management indicators places are calculated from known places using a weight were used for determining and mapping of the area environ- function – [34-36] was used to create a prediction of the mentally sensitive to desertification. A Digital Elevation ESAs using ArcGIS 10 software. Model (DEM) of the study area was derived from the topo- graphic map with a scale of 1:25,000. The DEM was used to The IDW formula 3 is: ∑ produce a slope and aspect map using ArcGIS [25, 26]. (3)

1196 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

Where, Z (s0) is value to be predicted for s0 location; N soil profile. Calcareous clay soil is formed over the parent ma- is sampled points around the prediction location; λi are the terial. The absolute and physiological depths are >120 cm. weights assigned measured point; and Z (si) is the observed Although the surface stony is 20%, the profile stony is more value at the location si. Determined weights for known val- than 35%. The clay, light clay, clay loam soil textures are ues are used in the following formula (4): formed in the field. The soil permeability and drainage is good. The field gradient is suitable for machine and hand ∑ 1 (4) tillage. Gullies were observed in the micro-topography. ∑ The organic layer is in a semi-decomposed form on the soil The quantity d is the distance between the predicted i0 surface. The soil organic matter content is sufficient (1.98- location, s , and each of the measured locations, s . The 0 i 5.15%). The pH level is neutral and indicates a slightly al- weighing of the sampled points depends on the parameter kaline reaction (7.65-8.45). The lime ratio is high (24.25- p and distance increases; the weight is reduced by a factor 39.8%). There was no problem of salinity (0.04-0.88%). of p [37]. The electrical conductivity was between 0.25x10-3 mmhos

and 0.65x10-3 mmhos. The decomposition of the organic 2.3 Soil tillage and planting technique layer is slow, and there is less mixture of the soil due to the The deep soil tillage was done with machines. The sub degradation of the field. soil tillage was done at a depth of 35-45 cm and a width of 80-100 cm in the form of Gradoni terraces; the land stoni- 3.2 Environmentally sensitive areas to desertification ness ratio was more than 25%, and the maximum slope was The soil, climate, vegetation and management quality 40%. indices were identified with 23 sampling points aimed at The soil tillage was done with worker force: The sub- establishing a new forest ecosystem in the research area. soil tillage was done at a depth of 35-40 cm and a width of SQI, VQI, CQI and MQI maps were prepared using geostatis- 0-80 cm in the form of Gradoni terraces and with a land tical methods and are presented in Figs. 2-5. slope over 40%. According to the results, the soil and climate quality indices were the most effective factors which could have 2.4 Planting operation caused the desertification of the environmentally sensitive For the conifer seeds, containerized seedlings were areas in the research field. The analyses of the map of soil used, and the pits for planting the seedlings were spaced at quality indices, high-quality areas 77.5 ha (49.18%), mod- 3-1.5 m. For the broadleaf seeds, bare-rooted seedlings erate quality areas 23.24 ha (14.75%), and low quality ar- were used, and the pits for planting those seedlings were eas 56.86 ha (36.08%) are listed in Table 3. The climate spaced at 3-2m. The ages of the seedlings were 2+0 years. quality index varies from 1.26 to 1.59. The aspect was the The total afforested area found was 140.0 ha and some in- most effective factor for determining the climate quality ternal network paths and real estate constructions were re- index in the study area. Considering the spatial distribution moved, as well. Some seedling species (Pinus sylvestris, Ro- of this small area, it was determined that many different binia pseudoacacia L., Fraxinus excelsior L., Acer negund., quality areas tend to experience desertification. Quercus petraea, Betula pendula, Amygdalus communis and Prunus mahleb) were selected according to the appropriate 3.3 Critical ESAs ecological conditions of the research area [38]. Subtypes C2, 5.4 ha (3.72 %) and C1 19.6 ha, (13.52 %) are areas with mainly very steep, (dominant slopes >35 and 2.5 Count technique of the healthy seedlings 30%, respectively), clay textured, stony, shallow to moder- For each 100 m in terraces, 30 pits were counted con- ately deep, badly drained soils formed on claystone parent taining healthy seedlings for each sensitive area. The follow- materials. The climate is characterized mainly as dry and ing formula (5) was used for counting the healthy seedlings: semiarid, with rainfall mainly <500 mm and a mainly very (5) dry bio-climatic index (BAI <75). The areas of this subtype are mostly found in the east-facing aspect. This area is where a1 is the sum of healthy seedlings and n repre- characterized mainly by low fire risk and no plant cover. sents the account points. These areas are mainly under high land-use intensity and a highly enforced environmental protection policy (Table 4 2.6 Statistical analysis and Fig. 6). Analyses of variance (ANOVA) were performed using the SPSS 16 package program [39]. 3.4 Fragile ESAs Subtype F3, 48.5 ha (33.45 %) has areas with an average to steep slope, clay textured, stony to slightly stony, 3. RESULTS AND DISCUSSION moderately deep to deep, badly drained soils formed mainly on claystone parent materials. The climate is char- 3.1 Evaluation of soil and land properties acterized mainly as dry and semiarid, with rainfall mainly The parent materials of the research areas are middle <500 mm and a mainly very dry bio-climatic index (BGI and upper Miocene. A-B and Cv horizons are seen in the <75). Areas of this subtype are mostly found in the north-

1197 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 2 - Soil quality map FIGURE 3 - Climate quality map

FIGURE 4 - Vegetation quality map FIGURE 5 - Management quality map

TABLE 3 - Soil quality index areas

Soil Quality Indices Soil Quality Indices Area (ha) High Quality 77.50 Moderate Quality 23.24 Low Quality 56.86 Total area 157.60

1198 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 4 - ESA types and areas and claystone parent materials. The climate is character- ESA Type Subtype Area (ha) ized mainly as dry and semiarid, with rainfall mainly <500 C2 5.4 mm and a mainly very dry bio-climatic index (BGI <75). Critical C1 19.6 Areas of this subtype are mostly found in the south and F3 48.5 north-facing aspects. This area is characterized mainly by Fragile F2 65.6 F1 4.8 low fire risk and no plant cover. These areas are mainly Potential P 1.1 under high land-use intensity and a highly enforced envi- Total area 145 ronmental protection policy (Table 4 and Fig. 6).

Subtype F1, 4.8 ha (3.31 %) has areas with a mainly gentle slope, moderately fine-textured, very stony, deep to deep, well-drained soils formed mainly on sandstone parent materials. The climate is characterized mainly as dry and semiarid, with rainfall mainly <500 mm and a mainly very dry bio-climatic index (BGI <75). Areas of this subtype are mostly found in the north-facing aspect. This area is characterized mainly by low fire risk and no plant cover. These areas are mainly under high land-use intensity and a highly enforced environmental protection policy (Table 4 and Fig. 6).

3.5 Potential ESAs Subtype P (potential areas) has 1.1 ha (0.76 %), with a mainly flat to gentle slope (slope <6%), moderately fine- textured, stone-free, very deep, well-drained that is formed mainly on sandy soils. The climate is characterized mainly as dry and semiarid, with rainfall mainly <500 mm and a mainly very dry bio-climatic index (BAI <75). Areas of this subtype are mostly found in the north-facing aspect. This area is characterized mainly by low fire risk and no plant cover. These areas are mainly under high land use intensity and a highly enforced environmental protection pol- FIGURE 6 - Map of environmentally sensitive areas (ESAs) to deser- icy (Table 4 and Fig. 6). tification (the soil erosion in the various sensitive sub types of ESAs increases as: C2 > C1 > F3 > F2 > F1 > P > no-threat [13]. 3.6 Resistance to reclamation of environmentally sensitive areas (ESAs) facing aspect. This area is characterized mainly by low fire In this project, a 240 km machine and a 40 km labor risk and no plant cover. These areas are mainly under high force form of Gradoni terrace were made. In the study field, land-use intensity and a highly enforced environmental 8 different varieties of trees (Pinus sylvestris, Robinia protection policy (Table 4 and Fig. 6). pseudoacacia L., Fraxinus excelsior L., Acer negund., Subtype F2, 65.6 ha (45.24 %) has areas with a mainly Quercus petraea, Betula pendula, Amygdalus communis gentle slope, moderately fine-textured, very stony, moder- and Prunus mahleb) were planted in ecologically sensitive ately deep, well-drained soils, formed mainly on sandstone areas (Table 5) according to the silvicultural features [38].

TABLE 5 - Number of seedlings according to ESA types

Number of seedlings ESA type Year Plant type Critical area Fragile area Potential C2 C1 F3 F2 F1 P Pinus sylvestris 2000 10905 22000 30000 2000 1000 Robinia pseudoacacia L. 5000 14800 38500 57517 6500 200 Fraxinus excelsior L. 1000 3300 5000 8600 100 200 Acer negundo 1000 5300 23090 28630 450 200 2009 Quercus petraea 910 200 1000 1500 100 200 Betula pendula 100 60 60 60 60 60 Amygdalus communis 100 2600 4000 400 200 200 Prunus mahleb 400 400 100 100 100 100 Total 10510 37565 93750 126807 9510 2160

1199 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

Number of the seedling was changed out of different areas been identified as ESA subtype “P” areas, respectively, P of the ESAs. These areas are determined especially accord- > F1 > F2 > F3 > C1 > C2, after the afforestation activities. ing to ecological requirements of the seedlings. The types of seedlings with the highest survival rates were observed to be Robinia pseudoacacia L. and Pinus syl- 3.7 Statistical analysis for seedling success in the years 2010, vestris. The most critical sensitive area (C2) seedlings in 2011, and 2012 the field with the highest success rates were identified as A one-way ANOVA test was performed on the seedling Robinia pseudoacacia L., Fraxinus excelsior L., Acer survival percentages for 2010, 2011, and 2012 (Table 6). All negundo, Betula pendula and Pinus sylvestris. The seed- plant species have shown a rapid decrease in numbers lings with the lowest success rates were identified as Quer- within a 3-years period. The best adapted species in the cus petraea, Prunus mahleb and Amygdalus communis study area are Robinia pseudoacacia and Pinus sylvestris. (Fig. 8). Quercus petraea, Prunus Amygdalus communis, and Prunus mahleb could not be adapted in the area (Fig. 7), possibly TABLE 6 - ANOVA test result of seedling survival percentages dur- because of their loss of excess water in the critical dry sea- ing 2010, 2011 and 2012. son, the high amount of lime content, and the excessive ex- Years Plant type posure to basic environments. Throughout the study pe- 2010 2011 2012 riod, not any significant change in climate parameters was Pinus sylvestris 56.17a 50.17a 33.83a observed. Robinia pseudoacacia L. 69.83a 66.50a 36.50b Fraxinus excelsior L. 64.00a 51.50a 20.17b 3.8 Statistical analysis for seedling success in environmen- Acer negundo 68.00a 39.00b 20.50c tally sensitive areas (ESAs) Quercus petraea 18.00a 13.83a 0.00b A one-way ANOVA test was performed to determine Betula pendula 63.33a 35.50b 10.00c the seedling survival success in the environmentally sensi- Amygdalus communis 39.67a 19.83b 0.00c tive areas (Table 7). The aim was to establish a new forest Prunus mahleb 39.50a 20.83b 0.00c ecosystem in the degraded soils of the research field according to ESA types. The highest plant survival rates have

80,00 70,00 Pinus Sylvestris 60,00 Robinia pseudoacacia L. 50,00 Fraxinus excelsior L 40,00 30,00 Acer negundo 20,00 Quercus petraea 10,00 Betula pendula 0,00

Seedling success % ratio Amygdalus communis 2010 2011 2012 Prunus Mahleb

Years FIGURE 7 -Seedling survival success percentages (%) during 2009-2012.

TABLE 7 - ANOVA test results of seedling survival success according to ESA types.

ESAs Type Plant type P F1 F2 F3 C1 C2 Pinus sylvestris 70.00a 63.00a 53.00b 47.00b 28.67a 18.67a Robinia pseudoacacia L. 73.33a 68.33a 58.33a 51.67a 49.33a 44.67a Fraxinus excelsior L. 63.33a 52.00a 45.67a 40.67a 37.33a 32.33a Acer negundo 53.00a 48.67a 45.33a 41.33a 37.33a 29.33a Quercus petraea 20.00a 16.00a 12.00a 8.67a 4.67a 2.33a Betula pendula 51.67a 43.67a 39.00a 35.00a 26.67a 21.67a Amygdalus communis 33.33a 25.00a 20.67a 17.67a 15.33a 7.00a Prunus mahleb 34.67a 27.33a 22.33a 21.33a 11.67a 3.33a

1200 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

80,00 70,00 60,00 Pinus Sylvestris 50,00 Robinia pseudoacacia L. Fraxinus excelsior L 40,00 Acer negundo 30,00 Quercus petraea 20,00 Betula pendula 10,00 Seedling success % ratio Amygdalus communis 0,00 Prunus Mahleb C2 C1 F3 F2 F1 P ESAs Type

FIGURE 8 - Seedling success ratios of environmentally sensitive types and tree species.

FIGURE 9 - Land preparation before planting seedlings FIGURE 10 - After planting seedlings

According to the research results, Robinia pseudoaca- of the seedlings species used to create the forest ecosys- cia L. were determined to have the highest seedling sur- tems were examined according to the MEDALUS environ- vival success rates in the Potential (P) and Critical (C2) mental sensitive areas and years. The environmentally sen- sensitivity areas, out of the success adaptability of the dif- sitive areas have affected the success rate of many plant ferent qualities of soil and climate [38]. Even though Pinus species, and especially in sensitive critical regions (C2 and sylvestris had the highest seedling survival success rate in C1) occurred resistance to rehabilitation made by conven- the potential risk (P) areas, it had the highest decrease of tional methods, and the seedling survival rates decreased in seedling survival success in the critical (C2) areas. Alt- these areas. According to the forestation types of ESAs, the hough Pinus sylvestris is more negatively affected by low highest seedling survival rate was identified in the Poten- soil quality areas, the high sand content and drainage prob- tial (P) areas described, respectively, as P > F1 > F2 > F3 lem areas have increased its seedling survival success rates > C1 > C2. The Critical (C2) areas were the most environ- in the Potential (P) risk areas. The land preparation before mentally sensitive but the highest seedling success rates planting seedlings in the study field is shown in Fig. 9, after were determined there. planting seedlings (Fig. 10). For a successful reclamation process, identification of spatial changes of environmentally sensitive areas (soils

4. CONCLUSIONS are fragile, critical and degraded) is as very important as the determination of appropriate seedling species in arid In the research, we planned to establish a new forest and semi-arid areas. ecosystem in degraded areas by applying classical methods Environmentally sensitive areas have different spatial- of soil preparation and planting. The survival success rates ity also in a small scale topography. Even in such a small

1201 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

area where soil quality index is negative, the other negative [7] Vitousek, P. M., D’Antonio, C. M., Loope, L.L. and ecological factors (vegetation quality, climate quality and Westbrooks, R. (1996) Biological invasions as global environmental change. American Scientist 84(5): 468-478. management quality) could not be tolerated by seedlings. To reduce the negative impacts of desertification, rehabili- [8] Puigdefábregas, J. and Mendizabal, T. (1998) Perspectives on desertification: western Mediterranean. Journal of Arid tation activities should begin in the areas where soil quality Environments 39(2): 209-224. index values are negative. Therefore, reclamation works will be dramatically successful with identification of envi- [9] Bakr, N., D. C. Weindorf, Bahnassy, M. H. and El-Badawi, M.M. (2012) Multi-temporal assessment of land sensitivity to ronmentally sensitive areas. The soil, climate, vegetation, desertification in a fragile agro-ecosystem: Environmental and management quality index should be identified. After indicators. Ecological Indicators 15(1): 271-280. that, proper land preparation and land reclamation tech- [10] Contador, J., Schnabel, S., Gutiérrez, A.G. and Fernández, niques should be done, and compatible types of plants M.P. (2009) Mapping sensitivity to land degradation in should be chosen, as well. Due to the fact that the soil and Extremadura. SW Spain. Land Degradation & Development air balance is damaged in these areas, organic matter 20(2): 129-144. should be added. In areas with high clay impermeable con- [11] Geymen, A. and Baz, I. (2008) Monitoring urban growth and tent layers, a deep tillage should be done to eliminate the detecting land-cover changes on the Istanbul metropolitan negative effects. If the plant species are selected according area. Environmental Monitoring and Assessment 136(1-3): 449-459. to the environmental sensitivity of areas, the selection will better increase the chances of the seedlings' successes. [12] Awasthi, K., Sitaula, B., Singh, B. and Bajacharaya, R. (2002) Land‐use change in two Nepalese watersheds: GIS and geomorphometric analysis. Land Degradation & Development 13(6): 495-513.

[13] Kosmas, C., Kirkby, M.J. and Geeson, N. (1999) The Medalus ACKNOWLEDGMENTS Project: Mediterranean Desertification and Land Use: Manual on Key Indicators of Desertification and Mapping The project was supported by the General Directorate Environmentally Sensitive Areas to Desertification, of Forestry in Turkey. The corresponding author would Directorate-General Science, Research and Development. like to express his gratitude to the General Directorate of [14] Kirkby, M., Abrahart, R., McMahon, M.J. and Thornes, J. Forestry, to give special thanks to members of the Erzurum (1998) MEDALUS soil erosion models for global change. Forest Service for establishing, maintaining, and measur- Geomorphology 24(1): 35-49. ing the research field plantings, and also like to thank the [15] Lavado Contador, J., Schnabel, S., Gomez Gutierrez, A., and East Anatolian Research Institute for laboratory analyses Pulido Fernandez, M. (2009) Assessing the environmental sensitivity to land degradation. A validation of the MEDALUS of the soils. method in SW Spain. EGU General Assembly Conference Abstracts. The author has declared no conflict of interest. [16] Salvati, L. and Zitti, M. (2009) Assessing the impact of ecological and economic factors on land degradation vulnerability through multiway analysis. Ecological indicators REFERENCES 9(2): 357-363. [17] Imeson, A. (1999) Land cover change, non-sustainable [1] Dregne, H. E. and Chou N.T. (1992) "Global desertification agriculture and desertification in mediterrranean Europe: the dimensions and costs." Degradation and restoration of arid transfer of experience from the Medalus project to other areas. lands: 73-92. [18] Cangir, C. and Boyraz, D. (2008) Climate change and [2] UNCCD (2002) United Nations Convention to Combat desertification or soil / land degradation in Turkey, and Desertification Secretariat. Fact Sheet 1. An introduction to the Combating Desertification Dimensions. Journal of Tekirdağ United Nations Convention to Combat Desertification and Agricultural Faculty 5: 169-186. Frequently Asked Questions. UNCCD, Bonn. [19] Öztaş, T. (1997) Soil Degradation. Ekoloji 22: 31-33. [3] Seager, R., Ting M., Held I., Kushnir Y., Lu, J., Vecchi, G., [20] Türkeş, M., Sümer U.M., and Çetiner, G. (2000) Flexibility Huang, H.P., Harnik, N., Leetmaa, A. and Lau, N.C. (2007) Mechanisms Under the Kyoto Protocol, Tesisat Dergisi 52: 84- Model projections of an imminent transition to a more arid 100. climate in southwestern North America. Science 316(5828): 1181-1184. [21] Sarıyıldız, T. and Ayan S. (2013) Analysis of Soil Properties During Desertification Process. [4] UNCCD (1994) United Nations convention to combat http://earsiv.kastamonu.edu.tr/jspui/handle/1/207 Date desertification in countries experiencing serious drought accessed: 09.11.2013. and/or desertification, particularly in Africa. A/AC.241/27, Paris. [22] Dindaroğlu, T. and Canbolat, M.Y. (2013) Evaluation of Soil Quality in Kuzgun Dam Lake Water Production Basin. Journal [5] UNCCD (1999) United Nations Convention to Combat of the Faculty of Agriculture 42(2), 145 Desertification. Secretariat for the Convention to Combat Desertification, Bonn. [23] M.G.M (2013) General Directorate of Meteorology. http://www.mgm.gov.tr/veridegerlendirme/il-ve-ilceler- [6] Okin, G. S., Parsons, A. Wainwright J., J., Herrick, J. E., istatistik.aspx?m=ERZURUM Date accessed: 09.10.2013 Bestelmeyer, B. T., Peters, D. C. and Fredrickson, E.L. (2009) Do changes in connectivity explain desertification? [24] Soil Survey Staff (2010) Keys to Soil Taxonomy, 11th ed. BioScience 59(3): 237-244. USDA-NRCS. GPO, Washington, DC.

1202 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[25] ESRI (2001) Environmental systems research institute. California, USA. [26] Amirian, P. (2013) Beginning ArcGIS for Desktop Development Using. NET, Wiley. com. [27] Bouyoucos, G. J. (1962) Hydrometer method improved for making particle size analyses of soils. Agronomy Journal 54(5): 464-465. [28] Soil Survey Division Staff (1993) Soil survey manual. Soil Conservation Service. U.S. Department of Agriculture Handbook 18. [29] McLean, E. (1982) Soil pH and lime requirement." Methods of soil analysis. Part 2. Chemical and microbiological properties (methodsofsoilan2): 199-224. [30] Nelson, R. (1982) Carbonate and gypsum. Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties (methodsofsoilan2): 181-197.

[31] Nelson, D. and Sommers, L. (1982) Total carbon, organic carbon, and organic matter, Methods of Soil Analysis, Part 2: Chemical and Microbiological Properties AL Page, RH Miller, DR Keeney, 539–580. Am. Soc. of Agron., Soil Sci. Soc. of Am., Madison, Wis. [32] M.T.A. (2002) Geological Map of Turkey. General Directorate of Mineral Reseach and Expolaration. Ankara. MTA, Map,.

[33] O.G.M. (2002) Forest Management Plan of Erzurum Forest Service (2002-2012). Forest General Management Ankara.

[34] Philip, G. and Watson, D.F. (1982) A precise method for determining contoured surfaces. Australian Petroleum Exploration Association Journal 22(1): 205-212.

[35] Hudson, G. and Wackernagel, H. (1994) Mapping temperature using kriging with external drift: theory and an example from Scotland. International journal of Climatology 14(1): 77-91. [36] Holdaway, M. R. (1996) Spatial modeling and interpolation of monthly temperature using kriging. Climate Research 6(3): 215-225. [37] Burrough, P. A., Burrough, P.A. and McDonnell, R. (1998) Principles of geographical information systems, Oxford university press Oxford. [38] Urgenc, S. (1990) Plantation Technique (Arboriculture). Istanbul University publication(ISBN: 975-404-220-9). [39] Norusis, M. (2008) SPSS 16.0 guide to data analysis, Prentice Hall Press. Received: May 26, 2014 [40] U.S.G.S. (1997) United States Geological Survey article Revised: August 11, 2014 “Desertification” Accepted: September 12, 2014 Maintaine d by Publications Service Center Last modified 10/29/1997. Date of Access: 08.03.2013 CORRESPONDING AUTHOR [41] U.N.E.P. (2007). Natural Disasters and Desertification. United Nations Environment Program. Post-Conflict Environmental Assessment. ISBN: 978-92-807-2702-9, p: 69 Turgay Dindaroğlu Kahramanmaraş Sütçü İmam University Faculty of Forestry Department of Forest Engineering 46100 Kahramanmaraş TURKEY

Phone: +90.344.2801815 Fax: +90.344.2801712 E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1195 - 1203

1203 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

DETERMINING THE TOTAL ANTIOXIDANT STATUS AND OXIDATIVE STRESS INDEXES OF HONEY SAMPLES OBTAINED FROM DIFFERENT PHYTOGEOGRAPHICAL REGIONS IN TURKEY

Ethem Akyol1,*, Zeliha Selamoglu1, Hamide Dogan1, Hasan Akgul2 and Adnan Unalan3

1 Department of Biology, Faculty of Arts and Science, Nigde University, 51200 Nigde, Turkey 2 Department of Biology, Faculty of Arts and Science, Gaziantep University, 27310 Gaziantep, Turkey 3 Nigde University, Faculty of Economics and Administrative Science, Department of Business Administration, 51240 Nigde, Turkey

ABSTRACT glucose oxidase, catalase, phosphatases), vitamins (ascorbic acid, niacin, pyridoxine etc.), organic acids (gluconic acid, This study aimed to determine the total antioxidant and acetic acid, etc.), flavonoids, phenolic acids, and other phy- oxidant status and oxidative stress index of honey samples tochemicals [3-5]. The content of honey depends on the plant which were collected from 11 different phytogeographical kinds visited by honeybees and external provisions (environ- regions in Turkey. Honey samples were kept at 4 ºC until ment, processing of honey and storage conditions) [6, 7]. analysis. Total antioxidant status, total oxidant status and Honey has both enzymatic (catalase, glucose oxidase, oxidative stress indexes were defined with in vitro analyses peroxidase) and non-enzymatic substances (ascorbic acid, in extracted honey samples, and then, the obtained data α-tocopherol, carotenoids, amino acids, proteins, Maillard were compared statistically. These characteristics of honey reaction products, flavonoids, phenolic acids). Therefore, were analysed by the most sensitive and reliable measure- honey has very important antibacterial and antioxidant ac- ment kits developed newly. In addition, biological activi- tivities. The vegetation of the area, from where the honey ties of honey samples originating from different regions is obtained, affects the amount and type of these antioxi- were also compared. As a result of the study, the highest dants. The antioxidant activity of honey depends on its to- total antioxidant capacity was observed (P<0.01) in the tal phenolic content [8, 9]. honey samples obtained from Duzce and Mugla regions. The results also showed that total antioxidant capacities of Oxidant substances are not only produced in vivo but honey samples could change according to various phytoge- also taken from the external environment. The amount of ographic properties of different regions in Turkey. Besides, various oxidant species can be determined in laboratories antioxidant properties of honey depended on phenolic separately, but these processes are time-consuming, labor- compounds that could change according to plant vegetation intensive, complicated and costly. When these procedures of honey obtained from different regions. are not practice for the measurement of different oxidant

molecules, the total oxidant status (TOS) of a material is measured, and this is named total peroxide, serum oxida- KEYWORDS: tion activity, reactive oxygen metabolites, or some other Antioxidant, honey, oxidant, in vitro analyses, Turkey synonyms [10]. Antioxidant molecules, such as albumin,

bilirubin, uric acid, vitamins C and E, and antioxidant en- 1. INTRODUCTION zymes, such as superoxide dismutase or glutathione peroxidase, are protecting the cells from the harmful effects of Turkey is suitable for all beekeeping activities due to oxidant agents. Concentrations of this antioxidant can be natural conditions, geographical location, favorable cli- measured individually but these analyses are time-consum- matic conditions, and rich vegetation. Honey is a natural ing, expensive and more complicated. The total antioxidant product collected from many plant nectars, secretion plants status (TAS), a recently developed method of measuring, or insects, and processed by honey bees [1]. Honey has been reflects concentration of enzymatic substances and non- used as both nutritional and therapeutic agent (treatment of availability of all antioxidants [11]. The oxidative stress in- burns, gastrointestinal disorders, asthma, skin ulcers and dex (OSI) is indicator of the degree of oxidative stress; a cataracts) [2]. Honey contains at least 200 substances in- new indicator of oxidative stress is the rate of TAS amount cluding mainly carbohydrates and water. It also contains of TOS values [12]. minerals, proteins, free amino acids, enzymes (invertase, The importance on human feeding of honey has increased recently. Therefore, investigators’ interest has been * Corresponding author attracted on this subject. Nowadays, studies on the importance

1204 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 1 - Locations of honey samples collected from Turkey: 1. Artvin, 2. Balikesir, 3. Duzce, 4. Edirne, 5. Karamanmaras, 6. Mersin, 7. Mugla, 8. Nigde, 9. Ordu, 10. Sivas, 11. Van.

of natural antioxidants are the term of human health. The at- 2.3 Biochemical assays tention of researchers is attracted by phenolic compounds 2.3.1 Total antioxidant status (TAS) and determination of their antioxidant properties. The TAS values were analysed using commercial kits (Rel amount and the chemical structures of these compounds assay, Turkey). The new automated method is based on the change according to the region where the honey is ob- bleaching of the characteristic colour of a more stable tained. In the light of this information, we aimed to deter- ABTS [2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic mine TAS, TOS and OSI values of honey samples obtained acid)] radical cation by antioxidants. The assay has excel- from different phytogeographical regions of Turkey with lent precision values, which are <3%. The results were ex- in vitro analyses. pressed as μmol Trolox equivalents/L [15]. TAS values were determined as mmol and converted to µmol.

2.3.2 Total oxidant status (TOS) 2. MATERIALS AND METHODS TOS were measured using commercially available kits 2.1 Honey samples (Rel assay, Turkey). In the new method, oxidants present in the sample oxidized the ferrous ion-o-dianisidine complex Honey samples were collected from 11 provinces to ferric ions. The oxidation reaction was enhanced by glyc- (Artvin, Balikesir, Duzce, Edirne, Kahramanmaras, Mer- erol molecules abundantly present in the reaction medium. sin, Mugla, Nigde, Ordu, Sivas and Van) that entirely rep- The ferric ions produced a colored complex with xylenol or- resent 7 geographic regions of Turkey (Fig. 1). Samples ange in an acidic medium. The color intensity, which could were collected from beekeepers at different months in be measured spectrophotometrically, was related to the total 2011. Flowering and maturation of honey samples show amount of oxidant molecules present in the sample. The as- differences depending on ecological, geographical and cli- say was calibrated with hydrogen peroxide, and the results matic conditions [13]. The samples were stored at 4° C un- were expressed in terms of micromolar hydrogen peroxide til they were analyzed. equivalents per L (μmol H2O2 equivalents/L) [10].

2.2 Preparation of honey extraction 2.3.3 Oxidative stress index (OSI) Honey samples (10 g) were added to 30 mL of 70% The oxidative stress index (OSI) was determined with ethanol, and they were mixed to maintain the homogeneity. the ratio of TOS to TAS. The unit of TAS was converted Then, the extracts were homogenated in a sonicator (Selec- to μmol/L from mmol/L, and then, the OSI was calculated tra Ultrasons) for 15 min. After then, solutions were added according to the following formula: OSI (arbitrary unit) = to 30 mL of 70% ethanol and again homogenated in a son- TOS (μmol H2O2 equivalents/L) / TAS×100 (μmol Trolox icator (Selectra Ultrasons) for 15 min. Obtained solutions equivalents/L) [12, 16, 17]. were filtered (Whatman no. 4 filter paper) and concentrated at 40 °C under reduced pressure in a rotary evaporator 2.4 Statistical analysis (Heidolph Heizbad HB 4000 digit). Prepared extracts were All data were analysed by comparing means with one- stored until analysis [14]. way ANOVA method using SPSS software (Chicago, IL,

1205 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

USA; version 16.0). The mean differences among the prov- Ordu and Sivas. Also between TAS values of honey sam- inces were compared with Duncan’s multiple comparison ples obtained from Edirne and Van no statistically signifi- test. cant differences had been observed (P>0.01).

3.2 Total oxidant status of honey 3. RESULTS Total oxidant capacities of honey samples were determined using commercial kits. These kits oxidize ferrous The results of variance analysis for each parameter (to- ion-o-dianisidine complex to ferric ions [10]. TOS values tal antioxidative status: TAS, total oxidant status: TOS, and of analysed honey samples are given in table 1. oxidative stress index: OSI) showed that mean differences among provinces were statistically significant (P<0.01). The TOS values of honey samples of Artvin, Balikesir, Duzce, Edirne, Kahramanmaras, Mersin, Mugla, Nigde, 3.1 Total antioxidative status of honey samples Ordu, Sivas and Van regions were 5.461±0.283 5.324±0.134 Total antioxidant capacities of honey samples were de- 5.818±0.170. 13.613±0.339. 8.193±0.143. 7.856±0.199. termined using commercial kits. TAS values of honey sam- 10.029±0.246. 6.388±0.251. 6.372±0.212. 3.922±0.163 ples are given in table 1. Honey samples obtained from dif- and 5.627±0.194 µmol H2O2 equivalents/L. respectively. As ferent floral sources had strong antioxidative properties, compared to the total oxidant capacities of honey samples and they had the scavenging activity of reactive oxygen obtained from different geographic regions: Honey sam- species [12]. Antioxidants are very important in cell de- ples obtained from Edirne, Karamanmaras and Mugla had fence, and in the prevention of certain illnesses, cardiovas- the highest TOS values (P<0.01). The lowest TOS value cular diseases, cancer, and diabetes, may be with oxidative was found from Sivas region's honey sample (P<0.01). damage [19]. There were not observed a statistically significant difference (P>0.01) among the TOS values of honey sam- The TAS values of honey samples obtained from ples collected from Duzce, Mersin, Nigde and Ordu. Also Artvin, Balikesir, Duzce, Edirne, Kahramanmaras, Mer- there were not found any statistically significant difference sin, Mugla, Nigde, Ordu, Sivas and Van were determined among the TOS values of honey samples collected from to be 713±0.050, 260±0.033, 2425±0.300, 1165±0.079, Artvin, Balikesir and Van (P>0.01). 1475±0.052, 1539±0.059 2913±0.037, 1362±0.074, 1373±0.070, 1665±0.055 and 1026±0.046 µmol Trolox 3.3 Oxidative stress index of honey equivalents/L, respectively. The highest TAS values (P<0.01) The oxidative stress index (OSI) was determined with were observed in the honey samples of Duzce and Mugla, the ratio of TOS to TAS. The OSI levels of honey samples compared to the total antioxidant capacities of honey samples obtained from the different regions Artvin, Balikesir, obtained from other geographical regions. The lowest TAS Duzce, Edirne, Kahramanmaras, Mersin, Mugla, Nigde, values were determined in honey samples obtained from Ordu, Sivas and Van were determined to be 0.792±0.094. Artvin and Balikesir (P<0.01). There were no statistically 2.196±0.299. 0.241±0.012. 1.180±0.073. 0.560±0.026. significant differences (P>0.01) among TAS values of honey 0.512±0.020. 0.345±0.015. 0.471±0.017. 0.512±0.028. samples obtained from Kahramanmaras, Mersin, Nigde, 0.237±0.013 and 0.557±0.043, respectively.

TABLE 1 - The TAS, TOS and OSI values of honey samples obtained from different regions.

Parameters TAS TOS OSI n (µmol Trolox Equiv./L) (µmol H2O2 Equiv./L) (TOS/TAS)×100 Locations 1.Artvin 5 713±50 f 5.461±0.283 e 0.792±0.094 c 2.Balikesir 5 260±33 g 5.324±0.134 e 2.196±0.299 a 3.Duzce 5 2425±30 b 5.818±0.170 de 0.241±0.012 d 4.Edirne 5 1165±79 e 13.613±0.339 a 1.180±0.073 b 5.Kahramanmaras 5 1475±52 d 8.193±0.143 c 0.560±0.026 cd 6.Mersin 5 1539±59 cd 7.859±0.199 c 0.512±0.020 cd 7.Mugla 5 2913±37 a 10.029±0.246 b 0.345±0.015 d 8.Nigde 5 1362±74 d 6.388±0.251 d 0.471±0.017 d 9.Ordu 5 1373±70 d 6.372±0.212 d 0.512±0.028 d 10.Sivas 5 1665±55 c 3.922±0.163 g 0.237±0.013 d 11.Van 5 1026±46 e 5.627±0.194 e 0.557±0.043 cd OVERALL 55 1447±96 7.146±0.357 0.687±0.078 For each parameter, different letters in the same column represent statistically significant mean differences (P<0.01).

1206 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

The highest OSI value was observed from Balikesir’ pacities of honeys. These analyses were tested with newly honey sample (P<0.01), and the lowest one sin honey sam- developed kits, being extremely reliable, sensitive, rapid ples obtained from Duzce, Mugla and Sivas. and simple, and according to other methods.

Statistically significant differences (P>0.01) among the The results of this study confirmed that antioxidant OSI values of honey samples obtained from Kahramanma- properties of honey samples are dependent on phenolic ras, Mersin, Nigde, Ordu and Van were not determined. The compounds. We observed diversities in the antioxidant ca- OSI level of Edirne’ honey sample was found to be signifi- pacities of honey samples obtained from different regions cantly lower (P<0.01) than that of Balikesir’ honey sample. of Turkey which had different ecologic and geographical The OSI level of Edirne’ honey sample was statistically structures. Phenolic compounds of honeybee products are higher significant (P<0.01) than Artvin honeys OSI level. excessively affected by the botanical and geographical origins and climatic properties of the regions [22, 24, 25]. The results of this work are parallel with literature [1, 26- 28]. The antioxidant activities of honey samples obtained 4. DISCUSSION AND CONCLUSION from different geographical localities may be changing as they depend on phytogeographic, climatic and ecological In this study, TAS values of 11 honey samples ob- conditions, such as soil components, moisture and temper- tained from different regions of Turkey were in the range ature [14]. of 260±0.033 and 2913±0.037 µmol Trolox equivalents/L.

TAS values of Indian honeys were in the range of 15.1±0.7 to 29.5±1.8 mg ascorbic acid equivalents/antioxidant con- ACKNOWLEDGEMENTS tent per 100 g honey [7]. Maximum TAS values of Argen- tinian honeys determined by DPPH and ABTS methods Nigde University Research Fund (FEB 2011/24) is were 10 and 2.73 µg/mL, respectively [20]. Antioxidant gratefully acknowledged for support of this work. status determined by DPPH, ABTS and FRAP assays in 40

Czech Republic honeys were in the ranges of 98.73 to The authors have declared no conflict of interest. 441.98, 431.38 to 1026.38, and 222.98 to 87.12 mg ascorbic acid equivalents per kg of honey evaluated [21]. Fer- reira et al. [1] determined the antioxidant levels of 3 Portu- REFERENCES gal honey samples by DPPH from 106.67±2.48 to 168.94±19.20 mg/mL, and reducing power values in the [1] Ferreira, I.C.F.R., Aires, E., Barreira, J.C.M. and Estevinho, range of 13.26±0.20 to 48.95±1.61 mg/mL. In the evalua- L.M. (2009) Antioxidant activity of Portuguese honey sam- tion of the antioxidant activity, the highest ABTS+ (2,2-az- ples: Different contributions of the entire honey and phenolic inobis-3-ethylbenzothiazoline-6-sulfonic acid) cation radi- extract. Food Chemistry, 114(4), 1438-1443. cal scavenging capacity was observed for the samples that [2] Devarajan, S. and Venugopal, S. (2012) Antioxidant and α- displayed the highest total phenolic contents [4]. Baltru- amylase inhibition activities of phenolic compounds in the ex- saityte et al. [22] have reported that the antioxidant activity tracts of Indian honey. Chinese Journal of Natural Medicines, of honey measured by reaction with ABTS+ were in the 10(4), 255-259. range of 50 to 95% of the free radical scavenging. Previous [3] Ashraf, M.W. and Akram, S. (2008) Physicochemical charac- studies on honey samples indicated that the presence of teristics and heavy metal contents of saudi arabıian floral honeys. Fresenius Environmental Bulletin, 17(7b), 877-881. compounds, such as polyphenols and flavonoids, may function as potential natural antioxidants [1, 5, 23]. [4] Silva, I.A.A., Silva, T.M.S., Camara, C.A., Queiroz, N., Mag- nani, M., Novais, J.S., Soledade, L.E.B., Lima, E.O., Souza, A.L. and Souza, A.G. (2013) Phenolic profile, antioxidant ac- TOS values of honey samples from 11 different re- tivity and palynological analysis of stingless bee honey from gions were determined to be between 3.922±0.163 to Amazonas, Northern Brazil. Food Chemistry, 141, 3552– 13.613±0.339 µmol H2O2 equivalents/L. OSI (TOS/TAS) 3558. levels of honey samples obtained from Duzce, Sivas and [5] Escuredo, O., Miguez, M., Fernandez-Gonzalez, M. and Seijo, Mugla localities were lower than OSI levels of honey sam- M.C. (2013) Nutritional value and antioxidant activity of hon- ples of other localities. Depending on these results, we ob- eys produced in a European Atlantic area. Food Chemistry, served that antioxidant activities of honey samples gath- 138, 851-856. ered from these regions were better than from other re- [6] Turhan, K. (2007) Chemical contents and some trace metals of gions. Additionally, OSI value of honey sample from Ordu honeys produced in the middle anatolia region of Turkey. Fresenius Environmental Bulletin, 16(5), 459-464. region was lower than that of honey sample from Artvin region. According to these data, we can say that antioxidant [7] Saxena, S., Gautam, S. and Sharma, A. (2010) Physical, biochemical and antioxidant properties of some Indian honeys. capacity of honey sample obtained from Ordu region is bet- Food Chemistry, 118(2), 391-397. ter than that from other regions. [8] Rosa, A., Tuberoso, C.I.G., Atzeri, A., Melis, M.P., Bifulco, E. and Dessì, M.A. (2011) Antioxidant profile of strawberry The data of our study showed that in vitro analyses tree honey and its marker homogentisic acid in several models could be used successfully to determine the antioxidant ca- of oxidative stress. Food Chemistry, 129, 1045-1053.

1207 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[9] Kowalski, S. (2013) Changes of antioxidant activity and for- [26] Küçük, M., Kolaylı, S., Karaoğlu, S., Ulusoy, E., Baltacı, C. mation of 5-hydroxymethylfurfural in honey during thermal and Candan, F. (2007) Biological activities and chemical and microwave processing. Food Chemistry, 141, 1378-1382. composition of three honeys of different types from Anatolia. Food Chemistry, 100, 526–534. [10] Erel, O. (2005) A new automated colorimetric method for measuringtotal oxidant status. Clinical Biochemistry, 38, [27] Silici, S., Sagdic, O. and Ekici, L. (2010) Total phenolic con- 1103-11. tent, antiradical, antioxidant and antimicrobial activities of Rhododendron honeys. Food Chemistry, 121, 238-243. [11] Yagcı, R., Özyurt, H., Akbaş, A., Aydın, B., Özlük, E., Ekşioğlu, M. and Totan, Y. (2007) Total Antioxidant Capac- [28] Tornuk, F., Karaman, S., Ozturk, I., Toker, O.S., Tastemur, B., ity, Total Oxidant Status and Dehydroepiandrosterone Sul- Sagdic, O., Dogan, M. and Kayacier, A. (2013) Quality char- phate Levels in Behcet’s Disease. Retina-Vitreus, 15, 263-266. acterization of artisanal and retail Turkish blossom honeys: [12] Yumru, M., Savas, H.A., Kalenderoglu, A., Bulut, M., Celik, Determination of physicochemical, microbiological, bioactive H. and Erel, O. (2009) Oxidative imbalance in bipolar disorder properties and aroma profile. Industrial Crops and Products, subtypes: a comparative study. Progress in Neuro-Psychophar- 46, 124-131. macology, 31, 33(6): 1070-1074. [13] Akyol, E. and Kaftanoğlu, O. (2001) Colony characteristics and the performance of Caucasian (Apis mellifera caucasica) and Mugla (Apis mellifera anatoliaca) bees and their recipro- cal crosses. Journal of Apicultural Research, 40, 3-4.

[14] Buratti, S., Benedetti, S. and Cosio, M.S. (2007) Evaluation of the antioxidant power of honey, propolis and royal jelly by amperometric flow injection analysis. Talanta, 71, 1387-1392. [15] Erel, O. (2004) A novel automated direct measurement method for total antioxidant capacity using a new generation, more stable ABTS radicalcation. Clinical Biochemistry, 37, 277-85. [16] Harma M, Harma M, Erel O (2003). Increased oxidative stress in patients with hydatidiform mole. Swiss Medcal Weekly, 133, 563-536.

[17] Kosecik, M., Erel, O., Sevinc, E. and Selek, S. (2005) In- creased oxidative stress in children exposed to passive smok- ing. International Journal of Cardiology, 100, 61-4.

[18] Nagai, T., Inoue, R., Kanamori, N., Suzuki, N. and Nagashima, T. (2006) Characterization of honey from different floral sources. Its functional properties and effects of honey species on storage of meat. Food Chemistry, 97, 256-262.

[19] Gasic, U., Keckes, S., Dabic, D., Trifkovic, J., Milojkovic- Opsenica, D., Natic, M. and Tesic, Z. (2014) Phenolic profile and antioxidant activity of Serbian polyfloral honeys. Food Chemistry, 145, 599-607. [20] Isla, M.I., Craig, A., Ordonez, R., Zampini, C., Sayago, J., Be- dascarrasbure, E., Alvarez, A., Salomon, V. and Maldonado,

L. (2011) Physico chemical and bioactive properties of honeys from Northwestern Argentina. LWT-Food Science and Tech- nology, 44, 1922-1930.

[21] Lachman, J., Orsak, M., Hejtmankova, A. and Kovarova, E. (2010) Evaluation of antioxidant activity and total phenolics of selected Czech honeys. LWT-Food Science and Technol- Received: June 05, 2014 ogy, 43, 52–58. Accepted: August 22, 2014

[22] Baltrusaityte, V., Venskutonis, P.R. and Ceksteryte, V. (2007) Radical scavenging activity of different floral origin honey and beebread phenolic extracts. Food Chemistry, 101, 502- CORRESPONDING AUTHOR 514. [23] Estevinho, L.M., Feas, X., Seijas, J.A. and Vazquez-Tato, Ethem Akyol M.P. (2012) Organic honey from Tras-Os-Montes region (Por- Department of Biology tugal): chemical, palynological, microbiological and bioactive Faculty of Arts and Science compounds characterization. Food and Chemical Toxicology, 50, 258-264. Nigde University Nigde, 5120 [24] Al-Mamary, M., Al-Meeri, A. and Al-Habori, M. (2002) An- tioxidant activities and total phenolics of different types of TURKEY honey. Nutrition Research, 22, 1041-1047. [25] Alves, A., Ramos, A., Goncalves, M.M., Bernardo, M. and Phone: ++90 5333599260 Mendes, B. (2013) Antioxidant activity, quality parameters E-mail: [email protected] and mineral content of Portuguese monofloral honeys. Journal of Food Composition and Analyses, 30, 130-138. FEB/ Vol 24/ No 4/ 2015 – pages 1204 - 1208

1208 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TRANSFORMATION OF NITROGEN IN BIOFILTERS USED IN DRINKING WATER TREATMENT

Huining Zhang*, Huihui Gan, Huixa Jin and Kefeng Zhang

Zhejiang University, Ningbo Institute of Technology, Ningbo, 315100, China

ABSTRACT ter sources and its removal during the water treatment process. Westerhoff et al. [4] reported that the average DON Dissolved organic nitrogen (DON) is the main precur- concentration in raw water entering U.S. drinking water sor for the formation of disinfection by-products (DBPs), plants is 0.186 mg/L. In China, many water sources are con- especially nitrogenous disinfection by-products (N-DBPs), taminated with micropollution. The average DON concen- in water treatment. Reducing the DON concentration in tration in raw water from the Huangpu River is 0.34 mg/L water before the disinfection process can help to minimize [5], and the DON concentration of Zhejiang Province ranges the formation of N-DBPs. However, new DON may be pro- from 0.9–1.8 mg/L [6]. Generally, water-bodies contami- duced during the bio-filtering process. In this study, the rule nated by wastewaters or agricultural pollution have a higher of mutual transformation of nitrogen in biofilters was ana- DON concentration. Besides, in wastewater from higher el- lyzed, and the relationship between nitrogen transformation evations or that contaminated by agricultural pollution, the and biomass in biofilters was preliminarily investigated. The DON in many rivers and drinking water sources is produced results show that non-dissolved organic nitrogen (NDON) by soluble microbial products (SMPs) and algae. could be removed by biofilters. The quartz sand filter showed the best NDON removal efficiency, followed by cer- Reducing the DON concentration in water before the amisite, activated carbon and anthracite coal filters. After fil- disinfection process facilitates the control of DBP formation. tration, total nitrogen (TN) and dissolved inorganic nitro- However, new DON may be produced during the bio-filter- gen (DIN) in water showed decreasing trends, whereas DON ing process. Particles or colloids in water are decomposed showed an increasing trend. Generally, TN was reduced, and into DON by microbes, and then DON is transformed into the reduction in DIN showed a weak correlation with the in- ammonia nitrogen under the action of ammonifying bacteria. crease in DON. Nitrification is not the only pathway of ni- SMPs are produced by microbes via degradation of the ma- trogen removal. Assimilation and anoxic denitrification also trix, or are the intermediate or final products of endogenous play a role in nitrogen transformation during the bio-treat- cell decomposition. SMPs and extracellular polymeric sub- ment process. The biomass at different depths in the filter stances (EPSs) constitute new DON [7]. showed a linear positive correlation with the decrease in TN + The major forms of nitrogen in raw water are NH4 -N, at the corresponding depth, and the correlation coefficient - - + - NO2 -N, NO3 -N, and organic nitrogen [8]. NH4 -N, NO2 - was highest for the top layer of the filter. The study results - N, and NO3 -N are commonly referred to as "3N" or dis- provide a basis for controlling DON concentration and DBP solved inorganic nitrogen (DIN). Organic nitrogen is the ge- formation in drinking water treatment plants. neric term for nitrogen-containing organic compounds, including DON and non-dissolved organic nitrogen (NDON) [9, 10]. There is mutual transformation between DIN, DON, KEYWORDS: Biofilter; DON; biomass; drinking water; disinfection by-products (DBPs) and NDON due to the influence of microbes in the biofilter. Understanding the rule of mutual transformation of nitro-

gen in biofilters is very important to achieve control over 1. INTRODUCTION the formation of DBPs in the subsequent disinfection process. Dissolved organic nitrogen (DON) is considered to be In this study, the DIN, DON, total nitrogen (TN), and the most important precursor in the formation of disinfec- corresponding biomass contents in 4 biofilters with various tion by-products (DBPs) during water treatment. Com- filling materials were monitored, and then a preliminary pared with DBPs formed without nitrogen, nitrogenated model for the mutual transformation of nitrogen was con- DBPs (N-DBPs) have stronger carcinogenicity and muta- structed. Based on the biomass in biofilters, the relation- genicity [1-3]. Several studies have examined DON in wa- ship between nitrogen transformation and biomass was examined, in order to provide a basis for the control of DON * Corresponding author and DBPs in raw water for drinking water treatment.

1209 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2. MATERIALS AND METHODS terminations were carried out according to the Chinese Na- tional Standard Methods [11]. The lipid phosphorus method 2.1 Study location was used to determine media biomass [12]. Ammonium mo- The experiments for this study were carried out at a lybdate spectrophotometry was used to determine the total Surface Water Treatment Plant in Zhejiang Province. The phosphorus (TP) concentration. All the chemical reagents filtering materials were quartz sand, activated carbon, cer- were supplied by Chemical Reagent Sinopharm Group Co. amisite, and anthracite coal. The testing unit was installed Ltd. TN represents the total nitrogen content of water sam- near the sedimentation tank, and the water samples tested ples without filtration through a 0.45-μm filter membrane, were taken from the effluent of the sedimentation tank. The and TDN represents the total dissolved nitrogen content of unit primarily consisted of 4 biofilters with identical spec- water samples with filtration through a 0.45-μm filter ifications (inner diameter = 10 cm, filtering material height membrane. DON was calculated as follows: DON = TDN- + - - = 100 cm, and total height of biofilter= 250 cm). To deter- DIN. DIN is the sum of NH4 -N, NO2 -N, and NO3 -N. mine the various concentrations in effluent samples and at NDON was calculated as follows: NDON = TN-DIN-DON. different height in the biofilters, sampling hatches were ar- The biomass was measured using the Lipid-P method, in ranged at the corresponding positions along the biofilters. which the amount of phosphorus in phospholipids, the A down-flow biofilter design was employed; that is, the main constituent of cell membranes, was used to represent water entered from the top and flowed out from the bottom. biomass. Units of nmol-P/cm3 were used for the filtering When the unit was in service, the effluent from the sedi- material, and 1 nmol-P was equivalent to 108 Escherichia mentation tank entered the 4 biofilters through separate si- coli cells [12]. The reagent solution was prepared by am- phon pipes. Under the control of a flow-meter, the influent monium molybdate spectrophotometry for TP determina- flow was kept constant at 60 L/h and the hydraulic retention [11]. tion time was 8 min. Therefore, the filtering speed was about 7–9 m/h. The biofilters were back-flushed with tap water, and the intensity and pressure head for back-flush- 3. RESULTS AND DISCUSSION ing was controlled by a valve. The tap water was dechlo- rinated by a huge activated carbon filter of 50 cm diameter 3.1 Nitrogen variation in biofilters and 100 cm height. A backflushing cycle was once per The variations in DIN, DON, TN, and NDON in the in- day, and the experiments lasted 37 days. fluent and effluent of the 4 biofilters were analyzed in our experiment (Fig. 1). The DIN and DON were monitored for 2.2 Analytical methods 37 days. The NDON content in the filtered water was lower Salicylic acid-hypochlorite, N-(1-naphthyl)-ethylene- than that in the influent, which means that NDON was re- diamine, ultraviolet, and alkaline potassium persulfate di- moved by the biofilter. The biofilter containing quartz sand gestion ultraviolet spectrophotometric methods were used removed the most NDON, followed by the ceramisite, the + - - to determine NH4 -N, NO2 -N, NO3 -N, and DIN concen- activated carbon and the anthracite coal filters. The specific trations (UV-2000 UNICO, China), respectively. These de- surface area and structural properties of the 4 types of me-

1.2 8 NDON DON TN DIN 1.0 6 0.8

0.6 4

N,0.4 mg/L 2 0.2

0.0 0 ent on nd ite oal ent on nd ite oal flu arb sa is e c flu arb sa is c In d c rtz ram cit In d c rtz ram cite ate ua Ce hra te ua Ce ra tiv Q nt tiva Q nth Ac A Ac A FIGURE 1 - Concentrations of different forms of nitrogen in influent and effluent of biofilters (n=3).

1210 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

dia were measured. There were differences in particle size, -- NO0.5 O  NO (2) porosity, and other surface properties among the media. Ac- 22 3 tivated carbon had the greatest porosity at 66.7%, whereas Nitrification cannot remove elemental nitrogen in wa- quartz sand had the lowest one at only 45.0%. The porosity ter. However, the above data did not indicate a linear cor- of ceramisite and anthracite were 53.8 and 52.9%, respec- relation between the decrease in DIN and increase in DON tively. Quartz sand could not easily retain the flocs or ag- in water, which means that the nitrogen was lost. There- glomerates. Thus, aggregated microorganisms and particles fore, nitrification is not the only pathway by which the ni- adhered to the microbial component. The biomass adhered trogen is removed. + to the media affecting the nitrogen variation in biofilters, and In addition to nitrification, NH4 -N can be transformed that containing quartz sand removed the most NDON. via another pathway during biological treatment. Gener- The TN in water decreased slightly with filtration. DIN ally, the non-nitration reactions that can remove nitrogen also decreased, and only DON increased with filtration. include assimilation and anoxic denitrification. According to the statistical analysis of t-test, the difference Assimilation is a process in which microbes absorb nu- values of DIN were extremely significant (P = 0.000, P = trients from the environment, metabolize them into new 0.000, P = 0.003, P = 0.003) in activated carbon, quatrz cellular materials or metabolites and, therefore, acquire es- sand, ceramisite and anthracite coal biofilters. The differ- sential energy for life activity. In assimilation, nitrogen ence values of DON in activated carbon biofilter (P = participates in the synthesis of microorganisms, and thus, 0.001) and quatrz sand biofilter (P = 0.009) were also sig- some inorganic nitrogen in water is exhausted. Microor- nificant, and difference values of DON in ceramisite (P = ganisms of inorganic nitrogen assimilation include photo- 0.041) and anthracite coal (P = 0.023) biofilters were ex- synthetic bacteria and algae. Some algae are used to re- tremely significant. It is well-documented that the DON move ammonia nitrogen, phosphorus and organic matter in concentration is elevated during filtering of water sources the wastewater treatment industry. After running the filters through sand filters in the conventional water treatment for a period of time, the amount of adhering microbes grad- process [6]. This is consistent with our findings. Due to the ually increased. Some of the microbes were retained when limitations of on-site testing conditions, the concentration the water sample was filtered through the filtering material, of elemental nitrogen dissolved in water samples was not and some were newly generated. These microbes were determined. Notably, the decrease in DIN and NDON com- flushed away after back-flushing. bined was greater than the increase in DON. Anoxic denitrification is a process in which, under an-

oxic conditions, denitrifying bacteria release NO, N2O, and 2.2 Relationship between decrease in DIN and increase in DON - N2 with organic matter as the electron donor, and NO3 -N - To investigate the mutual transformation of DIN and or NO2 -N as the electron acceptor. Because there is a gra- DON, the linear correlation between the decrease in DIN dient variation in dissolved oxygen in the biofilms of bio- and increase in DON was analyzed (Table 1). The results filters, an anoxic environment is created in the biofilms show that, in both the biofilm culturing and biofilm matu- [15]. Therefore, anoxic denitrification theoretically can oc- ration stages of the 4 types of filters, the decrease in DIN cur in the biofilter. and increase in DON were weakly correlated. The increase in DON in the effluent from biofilters is + Generally, NH4 -N in water can be removed by biolog- generally attributed to microbial activity (i.e., SMPs re- + ical nitrification, in which NH4 -N is oxidized to nitrate by leased by microbes and decomposed NDON). As seen + nitrifying bacteria. In nitrification, NH4 -N is firstly trans- from the above data analysis, the linear correlation between - formed into NO2 -N by ammonia-oxidizing bacteria the decrease in DIN and the increase in DON was weak. - - (AOB), and then NO2 -N is transformed into NO3 -N by ni- This means that when studying the transformation of nitro- trite oxidation bacteria (NOB). Nitrosation and nitration gen in the complete biofilter system, variations in the are expressed in formulas (1) and (2) [13, 14], respectively. amount of nitrogen in all forms should be comprehensively

- considered in order to accurately determine the rule of NH421.5 O NO 22 H O  2 H (1) DON transformation.

TABLE 1 - First-order linear correlation between reduction in DIN and increase in DON.

Coefficient Activated carbon Quartz sand Ceramisite Anthracite coal R2 0.0004 0.3132 0.4241 0.6539 Biofilm culturing stage a -0.0176 -0.4383 -0.6125 -0.8971 b 0.2059 0.0256 0.0008 0.0301 R2 0.5484 0.1802 0.3208 0.4763 Biofilm maturation stage a -1.9888 -0.3144 -0.5276 -0.4714 b -0.2282 0.1688 0.1181 0.0582 ΔDON = a*ΔDIN + b

1211 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.3 Relationship between decrease in TN and biomass in bio- ceramisite filters. In a previous study, the amount of biomass filters within the filtering material in an activated carbon filter was Comprehensive analysis of nitrogen transformation in greater than that in a quartz sand filter (610 nmol(P)/g and these biofilter systems indicated that the TN concentration 280 nmol(P)/g, respectively) [16]. The volumetric weights in the effluent was less than that in the influent. The of the activated carbon and quartz sand were measured to amounts by which the percentage of TN and the biomass be 0.45 g/cm3 and 2.62 g/cm3, respectively, and then, the decreased in each biofilter are shown in Fig. 2. The bio- biomass concentrations were calculated to be 274.5, and mass represented the amount of microbes adhering onto 733.6 nmol(P)/cm3, respectively. These findings are con- each unit volume of filling material at 5 cm from the sur- sistent with our experimental results. face of the filter. The microbial biomass at this site was the Notably, the distribution of the reductions in the per- representative value, and thus, it was used to measure the centage of TN among the 4 filters was similar to that of total biomass in the 4 filters. As the figure shows, the re- biomass. The relationship between the biomass at different duction in the percentage of TN was greatest in the quartz depths in each filter and the decrease in the percentage of sand filter. TN was analyzed, and the results indicated a positive linear The quartz sand filter had the highest amount of bio- correlation. Moreover, a linear analysis was conducted on mass, followed by the activated carbon, anthracite coal and the relationship between the biomass within the filtering

16 160

14 (b) (a) 120 12 (media) 3

10 80 8 TN, %

Δ 6 40 4 Biomass, nmol (P)/cm 2 0 Activated carbon Quartz sand Ceramisite Anthracite coal Activated carbon Quartz sand Ceramisite Anthracite coal

FIGURE 2 - Decrease in the percentage of TN in the effluent (a) and biomass (b) in the different biofilters.

15 1.0

12 (a) 0.8 (b)

9 0.6 2 R

TN, % 6 0.4 Δ y = 0.1114x + 0.1115 3 0.2 R2 = 0.8834

0.0 0 0 20406080100120 0 20406080100 Biomass, nmol(P)/cm3(media) Media depth, cm

FIGURE 3 - Change in the linear correlation coefficient R2 between the biomass in the filtering material at different depths and the decrease in the percentage of TN in the filters (a: linear relationship between the biomass in the filtering material at a depth of 15 cm and the decrease in the percentage of TN in each biofilter; b: linear correlation coefficient, R2, at each filtering layer).

1212 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

material at each depth and the decrease in the percentage anoxic denitrification also play a role in the transformation of TN in the biofilters (Fig. 3). The correlation was found of nitrogen during bio-treatment processes. to be relatively higher in the top section of the filter (0-30 (3) The amount of biomass at different depths within 2 cm), and the correlation coefficient, R , was highest the filters showed a positive linear correlation with the de- (0.8604) at 10 cm. With increasing filter depth, especially crease in TN at the corresponding depth, and the correla- from 30–40 cm, this correlation rapidly declined, indicat- tion coefficient was highest in the upper layer of the filters. ing that the decrease in TN was not well correlated with the biomass in the filter. Accordingly, the correlation coefficient was lower in the lower section of the filter. ACKNOWLEDGEMENTS There are several pathways leading to the decrease in nitrogen content. The DN can be transformed into organic The authors would like to acknowledge financial sup- nitrogen through assimilation by microbes, or into gaseous port provided by the National Natural Science Foundation nitrogen via the action of microbes. Assimilation is related of China (NSFC) (No. 50678080) and Ningbo Science and to the microbial biomass in the filtering material. The up- Technology Program (2014C50008) for this work. per layer of the biofilters showed a stronger linear correlation between the decrease in TN and biomass, whereas the The authors would like to acknowledge financial sup- lower layer showed a very weak correlation. This means port provided by the National Natural Science Foundation that the decrease in TN in the lower layer may be due to of China (No. 50678080), Ningbo Natural Science Foun- transformation of gaseous nitrogen. dation (No. 2014C50008, 2014A610093), Scientific Tech-

A previous study showed that N2 was transformed nological and Social Development Projects of Ningbo (No. + from NH4 -N by a biomembrane in an activated carbon fil- 2014C50008, 2014C50007), Zhejiang Provincial Natural ter [15]. The existence of an anoxic region inside the acti- Science Foundation (No. LQ14E090003) and China Spark vated carbon facilitated the nitrification and denitrification Program (No. 2014GA701006) for this work. + in the aerobic and anoxic environments, and thus, NH4 -N was transformed into N2 [15]. Bock et al. [17] found that The authors have declared no conflict of interest. when DO was limited, ammonia-oxidizing bacteria (AOB) + can utilize H2 and NH4 -N as electron donors to initiate the - NO2 -N reduction reaction. Kuai et al. [18] reported that REFERENCES under the conditions of limited DO, AOB utilize O2 as an + - electron acceptor and oxidize some NH4 -N into NO2 -N. [1] Lee, W., Westerhoff, P. and Croue, J.P. (2007) Dissolved or- + - Then, NH4 -N is used as an electron donor, and NO2 -N is ganic nitrogen as a precursor for chloroform, dichloroacetoni- reduced to N . trile, N-Nitrosodimethylamine, and trichloronitromethane. 2 Environmental Science & Technology. 41(15), 5485-5490. Besides the N2, nitrous oxide (N2O) was also produced [2] Gopal, K., Tripathy, S.S., Bersillon, J.L. and Dubey, S.P. during the nitrification process. Experiments suggest that (2007) Chlorination byproducts, their toxicodynamics and re- N2O production is significant during nitrification process, moval from drinking water. Journal of Hazardous Materials. and lower DO resulted in higher N2O production [19]. 140(1-2), 1-6. Measurements at lab-scale and full-scale wastewater treat- [3] Muellner, M.G., Wagner, E.D., McCalla, K., Richardson, ment plants (WWTPs) have demonstrated that N2O can be S.D., Woo, Y.T. and Plewa, M.J. (2007) Haloacetonitriles vs. emitted in substantial amounts during nitrogen removal in regulated haloacetic acids: Are nitrogen-containing DBPs WWTPs [20]. more toxic? Environmental Science & Technology. 41(2), 645-651. Therefore, it is concluded that the decrease in TN in [4] Lee, W., Westerhoff, P. and Esparza-Soto, M. (2006) Occur- the lower layer of the biofilters was related not only to as- rence and removal of dissolved organic nitrogen in US water similation by the microbes, but also to the generation of treatment plants. Journal American Water Works Association. gaseous nitrogen under an anoxic environment. 98(10), 102-110. [5] Xu, B., Ye, T., Li, D.P., Hu, C.Y., Lin, Y.L., Xia, S.J., Tian, F.X. and Gao, N.Y. (2011) Measurement of dissolved organic 4. CONCLUSION nitrogen in a drinking water treatment plant: Size fraction, fate, and relation to water quality parameters. Science of the Total Environment. 409(6), 1116–1122. (1) NDON was removed by filtration in biofilters. The quartz sand filter showed the best NDON removal effi- [6] Gu, L., Liu, B. and Yu, X. (2010) Dissolved organic nitrogen (DON) in the processes of polluted source water treatment. ciency, followed by ceramisite, activated carbon and an- Chinese Science Bulletin. 55(27-28), 3098-3101. thracite coal filters. After filtration, the TN and DIN concentrations in water decreased, whereas the DON concen- [7] Boero, V.J., Eckenfelder, W.W. and Bowers, A.R. (1991) Sol- uble Microbial Product Formation in Biological-Systems. Wa- tration increased. Overall, the TN content was reduced. ter Science and Technology. 23(4-6), 1067-1076. (2) A weak correlation was observed between the de- [8] Westerhoff, P. and Mash, H. (2002) Dissolved organic nitro- crease in DIN and the increase in DON. Nitrification is not gen in drinking water supplies: a review. Journal of Water the only pathway for nitrogen removal. Assimilation and Supply Research and Technology-Aqua. 51(8), 415-448.

1213 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[9] Sattayatewa, C., Pagilla, K., Pitt, P., Selock, K. and Bruton, T. (2009) Organic nitrogen transformations in a 4-stage Bardenpho nitrogen removal plant and bioavailability/biodeg- radability of effluent DON. Water Research. 43(18), 4507- 4516. [10] Krasner, S.W., Weinberg, H.S., Richardson, S.D., Pastor, S.J., Chinn, R., Sclimenti, M.J., Onstad, G.D. and Thruston, A.D. (2006) Occurrence of a new generation of disinfection byproducts. Environmental Science & Technology. 40(23), 7175- 7185. [11] Chinese SEPA (2002). Water and Wastewater Monitoring Methods. 4th ed Chinese Environmental Science Publishing House, Beijing , China.

[12] Yu, X., Zhang, X. and Wang, Z. (2002) Biomass examination by lipid-P method for drinking water bio-treatment. Water & Wastewater Engineering. 28(5), 1-6.

[13] Zheng, P., Xu, X. and Hu, B. (2004) New theory and techniques for biological nitrogen removal. Science Press. [14] van der Aa, L., Kors, L., Wind, A., Hofman, J. and Rietveld, L. (2002) Nitrification in rapid sand filter: phosphate limitation at low temperatures. Innovations in Conventional and Ad- vanced Water Treatment Processes. 2(1), 37-46.

[15] Liu, B. (2011) Study on the mechanism of the effect of drink- + ing water biofilter on NH4 -N and DON, in Institute of Urban Environment,. Chinese Academy of Sciences.

[16] He, Y.C., Yu, J., Zhang, C., and Ren W.H. (2004) Study on the performance of the treatment of slightly-polluted water with biofiltration. Industrial Water Treatment. 24(11), 29-32.

[17] Bock, E., Schmidt, I., Stuven, R. and Zart, D. (1995) Nitrogen loss caused by denitrifying Nitrosomonas cells using ammonium or hydrogen as electron-donors and nitrite as electron- acceptor. Archives of Microbiology. 163(1), 16-20. [18] Kuai, L.P. and Verstraete, W. (1998) Ammonium removal by the oxygen-limited autotrophic nitrification-denitrification system. Applied and Environmental Microbiology. 64(11), 4500-4506. [19] Zheng, H., Hanaki, K. and Matsuo, T. (1994) Production of nitrous oxide gas during nitrification of wastewater. Water Science and Technology. 30(6), 133-141.

[20] Kampschreur, M.J., Temmink, H., Kleerebezem, R., Jetten, M.S. and van Loosdrecht, M. (2009) Nitrous oxide emission during wastewater treatment. Water research. 43(17), 4093- 4103.

Received: June 15, 2014 Revised: September 22, 2014 Accepted: November 21, 2014

CORRESPONDING AUTHOR

Huining Zhang Zhejiang University Ningbo Institute of Technology Ningbo, 315100 P.R. CHINA

Phone & Fax: +86 574 88130283 E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1209 - 1214

1214 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

LIFE CYCLE ASSESSMENT OF WASTE TIRE PYROLYSIS

Müfide Banar

Department of Environmental Engineering, Faculty of Engineering, İki Eylül Campus 26555, Anadolu University, Eskişehir, Turkey

ABSTRACT the separate collection of tires from vehicle dismantlers, and encourages the recycling of tires. In addition, the EU In this study, in order to provide the input for decision- ‘Waste Landfill Directive’ has banned the landfilling of makers in Turkey, the environmental impact of a waste tire tires. These directives have changed direction of waste tire pyrolysis plant was researched using the Life Cycle As- treatment in the EU over the last 15 years. For example, in sessment (LCA) method. The functional unit is defined as 1996, approximately 50% of waste tires were sent to land- 1 t of waste tires entering the pyrolysis process. The system fills; however, currently the figure is only 4% (0.13 million boundary, including feedstock pretreatment and pyrolysis, tonnes/year) [2]. was illustrated and material/energy flows, including raw The management of waste tires is intended to follow a material, pyrolysis products were determined according to hierarchical approach, that is, to decrease the environmen- a pilot pyrolysis plant and literature data. The LCA calcu- tal impact according to the following order: waste minimi- lations were carried out using licensed SimaPro 8 software. sation, reuse, recycling, energy recovery, and landfilling. At the impact assessment step, the CML-IA baseline Legislation related to waste management requires the (V3.00) method was applied for selected impact categories search for economical and environmental mechanisms that (abiotic depletion, global warming, human toxicity, marine can contribute solving this waste disposal problem. Waste aquatic toxicity, acidification and eutrophication). The char- tire disposal in landfills is banned in the EU, with minimi- acterization results show that all the impact categories have sation and reuse being options with only limited applica- negative values, except for acidification. Negative values re- bility, and recycling being insufficient to mitigate the dis- sulting from avoided products were due to the valuable posal problem by itself. Therefore, energy recovery seems products of pyrolysis. The acidification impact resulted to have a high potential to process and to valorize waste mainly from SO2 emissions in combustion flue gases. As a tires. Thermochemical processes, such as pyrolysis, gasifi- summary, the results show that utilization of pyrolysis cation and combustion, offer important advantages from an products in sufficient quantities and the application of flue energy point of view to address this challenge. More than gas treatment systems play an important role in presenting 1.15 million tonnes of waste tires (>3.3 million tonnes) are pyrolysis as an environmentally effective solution for used as fuel in cement kilns each year in the EU. Other en- waste tires. ergy recovery options for tires include use in power plants and co-incineration with other wastes, which use approxi-

mately 0.1 million tonnes per year of tires. Approximately, KEYWORDS: life cycle assessment (LCA), pilot pyrolysis plant, py- 1.1 million tonnes of tires are used in material recovery op- rolysis, simapro 8, waste tire tions through the production of rubberised flooring for sports fields and playgrounds, paving blocks, roofing ma-

terials etc. A significant proportion of waste tires are used 1. INTRODUCTION in civil engineering applications, such as road and rail foundations and embankments (0.24 million tonnes), re- The remarkable worldwide increase in the number of treads (0.26 million tonnes), or are exported (0.33 million vehicles, and a lack of both technical and economical dis- tonnes) each year [1-3]. posal mechanisms made for waste tires are being consid- Other treatment alternatives are based on thermal tech- ered as a serious pollution problem [1]. It is estimated that nologies. Pyrolysis is becoming one of the best thermal al- 1.5 billion tires are produced worldwide each year, most of ternatives for waste tires. Pyrolysis of tyres is viewed as an which will eventually end up as waste. In terms of tonnage, environmentally attractive and viable technological route waste tires represent a significant proportion of the total for the recycling of scrap tyres that, depending on market solid waste stream. For example, approximately 3.3 million conditions, all products of the pyrolysis process, the char, tonnes of waste tires were generated annually within Europe oil, gas and residual steel may have an end use [4]. Table 1 (EU-27) in 2010, with an estimated stockpile of 5.7 million shows examples of commercial and semi-commercial tire tonnes of waste tires throughout Europe. The management pyrolysis systems around the world. The table shows some of waste tires in the European Union has been regulated examples, rather than an exhaustive list of the wide range of under the ‘End of Life Vehicle Directive’, which stipulates companies which have developed tire pyrolysis technologies.

1215 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 1 - Examples of commercial and semi-commercial tire pyrolysis systems [2]

Company Location Reactor Type Capacity (tons/day) Splainex Ltd The Hague, Netherlands Rotary kiln approx. 20 Xinxiang Doing Renewable Energy Equipment Co., Ltd Xinxiang, China Rotary kiln 6–10 RESEM Shangqui, China Rotary kiln 8–20 Kouei Industries Vancouver, Canada Fixed bed/Batch 16 DG Engineering Gummersbach, Germany Rotary kiln approx.10 FAB India Amedabad, India Rotary kiln 5–12 Octagon Consolidated Selangor, Malaysia Rotary kiln 2.4–120 No-Waste Technology Reinach, Germany Fixed bed/Batch 4 PYReco Teeside, UK Rotary kiln 200 Pyrocrat Systems Navi Mumbai, India Rotary kiln 2–10

Batch tire pyrolysis reactors have throughputs of typi- different processes for the end life treatment of exhausted cally between 1–2 tonnes per day, and for increased through- tires. For that reason, this study aims to determine the en- puts, additional modules could be added. For larger through- vironmental impact of waste tire pyrolysis in the context of puts, continuous tire pyrolysis reactors have been developed, a pilot plant using the LCA method. To the best of our the most common being rotary kiln reactors [2]. knowledge, this is the first LCA study on waste tire pyrolysis in Turkey. Therefore, the findings of this study should In Turkey, annual tire production numbers reach ap- help decision-makers and practitioners in Turkey, and also proximately 24 million units. In addition, annually, 8 mil- would make a genuine contribution with real plant data to lion units of waste tires are generated. Despite such large the related literature. quantities, the number of recovery and disposal facilities for waste tires are limited [5]. The Ministry of the Environ- ment and Urban has granted licenses to 22 recovery facili- 2. MATERIALS AND METHODS ties between 2006-2013. This corresponds to a total capacity of 113,500 tons/year. The recovery of waste tires as This study was conducted according to standard LCA granulated material is not sufficient for effective treatment. In relation to this, The Regulation on Control of End of Life guidelines (TSE EN ISO 14040:2006 and TSE EN ISO Tires came into force in 2006 (Official Gazette of 25 No- 14044:2006) developed by the International Organization for Standardization [10, 11]. The LCA study was consid- vember 2006, number 26357) [6]. This regulation aims to prevent direct and indirect delivery of waste tires to recep- ered under three sections: the goal and scope definition tor platforms which may harm the environment, the instal- (functional unit, system description and system boundaries), life cycle inventory (data collection and allocation lation of collection and carriage of tires for recycling or disposal, and to establish a management plan. According procedure), and life cycle impact assessment. to this regulation, plants, which produce carbon black from 2.1 Goal and scope definition waste tires via pyrolysis, have to have an environment li- cense. However, there are no environmental limit values The aim of the study is to evaluate the potential envi- for gaseous emissions resulting from the pyrolysis of waste ronmental impact of waste tire pyrolysis in Turkey using tires. Hence, unique regulations for pyrolysis plants do not pilot pyrolysis plant data. The functional unit is defined as exist in Turkey, and limit values for gaseous emissions re- 1 ton of waste tires entering the pyrolysis process. sulting from waste combustion plants are taken into account for pyrolysis plants (Large Combustion Plants Reg- 2.2 System description ulation, Official Gazette of 8 June 2010, number 27605, The system that was modeled in the LCA method is Regulation on Control of Industrial Air Pollution, Official based on a pilot pyrolysis plant. This plant (Aker Bioen- Gazette of 3 July 2009, number 27277) [7]. ergy, http://akerbioenerji.com) was established in the province of Adapazarı/Sakarya. Adapazarı is an industrial city LCA is a valuable tool for providing an overall evalu- in the Marmara Region of Turkey, 55 km from the Mar- ation or comparison of any potential environmental impact mara Sea. of various waste management technologies. However, a limited number of LCA studies on the pyrolysis of waste The process includes pre-treatment (shredding and tires have been investigated. Li et al. (2010) [8] compared grinding) and pyrolysis processes. In the pre-treatment pro- the potential environmental impact of four different end of cess, waste tires are separated into 3 groups: truck tires, life tire (ELT) treatment technologies in China. Corti and passenger tires and working machine tires. Before shred- Lombardi (2004) [9] used life cycle assessment to compare ding, the sidewall wires of the truck tires are removed with

1216 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

a drawing machine, and working machine tires are cut into  For the waste tire treatment process, with regards to large pieces using a hydraulic cutting device. A drawing the impact related to the life cycle of materials and machine is not applicable for working machine tires, since fuels required during the processes, only electric they have thicker wires, with a diameter of 9-10 cm. The power generation and heating oil consumption were hydraulic cutting device helps to divide a working machine considered. tire into 5 pieces to extract the wires. Next, truck tires without wires, cut working machine tires and raw passenger tires  Secondary waste transport and treatment (such as land- are fed into the shredder together. The chips enter a grinder filling) were excluded from the system boundaries. to obtain rubber granules of a suitable size for pyrolysis.  Infrastructure was also excluded from the system In the pyrolysis process, rubber granulates of 20 mm boundaries. are fed into a pyrolysis reactor. Pyrolysis is achieved by a horizontal rotary batch reactor. The pyrolysis temperature  Negative environmental effects were considered for is 400 °C. The reactor reaches this temperature after ap- the avoided materials. proximately 3 h, remains constant for 4 h, and returns to its initial temperature after a further 4 h. So, a batch pyrolysis As seen in Fig. 1, the total environmental impact, re- takes about 11 h. The output of this process consists mainly garding waste tire pyrolysis, consists of 3 components: in- of 3 fractions: gaseous fraction (hydrogen, methane and direct impact caused by energy and the material production carbon oxides), liquid fraction (water, tar and oils), and a stage, direct impact caused by waste tire pyrolysis, and solid product (char, ash and metals). The average yield avoided impact caused by valuable products (recycled ma- (weight %) of products as measured in the pilot plant was terial and energy). as follows: gas = 15%, pyrolysis oil = 41%, solid product 2.4 Life cycle inventory = 32%, and steel = 12%. The life cycle inventory (LCI) is an inventory analysis 2.3 System boundaries to identify input (materials and energy), output (emissions) The system boundaries that were considered in this and data quality, which play an important role in LCA re- LCA analysis are shown in Fig. 1, based on the system de- sults. Primary data on the input (electricity and heating oil scription and the assumptions given below: consumption) and output (pyrolysis gas composition, pyrolysis products and other products) were mainly obtained  Production and use phases were not considered in this from the pilot pyrolysis plant (Aker Bioenergy). SimaPro study, since the feed of the pyrolysis is waste. Simi- 8 software was used to develop and link primary unit pro- larly, the transport of waste tires was not considered, cesses. Secondary data were gathered from the EcoInvent since the aim was to assess the pyrolysis process. (v2.2-v.3.0.3) database which is embodied in SimaPro 8.

FIGURE 1 - System boundaries

1217 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.5 Energy and material requirement tion are given in Table 3. This calculated amount is con- According to the pilot pyrolysis plant electricity consistent with the amount (60kg CO2/t tires treated) reported sumption data, the average amount of electricity required by Li et al. (2010) [8]. for production is 155.51 kWh per ton of waste tires. This SO , NO and dust emission data (3.55 kg SO , 1.40 kg consumption quantity includes a pre-treatment (drawing 2 x 2 NO and 0.58 kg dust per 1 t of waste tire pyrolysis) were machine, hydraulic cutting device, shredder and convey- x obtained from Li et al. (2010) [8]. ors), cooling equipment, valves, scrubber, grinding, carbon treatment unit, gas storage and packaging unit. Turkey’s TABLE 2 - The averaged chemical composition of pyrolysis gas. grid mix electricity profile was compiled in SimaPro 8 using a breakdown of Turkey’s grid electricity primary en- Component % (volume) ergy sources in 2011 (45% natural gas, 29% lignite coal, Methane 32.93 23% hydraulic, 2.5% wind and 0.5% geothermal) by Banar Hydrogen 21.10 et al. (2013) [7]. In this study, this profile was used for Ethane 13.06 electricity data. Carbon dioxide 12.05 Heating oil is used for the start-up (for the first 3 h) Carbon monoxide 6.40 until the system produces its own gas. In this study, heating Propane 5.14 oil was considered as fuel oil no.4 that has an average heat- Propene 3.77 ing capacity of 9700 kcal/kg and a C content of 85%. In Ethylene 1.93 SimaPro 8, light fuel oil, at refinery/CH U data was used 1-Butene 1.26 for fuel oil no.4. 2-Fumaric acid 1.26 Butane 0.90 2.6 Emissions from pyrolysis gas combustion Nitrogen 0.85 Pilot pyrolysis plant gas product efficiency is normally Isobutane 0.30 15% by weight. Pyrolysis gas is used as a fuel by feeding Cis-2-butene 0.08 back into the pyrolysis process. The averaged chemical Pentane 0.06 composition of pyrolysis gas measured at the pilot plant is Hydrogen chloride <0.01 given in Table 2. The density of the pyrolysis gas was de- Hydrogen sulfite <3 (mg/m3) termined to be 0.895 kg/m3 using the volume percentages and densities of the components at Normal Temperature & 2.7 Allocation Pressure (NTP). In addition, the Gross Calorific Value There is a need for allocation to consider the benefits of (GCV) of pyrolysis gas was determined to be 42.54 MJ/kg pyrolysis processes modelled in this study. According to ISO 3 (38.10 MJ/m ) from Table 2. This GCV value of the pyrol- 14044, allocation should be avoided by dividing processes 3 ysis gas is in the range of 37.85–40.72 MJ/m reported by into subprocesses, or by expanding the system boundary so Islam et al. (2011) [12]. that co-products are included in the system. In this study, al-

The total CO2 emission amount was calculated to be location was avoided by expanding the system boundary with 68.06 (kg/ton waste tire pyrolysis). Details of the calcula- avoided products given in the following section.

TABLE 3 - CO2 emission calculation steps.

The capacity of batch reactor 12 t /batch Capacity of the burner 301,000 kcal/h Total operation time 11 h

Total CO2 generation for 1 t of waste tires pyrolysis 68.06 kg CO2 Heating oil combustion calculation Operation duration with heating oil 3 h GCV of heating oil 9700 kcal/kg Heating gas consumption for 1 t waste tires pyrolysis 93 kg C content of heating oil 85%

CO2 generation from combustion of heating oil 24.18 kg CO2 Pyrolysis gas combustion calculation Operation duration with pyrolysis gas 8 h GCV of pyrolysis gas 38.10 MJ/m3 Heating requirement for 1 t waste tire pyrolysis supplied by pyrolysis gas 838.79 MJ Pyrolysis gas consumption for 1 t waste tires pyrolysis 22.02 m3 Produced pyrolysis gas volume from 1 t waste tires pyrolysis 167.51 m3 The percentage of recovered pyrolysis gas for 1 t waste tires 13.14% The total C amount in the recovered pyrolysis gas 80.83 kg * CO2 generation from combustion of pyrolysis gas 43.88 kg CO2 *89% from hydrocarbons and 11% from pyrolysis gas own CO2

1218 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.8 Avoided products from valuable products of diesel fuel should be 40. At this point, it was not practi- There are 4 kinds of valuable products produced from cal to substitute 100% pyrolysis oil for diesel fuel, although waste tire pyrolysis: pyrolysis gas, pyrolysis oil, solid the GCV values of the pyrolysis oil and diesel fuel are product, and steel. Because the recycled material may not close. Based on this explanation, it was concluded that 1 kg be accepted as a full substitution of the same amount of of pyrolytic oil can substitute 0.5 kg of commercial diesel virgin material due to decline in its physical/chemical prop- fuel. Diesel, at refinery/RER U data, was used for the erties, a substitution factor between certain recovered and avoided diesel. virgin material should be assumed. Pyrolysis valuable  The solid product that exits in the pyrolysis reactor products and their substitutes are explained with their sub- contains both carbon black and the steel from the tires. stitution factors below: For the solid product, scrap steel and carbon black are  Pyrolytic oil can substitute diesel, which is used in sta- separated using a 3-step magnet process. The latter is tionary diesel generators. The GCV value is an im- then ground and packed. Carbon black obtained from portant parameter to utilize an alternative fuel. The el- pyrolysis only can substitute any commercial carbon emental composition and GCV value of the pyrolytic black in the case of a further acid treatment because of oil were determined using a LECO CHN and S and its high ash and sulphur contents (12.14 and 1.71%, IKA C200 calorimeter (ASTM D-5865), respectively. respectively) [15]. For this reason, it was assumed that The comparison of the results with diesel are listed in 1 kg of pyrolytic carbon black can substitute 0.5 kg of Table 4. The GCV of pyrolytic oil is slightly lower commercial carbon black, which is produced by the than that of diesel. The higher nitrogen in the pyrolytic oil-furnace method. Carbon black, at plant/GLO U oil comes from the initial introduction of nitrogen to data, was used for commercial carbon black. the pyrolysis system, and the sulfur content in the py-  The separated steel in the pre-treatment process was rolytic oil comes from the origin of tire. shipped to recycling, together with the scrap steel removed from the solid product. The substitution rates TABLE 4 - The comparison of pyrolytic oil with diesel. of this recovered steel were quantified as 1:1 for Parameter Pyrolytic oil Diesel avoided steel production. According to a web search Elemental analysis (% wt.) for steel wire producers for tires, it was found that high C 86.74 87.4 carbon steel wire rods are used to reinforce the dura- H 10.41 12.1 bility of automobile tires. For this reason, steel, low- N 0.52 370 ppm alloyed, at plant/RER U data, was used for avoided S 1.15 1.39 steel. Oa 1.18 2.1 GCV (MJ/kg) 40.37 45.5 2.9 Life cycle impact assessment a by difference The LCA calculations were carried out using licensed SimaPro 8 software. In the impact assessment step, the On the other hand, the cetane number (CN) is another CML-IA baseline (V3.00) method was applied. The CML- important parameter that has to be carefully considered IA baseline method is an update of the CML 2 baseline when an unconventional fuel is used inside a compression 2000, and elaborates the problem-oriented (midpoint) ap- ignited engine. When the traditional CN cannot be directly proach. Normalization results were also calculated using measured, it is possible to make a good estimation of it, by the EU25+3, 2000 calculation method under the CML-IA calculating the cetane index (CI) [14]. The CI is calculated baseline (V3.00) method. In addition, the modeled system from the formula given below (ASTM D976-06). was analyzed by excluding infrastructures. 454.74 1641.416 774.74 0.544 97.803 (1) 3. RESULTS AND DISCUSSION In this equation, D is the fuel density at 15 °C (g/ml) and T50 is the mid-boiling temperature (°C) corresponding The characterization and normalization results of all to a 50% point in the distillation curve. In this study, the CI impact categories in the CML-IA baseline (V3.00) method of the pyrolysis oil was calculated by combining the pilot are given in Table 5. As seen in Table 5, all the impact cat- plant data and the experimental data. The density (15 °C) egories have negative values, except acidification. Nega- of the pyrolytic oil was reported as 925.6 kg/m3 by the pilot tive values result from avoided products from valuable pyrolysis plant. Banar et al. (2012) [13] measured the T50 products. Normalization values can help to compare the of the tire pyrolysis oil as 250.1 °C for a pyrolysis heating impact of the different categories. According to the nor- rate of 5 °C/min. By using these values, the CI value of the malization results, it can be concluded that acidification pyrolysis oil was estimated to be 25.6. This calculated CI and marine aquatic ecotoxicity impact categories are the value is close to Frigo et al. (2014) [14] who found a CI of endpoints of the total impact that represent the environ- 27. According to ASTM D975-13a, the minimum CI value mental effects and avoided effects, respectively. Six impact

1219 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 5 - Characterization and normalization results.

Impact category Characterization Normalization Unit Amount Abiotic depletion (element) kg Sb eq. -2.39E-03 -1.44E+04 Abiotic depletion (fossil fuels) MJ -2.16E+04 -7.58E+17

Global warming (GWP100a) kg CO2 eq. -5.10E+02 -2.66E+15 Human toxicity kg 1,4-DB eq. -7.78E+02 -3.89E+14 Marine aquatic ecotoxicity kg 1,4-DB eq. -3.63E+05 -1.61E+19

Acidification kg SO2 eq. 2.92E+00 4.90E+10 --- Eutrophication kg PO4 eq. -3.91E-01 -7.23E+09 Fresh water aquatic ecotoxicity kg 1,4-DB eq. -2.54E+02 -5.30E+13 Terrestrial ecotoxicity kg 1,4-DB eq. -6.56E-01 -7.61E+10 Ozone layer depletion (ODP) kg CFC-11 eq. -2.66E-04 -2.71E+03

Photochemical oxidation kg C2H4 eq. 5.15E-03 8.91E+06

5,00E-04 Nickel Chromium 0,00E+00 Molybdenum -5,00E-04

kg Sb eq. kg Sb -1,00E-03

-1,50E-03

-2,00E-03

-2,50E-03

(a)

6,00E+03 Crude oil 3,00E+03 Natural gas 0,00E+00 Coal

MJ -3,00E+03 -6,00E+03 -9,00E+03 -1,20E+04 -1,50E+04

(b) FIGURE 2 - AD impact of 1 ton of waste tire pyrolysis: a) ADe impact, and b) ADff impact.

1220 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

categories were further investigated: abiotic depletion (el- impact. The production of heating oil and electricity con- ement and fossil fuel), global warming, acidification, eu- tributes 27.5 and 22.8% to the total positive GWP, respec- trophication, human toxicity, and marine aquatic toxicity. tively. Fresh water aquatic ecotoxicity and terrestrial ecotoxicity were excluded from further investigation, since marine 3.3 Human toxicity aquatic ecotoxicity was the dominant impact of the ecotox- Figure 4 shows the human toxicity impact of 1 t of icity impact. waste tire pyrolysis. From Fig. 4, it can be seen that there is a saving of the human toxicity impact, mainly from 3.1 Abiotic depletion avoided Chromium VI generation during the steel produc- Abiotic depletion (AD) was investigated under two tion. subcategories: the abiotic depletion element (ADe) (Fig. 2a) and abiotic depletion fossil fuel (ADff) (Fig. 2b). Fig- 3.4 Marine aquatic toxicity ure 2a shows that there is a significant saving for ADe from The marine aquatic toxicity impact of 1 t of waste tire steel alloy production elements due to avoided steel pro- pyrolysis is demonstrated in Fig. 5. As can be seen in Fig. 5, duction. From the point of view of ADff, as expected, there in the foreground, there is a remarkable negative impact are remarkable savings for crude oil and coal consumption due to avoided steel production. In the background, beryl- due to avoided diesel, carbon black and steel production lium, cobalt, hydrogen fluoride, nickel and vanadium emis- (Fig. 2b). On the other hand, heating oil production causes sions generated from steel production were avoided. Ex- abiotic depletion, entirely due to the crude oil cept for these, only heating oil and electricity production caused a positive marine aquatic toxicity impact at very 3.2 Global warming low values. The GWP (100a) impact of 1 t of waste tire pyrolysis is shown in Fig. 3. As can be seen, there are savings for the 3.5 Acidification GWP value due to the avoided production of diesel, carbon The distribution of the acidification impact resulting black and steel. CO2 accounts for over 95% of the total from 1 t of waste tire pyrolysis into the subprocesses is avoided impact. In the foreground, the highest saving is in shown in Fig. 6. Among the subprocesses, pyrolysis has the the carbon black production while in the background, the highest acidification impact, mainly due to the acidifica- saving is because of avoided crude oil used in the oil-fur- tion potential of SO2 and NOx emissions in combustion flue nace method. Beside the saving resulting from avoided gases. The negative impact is due to the avoided generation products, there are also positive impacts. The combustion of SO2 resulting from diesel, carbon black and steel. of pyrolysis gas and heating oil during the pyrolysis process accounts for 49.7% of the total positive GWP (100a)

1,00E+02 Carbon dioxide Methane 0,00E+00 eq.

2 -1,00E+02

kg CO -2,00E+02

-3,00E+02

-4,00E+02

FIGURE 3 - GWP (100a) impact of 1 ton of waste tire pyrolysis.

1221 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2,00E+02 Chromium VI Vanadium 0,00E+00 Barium Selenium -2,00E+02 Nickel kg 1,4-DB eq. kg 1,4-DB -4,00E+02

-6,00E+02

-8,00E+02

FIGURE 4 - Human toxicity impact of 1 ton of waste tire pyrolysis.

1,00E+05 Vanadium Selenium 0,00E+00 Nickel

-1,00E+05 Hydrogen fluoride Cobalt kg 1,4-DB eq. kg 1,4-DB -2,00E+05 Beryllium Barium -3,00E+05

-4,00E+05

FIGURE 5 - Marine aquatic toxicity impact of 1 ton of waste tire pyrolysis.

1222 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

6,00E+00 Sulfur dioxide Nitrogen oxides 4,00E+00 eq. 2

2,00E+00 kg SO

0,00E+00

-2,00E+00

FIGURE 6 - Acidification impact of 1 ton of waste tire pyrolysis.

2,00E-01 Phosphate 1,00E-01 Nitrogen oxides 0,00E+00 --- eq. 4 -1,00E-01 Nitrate

kg PO -2,00E-01 COD, Chemical Oxygen -3,00E-01 Demand -4,00E-01 -5,00E-01

FIGURE 7 - Eutrophication impact of 1 ton of waste tire pyrolysis.

1223 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

3.6 Eutrophication spectively [16]. Although the wet FGD system has experi- Figure 7 shows the eutrophication impact of 1 t of enced high SO2 and particulates removal efficiencies, that waste tire pyrolysis. In the foreground, the impact is mainly is not so for NOx. For that reason, instead of a post-com- due to NOx emissions in the combustion flue gases. There bustion system, an improved control technology for NOx is a noticeable negative impact on avoided steel produc- removal was considered in terms of flue gas treatment. In tion. This negative value was sustained by the avoided NOx that case, replacement of the conventional gas burner in the and phosphate emissions. The avoided NOx emissions also system with a Low NOx Burner (LNB) was investigated. sustain negative eutrophication values on avoided diesel The reasons for the selection of LNB system are very low and carbon black production. Additionally, diesel produc- nitrogen content of pyrolysis gas, low operational cost than tion is also significant for eutrophication, due to the other NOx removal systems, such as Selective Catalytic avoided pollutants which cause a COD. Reduction (SCR). LNBs can reduce NOx by 50% [17-19] On the other hand, to select a SCR system would not be so 3.7 Sensitivity analysis meaningful for a small plant that will only burn the excess In the sensitivity analysis, the parameters with poten- of pyrolysis gas. As a summary, a sensitivity analysis with tially impact on the overall results were investigated: diesel NOx emission reduction by 50% was performed to deter- substitution rate, carbon black substitution rate, and im- mine the improved flue gas treatment on the environmental proved flue gas treatment. Two different substitution rates profile of the system. (0.1 and 0.25), worse than baseline case (0.5), for diesel and carbon black were considered. The SOx, NOx and dust Each assumption was changed independently of all emission data regarding to waste tire pyrolysis represent others; therefore, the magnitude of its effect on the baseline the average technology tire pyrolysis plant that sustains the case could be assessed. The assumptions considered in the limits of air contol legislations. Most of the average tech- sensitivity analysis and their corresponding effects on the nology flue gas treatment systems have, at least, a flue gas baseline case are shown in Table 6. It is concluded from desulphurization (FGD) unit, and this FGD system in- Table 6 that decreasing the substitution factor for diesel cludes a wet scrubbing using a slurry of alkaline sorbent. and carbon black increases the impact categories values Wet scrubbers are also effective for particulates removal. (max. 37% for diesel and max. 58% for carbon black). SO2 and particulates (PM10) removal efficiencies using al- Substitution factor has remarkable influences on the abiotic kaline sorbents range between 95-99% and 80-95%, re- depletion, global warming, acidification and eutrophication

TABLE 6 - Results of sensitivity analysis for assumptions.

% Change in characterization results of impact categories* Impact categories Parameters** A B C

s.f.=0.1 s.f.=0.25 s.f.=0.1 s.f.=0.25 50% NOx red. Abiotic depletion (element) 0.04 0.02 0.00 0.00 0.00 Abiotic depletion (fossil fuels) 36.48 22.80 47.17 29.48 0.00 Global warming (GWP100a) 13.15 8.22 57.06 35.66 0.00 Human toxicity 1.71 1.07 1.01 0.63 -0.11 Marine aquatic ecotoxicity 5.90 3.68 2.36 1.48 0.00 Acidification 30.69 19.18 26.50 16.56 -11.99 Eutrophication 25.92 16.20 14.60 9.12 -23.28 * Changes in characterization results are expressed as a percentage of the baseline case; ** Parameters - A: Substitution of pyrolytic oil for diesel, substitution factors = 0.1 and 0.25, B: Substitution of pyrolytic solid product for carbon black, substitution factors = 0.1 and 0.25, C: Improved gas treatment system, 50% reduction on the emissions of NOx).

TABLE 7 - Comparison results

Impact categories Unit Amount (2):(1) This study Li et al. [8] (1) (2) Climate change DALY -1.07E-04 -1.17E-04 1.09 Respiratory organics DALY -5.21E-07 -2.00E-07 0.38 Respiratory inorganics DALY -1.70-04 -4.20E-05 0.25 Carcinogens DALY -1.01E-04 -1.14E-05 0.11 Acidification / Eutrophication PDF*m2yr 5.31E+00 1.00E+01 1.88 Ecotoxicity PAF*m2yr -5.38E+02 1.00E+00 -0.001 Fossil fuels MJ surplus -2.82E+03 -1.57E+03 0.56 DALY: Disability adjusted life years; PDF: Potentially disappeared fraction of plant species; PAF: Potentially affected fraction; * the rounded value of -0.00186.

1224 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

for both diesel and carbon black. On the other hand, NOx  The production of heating oil and electricity contrib- emission reduction influences acidification and eutrophi- utes 27.5 and 22.8% to the total positive GWP, respec- cation by the changes of 12 and 23%, respectively. tively.  The combustion of pyrolysis gas and heating oil during 3.8 Comparison of the results the pyrolysis process accounts for 49.7% of the total Life cycle assessments are always associated with un- positive GWP (100a) impact. certainties, especially when data is not obtained directly  Pyrolysis has an acidification impact, due mainly to from a specific plant [20]. However, in this study, most the acidification potential of SO emissions in combus- data represent the plant-specific data. For this reason, a 2 tion flue gases. sensitivity analysis was performed in terms of different impact categories using the EcoIndicator 99 method. The re-  NOx emissions in the combustion flue gases cause sults were compared with the findings of Li et. al. (2010) acidification and eutrophication. Sensitivity analysis [8], which were also based on real plant data (Table 7). As shows that a 50% NOx emission reduction resulted in can be seen in Table 7, the ratios of the results are in the a decrease of acidification and eutrophication by 12 range of 0.1-1.9, except for ecotoxicity. Ecotoxicity values and 23%, respectively. show a significant difference. This difference mainly re- As a consequence of this LCA study, in the light of the sults from power generation. Ranging of the negative val- sensitivity analysis, the results show that the sufficient uti- ues results from avoided products. lization of pyrolysis products and the application of flue

gas treatment systems play an important role in demon-

strating that pyrolysis is an environmentally effective solu- 4. CONCLUSIONS tion for waste tires.

In this study, the environmental impact of waste tire This is the first LCA study on waste tire pyrolysis in pyrolysis was determined in the context of a pilot plant us- Turkey. Therefore, the finding of this study could help de- ing the LCA method. The results of the study illustrate that cision-makers and practitioners by providing a framework there are large savings from waste tire pyrolysis in terms of to better understand the major environmental effects of material and emissions, due to the substitution of pyrolysis waste tire pyrolysis and conversion to useful products. products (solid product-carbon black and pyrolysis oil-diesel). Sensitivity analysis indicated the importance of sub- The author has declared no conflict of interest. stitution rate on the savings.

The savings are listed in more detail below: REFERENCES  Steel alloy production elements, crude oil and coal consumption due to avoided diesel, carbon black and [1]. Martinez JD, Puy N, Murillo R, Garcia T, Navarro MV, Mas- steel production. tral AM (2013) Waste tire pyrolsis-a review, Renewable and Sustainable Energy Reviews 23:179-213.  A GWP impact due to avoided production of diesel, [2]. Williams P (2013) Pyrolysis of waste tires, Waste Manage- carbon black and steel (CO2 is the main avoided emis- ment 33:1714-1728. sion. [3]. Sienkiewicz M, Kucinska-Lipka J, Janik H, Balas A (2012)  A human toxicity impact resulting from the steel pro- Progress in used tires management in the European Union: a duction (chromium VI generation is a mainly avoided review, Waste Management 32:1742-1751. emission). [4]. Williams PT, Brindle AJ (2002) Fluidised bed catalytic pyrol-  A marine aquatic toxicity impact generated from the ysis of scrap tyres: Influence of catalyst: tyre ratio ans catalyst steel production (beryllium, cobalt, hydrogen fluoride, temperature. Waste Management and Research 20:546-555. nickel and vanadium are the avoided emissions). [5]. Ministry of Environment and Urban (2011) Environmental Status Report [in Turkish]  An acidification generation from the diesel, carbon black and steel production (SO2 is the mainly avoided [6]. Ministry of Environment and Urban (2006) Regulation on emission). Control of End of Life Tires, Official Gazette No: 26357 [in Turkish]  An eutrophication impact due to diesel and carbon [7]. Banar M, Cokaygil Z, Ozkan A, Akyıldız V (2013) Character- black production (NOx and phosphate emissions and, ization of gas emissions resulted from tire derived fuel pyrol- in addition, a COD parameter are the avoided effects). ysis. Proceedings of the 3W Istanbul, Istanbul International Solid Waste, Water and Wastewater Congress, May 22-24, Is- tanbul, Turkey:1390-1396. On the other hand, with respect to the environmental savings, this study shows that waste tire pyrolysis also has [8]. Li X, Xu H, Gao Y, Tao YY (2010) Comparison of end-of-life treatment technologies: A Chinese case study. Waste Manage- an environmental impact: ment 30:2235-2246.  Heating oil production causes abiotic depletion en- [9]. Corti A, Lombardi L (2004) End life tyres: Alternative final tirely due to the crude oil. disposal processes compared by LCA. Energy 29:2089-2108.

1225 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[10]. Turkish Standards Institution (TSE) (2006) TSE EN ISO 14040: environmental management: life cycle assessment: principles and framework (Turkish Standard).

[11]. Turkish Standards Institution (TSE) (2006) TSE EN ISO 14044: environmental management: life cycle assessment: requirements and guidelines (Turkish Standard).

[12]. Islam MR, Joardder MUH, Hasan, SM, Takai K, Haniu H (2011) Feasibility study for thermal treatment of solid tire wastes in Bangladesh by using pyrolysis technology. Waste Management 31:2142-2149.

[13]. Banar M, Akyildiz V, Ozkan A, Cokaygil Z, Onay O (2012) Characterization of pyrolytic oil obtained from pyrolysis of TDF (Tire Derived Fuel). Energy Conversion Management 62:22-30. [14]. Frigo S, Seggiani M, Puccini M, Vitolo S (2014) Liquid fuel production from waste tyre pyrolysis and its utilization in a diesel engine. Fuel 116:399-408. [15]. Akyıldız V, Ozkan A, Cokaygil Z, Banar M, Baydar S (2010) Improvement of solid product quality in pyrolysis of tyre derived fuels (TDF). Chemical Engineering Transactions 21:775-780.

[16]. Muezzinoğlu A. (2000) The Principles of Air Pollution and Control. DEU Publications (in Turkish). [17]. EPA Technical Bulletin (1999) Nitrogen Oxides (NOx), Why and How They Are Controlled, EPA459/F-99-006R. [18]. Godish, T. (1997) Control of Emissions from Stationary Sources, Air Quality Boca Raton: CRC Press LLC. [19]. Kreith F, Goswami, DY (2005) The CRC Handbook of Me- chanical Engineering, CRC Press. [20]. Evangelisti SS, Lettieri PS, Borello DD, Clift R (2014) Life cycle assessment of energy from waste via anaerobic digestion: A UK case study. Waste Management 34:226-237.

Received: June 26, 2014 Revised: August 12, 2014 Accepted: September 29, 2014

CORRESPONDING AUTHOR

Müfide Banar Department of Environmental Engineering Faculty of Engineering İki Eylül Campus Anadolu University 26555 Eskişehir TURKEY

Phone: +90.222.321 35 50/6400 (ext.) Fax: +90.222.323 95 01 E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1215 - 1226

1226 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

PHYSICAL AND WATER PROPERTIES OF SOILS DEVELOPING FROM POST-MINING MATERIALS OF KONIN BROWN COAL MINE

Krzysztof Otremba1, Mirosława Gilewska1, Andrzej Mocek2, Wojciech Owczarzak2, Piotr Gajewski2,* and Zbigniew Kaczmarek2

1 Department of Soil Science and Recultivation, University of Life Sciences, Poznań, Poland 2 Department of Soil Science and Land Protection, University of Life Sciences, Poznań, Poland.

ABSTRACT restoration technologies. In the case of agricultural reclamation, the principal technology is based on the concept of The presented study discusses physical and water prop- “target plants”, according to which – after renovation of erties of soils developing in the course of the past 35 years physical and chemical properties of ground – the target from post-mining land of Konin Brown Coal Mine (KWB plants should directly introduced (without use of forecrops, Konin) as a result of diversified agricultural utilization. Ex- pioneer plants etc.) [4,5]. This approach assumes better- periments were carried out on an experimental field of the ment of the chemism of the post-mining materials (parent Department of Soil Science and Recultivation situated on rock) as well as improvement of their physical properties the Pątnów (central Poland) dumping ground established [6]. Certain biogenic elements such as carbon or nitrogen by Bender in 1978. Experimental samples were collected gradually accumulate during ecosystem development due from three different systems of utilization. The following to plant activity and during gradually accumulation of soil properties were determined in these samples: total carbon, organic matter [7]. In the period of 10-20 years, the post- specific density, bulk density, porosity (total, drainage and mining deposits are transferred into soil which, with re- differential), maximal hygroscopic water, saturated hy- spect to its nutrient availability for crop plants, equals soils draulic conductivity, soil water potential and readily and which belong to the 3rd soil valuation class (“good soils”) total available water. The research results presented in this [4,5]. However, it should be emphasised that availability of study corroborate the important role of lucerne in influenc- nutrient resources is a property which can be changed rel- ing physical and water properties of soils developing on atively quickly by the improvement of chemism, whereas post-mining land. This is probably determined by the biol- betterment of physical properties is considerably more dif- ogy of this plant which possesses a strong, well-branched ficult and lasts longer [8]. Many areas are reclaimed fol- root system. lowing specific long-term experience and comprehensive analyses of general restoration principles appropriate for post-mining landscapes [9]. These approaches can lead to the creation of new landscapes that are fully compliant to KEYWORDS: dumping ground, reclamation, soil properties, target plants. modern social and environmental requirements [10].

The aim of the article was to determine basic physical and water properties of soils which have been developing for the past 35 years from post-mining materials of Konin 1. INTRODUCTION Brown Coal mine as a result of various farming practices.

Open –cast brown coal mining which has been taking place in the region of the city of Konin (central Poland) for 2. MATERIALS AND METHODS over 60 years led to the development of an entirely new relief as well as soil cover [1,2]. The negative impact of Experiments were conducted in 2013 on an experi- coal mining on the soil environment also notes Charzyński mental field of the Department of Soil Science and Reculti- et al. [3]. The aim of land reclamation activities is to impart vation of Poznań University of Life Sciences located on new values and functions to post-mining deposits. The Pątnów (near Konin) dumping ground which was estab- above-mentioned objectives are achieved using different lished in 1978 by Bender [5]. Samples for investigations were collected from the following three different systems * Corresponding author of utilization (variants):

1227 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

A – fodder-cereal rotation consisting in a four-year mined at pF potential of 2.0), differential porosity – on the cultivation of lucerne followed by a two-year long cultiva- basis of water desorption curves by soil. The amount of tion of winter cereals. The following three fertilisation macropores was calculated as the difference between total combinations were analysed: porosity and pF 2,5; the amount of mesopores as the differ- A1 – 0 NPK, without fertilisation, ence between water content at pF 2,5 and pF 4,5; and quantity the of micopores is a valule of moisture at pF 4,5. The A2 – 1 NPK (380kg NPK), in accordance with the “law presented values are averaged values from five replica- of return” [6], tions. The statistical analysis was conducted with Tukey’s A3 – 2 NPK, two times higher than 1 NPK. test at the statistical significance of α = 0.05. B – rape-cereal rotation consisting in the alternate cultivation of winter wheat and winter rape. The following three fertilisation combinations were analysed: 3. RESULTS AND DISCUSSION B1 – 0 NPK, without fertilisation, The determined texture was uniform (Table 1). The an- B2 – 1 NPK (380kg NPK), in accordance with the “law alysed soil under all analysed variants and combinations of return” [6], exhibited sandy loam texture [17]. B3 – 2 NPK, two times higher than 1 NPK. The content of sand was contained within the following limits: sand: 62-64%, silt: 21-25%, clay: 11-15%. The C – Sand Lucerne (Medicago x varia T.Martyn ) (“per- recorded high content of the clay fraction was a character- manent lucerne”) (so called “conservation”) cultivation istic feature of these soils [18,19]. It should also be em- consisting in long-term growing of lucerne which enriches phasised that, in the case of the analysed materials, the soil in nitrogen and protects it from the establishment of same texture composition need not necessarily indicate the bushes and shrubs. No fertilisation was applied in this sys- same physical and physico-chemical properties [20]. The tem – 0 NPK. content of total carbon in the soils of individual variants D – control treatment, i.e. post-mining deposits which fluctuated from 4.0 g·kg-1 (variant D – fresh material) to were not subjected to any reclamation treatment (fresh ma- 15.0 g·kg-1 (variant C – “permanent lucerne”) (Table 2). terial) – combination with 0 NPK. The lowest total carbon (Ct) content was determined in Samples of intact (V = 100 cm3, of cylinder shape) the fresh post-mining material. In the case of A and B var- and disturbed structure were collected from the top layer iants, the lowest Ct content was determined in samples ob- (0-20 cm) for laboratory investigations. The following tained from the 0 NPK combinations. The amount of Ct in properties were determined in the collected samples: total the material collected from the fodder-cereal rotation was carbon – using Makro Elemental Analyzer (vario Max significantly higher than in variant “B” (rape-cereal rota- CNS), texture – employing sieve - sedimentation proce- tion) and significantly lower in comparison with variant C dure [11], specific density – using the pycnometric method (“permanent lucerne”). The relationship between organic [12], bulk density by the core method [13], total porosity – carbon content and land use also consider Shrestha and Lal calculated on the basis of specific and bulk densities deter- [21]. They emphasized in this case the role of pasture mination, maximal hygroscopic capacity – by Nikolaev plants. In the case of the rape-cereal rotation system, the method [14] saturated hydraulic conductivity – using the content of the soil organic carbon increased significantly method of constant pressure drop [15], soil water binding together with the increase of applied fertilisation. How- potential – by Richard’s pressure chamber method [16], ever, difference between B2 and B3 was insignificant. The total (TAW) and readily (RAW) available water – calcu- impact of mineral fertilisation on the accumulation of or- lated on the basis of pF determinations, effective (drainage) ganic matter in soils developing from post-mining deposits porosity – determined as the difference between total po- was emphasised by Gilewska and Otremba [22]. With re- rosity and moisture content at field water capacity (deter- spect to arable soils Stępień and Adamiak [23] link this

TABLE 1 - Texture of investigated soils

Fraction content (%) of diameter (mm) Texture acc. to: Combination 2.0-1.0 1.0-0.5 0.5-0.25 0.25-0.10 0.10-0.05 0.05-0.02 0.02-0.002 <0.002 (FAO 1977) [17] A1 1.3 7.4 18.5 25.8 9 10 15 13 SL A2 1.6 6.4 20.1 27.9 8 10 11 15 SL A3 1.6 8.0 15.9 23.5 15 6 16 14 SL B1 1.6 8.5 18.1 25.8 10 6 19 11 SL B2 1.9 6.4 18.8 24.9 12 9 15 12 SL B3 1.9 6,0 16.6 26.5 13 11 11 14 SL C 1.0 6.7 17.6 26.7 12 9 16 11 SL D 1.2 7.4 19.1 26.3 10 9 12 15 SL

1228 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 2 - Basic physical and chemical properties

Total carbon Specific density Bulk density Total porosity Combination (g·kg-1) (Mg· m-3) (Mg·m-3) (m3·m-3) A1 7.2e 2.65a 1.56f 0.4113c A2 11.7b 2.64b 1.64e 0.3788d A3 10.3c 2.64b 1.76c 0.3333e B1 4.3f 2.65a 1.79b 0.3245e B2 8.3d 2.65a 1.53g 0.4226b B3 8.8d 2.65a 1.65d 0.3774d C 15.0a 2.64b 1.50h 0.4318a D 4.0f 2.65a 1.90a 0.2830f

Macropores Mesopores Micropores (m3·m-3) A1 0.2000b 0.1496d 0.0617d A2 0.1479d 0.1612c 0.0697bc A3 0.1088e 0.1682ab 0.0563e B1 0.1033ef 0.1252f 0.0960a B2 0.2096a 0.1692a 0.0438f B3 0.1822c 0.1390e 0.0562e C 0.1972b 0.1635bc 0.0711b D 0.1022f 0.1150g 0.0658c

phenomenon with considerable quantities of after-harvest D) to 0.2096 m3·m-3 (variant B2). The same combinations residues left in farmland which interfere with the balance were also characterised by, respectively: the smallest – of mineralisation and humification processes. However, re- 0.1150 and the highest – 0.1692 m3·m-3 proportions of soil vegetation is not easy due to the fact that post mine soils mesopores. As evident from the data in Table 2, the amount are usually with low fertility, low content of nutrients [24]. of micropores was contained within limits ranging from 0.0438 m3·m-3 (B2) to 0.0960 m3·m-3 (B1). From agricul- Improving the physical properties of the soil is neces- tural point of view, soils with a considerable proportion of sary for successful revegetation [25] When nutrients are mesopores are characterised by the most advantageous wa- added to the soil, not only does plant growth improve, but ter properties. This group of soil interspaces is essential for the physical soil condition are also ameliorated [26]. Those the amounts of water available for plants. A high propor- reports were not confirmed by the results obtained in this tion of micropores indicates that a considerable quantity of paper. The specific density was not found to depend on the retained water remains unavailable to plants. The obtained applied utilization systems and fertilisation combination research results revealed slight impact of the applied rota- (Table 2). On the other hand, however, the significantly tion systems as well as of the employed fertilisation com- differences in bulk density and total porosity closely asso- bination on differential porosity. Nevertheless, it was pos- ciated with the former property were quite visible in the sible to observe an advantageous influence of the 35-year long analysed utilization combinations (Table 2). The highest period of agricultural utilization. Quantities of soil macro- and density (1.90 Mg·m-3) accompanied by the smallest poros- mesopores in all reclamated plots was significantly higher in ity (0.2830 m3·m-3) was observed in the “fresh post-mining comparison with the fresh material (variant D). material”. The high density of post-mining soils notes also Shrestha and Lal [27]. The above-mentioned two proper- In addition, an attempt was undertaken in this investi- ties were much more advantageous in the case of experi- gation to determine the impact of the applied individual ro- mental plots utilized agriculturally for 35 years. In these tation systems on the content of water in individual pF po- combinations, bulk density was contained within bounda- tentials as well as on soil retention capabilities (Table 3). ries ranging from 1.50 (C) to 1.79 Mg·m-3 (B1), while po- Retention capabilities of the discussed soils are particularly rosity fluctuated from 0.3245 (B1) to 0.4318 m3·m-3 (C) important because there are frequent atmospheric droughts (Table 2). The most advantageous values were recorded in as well as moisture deficits resulting from these droughts the soils of variant C (conservation). Porosity is character- in the area of Konin. Maximal water capacity (pF 0,0) was ised not only by the total pore volume but also by their slightly (about 2-3% v/v) lower than total porosity. Mois- qualitative distribution (differential porosity). The content ture content of the examined soils at field water capacity of of individual groups of pores in farmland soils depends, pF 2.0 varied usually significantly and fluctuated within among others, on: texture, organic matter content, soil the following boundaries: from 18.50 % v/v (variant D) to compaction as well as on its aggregate composition [28, 26.57 % v/v (variant C). Soil moisture content at pF 2.5 29]. The content of macropores in analysed soils varied was by about 1-3% lower in comparison with the above usually significantly and fluctuated from 0.1022 (variant values. Further, only slightly higher, decline of moisture

1229 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 3 - Soil water potentials and the readily and total available water

Water capacity at pF RAW TAW Combina-tion (% v/v) (% v/v) (% v/v) 0.0 2.0 2.5 3.7 4.2 4.5 2.0-3.7 2.0-4.2 A1 39.98b 22.45e 21.13e 17.59c 10.01d 6.17d 4.86e 12.44c A2 35.47d 25.91b 23.09b 20.26a 12.08b 6.97b 5.65d 13.83b A3 31.02e 24.40c 22.45c 17.09d 9.54e 5.63e 7.31b 14.86a B1 31.08e 23.58d 22.12d 18.90b 12.14b 9.60a 4.68e 11.44d B2 40.17b 22.42e 21.30e 16.14f 9.02f 4.38f 6.28c 13.40b B3 36.97c 20.73f 19.52f 16.68e 10.72c 5.62e 4.05f 10.01e C 41.19a 26.57a 23.46a 17.08d 12.71a 7.11b 9.49a 13.86b D 25.18f 18.50g 18.08g 14.49g 10.54c 6.58c 4.01f 7.96f

TABLE 4 - Saturated hydraulic conductivity and drainage porosity demonstrated about 57% increase in the saturated hydrau- Drainage lic conductivity on soils in which lucerne was cultivated. Saturated hydraulic conductivity Combination porosity Pietsch et al. [32] and Yang et al. [33] also lay emphasis on (µm . s -1) (%v/v) the role of lucerne in the process of shaping physical and A1 20.68e 18.68b water properties of arable soils. The research results pre- A2 18.81f 11.97d sented in this study prove that the above observation also A3 18.00g 8.93ef B1 28.13b 8.87f concerns soils developing from post-mining deposits and B2 27.10c 19.84a is probably determined by the biology of this plant. B3 9.15h 17.01c C 37.32a 16.61c D 25.63d 9.80e 4. CONCLUSIONS content (3-6%) was recorded at the potential corresponding 1. The most favourable values of the determined phys- to the limit of readily available water (pF 3.7). Water con- ical and water properties of the examined soils were ob- tent at the wilting point (pF 4.2) was typical for cultivated tained for the system of utilization “permanent lucerne” soils of similar texture [30]. At this potential, the soil under (variant C). B2 combination was characterised by the lowest moisture content (9.02% v/v), while soil samples collected from var- 2. Lucerne plays a significant role in the improvement iant C were found to contain the highest content of this of physical properties of the developing soils. kind of water (12.71 % v/v). Maximal hygroscopic capacity (pF 4.5) ranged from 4.38 (B2) to 9.60% v/v (B1). In 3. Available water of reclamated soils was greater in most cases the differences between tested combinations comparison with the “fresh post-mining material” (without were significantly, which indicates on the influence of reclamation). framing practise on water capacities. From the point of 4. The content of total carbon varied and depended on view of farming, water available for crop plants is the most the system of utilization and combination of the applied important property. Readily available water (pF 2.0-3.7) mineral fertilisation. The impact of the employed mineral ranged from 4.01% v/v (variant D - fresh material) to fertilisation - known in the restoration terminology as ‘im- 9.49% v/v (variant C). Readily available water was usually provement of the chemism of the post-mining deposits’ - significantly higher in soils developing from the recla- despite varying content of total carbon was, practically mated plots (A,B,C) in comparison with the fresh post- speaking, negligible with respect to the improvement of mining materials (variant D). The similar dependencies their physical properties. was confirmed in the case of TAW which was by about 4- 8% v/v higher than the readily available water (RAW). The The authors have declared no conflict of interest. most advantageous water properties were recorded in the case of cultivation of lucerne. Saturated hydraulic conductivity of the examined soils following the reclamation methods varied and ranged from 9.15 (B3) to 37.32 µm·s-1 REFERENCES (C) (Table 4). No correlation was determined between this feature and drainage porosity (Table 4). Significantly the [1] Gilewska, M. and Otremba, K. (2013) Seed quality of rapeseed plants obtained from cultivations on post-mining areas in the re- highest permeability of soil on which lucerne was culti- gion of Konin. Annual Set The Environment Protection 15, 505- vated was apparent. According to Gilewska [5], the strong 514. tap root system growing fairly deep into the soil improves [2] Jakubus, M., Gajewski, P. and Kaczmarek, Z. (2013) Physiochem- heavy textured soil physical and water properties and ical and chemical properties of top horizons of selected soils located in the neighborhood of prospective Tomisławice brown coal hence, filtration capabilities of the soil are increased. This opencast mine. Annual Set The Environment Protection 15, 2232- correlation is also observed on arable soils. Rasse et al. [31] 2248, (in Polish).

1230 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[3] Charzyński, P., Bednarek, R., Greinert , A., Hulisz, P. and Uza- [21] Shrestha, R.K. and Lal, R. (2007) Soil carbon and nitrogen in 28- rowicz, Ł. (2013) Classification of technogenic soils according to year-old land uses in reclaimed coal mine soils of Ohio. J. Environ. WRB system in the light of Polish experiences. Soil Science An- Qual., 36, 1775-1783. nual 64, 4, 145-150. [22] Gilewska, M. and Otremba, K. (2004) Properties of soils formed [4] Bender, J. (1995) Reclamation of post-mining areas in Poland. Re- from post-mining grounds. Soil Science Annual 55 (2), 111-121, search Bulletins on Agricultural Sciences Progress 418, 76-85, (in (in Polish). Polish). [23] Stępień, A. and Adamiak, J. (2007) Formation of various soil [5] Gilewska, M. (1991) Reclamation of post-mining areas on the ba- chemical properties under the influ-ence of different fertilization sis of KWB Konin opencast mine. Roczniki Akademii Rolniczej methods in rotation. Acta Agrophysica, 10(2), 465-472, (in Polish). w Poznaniu, 211, 59p. (in Polish). [24] Kochian, L.V., Hoenkenga , O. A. and Piñeros, M. A. (2004) How [6] Otremba K., Gilewska, M. and Owczarzak, W. (2013) Influence do crop plants tolerate acid soils? Mechanisms of aluminium tol- of agricultural utilization on select-ed properties of the soil devel- erance and phosphorus efficiency. Annual Review of Plant Biology oping from KWB Konin post-mining grounds. Annual Set The En- 55, 459-493. vironment Protection 15, 1738-1758, (in Polish). [25] O’Dell, R.E. and Classesen, V.P. (2009) Serpentine Revegetation: A Review. Northeastern Naturalist 16, 253-271. [7] Šouryková, M., Frouz, J and Šantrůčkowá, H. (2005) Accumula- tion of carbon, nitrogen and phosphorus during soil formation on [26] Clemente, R., Walker, D.J., Pardoa, T., Martinez-Fernández, D. alder spoil heaps after brown coal mining, near Sokolov (Czech and Bernala, M.P. (2012) The use of halophytic plant species and Republic). Geoderma 129, 73-80. organic amendments for the remediation of a trace elements – contaminated soil under semi-arid conditions. Journal of Hazardous [8] Gilewska, M. and Otremba, K. (2008) Influence of fodder utiliza- Material 223-224, 63-71. tion system on selected physical properties of soils developing on the basis of post-mining grounds. Zeszyty Problemowe Postępów [27] Shrestha, R. K. and Lal, R. (2011) Changes in physical and chem- Nauk Rolniczych 526, 163-170, (in Polish). ical properties of soil after surface mining and reclamation. Ge- oderma 161, 168-176. [9] Hendrychová, M. (2008) Reclamation success in post-mining landscapes in the Czech Republic: a review of pedological and bi- [28] Guber, A.K., Rawls, W.J., Shein, E.V. and Pachepsky, Ya.A. (2003) ological studies. Journal of Landscapes Studies 1, 63-78. Effect of soil aggregate size distribution on water retention. Soil Science 168, 223-233. [10] Kabrna, M., and Řehoř, M. (2007) Reclamation as an effective tool for post mining landscape regeneration. In: Kungolos A., Ara- [29] Pranagal, J., Lipiec, J. and Domżał, H. (2005) Changes in pore size vossis, K., Karagiannidis, A. and Samaras, P. (Eds.) Proceedings distribution and aggregate sta-bility of two soils under long term of SECOTEX Conference and the International Conference on En- tillage system. International Agrophysics 19, 165-174. vironmental Management Engineering Planning and Economics – [30] Lauenroth, W. K. and Brandford, J. B. (2012) Ecohydrology of volume I, 613-618. dry regions of the United States: water balance consequences of [11] Bouyoucos, G.J. (1927) The hydrometer as a new method for the small precipitation events. Ecohydrology 5, 46-53. mechanical analysis of soils. Soil Science 23, 343-354. [31] Rasse, D. P., Smucker, A. J. M. and Santos, D. (2000) Alfalfa Root and Shoot Mulching Effects on Soil Hydraulic Properties and Ag- [12] Soil Conservation Service. (1992) Soil Survey laboratory methods gregation. Soil Science Society of America Journal 64, 725-731. manual. Soil Survey Investigation Report 42, United States De- partment of Agriculture, Washington, DC. [32] Pietsch, G. and Friedel, J.K. (2007) Lucerne management in an organic farming system under dry site conditions. Field Crops Re- [13] Black, G.R. (1965) Bulk density. In. Methods of soil analysis, Part search 102, 104-118. I, Physical and mineralogical properties, including statistics of measurement and sampling C.A. Black (Ed)., American Society of [33] Yang, H., Unkovich, M., McNeill, A and Wang, X. (2011) Symbi- Agronomy, Madison WI, USA, 374-395. otic N2 fixation and nitrate utilization in irrigated lucerne (Medi- cago sativa) systems. Biology and Fertility of Soils, 47, 377-385. [14] Vadjunina, A.,F. and Kortchagina, Z., A. (1973) Investigation methods for physical properties of soils and grounds, 1-400, (in Russian). [15] Klute, A. and Dirksen, C. (1986) Hydraulic conductivity and dif- fusivity: Laboratory methods. In: Klute A. (Ed.). Methods of Soil Analysis , Part 1: Physical and mineralogical methods. 2nd ed. Agronomy Monographs 9 ASA and SSSA, Madison, Wi, USA, 687-734. Received: July 07, 2014 [16] Klute, A. (1986) Water retention: Laboratory Methods In: Klute Revised: November 04, 2014 A. (Ed.). Methods of soil analysis, Part 1: Physical and mineralog- Accepted: November 21, 2014 ical methods. 2nd ed. Agronomy Monographs 9 ASA and SSSA, Madison Wi, USA, 635-662. [17] FAO. (1977) Guidelines for soil profile description. Land and Wa- CORRESPONDING AUTHOR ter Development Division, FAO. Rome, 1-66. [18] Otremba, K. (2012) Influence of the addition of lignite on the pa- Piotr Gajewski rameters of soils structure de-veloping from post-mining grounds Department of Soil Science and Land Protection of KWB Konin brown coal opencast mine. Annual Set The Envi- ronment Protection 14, 695-707, (in Polish). University of Life Sciences ul. Szydłowska 50 [19] Otremba, K., Kaczmarek, Z. and Gajewski, P. (2012) Influence of the addition of brown coal dust on basic physical and water prop- 60-656 Poznań erties of the arable horizon of the soil developing from KWB Poland. Konin post-mining grounds. Annual Set The Environment Protec- tion, 14, 741-751, (in Polish). Phone +48 61 848 73 92 [20] Gilewska, M. and Otremba, K. (2002) Spatial changeability of se- E-mail: [email protected] lected properties of post-mining grounds. Roczniki Akademii Rolniczej w Poznaniu, CCCXLII, Melioracje i Inżynieria Środowiska, 23, 83-93, (in Polish). FEB/ Vol 24/ No 4/ 2015 – pages 1227 - 1231

1231 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

PHYSICAL PROPERTIES AND EXCHANGE SYSTEM OF SEAWATER IN AYLA LAGOONS IN THE NORTHERN GULF OF AQABA, RED SEA

Riyad Manasrah

Department of Coastal Environment, Faculty of Marine Sciences, The University of Jordan-Aqaba, PO Box 2595, Aqaba 77110, Jordan

ABSTRACT 1. INTRODUCTION

Spatial and temporal variation of temperature, salinity, Lagoon ecosystem is one of the shallow water ecosys- sigma-t, dissolved oxygen (DO) and residence time in the tems that are separated by barrier or coral from larger water artificial Ayla lagoons in the northern Gulf of Aqaba were systems, which is characterized by predominant sand sub- studied during the period June 2012-May 2013 to achieve, stratum [1]. Lagoons occupy about 15% of the earth's costal for the first time, a detailed description of water properties. zones [2]. Physical properties of the lagoons are important That’s for obtaining scientific bases for management and abiotic factors in ecosystem. Lagoons respond rabidly to sustainable use. Monthly base measurements were carried any changes in temperature and/or salinity of the water be- for the temperature, salinity, sigma-t, DO at 25 sites in the cause their areas are small with shallow depth [3]. Some three lagoons and for water flow at 7 sites in the tidal lagoon. physical and chemical aspects of the coastal lagoons of the The statistical tests of all parameters revealed significant dif- Red Sea have been studied [4-6]. ferences among months. There were no significant differ- The temporal variability and spatial distribution of nu- ences among sites and lagoons except for salinity among trients in estuaries and lagoons are controlled by complex sites and lagoons and for sigma-t among lagoons. The sea- physical, chemical and biological processes, non-linearly sonal trend of temperature, sigma-t and DO were observed linked to each other and associated with external loadings, but no distinct pattern was observed for salinity. The results tidal advection/dispersion, salinity stratification, wind mix- showed that temperature played a major role in thermoha- ing and groundwater inputs. Estuaries and lagoons support line structure and DO, where salinity had minor role. The significant ecological phenomena, but at the same time suf- annual range of seawater temperature, salinity, sigma-t and fer from excessive nutrient loadings, expressed with crea- DO, for all sites in the Ayla lagoons, were 16.24-27.76 °C, -3 tion of phytoplankton blooms and other adverse symptoms 39.95-41.2, 27.03-30.42 kgm and 6.21-8.23 mg/l, respec- of eutrophication [7]. tively. Besides, temporal variation of all parameters was in general comparable with open sea. The results of current Coastal lagoons are common landforms along the measurements revealed an excellent condition with ex- borders of most continents. They have restricted connections pected secure environment. The average residence time of to the ocean, are poorly flushed, exhibit long residence times, seawater in the lagoons during period June 2012-May 2013 and are ephemeral on a geologic time scale. They owe their was 2.58 ± 0.66 days. This period is assumed to be better than origin largely to eustatic sea level rise. With the onset of an the proper needed time for ensuring a complete flushing in interglacial stage near the end of the Pleistocence 15,000 the lagoons. In general, it is needed to increase water pump- years ago, sea level rose 130 m at a rapid but variable ing in wintertime because tidal current is weaker compared to rate [8] and flooded river valleys and low-lying coastal summer, particularly for the Tidal lagoon (TL) because it is depressions. Sea level reached its approximate present connected directly with the open sea. elevation 5000 years ago [9], but has undergone fluctuations of a few meters since then. As a result of coastal processes, barriers formed that now constitute the margins of coastal lagoons [10]. The combined action of marine and fluvial KEYWORDS: temperature, salinity; dissolved oxygen; residence processes caused trapping and infilling of semi-enclosed time; artificial lagoons; Gulf of Aqaba coastal systems, including coastal lagoons, and the reshaping of seaward boundaries. Coastal lagoons also formed in marginal depressions behind barriers of reworked deltaic sediments [11] in active river delta systems. Such lagoons are often oriented normal rather than parallel to the coast, as in

1232 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

the case of Lake Calcasieu, Louisiana, and Lagoa Feira, the sustainability of the environmental protection for dura- Brazil. ble development of the area is the high propriety of the Aq- Coastal lagoons experience the same forcing functions aba authorities. Therefore, the major objective of this study as coastal plain estuaries, yet differ from these in many re- is to achieve a detailed description of seasonal changes of spects. Estuaries and coastal lagoons are driven by tides, water properties (seawater temperature, salinity, sigma-t river input, wind stress, and heat balance at the surface, but and dissolved oxygen) and water exchange system inside respond unequally to these forcing functions because of the lagoons with the open sea in the northernmost Gulf of differences in geomorphology. Whereas circulation, mix- Aqaba. That is for obtaining scientific bases for manage- ing and exchange have been studied extensively in coastal ment and sustainable use. plain estuaries, these processes have been less well synthesized for coastal lagoons. However, the dependence of water quality and eutrophication on flushing, hydrodynamic 2. MATERIALS AND METHOD turnover and physical dynamics is of prime importance for planning and implementation of coastal management strat- 2.1 Study area egies in coastal lagoons [12]. The Gulf of Aqaba (180 km long, 20 km wide) is a semi- This paper focuses on studying the physical properties closed sea with many unique natural and physical features and exchange system of waters in Ayla lagoons (artificial (Fig. 1). It is geographically isolated by the narrow Strait of lagoons) that are located in the northernmost Gulf of Aq- Tiran. While much of the Gulf is deep (>1800 m), the north- aba. The study area is actually close to the center of two ern sector has a relatively shallow shelf adjacent to the major towns (Aqaba and Eilat). These two towns both are signif- population centers. The Jordanian coast (27 km) is fringed icantly important for the surrounding countries, because by discontinuous belt of reefs separated by sandy bottoms they are substantially under various active developments. that are usually covered by seagrass meadows [15]. The Ayla lagoons are the largest established artificial lagoons northern part of the Jordanian coast in front of Ayla lagoons in the Jordanian Gulf of Aqaba. The current rapid develop- is characterized by sandy bottoms that are covered by ment at the northern end of the Gulf of Aqaba is expected seagrass beds and absence of hard substrate and reef-build- to cause ocean pollutions at this area [13, 14]. In addition, ing corals [16].

FIGURE 1 - Map of study area and measurements sites in Ayla lagoons in the northern Gulf of Aqaba.

1233 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.2 Ayla Oasis project ture and Depth meter (CTD 19plusV2) at sites WS1-WS10 Ayla Oasis project is one of series of proposed devel- in the upper lagoon (UL), WS11-WS18 in the middle la- opments based on the Master plan of the Aqaba Special goon (ML), WS19-WS23 in the tidal lagoon (TL) and Economic Zone (ASEZ). It is located west of the main city WS24-WS25 in open sea (REF). The initial accuracy and of Aqaba at the most northern tip of the Gulf of Aqaba (Fig. resolution for the CTD sensors are ± 0.0005, 0.00007 for 1); the project site covers approximately 430 hectares, with conductivity; ± 0.005, 0.0001 for temperature; ± 0.1%, 235 m sea frontage onto the Gulf of Aqaba. The facilities 0.002% for pressure, respectively. of the Ayla Oasis project are centered on a triple lagoon system of 75 hectares [Upper Lagoon (UL) - 24 hectares, 2.4 Current data and residence time calculation Middle Lagoon (ML) - 16 hectares and Tidal lagoon (TL) The water flow was measured monthly at seven sites - 35 hectares]. The UL and ML (in the north and center of (Curr1-Curr7) inside the TL (Fig. 1) using Acoustic Dop- the project) have water surfaces at +6.0 m and +3.0 m pler Current Profiler (ADCP) workhorse 600 kHz. The above sea level respectively while the marina basin and ADCP measurers the horizontal current components of the navigation channel (TL) are directly connected to the open water column based on the prior setup the in situ survey. sea with a water surface same as the sea level. In order to Table 2 shows the setup of the ADCP. maintain reasonable water quality in the lagoons, a seawater replenishment system has been designed which con- TABLE 2 - The workhorse ADCP (600 kHz) deployment setup in the sists of a pumping station, associated pipelines, and dis- Ayla lagoons in the northern Gulf of Aqaba. charge weir structures. Water is being pumped up into the Bottom depth 5 m UL and ML, and then flowed under gravity over the vari- Number of depth cells 3 ous cascading weirs discharging at various points into the Depth cell size 1 m TL, and from there to the open sea (Table 1). These lagoons Time per Ensemble 20 sec will create remarkable seaside neighborhood that will add Pings per ensemble 250 17 km of new seafront to the Aqaba city. Time per ping 0.08 s

Transmit length 1.2 m TABLE 1 - The detailed operational scenarios of pumping system into the Ayla lagoons in the northern Gulf of Aqaba. Blank after transmit 0.8 m Distance to first bin 2.0 m Period Pump group Flow Total flow (m3s-1) (m3s-1) 1-May-12 to 15-Jul-12 Upper 10.2 The sites (Curr1-Curr7) were selected at the connected 6 Lagoon channel between the TL with the open sea on the one hand Middle 4.2 and between the TL with the UL and ML on the other hand. Lagoon 16-Jul-12 to 10-Jan-13 Upper 6.12 That is for calculating the residence time of waters inside 3.6 Lagoon all lagoons and inside the UL and ML. Middle 2.52 Lagoon The residence time of water in all lagoons was calcu- 10-Jan-13 to 1-Jun-13 Upper 6.5 3 lated using the results of current measurements at site Lagoon Curr1. However, the residence time of waters in the UL Middle 3.5 Lagoon and ML were calculated using the results of current data at sites Cuur3-Curr6. Water flow at sites Curr2 and Curr7 The bottom of the two perched lagoons (UL and ML) were used to ensure the systematic and sufficient flow in and TL consists of a layer of white sand with different thick- the TL. nesses. The UL and ML are lined with high-density polyeth- Water flow can be calculated based on the equation: ylene (HDPE), which is protected with a layer of blinding concrete. (1) There are 17 beaches constructed in the UL and ML where φ(m3s-1) is the water flow, V (m/s) is the wa- with barrier walls and having white sand spread at the flow ter current component parallel to the orientation of the con- beaches slopes and horizontal areas. To prevent erosion nected channel, and A (m2) is the cross section area at the and in order to protect the beaches, groynes have been con- site of measurements (Fig. 1). structed close to the beaches using armor rocks. As an erosion control system and scour protection, armor rocks have The residence time of water mass can be calculated been placed at seabed around vertical edge structures in the based on the relation: TL (anti scour rock). (2) 2.3 Data collection for physical properties of seawater Physical properties (temperature, salinity, sigma-t and where, T (days) is the residence time of the water mass, dissolved oxygen (DO)) of seawater were measured monthly Vol (m3) is the volume of the water mass, and φ(m3s-1) is from June 2012-May 2013 using Conductivity, Tempera- water flow.

1234 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.5 Statistical analysis non-normal distribution of n-independent samples, where The statistical analysis and comparison tests, as well as n could be the number of sites, lagoons, or months as interpretation of the contouring map for the physical prop- grouping variable. erties and current data of seawater in the lagoons were run TABLE 3: Test of normality of seawater temperature (°C), salinity, using Statview 5.0 [13] and SPSS statistics 17.0 and Surfer sigma-t and DO (mg/l) measurements during June 2012-May 2013 in 11.0 software. Ayla lagoons in the northern Gulf of Aqaba using Shapiro-Wilk test for (a) real data and (b) transformed data. The Kruskal-Wallis test is used for comparing ordinal or non-Normal variables for more than two groups. The (a) Shapiro-Wilk (real data) Kruskal-Wallis test is a nonparametric test, equivalent to Statistic df Sig. the one-way ANOVA [17]. The Shapiro-Wilk test for real Temperature (°C) 0.950 300 0.000 data and transformed data was used for normality test among more than two groups. Salinity (psu) 0.981 300 0.001 sigma-t (kgm-3) 0.947 300 0.000 DO (mg/l) 0.947 300 0.000 3. RESULTS AND DISCUSSION (b) Shapiro-Wilk (transformed data) Statistic df Sig. 3.1 Statistical analysis results Log (Temperature) 0.942 300 0.000 Test of normality for the seawater temperature, salin- 10 ity, sigma-t and DO measurements at all sites in Ayla la- Log10(Salinity) 0.980 300 0.000 goons, during the study period between June 2012 - May Log10 (Sigma-t) 0.948 300 0.000

2013, was performed using Shapiro-Wilk test. That is for Log10(DO) 0.951 300 0.000 checking whether the samples came from a normally distributed, so that afterwards we can determine the proper The results of Kruskal Wallis test for temperature, sa- comparison test of all parameters among sites, lagoons, and linity, sigma-t, and DO measurements revealed significant months. The results of the normality test (Table 3) revealed differences for all parameters among months (p=0.0). An that all parameters are not normally distributed (p ≤ 0.001). extended statistical test was performed to determine in A logarithmic transformation of the data was done, and a which months or seasons the significant differences for all test of normality was performed again. The results of the variables were occurred. The results revealed that the sig- transformed process also represented a non-normal distri- nificant difference was among seasons for temperature, bution (Table 3). sigma-t and DO. There is a significant differences among Based on the normality test results, the proper compar- summer, winter and autumn-spring months, which is due ison test was nonparametric test (Kruskal Wallis Test) for to the natural seasonal cycle of temperature, sigma-t and

TABLE 4 - Nonparametric comparison test (Kruskal Wallis test) of seawater temperature (°C), salinity, sigma-t and DO (mg/l) measurements during June 2012-May 2013 in Ayla lagoons in the northern Gulf of Aqaba among (a) months, (b) sites and (c) lagoons.

(a) Comparison test (Kruskal Wallis test) among months Temperature (°C) Salinity (psu) Sigma-t (kg m-3) DO (mg/l) Chi-Square 266.675 36.391 257.327 258.257 df 11 11 11 11 Asymptotic Sig. 0.000* 0.000* 0.000* 0.000*

(b) Comparison test (Kruskal Wallis test) among sites Temperature (°C) Salinity (psu) Sigma-t (kg m-3) DO (mg/l) Chi-Square 8.263 128.097 14.241 3.388 df 24 24 24 24 Asymptotic Sig. 0.999 0.000* 0.941 1.000

(c) Comparison test (Kruskal Wallis test) among lagoons Temperature (°C) Salinity (psu) Sigma-t (kg m-3) DO (mg/l) Chi-Square 5.489 84.815 8.455 1.374 df 3 3 3 3 Asymptotic Sig. 0.139 0.000* 0.037* 0.712 *: Significant if p<0.05 (confidence level is 95%).

1235 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 2 - Monthly spatial variation of seawater temperature (°C) during June 2012-May 2013 in Ayla lagoons in the northern Gulf of Aqaba.

1236 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

DO variation in the region. The test revealed that for salin- sigma-t values at these sites was recorded due to high evap- ity the significant difference among months was particu- oration and weak water exchange. larly in February due to the lowest evaporation rate compared to the other months. The same test was run for the 3.2 Temperature same parameters for testing statistical differences among The monthly spatial distribution of seawater tempera- sites and lagoons. The results detected that there are no signif- ture in Ayla lagoons (UL, ML and TL), during June 2012- icant differences of all parameter among sites and lagoons, ex- May 2013, revealed a slight monthly gradient among the cept for salinity among sites (p=0.0), lagoons (p=0.0), and for lagoons and open sea (REF) during summer season (May- sigma-t among lagoons (p =0.037) (Table 4). However, these October), while clear differences of temperature were de- significant difference for salinity and sigma-t among lagoons tected among the lagoons and REF during winter season were due to the differences values at sites WS19-WS23 in (November-April) (Fig. 2). The annual average of seawater the TL compared to the other sites. These sites, particu- temperature at all sites inside each lagoon (Table 5) re- larly, WS19 had very weak water flow compared to the vealed that all lagoons had mostly very close temperature other sites in all lagoons. Therefore, high salinity and values during summer compared to REF, where the maximum

TABLE 5 - Annual average and standard deviation of seawater temperature (°C), salinity, sigma-t and DO (mg/l) measurements during June 2012-May 2013 in Ayla lagoons in the northern Gulf of Aqaba.

Temperature (°C) Salinity (psu) Sigma-t (kg m-3) DO (mg/l) Season Lagoon Mean SD Mean SD Mean SD Mean SD UL 25.21 1.32 40.71 0.12 27.60 0.39 6.65 0.28 ML 25.38 1.33 40.74 0.14 27.56 0.41 6.64 0.30 Summer TL 25.49 1.54 40.93 0.09 27.67 0.48 6.68 0.26 REF 25.32 1.34 40.54 0.06 27.43 0.41 6.74 0.25 UL 20.27 1.66 40.73 0.23 29.06 0.56 7.16 0.45 ML 21.06 1.20 40.64 0.19 28.77 0.38 7.10 0.37 Winter TL 20.69 1.83 40.92 0.15 29.08 0.60 7.20 0.36 REF 22.70 0.90 40.69 0.05 28.29 0.35 7.10 0.54

FIGURE 3 - Monthly variation of seawater temperature anomaly (°C) during June 2012-May 2013 at all sites (WS1-WS25) in Ayla lagoons in the northern Gulf of Aqaba.

1237 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

difference of seawater temperature between the lagoons pumping should be higher than the 6.2 m3s-1, in order to and REF was about ±0.15°C. In contrast, the seawater tem- reduce the temperature differences between the lagoons perature during winter inside UL, ML and TL was in aver- and open sea in order to ensure a similar environmental age colder than REF of about 2.4, 1.6 and 2.0 °C, respec- conditions for flowed organisms through pumping between tively (Table 5). The maximum difference of seawater tem- open sea and the lagoons, therefore enrich the region with perature between all sites inside the lagoons and REF was extended coast that have the same ecosystem of the north- -6.06 °C, which was measured in January inside UL (site: ern Gulf of Aqaba without any possible effect on the or- WS7) (Fig. 3). ganisms due to temperature difference between open sea and the lagoons. Particularly, the lagoons was established The temperature measurements of all sites for each la- and connected with the open sea only since two years and goon (Fig. 3) revealed that the sites WS1 and WS2 in UL its very important in the beginning ensure perfect condi- and WS13-WS17 in ML (Fig. 1), which are near the area tions as could as possible. of pumps points, were relatively having the minimum differences compared to REF. On the other hand, the sites The physical properties of seawater in open sea, in the away from the pumping points showed higher differences northern Gulf of Aqaba and in vicinity of the Ayla lagoons, of measuring seawater temperature compared to REF. has been studied before establishing the project by several The seasonal trend of seawater temperature was ob- authors, e.g., [13, 14, 18, 19]. served clearly at all sites (Fig. 4). The maximum values All previous works have revealed a temperature annual were recorded in summer, where the highest (27.76 °C), range of the seawater in the Jordanian coast in the northern among all sites, was recorded in July at WS21 (TL). The Gulf of Aqaba between 20.5-27.5 °C. This definitely is due minimum values were recorded in winter, and the lowest to the natural annual cycle of stratification in summer and value (16.24 °C), among all sites, was recorded in January mixing in winter in the Gulf of Aqaba, which is attributed at WS6 in UL. The overall trend of seasonality appeared to driven by sea-surface cooling or net heat flux, weather differences in the annul range at each site that was clearly conditions, and seasonal differences in the temperature of due to its location to the pumping points, where tempera- the Red Sea water that flows into the Gulf of Aqaba [20- ture at sites away from the pumping points had larger an- 23]. The winter difference of seawater temperature in Ayla nual range. lagoons, compared to the open sea and previous works in Besides, the temperature anomalies of all sites in the the northern Gulf of Aqaba, is attributed the massive heat three lagoons compared to the REF in summer, in relation loss of water body of the lagoons, due to its small volume, to the total pump rates of 10.2 m3s-1 during May-July 2012 compared to the open sea that led to lose heat continuously and 6.5 m3s-1 during May 2013 in both the UL and ML, without sufficient substitutions from deeper water to main- revealed that the lowest pump rate (6.5 m3s-1) was suffi- tain this loss. The only source for maintaining the seawater cient for sustaining thermal distribution inside the lagoons temperature with the open sea is by pumping waters con- to be comparable with open sea. On the other hand, in gen- tinuously from the open sea to the lagoons. On the other eral, the temperature anomalies in winter at all sites in the hand, there was not a significant difference of seawater three lagoons with pump rate of 6.2 m3s-1 showed higher temperature during summer between the lagoon and open values compared to summer. This indicates that in winter sea in this study (Table 4), as well as with the previous season (particularly during November-February) the works, because the winds in summer is generally stronger

28 UL ML TL REF 26

20 Temperature (°C) 18

16 May-12 Jul-12 Aug-12 Oct-12 Dec-12 Jan-13 Mar-13 May-13 Jun-13 Month

FIGURE 4 - Monthly average of seawater temperature (°C) during June 2012-May 2013 at all sites (WS1-WS25) in Ayla lagoons in the northern Gulf of Aqaba.

1238 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 5 - Monthly spatial variation of seawater salinity during June 2012-May 2013 in Ayla lagoons in the northern Gulf of Aqaba.

1239 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

to cool the sea surface of the lagoons that reduced transfer- among the sites in UL and ML revealed fluctuation of in- ring heating from the atmosphere to the lagoons' water creasing and decreasing compared to REF, except at sites body. Besides, pumping seawater from the open sea in TL that showed in summer higher values compared to helped, as well, to maintain the seawater temperature to be REF (Fig. 5). comparable with open sea. The annual average of salinity at all sites inside each 3.3 Salinity lagoon (Table 5) during summer and winter was compara- The monthly spatial distribution of salinity at all sites ble with the average values of salinity at REF, where the in UL, ML and TL during summer (May-October) were maximum difference of the average salinity between the higher compared to REF, with gradual increase propor- lagoons and REF was about 0.40 (Fig. 6; Table 5). The tional to distance from the pumping points (Fig. 5). In win- maximum difference of salinity between all sites inside the ter (November-April) the spatial distribution of salinity lagoons, compared to REF, was -0.74 and +0.54. These re-

FIGURE 6 - Monthly variation of seawater salinity anomaly during June 2012-May 2013 at all sites (WS1-WS25) in Ayla lagoons in the northern Gulf of Aqaba.

41.4 UL ML TL REF 41.2

40.8

40.6 Salinity 40.4

40.2

40 May-12 Jul-12 Aug-12 Oct-12 Dec-12 Jan-13 Mar-13 May-13 Jun-13

Month FIGURE 7 - Monthly average of seawater salinity during June 2012-May 2013 at all sites (WS1-WS25) in Ayla lagoons in the northern Gulf of Aqaba.

1240 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 8 - Monthly spatial variation of seawater sigma-t (kgm-3) during June 2012-May 2013 in Ayla lagoons in the northern Gulf of Aqaba.

1241 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 9 - Monthly variation of seawater sigma-t (kgm-3) anomaly during June 2012-May 2013 at all sites (WS1-WS25) in Ayla lagoons in the northern Gulf of Aqaba.

sults were measured in February inside UL (WS3), and in mean value was recorded in summer. These values are August inside ML (WS15), respectively (Fig. 6). mostly comparable with this study. The seasonal trend of salinity was unclear and non- Although a value of salinity above 41, at many sites in dominant at all sites (Fig. 7). The maximum value (41.20) the lagoons (Fig. 5), was measured in all seasons, this was was recorded in January at WS8 in UL while the minimum not a significant difference compared to the open sea, value (39.95) was recorded in February at WS3 in UL. which might be attributed to the high transferring heat inside the lagoons compared to the open sea that led to in- In general, the spatial and temporal variation of sea- crease the evaporation at most of the sites inside the la- water salinity in the lagoons revealed no distinct pattern. goons, particularly at sites that are away from the pumping However, significant differences among sites and lagoons points with low circulation. were revealed (Table 4). On the other hand, salinity records at all sites were comparable to open sea with maximum an- 3.4 Sigma-t nual range of 39.95 in February to 41.2 in January. This Sigma-t measurements in all lagoons and REF sites, dur- mostly agrees with previous studies [13, 25-27] in the ing June 2012-May 2013, revealed a well inverse relationship northern Gulf of Aqaba that recorded a salinity range of with temperature. The monthly spatial distribution of sigma- 40.20-40.92. The salinity variation in the northern Gulf of t (Fig. 8) showed a slight monthly gradient among the la- Aqaba can be affected basically by the high evaporation goons and open sea (REF) during summer season (May-Oc- during summer and low saline water from the Red Sea, and tober) while clear differences of sigma-t were detected the mixing condition during winter [13, 22]. The general among the lagoons and REF during winter season (Novem- outcomes of the spatial and temporal variations of the sa- ber-April) (Fig. 8). The annual average of sigma-t at all sites linity in Ayla lagoons revealed a slight variation compared inside each lagoon (Table 5) revealed that all lagoons had to the open sea and previous works. Besides, the salinity mostly very close sigma-t values during summer compared played a minor role in thermohaline structure of the la- to REF. The maximum difference of sigma-t between the lagoons while temperature played the major role. goons, compared to REF, was less than 0.25 kgm-3. In con- Manasrah et al. [13] found that temporal variation of trast, the sigma-t during winter inside UL, ML and TL was average values of salinity in the coastal and offshore water greater than REF of about 0.77, 0.48, and 0.79 kgm-3, respec- column near the study area in the northern Gulf of Aqaba tively (Table 5). The maximum difference of sigma-t between revealed a fairly stable variation of salinity. Its maximum all sites inside the lagoons and REF was 1.93 kgm-3, which value of 40.91 was recorded in autumn while the minimum was measured in January inside UL (site: WS6) (Fig. 9).

1242 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

The sigma-t of all sites for each lagoon (Fig. 9) re- The sigma-t values in the lagoons were not signifi- vealed that the sites WS1 and WS2 in UL and WS13-WS17 cantly differed among all sites as well as compared to the in ML (Fig. 1), which are near the area of pumps points, open sea (Table 4). relatively had the minimum differences compared to REF. On the other hand, the sites away from the pumping points 3.5 Dissolved Oxygen showed higher differences of measuring sigma-t compared DO measurements in all lagoons and REF sites, during to REF. June 2012-May 2013, revealed clear change in monthly The seasonal trend of sigma-t was dominating obvi- spatial distribution (Fig. 11). On the one hand, very slight ously at all sites (Fig. 10). The maximum values were rec- monthly gradient was detected among the lagoons and orded in winter and the highest (30.42 kgm-3) among all open sea (REF), during summer season (May-October). In sites was recorded in January at WS6 (UL). The minimum addition, clear differences were observed among the la- values were recorded in summer and the lowest (27.03 goons and REF, during winter season (November-April) kgm-3), among all sites, was recorded in July at WS6 (UL). (Fig. 11). The annual average of DO at all sites inside each The overall trend of seasonality appeared differences in the lagoon (Table 5) revealed that all lagoons had mostly very annul range at each site. That was clearly due to its location close DO values during summer compared to REF. The to the pumping points. The sigma-t at sites, away from the maximum difference of DO among the lagoons, compared pumping points, had larger annual range. to REF, was less than 0.05 mg/l. In contrast, the DO during As well as what was explained previously for temper- winter inside UL, ML and TL was relatively less than REF ature, the sigma-t anomalies in winter inside the lagoons, of about 0.09, 0.17 and 0.18 mg/l, respectively (Table 5). in general, were higher compared to summer, associated The maximum difference of DO between all sites inside the with the total winter pump rate of 6.2 m3s-1. Again, this in- lagoons and REF was 1.05 mg/l, which was measured in dicates clearly that in winter season, (particularly during February inside UL (site: WS1) (Fig. 12). The DO of all November-February) the pumping should be higher than sites for each lagoon (Fig. 12) revealed that the monthly the 6.2 m3s-1 in order to reduce the sigma-t differences be- differences compared to REF did not relate to the geo- tween the lagoons and open sea. graphical distance of each site from the pumping points, where a clear fluctuation in DO differences compared to The sigma-t is calculated based on temperature and sa- REF dominated. linity, as well as pressure values [28]. But due to mostly stable salinity and pressure in the shallow lagoons' water The seasonal trend of DO was, in general, the domi- (Fig. 7), then sigma-t clearly reflected the direct inverse nant signal at all sites, as well as for the temperature and relationship with temperature variation (Figs 4, 10). sigma-t (Fig. 13). The maximum values were recorded in winter while the highest (8.23 mg/l), among all sites, was Sigma-t measurements in this study in all lagoons recorded in February at WS25 (REF). The minimum values agree well with many previous works [13, 18, 21, 29]. Ma- were, in general, recorded in summer. The lowest value nasrah and Badran [18] found that sigma-t at surface water was 6.25 mg/l, which was measured in July at WS11 (ML). in the northern Gulf of Aqaba ranged between 27.2-28.6 On the other hand, an exceptional minimum value of DO kgm-3, while Al-Najjar [25] reported a range of 26.87- was measured in December (6.21 mg/l) at WS2 in TL. 28.51 kgm-3. Besides, Manasrah et al. [13] observed maximum and minimum sigma-t between 27.62-28.88 kgm-3.

31 UL ML TL REF 30

Sigma-t 28

26 May-12 Jul-12 Aug-12 Oct-12 Dec-12 Jan-13 Mar-13 May-13 Jun-13 Month

FIGURE 10 - Monthly average of seawater sigma-t (kgm-3) during June 2012-May 2013 at all sites (WS1-WS25) in Ayla lagoons in the northern Gulf of Aqaba.

1243 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 11 - Monthly spatial variation of DO (mg/l) of seawater during June 2012-May 2013 in Ayla lagoons in the northern Gulf of Aqaba.

1244 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 12 - Monthly variation of DO (mg/l) anomaly of seawater during June 2012-May 2013 at all sites (WS1-WS25) in Ayla lagoons in the northern Gulf of Aqaba.

8.5 UL ML TL REF 8

7.5

6.5

Dissolvedf oxygen(mg/l) DO 6 May-12 Jul-12 Aug-12 Oct-12 Dec-12 Jan-13 Mar-13 May-13 Jun-13 Month

FIGURE 13 - Monthly average of DO (mg/l) of seawater during June 2012-May 2013 at all sites (WS1-WS25) in all Ayla lagoons in the northern Gulf of Aqaba.

The seasonal trend of DO might be attributed that the The variation in DO at all sites, during the study pe- oxygen saturation concentration depends on temperature riod, did not exhibit any significant difference among all and salinity [30], when salinity, as described previously, sites, where these values of DO were comparable with the had a very slight variation that led temperature to be the open sea (Table 4). This indicates the water quality in the major factor affect DO. In addition to these conservative lagoons can be considered environmentally secure. These parameters, the dissolved oxygen concentration depends values of DO during this study agree with previous works, on the photosynthetic rate and subsequently on nutrient e.g., Manasrah et al. [29] reported that oxygen concentra- concentrations. High temperature and salinity cause the ox- tions decreased gradually with increasing depth in the wa- ygen to be relatively low [31]: the higher the temperature, ter column, from 6.96 ± 0.17 mg/l in the surface water to the lower the solubility of oxygen in seawater. 5.45 ± 0.45 mg/l at 400 m. Besides, Badran [31] revealed

1245 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

that dissolved oxygen in the northern Gulf of Aqaba 3.6 Seawater exchange and residence time showed a maximum and homogeneous distribution in win- The main recovery to ensure the continuation of good ter and a minimum in summer (6.52 to 6.58 mg/l). Also, he quality water inside the lagoons is a good circulation and found that dissolved oxygen concentrations, both tempo- sufficient water exchange with the open sea to avoid stagna- rally and vertically, exhibited only small variations, which tion that could lead to affect the water quality and conse- were clearly temperature-dependent. In the southern Gulf quently to pollution. The monthly spatial distribution of wa- of Aqaba, Dorgham et al. [32] found that dissolved oxygen ter flow (m3s-1) was calculated in the TL that represents the attained slightly high concentrations (5.3-7.8 mg/l) in the final link with the UL and ML with the open sea (Fig. 14, whole water column, with slight seasonal variation. Table 6).

FIGURE 14 - Monthly spatial variation of seawater flow (m3s-1) and speed (ms-1) during June 2012-May 2013 in Ayla lagoons in the northern Gulf of Aqaba.

1246 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 6 - The average of flow (m3s-1) and residence time (days) of seawater during June 2012-May 2013 inside all the lagoons through the site Curr1 and inside the UL and ML through the sites Curr3-Curr6 in the Ayla lagoons in the northern Gulf of Aqaba.

Date All lagoons (UL, ML and TL) UL and ML Average flow Average residence time Average flow Average resi- (m3s-1) (days) (m3s-1) dence time (days) 28 June 2012 16.62 1.25 13.24 0.60 18 July 2012 10.50 1.98 10.99 0.73 22 August 2012 9.88 2.10 8.36 0.96 17 September 2012 9.17 2.26 10.35 0.77 17 October 2012 9.59 2.16 9.39 0.85 19 November 2012 5.94 3.49 5.08 1.58 17 December 2012 8.38 2.47 7.01 1.14 16 January 2013 7.25 2.86 7.38 1.08 17 February 2013 6.68 3.11 6.75 1.19 18 March 2013 7.33 2.83 7.40 1.08 21 April 2013 6.17 3.36 5.64 1.42 20 May 2013 6.71 3.09 7.71 1.04

The results of the average of flow rate at the main chan- to be better and more than a proper needed time for ensur- nel of the lagoons with the open sea at site Curr1 (Fig. 1), ing a complete flushing in the lagoons in order to secure during the period June 2012-May 2013, varied within the high quality environment. In general, it is needed to in- range of 5.94-10.50 m3s-1 except during June 2012, which crease water pumping in wintertime because tidal current was 16.62 m3s-1 (Fig. 14, Table 6). In general, the lowest in winter is weaker compared to summer, particularly for average value of the flow rate was recorded in November the TL, because it is connected directly with the open sea. 2012, which was in average lower of about 34% compared This agrees with the finding by Monismith and Genin [33], to the other months (Fig. 14, Table 6). This was attributed who found that when the Gulf of Aqaba is strongly strati- to shutting off the water pumps by the operator, before fied in summer, tidal currents are strong, and when stratifi- measuring water currents in November 19th, 2012, due to cation is weak, tidal currents are weak. the meteorological condition and high turbidity in the wa- In general, the water current in the open sea near the ter in this period. The average of residence time of the wa- Ayla lagoons has seasonal trends with vertical anticlock- ters inside all lagoons in November 2012 measurement was wise rotation [13]. However, in spring, the current direc- 3.49 days, which was in average longer of about 40% com- tion at 6, 12 and 18 m depth was firstly northwestward and pared to the other months of the study period. On the other then east- southeastward and then northwestward again, hand, the average of total water flow of the study period which revealed a mix of external forces dominated during through the channels at the sites Curr3-Curr6 between the UL spring on water movement. This current pattern might be 3 -1 and ML with the TL was in the range of 5.08-13.24 m s (Ta- related to the effect of bottom topography, which directed ble 6). In general, the lowest value of average of the flow rate currents to flow parallel to its shape, i.e. parallel to the through the channels at the sites Curr3-Curr6 (Fig. 1) was shoreline [13]. The main feature of current near the study recorded in November 2012, which was in average lower area during autumn, winter and spring seasons was the ex- of about 40% compared to the other months. Besides, the istence of multi temporal reverses of current direction from average residence time of 1.58 days in November 2012 was northwestward to southeastward, which may be related to calculated of the waters in the UL and ML, which was in the direction reflection of current density, due to differen- average longer of about 60% compared to the other months tial cooling between eastern and western parts near the of the study period, because of the same reason above re- study area. Other studies [34-36] reported that wind events garding shutting the water pumps prior and during the in the northern Gulf of Aqaba drive upwelling in the east- measurements of water currents (Table 6). ern side and downwelling in the western side. Besides, Nie- In general, the average of water flows and residence mann et al. [37] concluded that differential cooling of near time in the lagoons during period June 2012-May 2013 was – and offshore surface water, during cold winter nights, re- 8.69 ± 2.93 m3s-1 and 2.58 ± 0.66 days, respectively. Alt- sults in cross-shore gradient of density triggering gravity hough there are some semi-enclosed areas that did not ex- (density) currents. Manasrah et al. [13] found anticlock- change waters efficiently compared to the other sites, this wise current rotation with depth, during summer and au- did not represent an obstacle of overall water exchange in tumn seasons, near the study area, which may be related to the lagoons with the open sea. Therefore, the results of cur- the Ekman wind drift in the coastal region with vertical rent measurements revealed an excellent condition with ex- wall [38-39], which predicts that steady wind stress acting pected secure environment in the future. The results of av- together with the Coriolis force will produce a transport of erage residence time of seawater in the lagoons is assumed water to the right of the wind.

1247 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

These findings of current pattern in the open sea near REFERENCES the Ayla lagoons confirm that source of pumped water into the UL and ML, and the flowed water directly through the [1] Hutomo, M. and Moosa, M.K. (2005) Indonesian marine and TL by tidal current could be from either eastern or western coastal biodiversity: Present status. Indian J Mar. Sci, 34, 88- 97. side of the northern Gulf of Aqaba, i.e. it will be very dif- ficult to determine the source of any external contamina- [2] UNESCO (1981) Tenth Report of the Joint Panel on Oceano- tion might enter the lagoon. Therefore, continuous moni- graphic Tables and Standards. UNESCO Technical Papers in toring of physical, chemical and biological parameters in Marine Sciences, No. 36. Ayala lagoons is highly important in order to ensure the [3] Smith, N.P. (1977) Meteorological and tidal exchange be- sustainability of such a mega project. tween Corpus Christi Bay, Texas and the north-Western Gulf of Mexico. Estuarine, Coastal and Marine Science, 5, 511-520. [4] Al-Barakati, A. (2010) Some hydrographic features of Rabigh 4. CONCLUSIONS lagoon along the eastern coast of the Red Sea. Journal of King Abdulaziz University, 21(1), 123-132.

This study focused for the first time on the spatial and [5] Meshal, A.H. (1987) Hydrography of hypersaline coastal la- temporal variation of temperature, salinity, sigma-t, DO goon in the Red Sea. Estuarine, Coastal and Marine Science, and residence time in the artificial Ayla lagoons in the 24, 167-175. northern Gulf of Aqaba during the period June 2012-May [6] El-Sayed, M.Kh. (1987) Chemistry of modern sediments in 2013. The statistical tests of all parameters revealed signif- hypersaline lagoon, North of Jeddah. Estuarine, Coastal and icant differences among months due to the natural seasonal Shelf Science, 25, 467-480. cycle in the region. There were no significant differences [7] Gikas, G.D., Yiannakopoulou, T. and Tsihrintzis, V.A. (2009) among sites and lagoons except for salinity among sites Hydrodynamic and nutrient modeling in a Mediterranean coastal and lagoons and for sigma-t among lagoons, which, in gen- lagoon. Journal of Environmental Science and Health, Part A: Toxic/Hazardous Substances and Environmental Engineering, eral, all parameters inside all lagoons were comparable 44, 1400-1423. DOI: 10.1080/10934520903217336. with the open sea. The results of current measurements revealed an excellent condition with expected secure envi- [8] Nichols, M.M. and Biggs, R.B. (1985) Estuaries. In: R.A. Da- ronment. The average residence time of seawater in the la- vis Jr. (Editor), Coastal Sedimentary Environments. Springer, New York, pp.77-187. goons during period June 2012-May 2013 was 2.58 ± 0.66 days. This period is assumed to be better than the proper [9] Fairbridge, R.W (1980) The estuary: its definition and geody- needed time for ensuring a complete flushing in the la- namic cycle. In: E. Olausson and I. Cato (Editors), Chemistry and Biogeochemistry of Estuaries. Wiley, New York, pp.1-35. goons. Nevertheless, it is needed to increase water pumping in wintertime because tidal current is weaker compared [10] Lankford, R.R. (1976) Coastal lagoons of Mexico: their origin to summer, particularly for the TL because it is connected and classification. In: M.L. Wiley (Editor), Estuarine Pro- cesses. Academic Press, New York, Vol. 2, pp.182-215. directly with the open sea. This is to be sure the lagoons has the same environmental condition compared to the [11] Nichols, M.M. and Allen, G. (1981) Sedimentary processes in open sea, particularly, the lagoons was established and con- coastal lagoons. In: Coastal Lagoon Research: Present and Fu- nected with the open sea only since two years and its very ture. UNESCO, Paris, pp. 27-80. important in the beginning ensure perfect conditions as [12] Kjerfve, B. and Magill, K.E. (1989) Geographic and hydrody- could as possible to avoid any effect on the organisms and namic characteristics of shallow coastal lagoons. Marine Ge- the new ecosystem due to any abrupt change in temperature ology, 88, 187-199. between open sea and the lagoons. [13] Manasrah, R., Zibdah, M., Al-Ougaily, F., Yusuf, N. and Al- Najjar, T. (2007) Seasonal Changes of Water Properties and Current in the Northernmost Gulf of Aqaba, Red Sea. Ocean Science Journal, 42(2), 103-116.

ACKNOWLEDGMENT [14] Badran, M., Manasrah, R. and Rasheed, M. (2006) Sea-water seasonal changes at a heavy tourism investment site on the Jor- danian northern coast of the Gulf of Aqaba, Red Sea. Chemistry The author would like to thank the technical staff at the and Ecology, 22, 425-435, doi: 10.1080/02757540600917344. Marine Science Station and Ayla Oasis Company for their help in fieldwork. Special thanks are due to Maha Al-Refai, [15] UNDEP/IUCN (1988) Coral reefs of the world. UNEP Re- gional Seas Directories and Bibliographies. IUCN, Gland, Miran Manasrah, Prof. Mohammad Rasheed and Mr. Switzer-land and Cambridge, UK/UNEP, Nairobi, Kenya. Mysara Al-Amor, for their sincere help and support. [16] Al-Rousan, S., Rasheed, M., Khalaf, M.A. and Badran, M. (2005) Bottom habitat and biological characteristics of the Jor- The author has declared no conflict of interest. danian northern Gulf of Aqaba. Chemistry & Ecology 21/4,227-239.

[17] Chan, Y. and Walmsley, R.P. (1997) Learning and understanding the Kruskal-Wallis one-way analysis-of-variance-by-ranks test for differences among three or more independent groups. Phys Ther., 77(12), 1755-62.

1248 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[18] Manasrah, R and Badran, M. (2008) Inter-annual seasonal var- [36] Brenner, S., Rosentroub, Z. and Bishop, Y. (1991) Current iations in the seawater thermohaline structure in the northern measurements in the Gulf of Elat 1990/91. Report H12/91.Isr. Gulf of Aqaba. Dirasat, Pur Sciences, 35(2), 123-135. Oceanogr. Limnol. Res.

[19] Manasrah, R. (2006) Physical oceanography study in the [37] Niemann, H., Claudio, R., Jonkers, H. and Badran, M. (2004) coastal waters of the Jordanian sector of the northern tip of the Red Sea gravity current cascade near-reef phytoplankton to the Gulf of Aqaba (Hotels Area) during summer 2003. Mu'tah Lil- twilight zone. Mar. Ecolog. Prog. Ser., 269, 91-99. Buhuth Wad-Dirasat, 21(1), 81-98. [38] Price, J.F., Weller, R.A. and Schudlich, R.R. (1987) Wind- [20] Manasrah, R., Al-Horani, F., Rasheed, M., Al-Rousan, S. Driven Ocean Currents and Ekman Transport. Science, and Khalaf, M. (2006) Patterns of summer vertical and hor- 238(4883), 1534-1538. izontal currents in coastal waters of the northern Gulf of Aq- [39] Krauss, W. (1993) Ekman drift in homogeneous water. J. Ge- aba, Red Sea. Estuar. Coast. Shelf Sci., 69, 567-579. ophys. Res., 98, 20187-20210. doi:10.1016/j.ecss.2006.05.024.

[21] Manasrah, R., Badran, M., Lass, H.U. and Fennel, W. (2004) Circulation and winter deep-water formation in the northern Red Sea. Oceanologia, 46(1), 5-23. [22] Manasrah, R. (2002) The general circulation and water masses characteristics in the Gulf of Aqaba and northern Red Sea. Meereswissenschaftliche Berichte (Marine Science Report). 50, 1-120. [23] Genin, A., Lazar, B. and Brenner, S. (1995) Vertical mixing and coral death in the Red Sea following the eruption of Mount Pinatubo. Nature, 377, 507-510.

[24] Berman, T., Paldor, N. and Brenner, S. (2000) Simulation of wind driven circulation in the Gulf of (Aqaba). J. Mar. Syst., 26, 349-365. [25] Al-Najjar, T. (2000) The seasonal dynamics and grazing control of phyto- and mesozooplankton in the northern Gulf of Aqaba. Ph.D. thesis, Bremen University, Germany.

[26] Rasheed, M.Y. (1998) Assessment of Trace Nutrient and chlorophyll a concentration gradient within a Coral Reef of the Gulf of Aqaba, Red Sea. M.S. thesis, Yarmook University, Ir- bid, Jordan.

[27] Klinker, J., Reiss, Z., Kropach, C., Levanon, I., Harpaz, H., Halicz E. and Assaf, G. (1976) Observation on circulation pattern in the Gulf of Aqaba, Red Sea. Isr. J. Earth Sci., 25, 85- 103.

[28] Millero, F.J. and Poisson, A. (1981) International one-atmospheric equation of state of seawater. Deep-Sea Res., 28A, 625- 629.

[29] Manasrah, R., Rasheed, M. and Badran, M. (2006c) Relation- ships between water temperature, nutrients and dissolved ox- Received: July 02, 2014 ygen in the northern Gulf of Aqaba, Red Sea. Oceanologia, Revised: August 18, 2014; August 27, 2014 48(2): 237-253. Accepted: September 02, 2014 [30] Weiss, R.F. (1970) The solubility of nitrogen, oxygen and argon in water and seawater, Deep-Sea Res., 17 (4), 721–735. [31] Badran, M.I. (2001) Dissolved oxygen, chlorophyll a and nu- CORRESPONDING AUTHOR trients: seasonal cycles in waters of the Gulf Aqaba, Red Sea, Aquat. Ecosys. Health Manage., 4 (2), 139–150. Riyad Manasrah [32] Dorgham, M.M., El-Sherbiny and L., Hanafi, M.H. (2012) En- Department of Coastal Environment vironmental properties of the southern Gulf of Aqaba, Red Faculty of Marine Sciences Sea, Egypt M.M., Medit. Mar. Sci., 13(2), 179-186. The University of Jordan-Aqaba [33] Monismith, S.G. and Genin, A. (2004) Tides and sea level in PO Box 195 the Gulf of Aqaba (Eilat). J. Geophys. Res., 109, doi: Aqaba 77110 10.1029/2003JC002069 C04015. JORDAN [34] Brenner, S., Rosentroub, Z. and Bishop, Y. (1988) Current measurements in the Gulf of Elat. Report H3/88. Isr. Ocean- Phone: +962-77-5625380 ogr. Limnol. Res. Fax: +962-3-2013674 [35] Brenner, S., Rosentroub, Z. and Bishop, Y. (1989) Current E-mail: [email protected] measurements in the Gulf of Elat 1988/89. Report H8/89. Isr. Oceanogr. Limnol. Res. FEB/ Vol 24/ No 4/ 2015 – pages 1232 - 1249

1249 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

ELEMENTAL DISTRIBUTION AND CHEMICAL COMPOSITION OF AMADUMBE (COLOCASIA ESCULENTA) LEAF AND IMPACT OF SOIL QUALITY

Sihle Mngadi1, Roshila Moodley1 and Sreekanth B. Jonnalagadda1,*

1 School of Chemistry and Physics, University of KwaZulu-Natal, Westville campus, Chiltern Hills, Durban 4000, South Africa.

ABSTRACT of food by consumers, a great need exists for information on the nutritional content of food crops such as the Amadumbe The increased consumption of edible Amadumbe (Col- [3]. If the nutritional quality of food crops is assessed then ocasia esculenta) leaves and bulbs around the world, in- consumption of those with high nutritional value can be pro- cluding South Africa, has triggered investigations on its moted to prevent malnutrition in these communities. impact on ones nutritional needs and its potential toxicity. C. esculenta is known as Amadumbe in isiZulu and by The elemental distribution in Amadumbe leaves and bulbs the people of South Africa. It is an edible plant that belongs from eight sites in the KwaZulu-Natal region of South Af- to the family Araceae and normally grows in the subtropi- rica was investigated. The concentration of the elements in cal parts of South Africa such as KwaZulu–Natal (KZN), Amadumbe leaves was found to be in decreasing order of Mpumalanga Province and the Free State. Although it is Ca > Mg > Fe > Mn > Zn > Cu > Ni > Pb > Se ≈ Cr > Co. not extensively commercialized at present, the plant is At Tongaat, north of Durban, typical elemental concentra- -1 mostly grown in rural areas or on small farms in KZN as it tions (in µg g , dry mass) in the (Amadumbe leaves & soil) is a staple crop for these populations [4]. Studies have were Ca (11014 & 24671), Co (0.2 & 3.5), Cr (5.2 & 75), shown that the consumption of Amadumbe tubers is bene- Cu (16 & 40), Fe (453 & 21728), Mg (2368 & 1058), Mn ficial to diabetic and hypertensive patients [5]. This could (181 & 107), Ni (6.2 & 29), Pb (7.1 & 135), Se (6.7 & 5.7) also hold true for the leaves, as the medicinal value of dif- and Zn (62 & 39). Except for Zn, concentrations of the ferent parts of the same plant are generally similar due to studied elements were higher in the leaves than the bulbs. possibly containing similar phytocompounds. Amadumbe leaves appear to be a rich source of many essential elements, and macro elements, with low concentra- The nutritional value of leafy green vegetables is well tions of the toxic metals. known due to their high concentrations of essential elements, proteins and sugars. These essential elements are taken up by the plants from the soil, the quality of which is KEYWORDS: nutrition, toxic elements, contamination, synergy, affected by the surrounding environment. This makes antagonism, Colocasia esculenta, yams. plants an important link between the environment and humans and the elemental composition of plants and the food 1. INTRODUCTION chain therefore depend on soil quality. Literature abounds with research on the bulbs of the The consumption of edible plants like Colocasia escu- Amadumbe species in KZN [1], but there is little or no lit- lenta contributes significantly to the diet of many people in erature information on the nutritional value of Amadumbe many tropical and sub-tropical countries, especially those leaves and the impact of soil quality on its elemental uptake living in rural areas where there is a shortage of food [1]. and distribution. This study focused on Amadumbe leaves The low income people, especially the rural poor, tend to which are used in the preparation of various dishes such as subsist on staple crops and locally grown vegetables as pur- puri patha (Indian), imifino (Zulu) and isigwamba (Zulu) chased foods are not accessible or economically afforda- in KZN and in cooking with other vegetables in other parts ble. These communities rely on these food crops for food of the world. Plant samples from different geographic lo- security. Some of the food crops consumed as dietary sta- cations in KZN were investigated to determine the impact ples are sweet potato, millet, sorghum, cassava and of soil quality on elemental uptake by the leaves and to as- Amadumbe (C. esculenta) [2]. Nutritional value is of pri- sess for metal contamination. The elemental concentrations mary concern when a crop is being considered as a food in the leaves were also assessed for their nutritional value. source. Due to the emphasis placed on the nutritional value From eight sampling sites, the 12 elements selectively investigated were As, Ca, Cu, Co, Cr, Fe, Mg, Mn, Se, Ni, * Corresponding author Pb and Zn.

1250 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2. MATERIALS AND METHODS 2.4 Extraction of Bioavailable Metals An acidic solution of ammonium acetate and EDTA 2.1 Sampling was used to extract the nutrients potentially available to The eight chosen sites for plant and soil samples were: plants by the soil according to the method as described by Site A – Tongaat, Site B – Claremont, Site C – Adams Mis- Moodley et al. [6]. sion, Site D – Emakholweni, Site E – Umlazi, Site F – Um- gababa, Site G – Hibberdene and Site H – Margate. The 2.5 Digestion of Samples and Elemental analysis topography of sampling sites was flat and the soils were Both plant and soil samples were digested prior to sandy (Sites A, G, H) or sandy loam (Sites B, C, D, E, F) analysis using the microwave-assisted closed vessel diges- in texture. Irrigation at all sites was moderate; with most tion technique which facilitates rapid dissolution of the sites irrigated weekly. Due to the high fertile soils, the pro- sample matrix, requires low oxidizing reagent use and duction of Amadumbe is done without the use of fertilizers. causes minimal contamination of the sample. The diges- Samples were collected in June 2011, mostly during the tions were performed using the Anton Paar Multiwave Mi- harvest period. Approximately six Amadumbe plants crowave Sample Preparation System (1000 W) with 6 high- were uprooted from a randomly chosen section of the pressure tetrafluoromethaxil (TFM)-Ceramic Vessels (HF farm. Thereafter, the leaves and bulbs were removed with 50). The maximum temperature was 260°C and the maxi- a stainless steel knife. The soil immediately surrounding mum pressure was 75 bar. To ensure uniform sampling and these plants was sampled to evaluate its impact on ele- improved precision, three sub-samples of each sample, mental uptake by the plant. A thin walled stainless steel both plant tissue and soil were digested. Digestions were coring tube, 20 cm long and 3.8 cm wide, was used to performed according to the methods as described by extract short cores from the soil. The coring tube was Moodley et al. [6]. Elemental analysis was performed by driven into the soil with a wooden mallet to a depth of use of the Inductively Coupled Plasma-Optical Emission approximately 15 cm (plough depth) as the root system Spectrometer (ICP-OES) using the Perkin Elmer ICP-OES. that arises from the tuber is adventitious and shallow. Six The analytical wavelengths were selected based on mini- cores from around the plants were obtained in this way mum spectral interferences and maximum analytical per- from points that were approximately 50 cm apart. The formance. Method validation was performed by analysis of cores that were extruded were placed into a stainless steel certified reference material, lyophilized brown bread (BCR mixing bowl. Extraneous material such as leaves and 191), and comparing measured results to certified values. rocks were removed from the combined sub-samples before manual homogenization using a stainless steel spoon. 2.6 Soil organic matter (SOM), cation exchange capacity (CEC) Homogenization was done to ensure uniform distribution and soil pH of soil. The composited soil volume was reduced by the The pH of soil was determined by measuring a 1:1 method of coning and quartering. Soil samples from each (v/v) soil / water suspension using a pH meter fitted with a site were placed in separate polyethylene bags and refrig- glass electrode. SOM was estimated using a simple wet ox- erated at 4⁰C. Plant samples were also stored in separate idation method with the standard Walkley-Black titration polyethylene bags and refrigerated. (WB-T) method [7]. The pH 7.0 ammonium acetate method was used to determine the CEC of the soil [8]. 2.2 Reagents and Standards All chemicals used were of analytical-reagent grade 2.7 Bioaccumulation factors and exchangeable percentages (Merck or Sigma-Aldrich). Double distilled water was The relative accumulation of metals taken up by plants used in all the experiments. All glassware and other equip- can be calculated by dividing the concentration of the metal ment were cleaned with double distilled water to prevent in the plant by the concentration in the soil. This relative contamination. accumulation is known as the bioaccumulation factor (BF).

2.3 Preparation of Plant and Soil Samples [ Metal ] Plant Plant samples were washed with double distilled wa- BF  ter, cut into smaller pieces with a stainless steel knife and [ Metal ] Soil dried in an oven at 40°C, overnight. Dry leaf samples were ground in a food processor (Braun range) to obtain a fine Bioaccumulation factors can be obtained for both total powder. Each of the soil samples from a site was air dried and bioavailable amounts of metals found in soil. The ex- then passed through a 2 mm mesh sieve. For each site, three changeable percentage can be calculated by dividing the sieved fractions were obtained. In total, there were eight concentration of the metal that is exchangeable (Ex) by the sampling sites, each with three sieved soil replicates for total (T) concentration in the soil. analysis. All samples were stored in polyethylene bags and kept in a refrigerator at 4⁰C. For microwave digestion, soil [ Soil ] Ex aliquots were crushed with a mortar and pestle to reduce x100 the particle size prior to digestion. [ Soil ] T

1251 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.8 Statistical analysis If present in leaves, As concentrations were below the Significance of plant-soil relationships and soil com- instrument detection limit and is therefore omitted from the petition effects were established by obtaining Pearson’s discussion. Calcium was one of the major elements found -1 product-moment correlation coefficients for soil parame- in the leaves. Total soil Ca at site C was high (31940 µg g ) -1 -1 ters (exchangeable concentration of elements, pH, SOM with 1643 µg g being exchangeable and 21284 µg g being and CEC) and concentration of elements in plant. A com- taken up by the plant (BF = 13). Total soil Ca at site H was -1 -1 plex inter-item correlation matrix was obtained and the in- low (1479 µg g ) but high in the leaves (17897 µg g ) which formation extracted from the correlation matrix was used showed the plants’ tendency to accumulate this element (BF to discuss the positive and negative correlations that exist = 42). At most sites where total or exchangeable soil Ca amongst the soil and plant parameters. To understand the was low, Ca in the plant was found to be high thereby pro- correlation analysis, the significant relationships were de- ducing high BFs. The BFs suggest that when the soil con- scribed based on the correlation coefficients. Interactions centration of Ca is below the physiological requirement were considered positive if the correlation coefficient was level of the plant, it tends to accumulate the element until greater than or equal to 0.6 and negative if the correlation the required level is reached. Calcium in the leaves ranged -1 -1 coefficient was less than or equal to -0.6. A strongly syner- from 11014 µg g to 21284 µg g . gistic interaction produced a correlation coefficient greater Cobalt concentrations in both the leaves and soil were than or equal to 0.8 and a strongly antagonistic interaction low with concentrations in leaves ranging from 0.2 – 1.2 produced a correlation coefficient greater than or equal to µg g-1. Chromium is a metal not generally considered of -0.8. The impact of soil quality on elemental uptake by the high environmental risk as most Cr in soils occur as Cr(III) plant was determined based on these relationships that ex- which is generally strongly adsorbed by soils and regarded ist between soil and plant and control on uptake by the as relatively inert. The results showed that on average only plant was evaluated. All statistical analyses were done us- 2.3% of total soil Cr was exchangeable. Sites G and H ing the Statistical Package for the Social Sciences (PASW tended to exclude Cr with Cr in leaves being below the in- Statistics 19, IBM Corporation, Cornell, New York). strument detection limits. If detected Cr concentration in leaves were below 10.2 µg g-1.

3. RESULTS AND DISCUSSION Copper concentrations in the leaves were in the narrow range of 16 – 26 µg g-1. The plant appeared to control the 3.1 Quality Assurance uptake of Cu by accumulating it when exchangeable concentrations were low as in site D. This also holds true for The accuracy of the method for elemental analysis was Fe, where BFs were as high as 36 to make up for low ex- measured by comparing results obtained with certified re- changeable concentrations. For Mg, soil concentrations sults (Table 1). Values for Cu, Fe, Mn and Zn are certified (total and exchangeable) were lower than that in the leaves; and those for Ca and Mg are indicative since these were BFs were between 63 and 301. This showed the plants con- additional tests, to those required for certification, and were trol on uptake to meet its metabolic needs. determined by a few laboratories. Values for As were below the instrument detection limits. For the other elements, Similar to Ca and Co, the exchangeable % for Mn was measured values compared well with certified results. high. When exchangeable concentrations were high, their concentrations in leaves were high (site E) and when ex- 3.2 Elemental Distribution in Amadumbe and Soil Samples changeable concentrations were low, leaf concentrations from Different Sites were low (site H). This shows a positive relationship be- The data provided in Table 2 represents the concentra- tween soil and plant concentrations for Mn. For Ni, alt- tions of selected elements in soil (total and exchangeable) hough the plant tends to accumulate this metal, its concen- and plant (leaves and bulbs) as well as the exchangeable tration is restricted to a narrow range of 6.2 – 33.8 µg g-1. percentages and bioaccumulation factors (BFs) calculated. No observable trend was noticed for Pb. Six of the eight

TABLE 1 -Comparison of measured and certified/indicative values (Mean ± SD; at 95% confidence interval; n = 3), based on dry mass, in the certified reference material.

Element Wavelength (nm) Certified/Indicativea concentration Measured concentration As 193.70 23.0 ng·g-1 NDb Ca 317.94 0.41 mg·g-1 0.49 ± 0.09 mg·g-1 Cu 324.76 2.60 ± 0.1 µg·g-1 2.58 ± 0.99 µg·g-1 Fe 238.21 40.7 ± 2.3 µg·g-1 38.7 ± 8.97 µg·g-1 Mg 279.08 0.50 mg·g-1 0.47 ± 0.04 mg·g-1 Mn 259.37 20.3 ± 0.7 µg·g-1 18.9 ±2.19 µg·g-1 Zn 206.20 19.5 ± 0.5 µg·g-1 20.7 ± 3.21 µg·g-1 a Indicative values are those without uncertainties. b ND = Not detectable

1252 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 2 - Concentrations of elements in µg g-1 dry mass (mean (SD), n = 3) in soil (total (T) and exchangeable (Ex)) and plant (leaves and bulbs). Element Sitea Soil (T)b Soil (Ex)c (Ex%)f Leaves BFe Bulbs Ca A 24671(10) 624(5) 2.5 11014(22) 18 5656(7) B 25870(56) 2176(7) 8 12111(38) 6 3057(5) C 31940(101) 1643(9) 5 21284(45) 13 1628(6) D 11122(112) 344(6) 3 13062(57) 38 3283(6) E 20162(112) 700(6) 3 19684(58) 28 915(5) F 9474(60) 273(7) 3 11477(60) 42 935(12) G 2413(12) 752(9) 31 20014(30) 27 664(7) H 1479(34) 355(7) 24 17897(90) 50 1742(8) Co A 3.5(0.76) 0.2(0.08) 6 0.2(0.03) 1 0.68(0.02) B 10.0(0.97) 0.7(0.03) 7 0.7(0.02) 1 0.68(0.01) C 9.5(0.77) 3.6(0.07) 38 0.8(0.04) 0.2 0.63(0.03) D 2.4(0.56) 2.0(0.06) 84 0.9(0.02) 0.5 0.76(0.01) E 10.6(0.72) 3.9(0.03) 37 1.2(0.01) 0.3 0.42(0.02) F 4.6(0.77) 1.2(0.02) 27 NDd 0.90(0.04) G 10.3(0.007) 3.6(0.07) 35 0.3(0.02) 0.1 0.55(0.01) H 26.0(0.09) 0.8(0.04) 3 0.3(0.01) 0.4 0.30(0.01) Cr A 75(0.76) 3.5(0.01) 5 5.2(0.01) 1 3.14(0.08) B 135(0.45) 0.5(0.01) 0.4 5.9(0.06) 12 4.79(0.05) C 88(20.3) 0.23(0.06) 0.3 2.2(0.01) 10 3.57(0.04) D 35(0.34) 2.7(0.78) 8 8.4(0.02) 3 2.48(0.03) E 75(0.54) 0.9(0.03) 1 10.1(0.05) 11 4.56(0.02) F 54(0.003) 0.93(0.01) 2 5.3(0.07) 6 1.98(0.02) G 85(3.34) 0.66(0.02) 1 ND 3.96(0.04) H 89(0.67) 0.97(0.08) 1 ND 2.18(0.05) Cu A 40(2) 2.2(0.58) 6 16(0.08) 7 6(0.01) B 51(2) 7.3(0.01) 14 19(0.34) 3 26(0.35) C 37(1) 7.0(0.04) 18 20(0.07) 3 10(0.56) D 24(1) 1.2(0.01) 5 20(0.56) 17 11(0.48) E 44(1) 3.7(0.05) 8 26(0.07) 7 6(0.74) F 30(2) 1.6(0.02) 5 21(0.80) 13 11(0.52) G 34(1) 2.6(0.03) 8 19(0.89) 7 12(0.54) H 33(1) 5.3(0.07) 16 24(0.6) 4 12(0.78) Fe A 21728(119) 132(5) 1 453(12) 3 202(1) B 30573(234) 244(6) 1 1288(20) 5 77(5) C 17306(101) 70(4) 0.4 435(10) 6 82(3) D 4169(85) 45(4) 1 1606(15) 36 101(5) E 11480(96) 47(3) 0.4 978(12) 21 53(2) F 8389(89) 37(2) 0.4 1121(13) 30 48(2) G 7822(90) 44(3) 0.6 278(6) 6 74(4) H 8286(85) 29(2) 0.4 240(5) 8 58(0.05) Mg A 1058(17) 38(0.3) 4 2368(0.7) 63 1539(12) B 1466(5) 29(0.09) 2 3479(0.7) 122 3067(45) C 3152(16) 29(0.7) 1 5303(0.6) 183 1430(7) D 424(16) 43(0.5) 10 3645(0.7) 86 2063(5) E 3861(13) 28(0.7) 1 8442(0.8) 301 1229(4) F 1970(18) 35(0.9) 2 6573(0.3) 191 1474(6) G 733(12) 44(0.5) 6 43702(0.7) 100 1733(12) H 3861(19) 52(0.6) 1 3714(0.8) 71 1708(12) Mn A 107(12) 31(0.5) 29 181(6) 6 24(4) B 121(11) 48(0.6) 39 202.6(9) 4 33(6) C 249(12) 110(0.5) 44 968(5) 9 60(6) D 73(0.7) 48(0.01) 66 236(11) 5 24(3) E 335(0.7) 147(0.8) 44 1299(34) 9 43(2) F 119(0.8) 30(0.7) 25 239(1) 8 20(1) G 387(0.3) 137(0.7) 35 425(2) 3 38(2) H 86(0.6) 19(0.07) 22 91(1) 5 7(1) Ni A 29(0.8) 0.18(0.09) 1 6.2(0.01) 34 2.9(0.01) B 57(0.3) 0.47(0.006) 1 10.5(0.50) 22 4.4(0.06) C 31(0.2) 2.57(0.007) 8 10.1(0.01) 4 2.5(0.01) D 6(0.01) 0.31(0.04) 5 14.3(0.06) 46 4.1(0.65) E 34(0.8) 1.51(0.09) 4 15.9(0.06) 11 2.1(0.01) F 14(0.8) 0.19(0.01) 1 7.5(0.71) 39 2.1(0.01) G 18(0.05) 0.69(0.07) 4 33.8(0.52) 49 2.3(0.01) H 20(0.7) 0.32(0.01) 2 6.7(0.01) 21 5.8(0.56) Pb A 135(0.78) 2.2(0.01) 2 7.1(0.84) 3.2 7.40(0.4) B 194(0.76) 8.0(0.78) 4 8.2(0.56) 1.0 1.70(0.6) C 91(0.08) 6.4(0.78) 7 4.7(0.89) 0.7 7.3(0.4) D 21(0.01) 0.7(0.06) 3 7.5(0.01) 10.7 1.7(0.01) E 847(0.08) 5.6(0.89) 7 9.3(0.89) 1.6 2.4(0.07) F 72(0.89) 2.9(0.09) 4 1.4(0.01) 0.48 2.4(0.01) G 630(0.33) 47.1(0.18) 39 1.3(0.01) 0.01 5.0(0.67) H 51(0.77) 5.7(0.07) 11 1.7(0.01) 0.30 1.3(0.01)

1253 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

Se A 5.7(0.66) ND - 6.7(0.46) 1.9(0.05) B 5.6(0.70) ND - 3.3(0.01) 0.9(0.08) C 5.2(0.01) ND - 4.9(0.57) 0.9(0.07) D 4.8(0.01) 1.7(0.01) 35 3.1(0.08) 2 1.1(0.07) E 5.2(0.67) ND - 3.7(0.04) 0.4(0.05) F 4.1(0.01) ND - 3.0(0.07) 0.4(0.05) G 4.0(0.01) 0.20(0.01) 5 4.0(0.01) 20 0.6(0.01) H 3.7(0.78) ND - 3.6(0.08) 0.6(0.02) Zn A 39(0.01) 4(0.01) 10 62(0.04) 16 122(0.05) B 111(0.53) 26(0.34) 24 69(0.01) 3 273(0.07) C 81(0.65) 13(0.08) 16 51(0.06) 4 170(0.05) D 18(0.78) 3(0.56) 14 49(0.01) 19 155(0.58) E 55(0.05) 8(0.65) 14 74(0.01) 10 76(0.07) F 61(0.87) 8(0.91) 13 63(0.01) 8 45(0.57) G 161(0.13) 21(0.04) 13 62(0.78) 3 133(0.56) H 39(0.09) 15(0.55) 37 144(0.76) 10 150(0.01) a Site: A = Tongaat, B = Claremont, C = Adams Mision, D = Emakholweni, E = Umlazi, F = Umgababa, G = Hibberdene and H = Margate; b c d e f T = Total; Ex = Exchangeable; ND = not determinable; Bioaccumulation factor, BF = [Leaf]/[Soil]Ex; Exchangeable %, Ex% = [Soil]Ex/[Soil]T X100.

TABLE 3 - pH, soil organic matter (SOM), and cation exchange capacity (CEC) of soil samples from each site (n = 3).

Sitea Soil pH SOM (%) CEC(meq/100 g) A 4.19 ± 0.04 4.85 ± 0.30 11.87 ± 0.03 B 4.05 ± 0.04 2.23 ± 0.04 3.99 ± 0.13 C 4.72 ± 0.18 3.06 ± 0.10 7.99 ± 0.08 D 6.82 ± 0.11 6.82 ± 0.61 12.05 ± 0.18 E 4.65 ± 0.04 6.48 ± 0.77 6.87 ± 0.03 F 4.69 ± 0.09 5.29 ± 0.56 9.77 ± 0.12 G 6.79 ± 0.12 4.30 ± 0.02 12.02 ± 0.03 H 4.56 ± 0.06 6.01 ± 0.03 4.97 ± 0.32 a Site: A = Tongaat, B = Claremont, C = Adams Mision, D = Emakholweni, E = Umlazi, F = Umgababa, G = Hibberdene and H = Margate.

sites had exchangeable Se concentrations below the instru- particles are more available for cation exchange within a ment detection limit, but Se was detected in leaves at all pH range of 6.5 to 8 as evident by Site D and G where the these sites indicating an accumulation of the element. The pH is in this range and CEC values are high. A positive accumulation of Zn to meet the physiological requirement relationship was observed between the CEC of the soil and levels of the plant was also observed. SOM. In most cases, as the CEC increased so too did the SOM, although not to the extent of the CEC. An increase With the exception of Zn, concentrations of the studied in SOM would increase the soils CEC since mineral cations elements at most sites were higher in the leaves than the adsorb to the negative surface charges of organic soil par- bulbs. Generally, Zn concentrations were higher in the ticles. To understand this further, correlation analysis will bulbs than leaves. A previous study also found high con- be done on this data. centrations of Zn in root crops [9]. A possible explanation for this could be that Zn is the only metal to be represented 3.3 Estimated Contribution of Amadumbe Leaves to the Diet in six enzyme classes in the plant (oxidoreductases, trans- Table 4 compares the estimated contribution of 10 g of ferases, hydrolases, lyases, isomerases and ligases) and Amadumbe leaves (the average serving size) to the RDA. functions as an enzyme activator in carbohydrate metabo- About 10 g of leaves contributes more than 12%, 23%, lism and protein formation, which the bulb is rich in [10]. 44%, 15% and 7% towards the RDA for Ca, Cu, Fe, Mg, The elements in the leaves were found to be in decreasing and Zn, respectively in most adults. Chromium is consid- order of Ca > Mg > Fe > Mn > Zn > Cu > Ni > Pb > Se=Cr ered to be an important microelement for normal carbohy- > Co. drate, lipid and protein metabolism in humans [11]. The Table 3 compares pH, SOM and CEC of soils from the amount of Cr exceeds the RDA for this element. However, eight different sites. The results show that pH of the soil high intake of Cr is not linked to adverse health effects; the ranged from 4.05 to 6.82. This indicates that the Institute of Medicine has not established a tolerable upper Amadumbe plant thrives in slightly acidic soil conditions. intake level (UL) for this element [12]. Amadumbe leaves The SOM ranged from 2.23 to 6.82% and CEC ranged be- appear to be a rich source of Se. Selenium is an essential tween 3.99 and 12.05 meq/100g. Site D had the highest element in the human diet, being a vital component of a CEC and SOM values whilst site B had the lowest. Cations number of functional seleno-proteins, and low Se status has like Ca and Mg that are usually bound to the surface of soil been associated with a wide range of diseases since it is

1254 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

needed for proper functioning of the immune system, for Cr. These interactions in soil, whether synergistic or antag- counteracting the development of virulence and inhibiting onistic, can alter the balanced supply of nutrients to plants, the human immunodeficiency virus (HIV) progression to thus affecting their nutrition [17]. The interactions noticed acquired immune deficiency syndrome (AIDS) [13]. Con- in the plants are complex in nature and linked to the phys- sumption of Amadumbe will make valuable contributions iological needs of the plants. Same two elements may yield to human health especially in South Africa, the country be- synergistic relationships in one plant while they could be lieved to have more people infected with HIV/AIDS than antagonistic in another [16]. Observed synergies could be any other country. The average concentration of Mn ex- due to an element blocking the uptake of another element ceeds the RDA for this element but not the UL. It is advis- which might be competing with the element in question. able for one not to consume too much of Amadumbe leaves An antagonistic relationship occurs in plants when the to avoid toxic effects associated with high Mn intake. plant takes up two different elements by the same mecha- Leafy green vegetables have been shown to contain one of nism [18]. Generally, plant roots favour the uptake of the the highest amounts of Mn in the normal adult male diet as metal found in higher concentrations in the soil [19]. The reported by the Total Diet Study [14]; this study confirms greatest number of antagonistic reactions have been ob- this finding. The maximum limit for metals in vegetables served for Fe, Mn, Cu, and Zn which are, obviously, the -1 set by the Department of Health, South Africa is 1 µg g key elements in plant physiology [16]. The only notable -1 for As and 0.5 µg g for Pb [15]. Amadumbe leaves exhib- antagonistic relationship observed in this study is between ited safe levels of these toxic metals. Cr and Ca (-0.6) where an increase in the availability of Cr reduced the uptake of Ca. Negative correlations between 3.4 Statistical Analysis Cr in the leaves with exchangeable Mg, Pb and Zn were Concentrations of elements in leaves were, to varying also noted which could explain, in part, the exclusion of Cr degrees, correlated with total and exchangeable concentra- from leaves at sites G and H (Table 2). Synergism in the tions in soil (Table 5). Metals in soil interact with each plant occurs when an increase in the concentration of one other and their availability can be influenced by the pres- element in the plant increases the concentration of another ence of other metals in soil or even by the effect of some element in the plant. Synergistic relationships were ob- chemical properties of the soil [16]. The interactions can served between Co and the following metals (Pb (0.7), Mn either be synergistic which is characterized by a mutual ef- (0.7) and Cr (0.8)); between Ca and Mn (0.7); between Cu fect of the interacting elements on each other or antagonis- and Mg (0.7) and between Cr and Pb (1). tic, where, an increase in the concentration of one element A strong correlation between an exchangeable cation decreases the concentration of the other. in the soil with its concentration in the leaves would indi- Synergistic relationships in soil were observed be- cate that soil concentrations influence uptake by the plant tween Ca and Cu (0.8), Fe (0.8) and Zn (0.7). A three-way for that element. In this study, with the exception of Mn synergy was observed between Co, Mn and Ni i.e. Co was (0.7) no strong correlations were observed between ex- synergistic with Mn (0.9) and Ni (0.7) and Mn and Ni were changeable levels in the soil with their corresponding lev- synergistic with each other (0.7). The interaction observed els in the leaves, suggesting good inherent controls by the between Zn and Cu (0.7) was synergistic but with Cr (-0.7), plant on elemental uptake to meet physiological require- was antagonistic. This could be due to the similarity in ment levels and to protect cells from toxic effects. The up- charge and size of Zn and Cu as opposed to that of Zn and take mechanisms could be physiologically controlled and

TABLE 4 - Dietary Reference Intake (DRIs), Recommended Dietary Allowance (RDA) and Tolerable Upper Intake Levels (ULs) of elements for most individualsa and average concentration of elements (n = 3) in leaves of Amadumbe.

Element Average Concentration DRI (mg/day) Estimated Contribution to (mg/10 g dry mass) RDA (%) RDA UL Ca 158.2 1000-1300 2500 12 Co 0.005 Cr 0.064 0.024-0.035 ND 183 Cu 0.205 0.9 8 23 Fe 8.00 8-18 45 44 Mg 47.7 310-320 350 15 Mn 4.55 1.6-2.3 11 198 Ni 0.13 ND 1.0 NDb Pb 0.052 Se 0.040 0.055 0.4 73 Zn 0.717 8-11 34 7 a Institute of Medicine of the National Academies: Dietary Reference Intakes (2001). b ND = not determined due to lack of data.

1255 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 5 - Correlation matrix for concentrations of elements in Soil (Exchangeable (E)) and Amadumbe leaves (L).

CaE CaL CoE CoL CrE CrL CuE CuL FeE FeL MgE MgL MnE MnL NiE NiL PbE PbL SeL ZnE ZnL pH SOM CaEa CaLb 0.1 CoE 0 0.8 CoL 0.3 0.4 0.6 CrE -0.5 -0.6 -0.4 -0.1 CrL 0.1 -0.2 0.2 0.8 0.4 CuE 0.8 0.4 0 0.3 -0.6 -0.1 CuL -0.3 0.5 0.4 0.4 -0.5 0.2 0.1 FeE 0.8 -0.5 -0.5 0.1 0 0.3 0.5 -0.5 FeL 0.1 -0.5 -0.1 0.4 0.2 0.6 -0.2 0 0.2 MgE -0.6 0 -0.3 -0.5 0.3 -0.4 -0.3 0 -0.5 -0.4 MgL -0.1 0.4 0.6 0.4 -0.5 0.2 0 0.7 -0.4 0.2 -0.5 MnE 0.2 0.8 0.9 0.5 -0.4 0.2 0.1 0.3 -0.2 -0.2 -0.4 0.6 MnL 0.2 0.7 0.8 0.7 -0.4 0.4 0.3 0.5 -0.2 -0.1 -0.6 0.7 0.8 NiE 0.5 0.8 0.7 0.6 -0.5 0.1 0.5 0.3 -0.1 -0.2 -0.5 0.5 0.7 0.9 NiL 0 0.5 0.7 0.1 -0.3 -0.1 -0.2 0 -0.2 -0.2 0.1 0.1 0.7 0.2 0.1 PbE 0 0.4 0.4 -0.2 -0.2 -0.4 -0.2 -0.1 -0.2 -0.4 0.3 -0.1 0.5 0 0 0.9 PbL 0.4 -0.2 0 0.7 0.3 1 0.1 0 0.5 0.5 -0.6 0 0.1 0.3 0.2 -0.2 -0.5 SeL 0.1 0 -0.2 -0.2 0.5 0 0 -0.6 0.2 -0.6 -0.1 -0.4 0 0.1 0.2 -0.2 0 0.2 ZnE 0.7 0.2 0 -0.1 -0.7 -0.4 0.7 0 0.5 -0.2 -0.1 -0.2 0.2 -0.1 0.1 0.3 0.5 -0.2 -0.2 ZnL -0.2 0.2 -0.4 -0.2 -0.2 -0.3 0.3 0.5 -0.2 -0.4 0.6 -0.1 -0.3 -0.3 -0.3 -0.3 -0.1 -0.3 -0.2 0.2 pH -0.4 0.2 0.5 0.1 0.1 0 -0.5 0 -0.5 0.1 0.4 -0.1 0.3 -0.1 -0.1 0.7 0.6 -0.2 -0.3 -0.1 -0.3 SOMc -0.9 0 0.1 0.1 0.4 0.3 -0.7 0.5 -0.7 0.2 0.5 0.3 -0.1 0 -0.3 0 -0.2 0 -0.2 -0.7 0.3 0.3 CECd -0.5 -0.2 0.2 -0.3 0.6 0 -0.8 -0.5 -0.4 0 0.2 -0.2 0.1 -0.1 -0.2 0.4 0.4 -0.2 0.3 -0.5 -0.6 0.6 0.3 a b c d XE – [X]Exchangeable where X=the various elements; XL – [X]Leaves where X=the various elements; SOM – Soil organic matter; CEC – Cation exchange capacity.

the membrane transporters could transport heavy metals in ACKNOWLEDGEMENTS the plant [20]. The strong antagonistic relationship observed between SOM and cations like Ca, Cu, Fe and Zn is The authors are thankful to UKZN for financial sup- expected as increased organic carbon binds cations port. strongly to soil. The authors have declared no conflict of interest.

4. CONCLUSIONS REFERENCES The elemental distribution in Amadumbe leaves and bulbs from eight sites was investigated. The concentration [1] Reddy M, Moodley R, Kindness A, Jonnalagadda SB. (2011). of the elements in Amadumbe leaves was found to be in Impact of soil quality on the elemental uptake and distribution the decreasing order of Ca > Mg > Fe > Mn > Zn > Cu > of Colocasia esculenta (Amadumbe), an edible root. Journal of Environmental Science and Health Part B, 46: 247–256. Ni > Pb > Se ≈ Cr > Co. Except for Zn, concentrations of the studied elements were higher in the leaves than the [2] Ewell PT, Matuura J. (1991). Tropical root crops in a develop- th bulbs. Although statistical analysis indicated some correla- ing economy. Proceedings, 9 Symposium of the International Society for Tropical Root Crops: Accra, Ghana. pp 20–26. tion between metals in soil and uptake of certain elements, the Amadumbe plant accumulated and excluded essential [3] Huang CC, Chen WC, Wang CCR. (2007). Comparison of Taiwan paddy and upland-cultivated taro (Colocasia esculenta elements according to its physiological needs. Amadumbe L.) cultivars for nutritive values. Food Chemistry, 102: 250– leaves were found to be a rich source of Ca and Mg and 256. essential elements that conform to required RDAs. [4] Brown AC, Valiere A. (2004). Probiotics and medical nutrition therapy. Nutrition in Clinical Care, 7: 56–68.

[5] Otari KV, Bhalsing GV, Shete RV, Nandgude TD, Mali VR, Bodhankar SL. (2012). Antihypertensive and diuretic effects of the aqueous extract of Colocasia esculenta L. leaves in experimental paradigms. Iranian Journal of Pharmaceutical Re- search, 11: 621–634.

1256 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[6] Moodley R, Kindness A, Jonnalagadda SB. (2007). Chemical composition of edible Macadamia nuts (Macadamia integrifo- lia) and impact of soil quality. Journal of Environmental Sci- ence and Health Part A, 42: 2097–2104.

[7] Walkley A, Black IA. (1934). An examination of Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. Soil Sci- ence, 37: 29–38. [8] Chapman HD. (1965). Cation Exchange Capacity. In: Black CA (ed.), Methods of soil analysis Part 2– chemical and microbiological properties. American Society of Agronomy: Madison. pp 891–901.

[9] Mahlangeni N, Moodley R, Jonnalagadda SB. (2012). Soil nutrient content on elemental uptake and distribution in sweet potatoes. International Journal of Vegetable Science, 18: 245– 259.

[10] Auld, DS. (2001). Zinc coordination sphere in biochemical zinc sites. Biometals, 14: 271–313. [11] Kobla HV, Volpe SL. (2000). Chromium, exercise and body composition. Critical Reviews in Food Science and Nutrition, 40: 291–308.

[12] Institute of Medicine. (2001). Dietary reference intakes (DRIs): Tolerable upper intake levels, elements. National Academies Press, Food and Nutrition Board: Washington D.C.

[13] Chaturvedi UC, Shrivastava R, Upreti RK. (2004). Viral infec- tions and trace elements: A complex interaction. Current Sci-

ence, 87: 1536–1554.

[14] McCauley A, Jones C, Jacobsen J. (2003). Soil pH and organic matter. Available at www.landresources.montana.edu (accessed 27 September 2011).

[15] Department of Health. (1994). Foodstuffs, Cosmetics and dis- infectants Act (Act No. 54 of 1972). Government Gazette: South Africa GN. R. 1518. [16] Kabata-Pendias A, Andriano DC. (1995). Trace Metals. In J. E. Rechcigl (Ed.) Soil amendments and environmental quality.

CRC Press: Boca Raton, Florida. [17] Arnon, J. (1975). Mineral Nutrition of Maize. International Potash Institute: Bern-Worblaufers, Switzerland. [18] Kalavrouziotis IK, Koukoulakis PH, Robolas P, Papadopoulos AH, Pantazis V. (2008). Interrelationships of heavy metals, macro and micro nutrients and properties of a soil cultivated with Brassica oleracea var. Italica (broccoli) under the effect of treated municipal wastewater. Journal of Water, Air and Soil Pollution, 190: 309–321. Received: July 18, 2014 [19] Prasad MNV, Sajwan KS, Naidu R. (2006). Trace Elements in Revised: August 22, 2014 the Environment: Biogeochemistry, Biotechnology, Bioreme- Accepted: October 14, 2014 diation. CRC Press: Boca Raton, Florida. [20] Williams LE, Pittman JK, Hall JL. (2000). Emerging mechanisms for heavy metal transport in plants. Biochimica et Bio- CORRESPONDING AUTHOR physica Acta (BBA) - Biomembranes, 1465: 104–126. Prof. S.B. Jonnalagadda University of KwaZulu-Natal Private Bag X 54001

Durban 4000 SOUTH AFRICA

Phone: + 2731 260 7325.

Fax: + 2731 260 3091. E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1250 - 1257

1257 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

EFFECTS OF AZOXYSTROBIN AND FLUSILAZOLE ON GROWTH AND PROTEIN AMOUNT OF Scenedesmus acutus

Burcin Bedil, Gokce Kendirlioglu, Nur Agirman and A.Kadri Cetin*

Department of Biology, Faculty of Science, Firat University, 23119 Elazig, Turkey

ABSTRACT Although the instructions for the use of these chemicals are aimed at minimizing the risk of contamination to In this study, the effects of azoxystrobin and flusilazole aquatic environments, residuals of pesticides can be de- on algae were investigated using Scenedesmus acutus. The tected in water courses draining agricultural areas. Alt- green alga was exposed to different concentrations of hough pesticides are designed specifically to destroy un- azoxystrobin and flusilazole (1, 3, 6, 10 and 15 μl) in the wanted target organisms, their application may cause laboratory, maintained at 23 ± 1 °C and 16:8-h light:dark re- many diverse problems to non-target organisms like fish, gime. Cell numbers were determined daily and growth rates birds, and even human beings. The effects of pesticides on were calculated for a period of 4 days. The results showed the aquatic environment were frequently evaluated using toxic effects of azoxystrobin and flusilazole on S. acutus organisms, such as fish or water flea [1]. The effects of pes- on the second day (24 h). In general, the growth rate and ticides in the ecosystems do not remain restricted to target protein amount of S. acutus in control cultures treated with organisms but rather extend to non-target organisms, such azoxystrobin and flusilazole were suddenly decreasing be- as algae, which play an important role in the primary pro- tween 2-4 days. In contrast, in all control cultures, the growth duction of the aquatic ecosystems [2, 3]. Numerous studies rate and protein amount of S. acutus was always positive be- have been conducted to determine the harmful effects of tween 1-4 days. The results of this study indicate that sensitiv- these pollutants on living organisms in aquatic ecosystems ities to azoxystrobin and flusilazole of S. acutus began in the [4-8]. first day of inoculation, and in this time, the algae decreased The significance of phytoplankton species as primary in cultures treated with azoxystrobin and flusilazole. producers as well as their ability to intrinsically alter the balance of the aquatic ecosystem has warranted greater concern, due to the toxic effects of widely accepted pesti- KEYWORDS: Scenedesmus acutus, azoxystrobin, flusilazole, protein amount cides. Algae are essential components of aquatic ecosystems. They produce oxygen and organic substances on

which most other life forms depend to provide food for 1. INTRODUCTION other organisms, including fish and invertebrates. Chemi- cal effects on algae can directly affect the structure and It is a known fact that, chemical pollution is one of the function of an ecosystem, resulting in oxygen depletion, most serious environmental problems. The increasing pol- decreased primary productivity, increased surface runoff lution of aquatic systems by chemicals has raised public and soil erosion, and degradation. The effects of pesticides concerns. Generally, contamination of water-bodies very on growth, photosynthesis, survival, reproduction, mem- often appears to be the unavoidable consequence of agri- brane permeability and other metabolic activities of algae cultural activities, mainly related to the use of plant protec- were studied by different authors. tion products. The use of pesticides has increased substan- Scenedesmus acutus is a freshwater alga found in tially in the last four decades, and has contributed to both many aquatic habitats. The organism has a fast growth and increased crop yields and decreased production costs. We can be cultured easily and economically under laboratory face the task of understanding the effects of nearly 5000 reg- conditions. Scenedesmus acutus has been successfully ap- istered contaminants in Turkey, including hundreds of pesti- plied for the bioassessment of various water pollutants like cide active ingredients that come in thousands of commer- heavy metals, pesticides and other biologically active com- cial formulations. Moreover, the effects of contaminants on pounds. communities depend on many factors including the concen- Azoxystrobin and flusilazole are commonly used pes- tration of the contaminant, the timing of the exposure, and ticides for pest control in agricultural fields. The objective the number of exposures. of this study was to determine the effects of azoxystrobin

and flusilazole on growth and protein amount of Scenedes- * Corresponding author mus acutus.

1258 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2. MATERIALS AND METHODS experimental cultures were grown in the same liquid medium at 23± 1 ºC and a light intensity of 2000 lux, for a 16-h light 2.1 Pesticide background and 8-h dark photoperiod. The inocula were prepared from Azoxystrobin used in this study is a broad spectrum these cultures to provide an initial cell density of 1.145.000 fungicide of the class of synthetic compounds called β- individual/ml of S. acutus in treated and control culture. methoxyacrylates. Azoxystrobin can be applied as a foliar Azoxystrobin was added to the S. acutus culture just after spray or a soil-treatment prior to planting, as it has a resid- inoculation. Flusilazole was also added separately to other ual effect. It acts as a systemic fungicide which has cura- cultures. Control cultures were incubated in the same me- tive, translaminar and preventative action. Azoxystrobin is dium without pesticides. The experimental sets were run in approved for use on a wide variety of crops in Turkey. triplicate, and all cultures were hand-shaken twice daily. Af- Flusilazole is a systemic fungicide with protective and ter the cultures were incubated, S. acutus cells were counted curative action. Its resistance to wash-off, redistribution by with an inverted microscope at 0, 24, 48, 72 and 96 h (1, 2, rainfall and vapour phase activity are important compo- 3 and 4 days). nents in its biological activity. Flusilazole is a broad-spec- Total protein was measured by Lowry methods. Sam- trum, systemic, preventive and curative fungicide effective ple processing was carried out in the dark to prevent deg- against many pathogens (Ascomycetes, Basidiomycetes radation of the Folin reagent. The spectrophotometric ab- and Deuteromycetes). sorbance was converted to protein concentration using a calibration curve. 2.2 Cell culture and growth conditions Scenedesmus acutus, chosen as experimental organism, is a member of green algae which generate most of the 3. RESULTS AND DISCUSSION phytoplankton in fresh water. S. acutus was collected from a fishpond using plankton ladle and grown in Jaworski liquid The objective of this study was to determine the effects medium [9]. The media was composed of distilled water and of azoxystrobin and flusilazole on growth as well as protein the following chemical ingredients: Ca(NO3)2.4 H2O, amount of Scenedesmus acutus. The effect of the pesticides KH2PO4, MgSO4.7 H2O, NaHCO3, EDTA Fe/Na, EDTA was investigated when alga S. acutus had been exposed to Na2, H3BO3, MnCl2.4 H2O, (NH2)6Mo7O2H2O, NaNO3, various concentrations at 24, 48, 72 and 96 h of exposure. Na2HPO4, cyanocobalamin, thiamine and biotin. The Control cultures with the same cellular density as the culture medium was sterilized at 121 °C and 1 atm pres- treated ones were prepared, in order to determine the effect sure for 15 min. S. acutus was exposed to various concentra- of azoxystrobin and flusilazole on population growth of S. tions of azoxystrobin and flusilazole (1, 3, 6, 10 and 15 μl). acutus. Green algae were affected in a different manner Algal toxicity test of 96 h (4 days) of exposure was con- from each concentration of these pesticides. Fig. 1 shows ducted following the general design of Environmental Pro- the population growth of the cultures exposed to different tection Agency [10]. After the inoculation, the stock and concentrations of azoxystrobin.

4500000 4000000 3500000 Control 1µL 3000000 3µL 2500000 6µL 10µL 2000000 15µL 1500000

Individual/ml 1000000 500000 0 Inoculation 24 h48 h72 h96 h

Times (Hours)

FIGURE 1 - Effect of azoxystrobin to population growth of Scenedesmus acutus.

1259 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

4500000 4000000 Control 3500000 1µL 3000000 3µL 2500000 6µL 2000000 10µL

Individual/ml 1500000 15µL 1000000 500000 0 Inoculation 24 h48 h72 h96 h Times (Hours)

FIGURE 2 - Effect of flusilazole to population growth of Scenedesmus acutus.

The observation of toxic effects of these pesticides on At the second day, S. acutus was counted in control S. acutus indicated a significant decrease in population culture as 2.755.000 individuals/ml. During this day, a growth, with respect to the control in the treated cultures, non-significant decrease of S. acutus growth in 1 μg/L-1 at all assayed concentrations at the first day. S. acutus in treated cultures could be observed, when compared to the control culture was counted as 2.125.000 individuals/ml. In control (Fig. 2). The effects of flusilazole on S. acutus were treated cultures (1, 3, 6, 10 and 15 µg/L of azoxystrobin), noticed when growth ratios of some treated cultures de- S. acutus was counted as 2.125.000, 1.915.000, 1.635.000, creased significantly with respect to the control after 96 h 1.425.000 and 1.215.000 individuals/ml, respectively. of exposure. At concentrations of 15 µg/L flusilazole, the Growth rate of S. acutus decreased in treated cultures with growth of S. acutus was almost completely finished. 6, 10, and 15 µg/L azoxystrobin 3.1 Protein amount At the second day, S. acutus was counted in the control culture as 2.755.000 individiuals/ml. The effect of Effects of azoxystrobin on protein amount of Scenedes- azoxystrobin on population growth of S. acutus was noticed mus acutus are shown in Fig. 3. Changes in protein amount when growth of all treated cultures significantly decreased of S. acutus were investigated during a 4-days period. The with respect to the control at the second day of inoculation. effects of azoxystrobin were investigated when S. acutus S. acutus declined by 1.845.000, 1.635.000, 1.355.000, algae had been exposed to different concentrations of 1.144.000 and 935.000 individuals/ml at 1, 3, 6, 10 and azoxystrobin (1, 3, 6, 10 and 15 µg/L). The observation of 15 µg/L, respectively. However, at the third day, there toxic effects of this pesticide on S. acutus indicated a sig- were continued decreases of S. acutus growth in treated nificant decrease in protein amount, with respect to the cultures, but green algae were significantly increased in the control in the treated cultures, at all assayed concentrations control culture. At the fourth day, population growth of S. at the first day. acutus was significantly decreased in cultures treated with Protein amount in control culture was measured as azoxystrobin. 48.698 μg/ml, but in cultures treated with 1, 3, 6, 10 and 15 µg/L of azoxystrobin, protein was measurement as 39.592, Scenedesmus acutus was affected in a different manner 36.991, 27.885, 23.332 and 20.730 μg/ml, respectively. from each concentration of flusilazole. Figure 2 shows the population growth of the cultures exposed to different con- At the second day, protein amount in control culture centrations of flusilazole. The observation of toxic effects was determined as 51.950 μg/ml. Protein amount in cul- of flusilazole on S. acutus indicated a significant decrease tures exposed to azoxystrobin declined by 34.389, 31.137, in population growth, with respect to the control in the 24.633, 20.730 and 13.576 μg/ml at 1, 3, 6, 10 and 15 µg/L, treated cultures, at all assayed concentrations at the first respectively. At the third day, protein amount was meas- day. S. acutus in control culture was counted as 2.125.000 ured in the control culture as 56.503 μg/ml. However, at individuals/ml. S. acutus declined by 2.054.900, the third and the fourth days, a decrease of protein amount 1.775.000, 1.495.000, 1.214.950 and 865.000 individu- in cultures treated with azoxystrobin could be observed, als/ml at 1, 3, 6, 10 and 15 µg/L, respectively. when compared to the control (Fig. 3).

1260 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

60 Control 50 1 µl 3 µl 40 6 µl g/ml

μ 30 10 µl 20 15 µl 10

0 Inoculation 24 h48 h72 h96 h Times (Hours)

FIGURE 3 - Effect of azoxystrobin to protein amount of Scenedesmus acutus.

70 Control 60 1 µl 50 3 µl 40 6 µl

µg/ml 30 10 µl

20 15 µl

0 Inoculation 24 h48 h72 h96 h Times (Hours)

FIGURE 4 - Effect of flusilazole to protein amount of Scenedesmus acutus.

Flusilazole is a systemic fungicide with protective and at 1, 3, 6, 10 and 15 µg/L, respectively. The effects of flusi- curative action. Effects of flusilazole on protein amount of lazole on protein amount were noticed when protein amount Scenedesmus acutus in cultures are shown in Fig. 4. of some treated cultures decreased significantly, with respect The observation of toxic effects of flusilazole on S. to the control after 48 h of exposure. There was a significant acutus indicated a significant decrease in protein amount, inhibition of protein amount with respect to the control, in with respect to the control in the treated cultures, at all as- the cultures exposed to >6 µg/L after 48 h. At the third and sayed concentrations at the first day. Protein amount was de- fourth days, a decrease of protein amount in treated cul- termined as 48.698 μg/ml, and in cultures treated with 1, 3, tures with flusilazole could be observed, when compared 6, 10 and 15 μg/L flusilazole, S. acutus was determined as to the control (Fig. 4). 44.795, 40.243, 33.739, 17.478 and 12.925 μg/ml, respec- The effects of pesticides on reproduction, photosyn- tively. thesis, growth and other metabolic activities of algae were At the second day, protein amount in control culture was studied by different workers [11-13]. Investigations with determined as 51.950 μg/L. At the same day, protein amount different green algal species have shown that algae vary declined by 33.088, 27.234, 20.008, 9.673 and 8.373 μg/ml greatly in their response to chemicals. Differential sensitiv-

1261 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

ity of the green algae to the compounds could induce spe- [8] Assis, C.R.D., Amaral, I.P.G., Castro, P.F., Carvalho, L.B., cies shifts within communities [14]. The loss of a few, par- Bezerra, R.S. (2007) Effects of Dichlorvos on the acetylcholinesterase from tambaqui (Clossoma macropomum) brain. ticularly sensitive algal species from a community contain- Environmental Toxicology and Chemistry, 26, 7, 1451-1453. ing hundreds of species may not be considered to be sig- [9] Thompson, A.S., Rhodes, J.C., Pettman, I. (1988) Natural En- nificant, as long as the function of the community remains vironmental Research Council Culture Collection of algae and unchanged. Most of the work in mixed cultures of algae protozoa: catalogue of strains. Freshwater Biology Associa- showed that application of pesticides resulted in elimination, Ambleside, 164 pp. tion of sensitive species [2]. Sensitivity to toxicants is im- [10] USEPA. (2002) Short-term Methods for Estimating the portant in determining the suitability of a test for adoption Chronic Toxicity of Effluents and Receiving Waters to Fresh- into chemical regulations. The toxicities of some pesticides water Organisms, 4th edition. EPA821/R-02/013. to species of algae and toxicity data published for several [11] Ma, J., Lin, F., Wang, S., Xu, L. (2004) Acute toxicity assess- species of algae with other pesticides have shown that the ment of 20 herbicides to the green alga Scenedesmus quadri- variations in sensitivity may be considerable [3, 15]. In the cauda (Trup.) Breb. Bulletin of Environmental Contamination present study, different concentrations of azoxystrobin and and Toxicology, 72, 1164-1171. flusilazole decreased the population growth and protein [12] Ma, J., Wang, P., Huang, C., Lu, N., Quin, W., Wang, Y. amount of S. acutus. (2005) Toxicity of Organophosphorous Insecticides to Three Cyanobacterial and Five Green Algal Species. Bulletin of En- The results of this study indicate that sensitivities to vironmental Contamination and Toxicology, 75, 490-496. azoxystrobin and flusilazole of S. acutus began in the first [13] Ma, J., Wang, S., Wang, P., Ma, L., Chen, X., Xu, C. (2006) days of inoculation, and in this time, the algae decreased in Toxicity assessment of 40 herbicides to the green alga Raphi- cultures treated with azoxystrobin and flusilazole. Growth docelis subcapitata. Ecotoxicology and Environmental Safety, rates of S. acutus were always high between 0 to 4 days in 63, 456-462. control cultures whereas those in pesticide-treated cultures [14] Balton-Warberg, M., Coen, L.D., Weinstein, J.E. (2007) Acute continuously decreased at all concentrations between 1 to toxicity and acetylcholinesterase inhibition in grass shrimp 4 days. In contrast, in the control culture, the growth rate (Palaemonetes pugio) and oysters (Crassostrea virginica) exposed to the organophosphate dichlorvos: Laboratory and field and protein amount of S. acutus was always positive at in- studies. Archives of Environmental Contamination and Toxi- tervals of 1 to 4 days. In general, the growth rates and pro- cology, 52, 207-216. tein amounts of S. acutus were found to be negatively cor- [15] Katsumata, M., Koike, T., Nishikawa, M., Kazumura, K., related with high concentrations of azoxystrobin and flusi- Tsuchiya, H. (2006) Rapid ecotoxicological bioassay using de- lazole. layed fluorescence in the green alga Pseudokirchneriella subcapitata. Water Research, 40, 3393-3400. The authors have declared no conflict of interest.

REFERENCES

[1] Yeh, H.J., Chen, C.Y. (2006) Toxicity assessment of pesti-

cides to Pseudokirchneriella subcapitata under air-tight test environment. Journal of Hazardous Materials, A131, 6-12. [2] Ma, J., Liang, W., Xu, L., Wang, S., Wei, Y., Lu, J. (2001) Acute toxicity of 33 herbicides to the green alga Chlorella pyr- enoidosa. Bulletin of Environmental Contamination and Tox- icology, 66, 536-541.

[3] Cetin, A.K., Mert, N. (2005) The Effects of Paraquat on Pop- Received: July 11, 2014 ulation Growth of Scenedesmus acutus. Fresenius Environ- Accepted: September 18, 2014 mental Bulletin, 14, 7, 634-636.

[4] Cetin, A.K., Mert, N. (2006) Growth Rate of Scenedesmus acutus in Cultures Exposed to Trifluralin.Polish J. of Environ. CORRESPONDING AUTHOR Stud. 15, 4, 631-633. A. Kadri Cetin [5] Agirman, N., Kendirlioglu, G., Cetin, A.K. (2014) The Effects of Four Pesticides on the Growth of Chlorella vulgaris. Frese- Firat University nius Environmental Bulletin, 23, 6, 1418-1422. Science Faculty Department of Biology [6] Sabater, C., Cuesta, A., Carrasco, R. (2002) Effects of bensul- furon-methyl and cinosulfuron on growth of four freshwater 23119 Elazig species of phytoplankton. Chemosphere, 46, 953-960. TURKEY

[7] Gómez De Barreda, D., Sabater, C., Carrasco, J. M. (2004) Ef- fects of Propanil, Tebufenozide and Mefenacet on Growth of E-mail: [email protected] Four Freshwater Species of Phytoplankton: A Microplate Bi- oassay. Chemosphere, 56, 315-320. FEB/ Vol 24/ No 4/ 2015 – pages 1258 - 1262

1262 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

IN VITRO PLANTLET REGENERATION FROM NODAL SEGMENTS OF CREEPING JENNY (LYSIMACHIA NUMMULARIA L.) - A MEDICINAL AQUATIC PLANT

Mehmet Karataş* and Muhammad Aasim

Department of Biology, Kamil Ozdag Faculty of Science, Karamanoglu Mehmetbey University, Yunus Emre Campus, 70200, Karaman, Turkey

ABSTRACT ducers of water ecosystems that convert radiant energy into chemical energy in the presence of nutrients like P, N, Fe, Creeping jenny (Lysimachia nummularia L.) is an Mn, Mo and Zn [1]. Besides that, these plants are used for aquatic medicinal plant of Myrsinaceae family. The study phytoremediation [2, 3], bio-monitoring [4, 5] and as medic- presents the axillary shoot regeneration from shoot tip and inal plants [6, 7]. st nd 1 and 2 nodal segment explants cultured on agar solidi- Creeping jenny (Lysimachia nummularia L.) is an fied MS (Murashige and Skoog) medium containing 0.25- aquatic flowering plant of Myrsinaceae family, native to Eu- 1.25 mg/L BA (6-benzylaminopurine), with or without 0.25 rope. It is a perennial prostrate plant that can tolerate very mg/L NAA (α-naphthaleneacetic acid). Shoot regeneration low temperatures and can be used for planting in temperate st nd frequency of shoot tip, 1 and 2 nodal segment explants regions as horticultural plant. The plant can also grow as bog ranged between 58.33-83.33%, 33.33-83.33% and 41.67- garden or marginal aquatic plant [8]. The cultivated cultivar 91.67%, respectively, whereas shoots per explant of shoot Aurea has yellow leaves and is comparatively less aggressive st nd tip, 1 and 2 nodal segments were recorded to be 2.75- than the wild one. The plant is also one of the popular aquatic 9.30, 2.9-8.94 and 2.13-8.11, respectively. Maximum num- plants in the aquarium industry. ber of shoots from all explants were scored on MS medium supplemented with 1.25 mg/L BA. Contrarily, minimum L. nummularia (L.) is an important medicinal plant con- number of shoots per explants were scored on MS medium taining a number of phenolic acids [9] that can be used for with 1.25 mg/L BA plus 0.25 mg/L NAA. Shoot length treating healing wounds and stone lin syndrome. To date, a ranged between 1.10-2.12 cm, 1.04-2.25 cm and 0.96-2.33 cm lot of phenolic compounds has been isolated from different (Table 3) for shoot tip, 1st and 2nd nodal segment, respec- parts of the plants like flavonol glycosides from whole plant tively. Shoot length of all explants decreased with in- extract [10]. Recently, a new saponin (glycosylated triter- creased BA concentration, irrespective of NAA in the cul- pene 1) from underground parts of L. nummularia L. has ture medium. All explants generated plantlets (shoots with been isolated that showed significant activity against pros- rooting) directly on all regeneration media, irrespective of tate cancer cells (DU145 and PC3) without affecting normal explant type. Regenerated plantlets were successfully ac- cells. The compound also affected Glioblastoma cells and climatized in the aquariums without showing any negative showed moderate activity against melanoma cells [11]. sign on the growth of plants. Multiple shoot regeneration under in vitro conditions

has been employed for many aquatic plants with different objectives. However, there is still need of extensive work KEYWORDS: Aquatic, creeping jenny, nodal segments, shoot tip, medicinal on the in vitro shoot regeneration of aquatic plants in order to meet local requirements for the aquarium industry, or for isolation of secondary metabolites from medicinal aquatic 1. INTRODUCTION plants. Keeping this in view, the present study was designed to propagate L. nummularia (L.) plants under in Water is the basic requirement for the survival of all vitro conditions. living organisms, and all water-bodies on the earth com- prise aquatic plants, along with other living organisms. Aquatic plants, also known as hydrophytic plants or hydro- 2. MATERIALS AND METHODS phytes, are the group of plants that occupy different ecological niches in the aquatic environment, grow permanently in Plants of creeping jenny (L. nummularia L.) were pur- water or saturated soils. These plants are the primary pro- chased from the local aquarium of Karaman province, Tur- key. Twigs containing ≈4-6 nodes (shoot tip plus nodal * Corresponding author segments) with attached leaves were subjected to steriliza-

1263 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

tion. These twigs were washed under running tap water for 3. RESULTS AND DISCUSSION a few min, prior to sterilization. Thereafter, these twigs were surface-sterilized with 15 % (v/v) hydrogen peroxide Isolation of secondary metabolites from medicinal (H2O2) for 10 min, followed by rinsing thrice with steri- plants, followed by screening of metabolites as antimicro- lized distilled water for 5 min, each by continuous stirring. bial, antioxidant, anticancer agents etc., is gaining popular- After sterilization, leaves were detached from the twigs and ity all over the world. Aquatic plants are generally consid- explants (shoot tip, 1st and 2nd nodal segment) were isolated ered as ornamental plants in aquariums or backyard gar- under aseptic conditions. All explants were placed individ- dens, or invasive weeds in the water-bodies. However, ually on culture tubes supplemented with MS [12] medium some of them are also used as medicinal plants. These without any growth variants, in order to screen out contam- plants are normally vegetative-propagated, possibly limit- ination-free explants for 2 weeks. After two weeks , sterilized ing the required number of plants used for metabolite iso- shoot tip, 1st and 2nd nodal segment explants were cultured on lation. Besides that, contamination of water-bodies with 0.65 % agar solidified MS medium supplemented with 3% different pollutants and heavy metals may also affect the sucrose and 0.25-1.25 mg/L BA (6-Benzylaminopurine), isolation of purified metabolites. Therefore, plant tissue with or without 0.25 mg/L NAA (1-Naphthaleneacetic culture provides an alternate and rapid way to propagate Acid) for shoot regeneration. After 8 weeks on culture me- plants under in vitro conditions. Keeping this in view, the dium, data pertaining shoot regeneration frequency, shoots present study was designed to propagate creeping jenny (L. per explant and mean shoot length were taken and sub- nummularia L.) plant using shoot tip and 1st and 2nd nodal jected to statistical analysis. Regenerated shoots were di- segment explants cultured on MS medium supplemented rectly transferred to aquariums provided with tap water and with different concentrations of BA-NAA, followed by ac- sand for acclimatization. climatization in the aquariums. The experiment was run in triplicate, and the pH of Shoot tip and nodal segment explants are commonly all culture media was adjusted to ≈5.8 before autoclaving used explants for axillary shoot regeneration of aquatic (118 kPa atmospheric pressure, 121 °C for 20 min). All plants like Mentha viridis [13], Steviare baudiana [14], Vi- cultures were incubated under a 16-h light photoperiod tex negundo [15], Marsdenia brunoniana [16], and Cera- (5000 lux) using white fluorescent lamps; 8 explants were tophyllum demersum [17], using different growth variants. used in each treatment and experiment were repeated Our results revealed that shoot tip explant responded well twice. Statistical analysis of the data was performed as to growth variants resulting in earlier shoot buds initiation 1-way ANOVA using SPSS 17 for Windows whereas post compared to 1st and 2nd nodal segment explant, with clear hoc test was performed using Duncan`s test. Before statis- shoot regeneration started within 2 weeks, in line with the tical analysis, all data in the experiment were subjected to results of Karatas et al. [17] in C. demersum. arcsine transformation [13].

‘ FIGURE 1 - In vitro shoot regeneration and acclimatization of creeping jenny (L. nummularia): (a) Shoot regeneration from shoot tip, (b) 1st and (c) 2nd nodal segment explant, (d) rooted plantlets, and (e) acclimatized plants in aquarium.

1264 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

TABLE 1 - Effects of different BA-NAA concentrations on shoot regeneration frequency of creeping jenny (L. nummularia).

BA NAA Shoot Tip 1st nodal 2nd nodal segment segment 0.25 - 75.00ns 66.67ab 50.00c 0.50 - 66.67 58.33ab 58.33bc 0.75 - 83.33 41.67ab 83.33ab 1.00 - 58.33 83.33a 75.00abc 1.25 - 66.67 75.00ab 91.67a 0.25 0.25 66.67 83.33a 75.00abc 0.50 0.25 75.00 75.00ab 66.67bc 0.75 0.25 58.33 33.33b 58.33bc 1.00 0.25 66.67 50.00ab 41.67d 1.25 0.25 50.00 58.33ab 50.00c Means followed by different small letters within columns are significantly different using Duncan p<0.01.

TABLE 2 - Effects of different BA-NAA concentrations on shoot per explants of creeping jenny (L. nummularia).

BA NAA Shoot Tip 1st nodal 2nd nodal segment segment 0.25 - 3.50efg 2.92h 3.25f 0.50 - 4.00def 4.04d 3.45e 0.75 - 4.60cd 4.16d 4.66d 1.00 - 5.53bc 4.38c 6.04b 1.25 - 9.30a 8.94a 8.11a 0.25 0.25 6.28b 3.38f 5.16c 0.50 0.25 4.53cde 4.59b 3.39ef 0.75 0.25 3.75defg 3.53e 2.63g 1.00 0.25 3.40fg 3.21g 2.17h 1.25 0.25 2.75g 2.94h 2.13h Means followed by different small letters within columns are significantly different using Duncan p<0.01.

Earlier shoot regeneration from shoot tip might be due NAA) concentration in the culture medium. Contrarily, to the presence of relatively young and more actively di- Gnanaraj et al. [19] reported increased shoot regeneration viding cells [18] compared to other nodal segments ex- frequency with increased BA concentration in Alternanthera plants. However, after 4 weeks of culture, multiple shoot sessili whereas results of Hassan and Roy [20] showed de- regeneration was observed on all explants with no callus creased shoot regeneration frequency from shoot apex and induction on all explants, irrespective of presence of NAA nodal segments of Gloriosa superba L. Presence of NAA in the culture medium, along with different concentrations in the culture medium also exerted variable response to of BA. These results are in line with findings of Karatas et shoot regeneration frequency of each explant. Contrarily, al. [17], who also reported no callus induction from shoot Karatas et al. [21] reported 100% shoot regeneration fre- tip and nodal segment explants of C. demersum. After 8 quency of different internodal segments and leaf explant of weeks of culture (Figs. 1 a, b, c), data regarding shoot re- Bacopa monnieri. Similarly, Karatas et al. [22] reported generation frequency, shoots per explant and shoot length 100% shoot regeneration frequency when culturing leaf ex- were subjected to statistical analysis. plants on MS medium containing IBA (indole-3-butyric acid) with Kinetin or TDZ in dwarf hygro (Hygrophila Analysis of variance results showed the clear bearings polysperma). of BA-NAA on shoot regeneration of different explants used in the study. Results showed that BA-NAA had no Results on shoots per explants clearly revealed the ef- effects on shoot regeneration frequency of shoot tip ex- fects of BA-NAA concentration on all explants, and were plants ranging between 58.33-83.33 % (Table 1). Contra- found to be statistically significant (p<0.01). Shoots per rily, shoot regeneration frequency of 1st and 2nd nodal seg- explant of shoot tip, 1st and 2nd nodal segments were rec- ments were found to be statistically significant (p<0.01) to orded as 2.75-9.30, 2.9-8.94 and 2.13-8.11, respectively BA-NAA, and were recorded as 33.33-83.33% and 41.67- (Table 2). Shoots per explants of all explants showed in- 91.67%, respectively. (Table 1). These results are contra- creasing pattern with increase in BA concentration, in line dictory to the findings of Karatas et al. [17], who reported with Jo et al. [23] in Alocasia amazonica and Karatas et al. insignificant effects of different BA concentrations on [17] in C. demersum. Contrarily, negative effects of in- shoot regeneration frequency of shoot tip, 1st and 2nd nodal creased BA concentartion have been reported in B. mon- segments cultured on agar-solidified medium in C. demer- nieri [24]. Maximum number of shoots from all explants sum, which might be due to the presence of NAA in the were scored on MS medium supplemented with 1.25 mg/L culture medium in this study. Results further illustrated the BA, revealing the need of higher BA concentration irre- variable response of each explant to growth variant (BA- spective of explant type in line with Karatas et al. [17] in

1265 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

C. demersum. However, other studies revealed the require- type or NAA in the culture medium. The results are con- ment of low BA concentration for maximum shoots per ex- tradictory to the findings of Karatas et al. [17] who gained plants as Gnanaraj et al. [19] obtained a maximum number stunted shoots with increase of BA concentration in C. de- of shoots on 0.20 mg/L BA from shoot meristem explants mersum. Positive effects of increased BA concentration on of A. sessilis. Contrarily, lowest number of shoots per ex- shoot length have been reported by Vijaykumar et al. [31]. plants of all explants were obtained from MS medium sup- Results also showed that presence of 0.25 mg/L NAA with plemented with 1.25 mg/L BA plus 0.25 mg/L NAA. Addi- 0.25 and 0.50 mg/L BA in the culture medium exerted pos- tion of NAA in the culture medium exerted variable effects itive effects on shoot length of shoot tip and 2nd nodal seg- on shoots per explants. 0.25 mg/L NAA with 0.25 and ment explant, compared to BA used singly, in line with 0.50 mg/L BA resulted in an increased number of shoots per Karatas et al. [22]. Contrarily, NAA with all other concen- explants compared to the remaining concentrations (0.75- trations resulted in relatively shorter shoots compared to 1.25 mg/L BA) which resulted in sharp decline of shoots per BA used singly. Shorter shoots of all explants were rec- explants. This difference was more prominent at 1.25 mg/L orded on MS medium supplemented with 1.25 mg/L IBA BA plus 0.25 mg/L NAA which resulted in 3-fold less num- plus 0.25 mg/L NAA. bers of shoots per explants, compared to BA used singly Results also revealed direct plantlet regeneration as re- for all explants. Cytokinin + auxin combinations irrespec- generated shoots rooted directly (Fig. 1d) in the regenerative of type, concentrations and explant type have been re- tion medium, irrespective of explant type, BA concentra- ported to obtain maximum numbers of shoots per explant tion with or without NAA in the culture medium, in line in other aquatic plants like Ludwigia [25]), Leucopogon with Karatas et al. [32] in R. rotundifolia. Therefore, rooted verticillatus [26], H. auriculata [27], H. difformis [28], plantlets were transferred directly to the aquariums (Fig. 1e) Rotala macrandra [29], B. monnieri [21], and H. poly- containing tap water and sand. These plants adapted well and sperma [22]. Comparing explant types, shoot tip explants continued their growth in the aquariums without external generated more numbers of shoots compared to other nodal provision of oxygen or minerals. Successful establishment segments at all concentrations of BA-NAA used in the study, of in vitro regenerated aquatic plants in the aquariums is the in line with Çınar et al. [30]. This might be due to relatively prerequisite of a successful tissue culture protocol for clonal younger cells which divided actively at all combinations of propagation. Successful acclimatization of in vitro regener- BA-NAA [18, 21]. Contrarily, 1st nodal explant has been re- ated aquatic plants has been reported for Nymphoides indica ported to be the best for maximum numbers of shoots com- [33], B. monnieri [34], R. macrandra [29], Veronica ana- pared to shoot tip meristem, leaf, and petiole explants of H. gallisaquatica [35], A. sessilis [19], Cryptocoryne wendtii difformis [28]. and Cryptocoryne beckettii [36], H. polysperma [22, 30], C.

demersum [17], and R. rotundifolia [32]. Results of shoot length also showed the clear impact of growth variants (p<0.01) on all explants which ulti- The study presents the repeatable and efficient proto- mately resulted in statistically significant values. Shoot col for direct plantlet regeneration from different nodal ex- length ranged between 1.10-2.12 cm, 1.04-2.25 cm and plants, followed by successful acclimatization of creeping 0.96-2.33 cm (Table 3) for shoot tip, 1st and 2nd nodal seg- jenny. This protocol can be used for the application of bio- ment, respectively. Results of shoot length further showed technological tools in order to multiply the plants for me- similar patterns of response of explants to different growth dicinal purposes and as aquarium plants. This protocol can variants, and decreased gradually with increase of BA con- also provide the propagation of plants for the extraction of centration in the culture medium, irrespective of explant medicinally important compounds used for further studies.

TABLE 3 - Effects of different BA-NAA concentrations on shoot length of creeping jenny (L. nummularia).

BA NAA Shoot Tip 1st nodal 2nd nodal segment segment 0.25 - 2.02b 2.25a 2.10ab 0.50 - 1.76c 2.21a 1.80abc 0.75 - 1.74c 1.98b 1.85abc 1.00 - 1.55b 1.80c 1.60bc 1.25 - 1.46e 1.48e 1.45cd 0.25 0.25 2.06b 1.79c 2.33a 0.50 0.25 2.19a 1.65d 2.30a 0.75 0.25 1.74c 1.56de 1.41cd 1.00 0.25 1.19f 1.25f 1.35cd 1.25 0.25 1.10f 1.04g 0.96d

Means followed by different small letters within columns are significantly different using Duncan p<0.01.

1266 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

ACKNOWLEDGMENT [13] Raja, H.D. and Arockiasamy, D.I. (2008). In vitro Propagation of Mentha viridis L. from nodal and shoot tip explants. Plant Tissue Culture and Biotechnology, 18, 1-6. The authors acknowledge the financial assistance by the Karamanoglu Mehmetbey University through the Sci- [14] Janarthanam, B., Gopalakrishnan, M., Lakshmi Sai, G. and Sekar, T. (2009). Plant regeneration from leaf derived callus entific Research Project commission (BAP) for funding of Stevia rebaudiana Bertoni. Plant Tissue Culture and Bio- project number 50-M-12. technology, 19, 133-141. [15] Islam, S.A.M.N., Banik, H., Alam, S., Tarek, M. and Rahman, The authors have declared no conflict of interest. M. (2009). In vitro propagation of Holarrhena antidysenterica Wall., Wedelia chinensis (Osb.) Merr. and Woodfordia fruticosa (L.) Kurz. Plant Tissue Culture and Biotechnology,19, 253-255.

REFERENCES [16] Ugraiah, A., Karuppusamy, S. and Pullaiah, T. (2010). Micro- propagation of Marsdenia brunoniana Wight & Arn. A rare antidiabetic plant. Plant Tissue Culture and Biotechnology, 20,

[1] Oyedeji, A.A. and Abowei, J.F.N. (2012). The classification, 7-12. distribution, control and economic importance of aquatic plants. International Journal of Fisheries and Aquatic Sci- [17] Karataş, M., Aasim, M., Dogan, M. (2014). Multiple shoot re- ences, 1, 118-128. generation of Ceratophyllum demersum L. on agar solidified and liquid mediums. Fresenius Environmental Bulletin, 24: 3- [2] Hegazy, A.K., Abdel-Ghani, N.T. and El-Chaghaby, G.A. 9. (2011). Phytoremediation of industrial wastewater potentiality by Typha domingensis. International Journal of Environmental [18] Celiktas, N., Can, E., Hatipoglu, R. and Avci, S. (2006). So- Science and Technology, 8, 639-648. matic embryogenesis, callus production, and plantlet growth in sainfoin (Onobrychis viciifolia Scop.). New Zealand Journal [3] Farid, M., Irshad, M., Fawad, M., Ali Awan, Z., Egrinya Eneji, of Agricultural Research, 49, 383-388. A. and Aurangzeb, N. (2013). Effect of cyclic phytoremedia- [19] Gnanaraj, W.E., Marimuthu, J., Subramanian, K.M. and Nal- tion with different wetland plants on municipal wastewater. lyan, S. (2011). Micropropagation of Alternanthera sessilis International Journal of Phytoremediation, 16, 572-581. (L.) using shoot tip and nodal segments. Iranian Journal of Bi- [4] Nirmal Kumar, J.I., Soni, H. and Kumar, R.N. (2008). Evalu- otechnology, 9, 206-212. ation of biomonitoring approach to study lake contamination [20] Hassan, A.K.M.S. and Roy, S.K. (2005). Micropropagation of by accumulation of trace elements in selected aquatic macro- Gloriosa superba L. through high frequency shoot prolifera- phytes: A case study of kanewal community reserve, Gujarat, tion. Plant Tissue Culture & Biotechnology, 15, 67-74. India. Applied Ecology and Environmental Research, 6, 65- 76. [21] Karatas, M., Aasım, M., Dogan, M. and Khawar, K.M. (2013). Adventitious shoot regeneration of the medicinal aquatic plant [5] Skorbiłowicz, E. (2009). Aquatic plants as bioindicators of water hyssop (Bacopa monnieri L. Pennell) using different in- contamination of upper Narew river and some of its tributaries ternode. Archives of Biological Sciences, 65, 297-303. with heavy metals. Environmental Protection Engineering, 35, 65-77. [22] Karataş, M., Aasim, M., Çınar, A. and Dogan, M. (2013). Ad- ventitous shoot regeneration from leaf explant of dwarf hygro [6] Shanmugasundaram, P. and Venkataraman, S. (2005). Hepa- (Hygrophila polysperma (Roxb.) T. Anderson). The Scientific toprotective and antioxidant effects of Hygrophila auriculata World Journal, DOI:http://dx.doi.org/10.1155/2013/680425 (K. Schum) Heine Acanthaceae root extract. Department of [23] Jo, U.A., Murthy, H.N., Hahn, E.J. and Paek, K.Y. (2008). Mi- pharmacology and environmental toxicology, Dr. ALM Post cropropagation of Alocasia amazonica using semisolid and Graduate Institute of Basic Medical Sciences University of liquid cultures. In Vitro Cellular & Developmental Biology- Madras, Chennai, India. Plant, 44, 26-32. [7] Kshirsagar, A.D., Ingale, K.G., Vyawahare, N.S. and Thorve, [24] Sharma S, Kamal B, Rathi N, et al. (2010). In vitro rapid and V.S. (2010). Hygrophila spinosa: A comprehensive review. mass multiplication of highly valuable medicinal plant Bacopa Pharmacogonsy Reviews, 4, 167-171. monnieri (L.) Wettst. African Journal of Biotechnology, 9, [8] Brickell C. (2008). RHS A-Z encyclopedia of garden plants. 8318–8322. Dorling Kindersley, United Kingdom, p. 1136 [25] Öztürk, M. (2002). The effects of different rates of plant growth regulators on in vitro micropropagation of an aquarium [9] Luczak, S., Swiatek, L. and Daniewski, M. (1989). "Phenolic plant Ludwiga sp, M.S. Thesis, Ankara University, Ankara, acids in herbs Lysimachia nummularia L. And L. Vulgaris Turkey. L". Acta poloniae pharmaceutica, 46, 381-385. [26] Anthony, J. M., Senaratna, T, Dixon, K.W. and Sivasitham- [10] Yasukawa, K., Ogawa. H. and Takido, M. (1990). Two flavo- param, K. (2004). Somatic embryogenesis for mass propaga- nol glycosides from Lysimachia nummularia. Phytochemistry, tion of Ericaceae: a case study with Leucopogon verticillatus. 29, 1707-1708. Plant Cell, Tissue and Organ Culture, 76, 137-146. [11] Podolak, I., Koczurkiewicz, P., Michalik, M., Calanty, A., Zaj- [27] Panigrahi, J., Mishra, R.R. and Behera, M. (2006). In vitro del, P. and Janeczko, Z. (2013). A new cytotoxic triterpene multiplication of Asteracantha longifolia (L.) Nees: a medici- saponin from Lysimachia nummularia L. Carbohydrate Re- nal herb. Indian Journal of Biotechnology, 5, 562-564. search, 37, 16-20. [28] Öztürk M, (2008). In Vitro Propagation of Aquarium Plants [12] Murashige, T. and Skoog, F. (1962). A revised medium for Hygrophila difformis and Microsorium pteropus, M.S. thesis, rapid growth and bioassays with tobacco tissue cultures. Phys- Ankara University, Biotechnology Institute, Ankara, Turkey. iologia Plantarum, 15, 473-497. [29] Sumlu, S. (2009). In Vitro Micropropagation and Genetic [12] Snedecor, G.W. and Cochran, W.G. (1967). Statistical meth- Transformation of Aquatic Plant Rotala macrandra [M.S. the- ods. The Iowa State University Press, Iowa, USA. sis], Ankara University, Ankara, Tuekry.

1267 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[30] Çınar, A., Karataş, M. and Aasim, M. (2013). High frequency plant regeneration of dwarf hygro (Hygrophila polysperma [Roxb.] T. Anderson) on liquid culture. Journal of Applied Bi- ological Sciences, 7, 75-78.

[31] Vijayakumar, M., Vijayakumar, R. and Stephen, R. (2010). In vitro propagation of Bacopa monnieri L. - a multipurpose medicinal plant. Indian Journal of Science and Technology, 3, 781-786. [32] Karataş, M., Aasim, M. and Çiftçioğlu M. (2014). Adventi- tious shoot regeneration of Roundleaf toothcup-Rotala rotundifolia [(Buch-Ham. ex Roxb) Koehne]. Journal of Animal and Plant Sciences, 24, 838-842. [33] Jenks, M.A., Kane, M.E., Dennis, B. and McConnell, D.B. (2000). Shoot organogenesis from petiole explants in the aquatic plant Nymphoides indica. Plant Cell, Tissue and Organ Culture, 63, 1-8.

[34] Banerjee, M. and Shrivastava, S. (2008). An improved protocol for in vitro multiplication of Bacopa monnieri (L.). World Journal of Microbiology and Biotechnology, 24, 1355-1359.

[35] Shahzad, A., Parveen, S. and Fatema, M. (2011). Development of a regeneration system via nodal segment culture in Veronica anagallis-aquatica L.- An Amphibious medicinal plant. Jour- nal of Plant Interactions, 6, 61-68. [36] Stanly, C., Bhatt, A. and Keng, C.L. (2011). An efficient in vitro plantlet regeneration of Cryptocoryne wendtii and Cryp- tocoryne becketti through shoot tip culture. Acta Physiologiae Plantarum, 33, 619-624.

Received: July 22, 2014 Accepted: September 18, 2014

CORRESPONDING AUTHOR

Mehmet Karataş

Department of Biology

Kamil Ozdag Faculty of Science,

Karamanoglu Mehmetbey University

Yunus Emre Campus

70200 Karaman

TURKEY

Phone: +90 338 226 2151-3827

E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1263 - 1268

1268 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FLUORIDE REMOVAL FROM AQUEOUS SOLUTIONS USING MORINGA OLEIFERA SEED ASH AS AN ENVIRONMENTAL FRIENDLY AND CHEAP BIOSORBENT

Sina Dobaradaran1,2,3,*, Maryam Kakuee4, Iraj Nabipour5, Abdolrahim Pazira4, Mohammad Ali Zazouli6, Mozhgan Keshtkar2 and Maryam Khorsand4

1The Persian Gulf Marine Biotechnology Research Center, Bushehr University of Medical Sciences, Bushehr, Iran 2 Department of Environmental Health Engineering, Faculty of Health, Bushehr University of Medical Sciences, Bushehr, Iran 3Systems Environmental Health, Oil, Gas and Energy Research Center, Bushehr University of Medical Sciences, Bushehr, Iran 4 Islamic Azad University, Bushehr Branch, Iran 5The Persian Gulf Tropical Medicine and Infectious Research Center, Bushehr University of Medical Sciences, Bushehr, Iran 6Department of Environmental Health Engineering, Faculty of Health, Health Sciences Research Center, Mazandaran University of Medical Sciences, Sari, Iran

ABSTRACT fects in soft tissues [4-7]. With view to the special concerns of F, various studies have been done in relation to F concen- Adsorption is considered as the most promising treat- trations in drinking water, air, tea, fish, and sea as well as ment technology for fluoride (F) removal from aqueous so- in connection with its removal from high-F waters [6-23]. lutions. The aim of this study was to determine the effi- Different methods such as adsorption, precipitation, ion- ciency of seed Moringa oleifera ash in removal of F from exchange, electro-dialysis and electro-chemical, were de- aqueous solutions. After determining optimum pH (pH=7) veloped to remove extra F from water [24] that among and ashing temperature (650º C) by pretests, the adsorp- these processes, adsorption is a widely used method for F tion experiments were studied in batch systems at room removal of water [25]. Currently, considerable interests were temperature. The effects of experimental parameters such observed on the utilization of biosorbent materials for re- as adsorbent dose (0.8 – 64 g/l), contact time (5 -120 min) moval of different pollutants [26]. Biosorption advantages and initial F concentration (2 -8 mg/l) were studied. The over conventional treatment methods include low cost, less highest removal adsorption was at 64 g/l adsorbent, 10 min sludge production, high efficiency in dilute effluents, no nu- contact time and initial F concentration at 8 mg/l (81.14 % F trient needs, regeneration of biosorbent, and environmental removal). The results showed that the Moringa Oleifera ash favorable and economical viable [26-27]. Moringa oleifera can be used as an environmental friendly, cheap and ef- is grown in tropical areas and has been consumed as food in fective adsorbent from aqueous solutions. some African regions. The Coagulating demeanor of its seed

powder has been studied for different aspects of water treatment such as turbidity, alkalinity, total dissolved solids and hardness [28] but very small studies have been directed towards its sorption demeanor for the removal of pollution KEYWORDS: Moringa oleifera seed; biosorption; Fluoride removal; Environmentally friendly. from water. In the present study, the efficiency of Moringa oleifera seed as a biosorbent for F removal of aqueous solution was studied and the results can be useful in areas suffer-

ing from high F levels that mainly located in rural and remote areas. 1. INTRODUCTION

In the environment, fluoride (F) occurs naturally 2. MATERIALS AND METHODS through the earth’s crust and anthropogenic activities such as steel, semiconductor, electroplating, aluminum, bricks, 2.1 Adsorbent preparation glass, ceramic and fertilizers industry. These industries play After collection and transfer to laboratory, the an important role in increasing water pollution by F [1-3]. F Moringa oleifera seeds were washed by tap water and then can cause a wide range of adverse health effects including two times washed by distillated water to remove sand, clay, teeth and damaging and bone, as well as adverse health ef- and other impurities. The washed Moringa Oleifera seeds were heated in an oven at 650 ˚C, and finally ground and * Corresponding author sieved through a 0.71mm screen.

1269 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

2.2 Batch studies value of 7 has been selected as the optimum pH to perform A stock solution of 100 mg/L F was prepared by dis- all experiments. solving sodium fluoride (NaF) in ultrapure water. F solutions were prepared at 2, 3, 5 and 8 mg/L concentrations. 3.2 Effect of ashing temperature At each stage of the experiments, 100 ml of F solution with In order to determine the optimal activation tempera- a specific initial F concentration at neutral pH (pH=7) was ture for adsorbent, pretests of the experiment were per- agitated at 120 rpm. The effects of five contact times (5, formed at different temperatures (350, 450, 500, 550, 600, 10, 25, 60, and 120 min), four initial F concentration (2, 3, 650˚C), two adsorbent doses (1, and 2 g/L), three contact 5, and 8 mg/L), and various ratios of bio-sorbent to the F times (30, 45, and 60 min), and initial F concentration of levels (five ratios within the range of 400-8000) were in- 2 mg/L. The results showed that by increasing temperature, vestigated in the batch experiments. The standard the F removing efficiency will be increased so 650˚C tem- SPANDS method was used by using a spectrophotometer perature was selected as the ashing temperature to perform (Model CAM Spec M501) for analysis of the remaining F all experiments. concentration in the aqueous solution after each run. 3.3 Effect of adsorbent dose The effect of biomass dosage on the F removal by ad- 3. RESULTS AND DISCUSSION sorbent was studied using various biomass doses of Moringa oleifera seed in the range of 2– 40 g/L (Fig. 1). 3.1 Effect of pH The results showed that the adsorption efficiency depend The effect of pH on F adsorption in a range from 5 to on the increasing biomass dosage in the solution. The maxi- 11 (5, 7, 9, 11) was investigated as pretests. In the pretest mum adsorption of F was obtained at biomass dosage of experiments, adsorbent dose of 2 g/L, contact time of 35 min 30 g/L. This trend could be elucidated as a consequence of and initial F concentration of 2 mg/L were used to determine a partial compression of biomass at higher biomass con- the effect of pH. The results suggested a pH value of 7 as the centration, which results in a decrease in effective surface best pH value for removal of F by Moringa oleifera seed. area for the adsorption [31]. Decrease in F removal at lower pH value in this study Similarly, Jagtap et al. [32] found that F removal ca- (pH=5) may be due to the formation of the weakly ionised pacity distinctly increases from 19 to 81.98% with the in- HF (pKa = 3.2) and because of the competitiveness of the creasing adsorbent dosage from 0.2 to 1 g/L and after a OH− and F− ions at higher pH values (pH values of 9 and dose of 1 g/L there was no significant improvement in the 11) [24]. The same results have been reported for F re- F removal. In another study, Dobaradaran et al. [20] removal by different adsorbents [24, 29, 30]. Therefore a pH ported that for an initial F level of 8 mg/L, the F removal

100 90 80 70 (%)

60 5 min 50 efficiency 10 min 40 25 min

Removal 30 60 min 20 10 0 0 5 10 15 20 25 30 35 40

Biomass dosage (g/l)

FIGURE 1 - F adsorption as a function of adsorbent dose (initial F concentration = 5 mg/L).

1270 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

100 90 80 (%) 70

60 2 mg/l 50 efficiency

3 mg/l 40 5 mg/l 30

Removal 8 mg/l 20 10 0 0 20406080100120 Time (min)

FIGURE 2- F adsorption, using a fixed mass ratio of biosorbent to initial F concentration, as a function of initial F concentration (g L-1 biomass/g L-1 F=1000)

percentage increased with increasing shrimp shell waste where qe is the mass of F adsorbed per unit weight of dose as biosorbent from 3.2 g/L to 64 g/L but there were the sorbent (mg/g), kF is the Freundlich capacity factor and no significant differences in the F removal percentage be- a measure of adsorption capacity, 1/n is the Freundlich in- tween adsorbent doses of 48 and 64 g/L. Sivasankar et al. tensity parameter, and Ce is the equilibrium concentration [24] examined F removal with MnO2-coated Tamarind of F in solution (mg/l) after adsorption. The linear plot of Fruit (Tamarindus indica) shell at various doses and ob- log qe vs log Ce show the applicability of Freundlich iso- served that the percentage removal increased with increas- therm. The values of 1/n and kF for the sorbent were ac- ing dosage of the adsorbent and then remained constant. In counted from the slope and the intercept of the linear plot contrast, Thakre et al. [33] reported that biomass dosage of log qe vs log Ce. chitosan has no significant F removal capacity. The Langmuir adsorption isotherm is defined as: 3.4 Effect of contact time and initial F concentration qe = (3) The effect of initial F concentration on the removal of F is shown in Fig.2. It was observed that by increasing the and can be rewritten as [30]: F concentration from 2 to 8 mg/L the removal efficiency increased from 33.14 to 80.84 percent. This could be due (4) to high adsorption capacity of Moringa olifera ash. Viswa- nathan et al. [29] and Mahramanlioglu et al. [34] reported where qe is the mass of F per unit mass of sorbent similar results in the defluoridation of aqueous solutions by (mg/g), qmax is the monolayer sorption capacity, b is the protonated chitosan beads and poly aluminum chloride, re- Langmuir constant related to the free energy of sorption spectively. In contrast, Ramanaiah et al. [35] found that the equilibrium concentration of F in solution (mg/l) after ad- removal efficiency of F by waste fungus decrease with in- sorption. The Langmuir constant can be determined by creasing initial F concentration. plotting versus Ce. 3.5. Sorption isotherms To quantify the sorption capacity of Moringa Oleifera Figure 3 shows that Langmuir model is better fitted for the removal of F from aqueous solutions, two usually than Freundlich model. Langmuir isotherm assumes mon- used isotherms like Freundlich and Langmuir have been olayer adsorption on homogenous flat surface; due to the adopted. greater tendency of F to be adsorbed onto the adsorbent surface instead undergo heterogeneous adsorption. The Freundlich isoterm can be written as: 3.6.1. Biosorption kinetics and modeling 1/n qe= kf ce (1) The sorption kinetics is important in the treatment of and written in linear form as [30]: aqeous solution, as it presents valuable insights into reac- ion and mechanisms of sorption reactions. The experi- log qe = log kF + logCe (2) mental biosorption kinetic were defined by using pseudo-

1271 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 3- (a) Freundlich and (b) Langmuir isotherm investigation of F adsorption by Moringa Oleifera.

FIGURE 4- (a) Pseudo-first-order model and (b) Pseudo-second-order model of F adsorption by Moringa Oleifera.

first and pseudo-second- order kinetics. These kinetics can 4. CONCLUSION be represented in their nonlinear forms, as follows: Biosorption is an effective method for removal of ex- Pseudo-first- order model, cess F from water or wastewater. This study showed that Moringa Oleifera seed ash is a suitable adsorbent for the , log(qe − qt ) = log qe − t, (5) . removal of F from aqueous solutions. The removal percentage of F was in the range of 7.23 –81.14 % and it was de- Pseudo-second- order model, pended on the initial F concentration, pH, ashing temperature, contact time and the biomass dosage. The highest re- + , (6) moval was reached only after 10 min contact time in opti- , mum operation condition. The applied biosorbent in this Where qe is the mass of solute sorbed at equilibrium study, Moringa Oleifera seed, is easy to source, inexpen- (mg/g), qt the mass of solute sorbed at time t (mg/g), K1 the sive, and renewable. Finally, it should be noted that first-order equilibrium rate constant (g/g/min), and K2 is Moringa Oleifera seed can be used as an environmental the second-order equilibrium rate constant (g/g/min). The friendly, effective and cheap adsorbent for removal of F first-order equilibrium rate constant (K1) for F sorption was from water especially in rural and remote areas due to its calculated from the slop of the linear plot of log (qe − qt) easy operation as well as it can be used for removal of F versus time. In the case of the seconed-order equilibrium from industrial effluent containing high level of F. rate constant (K2), kinetic data were plotted between t/qt against time. As show in Fig.4, it can be concluded that F sorption onto Moringa Oleifera seed ash seems to be pseudo-second- order. ACKNOWLEDGMENTS

The authors are grateful to the Bushehr University of Medical Sciences for their financial support.

1272 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

The authors have declared no conflict of interest. [18] Ostovar, A. Dobaradaran, S. Ravanipour, M. Khajeian, A. (2013) Correlation between fluoride level in drinking water and the prevalence of hypertension: an ecological correlation study. Int J Occup Environ Med 494, 216-7. REFERENCES [19] Shams, M. Dobaradaran, S. Mazloomi, S. Afsharnia, M. Ghasemi, M. Bahreini, M. (2012) Drinking water in Gonabad, [1] Toma, S. Kreidman, J. Vedina, O. Veliksar, S. (1999) Some Iran: fluoride levels in bottled, distribution network, point of observations on fluoride problems in the Moldova Republic. use desalinator, and decentralized municipal desalination plant Fluoride 32(2), 67-70. water. Fluoride 45(2), 138- 41. [2] Jezierska-Madziar, M. Pińskwar, P. Przybył, A. (2001) Reduc- [20] Dobaradaran, S. Nabipour, I. Mahvi, A.H. Keshtkar, M. Elmi, tion in fluoride levels in the old Warta reservoir near Luboń, F. Amanollahzade, F. Khorsand, M. (2014) Fluoride removal Poland. Fluoride 34(1), 51-54. from aqueous solutions using Shrimp Shell waste as a cheap [3] Morra, P. Lisi, R. Spadoni, G. Maschio, G. (2009) The assess- biosorbent. Fluoride 47(3), 253–257. ment of human health impact caused by industrial and civil ,activities in the Pace Valley of Messina. Sci Total Environ. [21] Kocakerim, M. M. Kocadagistan, R.B.B. Yartali, A. K◌ ِ sklü 407, 3712-3720. A. (2003) Kinetics Of Fluoride Removal From Drinking Water By Ion Exchange Method. Fresen Environ Bull 12(11), 1394- [4] Spittle, B. (2008) Dyspepsia associated with fluoridated water. 1399. Fluoride 41(1), 89-92. [22] Yoshioka1, T. Kameda1, T Miyahara, M. Uchida, Mizoguchi, [5] Shivarajashankara, YM. Shivashankara, AR. Rao, SH. Bhat, T. Okuwaki1, A. (2007) Uptake of fluoride and borate ions PG. (2001) Oxidative stress in children with endemic skeletal from aqueous solution by magnesium–aluminum oxide. fluorosis. Fluoride 34(2), 103-7. Fresen Environ Bull 16 (8), 928- 933. [6] Dobaradaran, S. Mahvi, AH. Dehdashti, S. Abadi, DRV. [23] Dongyun Du, Zhongshan Yu, Jinzhao Liu, and Xiaohua Lu. (2008) Drinking water fluoride and child dental caries in (2005) Adsorption of fluoride from aqueous solution By alu- Dashtestan, Iran. Fluoride 41(3), 220-6. minum pillared rectorite. Fresen Environ Bull, 14 (11), 972- 975. [7] Rahmani, A. Rahmani, K. Dobaradaran, S. Mahvi, AH. Mo- hamadjani, R. Rahmani, H. (2010) Child dental caries in rela- [24] Sivasankar, V. Ramachandramoorthy, T. Chandramohan, A. tion to fluoride and some inorganic constituents in drinking (2010) Fluoride removal from water using activated and water in Arsanjan, Iran. Fluoride 43(3), 179-86. MnO2-coated Tamarind Fruit (Tamarindus indica) shell: Batch and column studies. Journal of Hazardous Materials [8] Zazouli, MA. Mahvi, AH. Dobaradaran, S. Barafrashtehpour, 177, 719–729. M. Mahdavi, Y. Balarak, D. (2014) Adsorption of fluoride from aqueous solution by modified Azolla Filiculoides. Fluo- [25] Thakre, D. Jagtap, S. Sakhare, N. Labhsetwar, N. Meshram, S. ride 47 (4), 349-358. Rayalu, S. (2010) Chitosan based mesoporous Ti–Al binary [9] Mahvi AH, Zazoli MA, Younecian M, Esfandiari Y. (2006) Flu- metal oxide supported beads for defluoridation of water. oride content of Iranian black tea and tea liquor. Fluoride Chemical Engineering Journal 158, 315–324. 39(4), 266-8. [26] Venkata Mohan, S. Ramanaiah, S.V. Rajkumar, B. Sarma, [10] Dobaradaran, S. Mahvi, AH. Dehdashti, S. (2008) Fluoride P.N. (2007) Removal of fluoride from aqueous phase by bio- content of bottled drinking water available in Iran. Fluoride sorption onto algal biosorbent Spirogyra sp.-IO2: Sorption 41(1), 93-4. mechanism elucidation. Journal of Hazardous Materials 141, 465–474. [11] Dobaradaran, S. Mahvi, AH. Dehdashti, S. Dobaradaran, S. Shoara, R. (2009) Correlation of fluoride with some inorganic [27] Volesky, B. and Holan, Z.R. (1995) Biosorption of heavy met- constituents in groundwater of Dashtestan, Iran. Fluoride als. Biotechnol Prog 11, 235–250. 42(1), 50-3. [28] Sharma, P. Kumari, P. Srivastava, M.M. Srivastava, S. (2006) [12] Dobaradaran, S. Fazelinia, F. Mahvi, AH. Hosseini, SS. Removal of cadmium from aqueous system by shelled (2009) Particulate airborne fluoride from an aluminium pro- Moringa oleifera Lam. seed powder. Bioresource Technology duction plant in Arak, Iran. Fluoride 42(3), 228-32. 97, 299–305. [13] Boldaji, MR. Mahvi, A. Dobaradaran, S. Hosseini, S. (2009) [29] Viswanathan, N. Sairam Sundaram, C. Meenakshi, S. (2009) Evaluating the effectiveness of a hybrid sorbent resin in re- Removal of fluoride from aqueous solution using protonated moving fluoride from water. Int J Environ Sci Technol 6(4), chitosan beads. Journal of Hazardous Materials 161, 423–430. 629-32. [30] Viswanathan, N. and Meenakshi, S. (2008) Enhanced fluoride [14] Dobaradaran, S. Abadi, DRV. Mahvi, AH. Javid, A. (2011) sorption using La(III) incorporated carboxylated chitosan Fluoride in skin and muscle of two commercial species of fish beads. Journal of Colloid and Interface Science 322, 375–383. harvested off the Bushehr shores of the Persian Gulf. Fluoride 44(3), 143-6. [31] Sarı, A. and Tuzen, M. (2008) Biosorption of Pb(II) and Cd(II) from aqueous solution using green alga (Ulva lactuca) bio- [15] Shams, M. Qasemi, M. Dobaradaran, S. Mahvi, AH. (2013) mass. Journal of Hazardous Materials 152, 302–308. Evaluation of waste aluminum filling in removal of fluoride from aqueous solutions. Fresen Environ Bull 22(9), 2604-9. [32] Jagtap, S. Thakre, D. Wanjari, S. Kamble, S. Labhsetwar, N. Rayalu, S. (2009) New modified chitosan-based adsorbent for [16] Nabipour, I. and Dobaradaran, S. (2010) Fluoride and chloride defluoridation of water. Journal of Colloid and Interface Sci- levels in the Bushehr coastal seawater of the Persian Gulf. Flu- ence 332, 280–290. oride 46(4), 204-7. [33] Thakre, D. Jagtap, S. Bansiwal, A. Labhsetwar, N. Rayalu, S. [17] Nabipour, I. and Dobaradaran, S. (2013) Fluoride concentra- (2010) Synthesis of La-incorporated chitosan beads for fluo- tions of bottled drinking water available in Bushehr, Iran. Flu- ride removal from water. Journal of Fluorine Chemistry 131, oride 46(2), 63-4. 373–377.

1273 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[34] Mahramanlioglu, M. Kizilcikli, I. Bicer, I.O. (2002) Adsorp- tion of fluoride from aqueous solution by acid treated spent bleaching earth. Journal of Fluorine Chemistry 115, 41–47.

[35] Ramanaiah, S.V. Venkata Mohan, S. Sarma, P.N. (2007) Ad- sorptive removal of fluoride from aqueous phase using waste fungus (Pleurotus ostreatus 1804) biosorbent: Kinetics evaluation. ecological engineering 31, 47–56.

Received: August 10, 2014 Revised: October 28, 2014 Accepted: December 08, 2014

CORRESPONDING AUTHOR

Sina Dobaradaran

The Persian Gulf Marine Biotechnology

Research Center

Bushehr University of Medical Sciences

Faculty of Health

Department of Environmental Health Engineering

Bushehr

IRAN

Phone/ Fax: +987733450134

E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1269 - 1274

1274 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

THE TOXIC EFFECT OF CELL MEMBRANE OF E. COLI CAUSED BY CdTe/MPA QUANTUM DOTS

Jinhui Zhou1,*, Ling Ding1,*, Zhenyu He2, Ling Jin1, Maolan Liu1, Xiujin Han1, Qingzhu Zhao1, and Xusheng Jia1

1College of Chemical Engineering and Technology, Coal Conversion and New Carbon Materials Hubei Key Laboratory, Wuhan University of Science and Technology, Wuhan 430081, China 2Wuhan Centers for Disease Prevention and Control, Wuhan 430022, China

ABSTRACT which are toxic to cells [6-11]. Clearly, it is critically important to systematically investigate the toxicity of QDs in The biological effects of CdTe/MPA quantum dots order to evaluate the feasibility of their widespread appli- (QDs) were studied on the membrane of the Escherichia cations in biology. Thus, the potential toxicity and bio-ef- coli (E. coli) cells. The free radicals produced by QDs lead fects of QDs have become a topic of considerable im- to the rupture of lipopolysacchride (LPS) patches on cell portance and discussion, and the biosafety problem for surface. Ca2+ and Mg2+ also released to the solution with QDs has drawn much attention. For instance, Su et al. [9] the broken LPS patches. An increase in the fluorescence thought that the cytotoxicity of CdTe QDs cannot attribute polarization of 1,6-Diphenyl-1,3,5-hexatriene(DPH) was solely to the toxic effect of free Cd2+ for HEK293 cells. observed, indicating a significant decrease in membrane Similarly, other researchers [10] reported that QDs parti- fluidity of E. coli. The alteration of functional group on the cles dissolved in aqueous solution release Cd2+ ions and the E. coli surface determined by attenuated total reflection concentration of the Cd2+ ions directly correlates with cy- fourier transform infrared (ATR-FTIR) has confirmed the totoxic effects. Lovric et al. [8] and Cho et al. [12] thought ion analysis and fluorescence polarization results. By mi- that the cytotoxicity of QDs was due to the synergistic re- crocalorimetry, the thermal curves showed the growth of action of free radicals, ROS, and free Cd2+. E.coli was inhibited significantly in the presence of QDs Actually, the cell out-membrane as the outermost layer under UV light. This phenomenon could also be caused by is vulnerable to environment (QDs). To our knowledge, the free radical produced by QDs, which would inhibit the there have also been a few reports about how the QDs af- synthesis of bio-molecules in cells. fect the cell membrane. Furthermore, the toxic effects of

QDs are evaluated usually by cell viability (MTT assay) [9, 13]. This assay requires toxic reagents which are toxic to KEYWORDS: the organism. It would be of great interest to develop a new membrane, CdTe/MPA QDs, UV light, fluidity, spectra method to evaluate the QDs biological effects. In this study, the CdTe/MPA QDs were used to treat

1. INTRODUCTION the E. coli cells. The alteration of cell membrane of E. coli was studied in detail. Inductively Coupled Plasma-atom Quantum dots (QDs, semiconductor nanocrystals) are emission Spectrometry (ICP-AES) has determined the ion versatile inorganic probes with unique photophysical prop- released because of LPS rupture. Fluorescent microscope, erties, such as narrow and size-dependent fluorescence fluorescent polarization and microcalorimetry were used to (FL) with broad absorption spectra, strong FL intensity, study the change of permeability and fluidity. The biologi- and excellent anti-photobleaching. Since Chan and Nie [1] cal effect of QDs was also evaluated on the cell growth. and Bruchez et al. [2] demonstrated the application of QDs for biology, QDs have been used in a diverse range of biological applications such as cell labeling, genomic detec- 2. MATERIALS AND METHODS tion, optical sensors, bioimaging [3-5]. 2.1 Materials As we know, QDs are composed of some harmful constituents such as cadmium and produce some by-products E. coli (K12) was provided by the Chinese Center for (e.g. free radicals and reactive oxygen species (ROS), etc), Type Culture Collections of Wuhan University. The peptone culture is consist of NaCl 5 g, peptone 5 g, beef extract 5 g per 1000ml (pH=7.2). The culture was sterilized in * Corresponding author high-pressure steam at 120 for 30 min.

1275 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

The preparation of mercaptoacetic acid (MPA) capped The three groups of E. coli cells were prepared with CdTe QDs were adopted from literature. CdTe/MPA QDs the different incubation time (0, 30, 60, 120mins) as fol- with yellow emission were applied in the following toxico- lows. A (the control): the native cells; B: the cells in PBS logical assays. 1,6-Diphenyl-1,3,5-hexatriene (DPH) was in the presence of 0.3 g/L QDs with shaking in dark; C: the purchased from Sigma, and dissolved in THF (tetrahydrofu- cells in PBS in the presence of QDs with UV light. ran). Then the DPH solution was diluted to 0.004 mmol/L in phosphate buffer solution (PBS, pH 7.2). The mixture was 2.6 ATR-FTIR well mixed with vigorous shaking for 15 min. The ATR-FTIR spectra of seven groups of E. coli were 1,6-diphenyl-1, 3, 5-hexatriene (DPH) was purchased measured by FT-IR spectroscopy (AVA TAR 370, Thermo from Sigma, and dissolved in THF. DPH was add into Nicolet Co. of America). The cells in PBS in the presence phosphate buffer solution (PBS, pH=7.2) and made into of the QDs with UV light for A: 0min; B: 30min C: 60min 0.004mmol/L. The mixture was well mixed with vigorous D: 120min. Then the samples were prepared by the follow- shaking for 15min. ing way: 0.5mL of the liquid from the samples was evaporated on the glass vessel and dried in a vacuum chamber. 2.2 Photo-catalytic reaction procedures Spectra were the results of 64 scans with a resolution of When E. coli was grown to primary-log phase at 37 4 cm-1 in the spectra range 4000-600cm-1. Either hydrated in a peptone culture, three groups (labeled A, B and C, re- or dry samples showed the same peaks irrespective of the spectively) of bacteria suspensions were centrifuged at degree of the hydration attained. 4000r/min for 5 minutes, then the precipitate was resuspended in PBS buffer. The bacteria were washed for 2 times 2.7 Microcalorimetric Measurement to thoroughly remove the culture medium. After that, three TAM Air (Thermometric AB, Sweden), which is a heat groups of cells were treated by different processes, which conduction microcalorimeter for measurement in the milli- were shown as the follows. A (the control): the native cells watt range, was designed to monitor continuously heat re- in PBS with shaking for 1 h; B: the cells in PBS in the pres- leased or absorbed in a series of processes, such as metab- ence of 0.3 g/L QDs with shaking for 1 h in dark; C: the olism of cells [14]. cells in PBS in the presence of 0.3 g/L QDs with UV light for 1 h. After the E. coli was inoculated, the solutions (A: LB medium; B: LB medium in the presence of QDs in dark; C: 2.3 SEM observation the cells in PBS in the presence of 0.3 g/L QDs with UV After microcalorimetric measurement, the E. coli cells light) were put into the calorimeter to monitor the growth samples (1, 2, 4, and 6) were treated as follows before SEM of E. coli cells. The metabolic thermogenic curves were recorded in real time. observation. Glutaraldehyde and osmium tetraoxide (OsO4) were used to fix and dehydrate the protein (or lipid) in cells. Double fixation by oxidization after reduction made the most of the characteristics of two different fixatives. Then 3. RESULTS AND DISCUSSION the cells were dehydrated in an increasing concentration of ethanol (50, 60, 70, 80, 90, 95, and 100 %). After the E. As described above, the outer membrane of E. coli is coli cells were fixed and dehydrated, they were observed composed of LPS, peptidoglycan and periplasm, as shown by SEM (JSM-5610LV, JEOL Ltd., Japan). in Figure 1. The strength of cell wall depends largely on peptidoglycan, whose structure is a mechanically strong 2.4 Analyses of Ca2+ and Mg2+ network. Thus, the cell wall, firm and resilient as it is, plays an important role in the maintenance of cell morphology. After photo-catalysis, three groups of E. coli were cen- LPS, as the outermost layer, is firstly exposed to the sub- trifuged at 4000 r/m for 5 minutes. The supernatants were stance or energy in environment, such as free radical and analyzed by ICP-AES (TE, IRIS Intrepid II) in order to de- 2+ 2+ UV light. QDs under irradiation will produce the charge termine the concentration of Ca and Mg . carriers (surface trapped holes h+ and the generated electrons e-). The reaction of the cell wall organics with h+ 2.5 Fluorescence polarization method leads to the formation of free radicals. Laser photolysis The E. coli membrane fluidity was studied with the flu- shows that E. coli, LPS, and PE compete in the scavenging orescence polarization technique (Shimadzu, RF-5301), in of h+ with the recombination reaction of h+ with e-, which which 1,6-diphenyl-1,3,5-hexatriene (DPH) was used as a leads to the formation of organic radicals initiating the rad- fluorescence probe. DPH was dissolved in THF and then ical chain peroxidation of organic molecules on surface diluted to 0.004 mmol/L in PBS. After vigorous shaking [15-17]. for 15 min, this solution was mixed sufficiently. E. coli were incubated in DPH solution at 25 for 30 min, then In this study, the surface changes of E. coli cell were washed twice by PBS, and resuspended in PBS (vide supra) investigated by FE-SEM. The microstructures of the four for measuring FL polarization. The excitation and emission samples were obtained, as shown in Figure 2. In Figure 2a, wavelength were 362 nm and 432 nm, respectively. SEM image revealed that the surface structure of the native

1276 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

FIGURE 1 - Schematic molecular representation of E. coli envelop

E. coli was intact, with regular wrinkles at nano-level sence of metals in the LPS assembly facilitates dissociation resolution. However, cells after reaction with QDs (Figure of LPS from the assembly [19]. 2b) became twisted and rougher. Regular wrinkles were re- placed by depressed space on the surface. The morpholog- Much of Ca2+ and Mg2+ were found in group C. The ical character of the cells also changed, for example, the concentration of group B was similar to the native cells in cells bulged at two ends and became wider and shorter. As medium, which was shown in Figure 2. During the process for figure 2c, the cells were almost completely deformed: of the photocatalysis, the surface of E. coli was exposed to the normal cell kept rod shape while cells after treatment the free radical. As a result, the LPS molecule was attacked with QDs under UV light became elliptical (Figure 2c). by the free radical. With LPS patches broken apart, Ca2+ and Mg2+ on the binding sites would released to the solu- For Figure 2b, it could be inferred that LPS ruptured to tion. The ion concentration could be regarded as an indica- some extend in the presence of QDs. However, the pepti- tion of the degree of LPS damage. doglycan and periplasm was not badly destroyed since the cells still kept their rod shape. As to the E. coli treated by From the data in the Figure 3, compared with the con- QDs under UV light, the cell outer membrane was almost trol, much amount of Ca2+ and Mg2+ were found in the so- completely damaged because cells could no longer main- lution of the cells treated by QDs under UV light. The cells tain the cell morphology. It is proposed that the outer mem- in group B were exposed to QDs without light. Under this brane including LPS, peptidoglycan and periplasm was de- condition, QDs could not produce the free radical. The con- stroyed. Without a mechanically strong network, the cell centration of Mg2+ and Ca2+ increased a little compared became elliptical or round. Of course, the inner membrane with the native group, which indicated that QDs alone of cells still exist, otherwise the cells would break apart. could cause damage to cells outer membrane.

3.1 Ca2+ and Mg2+ analysis from the E.coli medium 3.2 Decrease of cell membrane fluidity after treatment It has been reported that the metal ions such as Ca2+ Membrane fluidity is an important physical character and Mg2+ on the binding sites, such as Ca2+ and Mg2+, can of cell membrane. Many cell functions, including energy maintain the structure of the LPS patches of E. coli [18]. transformation, nutrients transportation and transferring Molecular dynamics simulation also suggests that the ab- signals, are all tightly relevant with cell membrane fluidity.

1277 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

Therefore, stability of membrane fluidity plays an im- DPH polarization reflects the average fluidity of all cellular portant role in keeping normal cell functions and resisting membrane lipids [21]. An inverse relationship exists be- various environmental stresses. In cells, DPH is distributed tween membrane fluidity and polarization. In Figure 4, the within the hydrophobic region of lipid membranes [20] and cells of group C were irradiated under UV light for different

(A) (B)

0.70 2.5 0.65

2.0 0.60

0.55 A 1.5 B 0.50 C

1.0 0.45

A 0.40 B Polarization Values 0.5 0.35 concentration of of ions (ppm) concentration

0.30 0.0

0 20406080100120 Native dark UV light t (min) different types of lighting FIGURE 4 - Polarization value of for three groups of E.coli cells with FIGURE 3 - The concentration of A (Ca2+) and B(Mg2+) in the super- the different irradiation time natant from the native cells and the treated cells by QDs under dark and UV light

1278 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

time. Obviously, QDs caused an increase in the fluores- peaks. Peaks at ~2923 cm-1 and 1029 cm-1 were also prom- cence polarization of DPH, reflecting a significant decrease inent, and these groups were attributed to va(CH2), phos- in membrane fluidity of E. coli. The free radical could at- phoric acid asymmetric vibration of and the -C-O-C- in ol- tack lecithoid group, causing that the alignment of lecithoid igosaccharide which was bond with protein to form glyco- molecular were closer, the sport the fat chain were re- protein. Small peaks at ~1392 cm-1 correspond to –COO- in stricted and the rigidity of the whole membrane increased hydrophobic glycerol end, 1230 cm-1 to symmetric vibra- - [22, 23]. The cells of group B in dark have not much in- tion of PO2 . All the peaks are summarized in Table 1 in crease in the fluorescence polarization. QDs alone damage accordance with their groups. the membrane to some extent, which confirmed the result Changes of the spectral profile of E. coli for the de- from the concentration analysis. crease of amide I~1639cm-1,II ~1538 cm-1 and for the significant decay of amide A ~3270 cm-1 were observed, 3.3 Functional group changes determined by ATR-FTIR which suggested the outer leaflet damages of amide groups Figure 5 shows the changes in the spectral profile of in the hydrophobic end of the phospholipids. The symmet- − -1 − ATR-FTIR spectra of E. coli for the E. coli in the presence ric vibration of PO2 at ~1230 cm also decreased. PO2 of QDs with the different irradiation time. It was obvious that decrease indicated the peroxidation of the hydrophobic the peak intensity of the broad band 4000-600cm-1 decrease end. The intensity decrease at 2923 cm-1, indicated the during the photocatalytic treatment as a function of time. changes in C-H of the fatty-tail structure. The results indicated that during the photocatalytic process, the outermost groups which exposed to the environment were the easiest to be oxidized. The groups like CH2, CH3 also were involved, but at a slower pace, showing a higher resistance to photocatalytic peroxidation. Other factors for the differences can also be attributed to the bond energy as well as the ordering and position of the functional groups.

3.4 Growth Curves Determined by Microcalorimetry Heat produced by metabolism of microorganism is very small. Therefore, it is necessary to construct a calorimeter which is sensitive and accurate enough to monitor the metabolism of microorganism. Isothermal heat-conducting is particularly valuable for monitoring such a process because it is a non-destructive and non-invasive technique.

FIGURE 5 - The ATR-FTIR spectra of the native E.coli and cells treated by the QDs with the different irradiation time. A: 0min; B: 30min C: 60min D: 120min

The ATR-FTIR spectra of the native E. coli are shown in Figure 5. The envelop of E. coli consists of LPS, phospholipids bilayer, protein. The protein was attached to the membrane. According to Figure 5 and Table 1, the sharpest and most prominent peaks at ~3290cm-1, 1639 cm-1, 1538 cm-1 in Figure 5 belonged to the amide groups [21]. As we know, the membrane consists of lipids and protein. Phospholipid, which forms the bilayer structure, has a hydrophilic circular head and two hydrophobic fatty tails. The out layer of the phospholipid bilayers is the hydrophilic end FIGURE 6 - The thermogenic curve of E.coli in the process of growth. and the hydrophobic end is in between. Because the amide A (the control): the native cells in PBS with shaking; B: the cells in groups connected to the hydrophobic head are exposed to PBS in the presence of the QDs with shaking in dark; C: the cells in the outmost, these groups showed the most significant PBS in the presence of QDs with UV light

TABLE 1 - The ATR-FTIR peaks of E. coli in the spectra are summarized

Wavenumber /cm-1 1029 1230 1392 1538 1639 2923 3290 - - Functional groups C-O-C va(PO2 ) -COO Amide Ⅱ Amide Ⅰ va(CH2) Amide A

1279 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

All biological processes coincide with heat release or REFERENCES absorption, e.g. the metabolism of bacteria. Microcalorim- etry can monitor the bacteria growth by measuring very [1] Chan W.C.W. and Nie S. (1998) Quantum dot bioconjugates small heat flow as a non-destructive and non-invasive tech- for ultrasensitive nonisotopic detection. Science 281 (5385): 2016-2018. nique. Figure 6 has shown the growth process of E. coli in the presence of the QDs under UV irradiation. [2] Bruchez M., Moronne M., Gin P., Weiss S. and Alivisatos A.P. (1998) Semiconductor nanocrystals as fluorescent biological labels. Science 281(5385):2013-2015. According to thermogenic curves in Figure 6b, the [3] Zhang W.J., Lou Q., Ji W.Y., Zhao J.L. and Zhong X.H. presence of QDs in dark has a little effect on E. coli growth. (2013) Homogeneously alloyed CdSe1–xSx Quantum dots (0 Lag phase was prolonged a little. No obvious change was ≤ x ≤ 1): an efficient synthesis for full optical tunability. observed about the maximum heat power and the metabo- Chemistry of Materials 26 (2):1204–1212. lism in growth process. While in Figure 6c, the presence of [4] Liu P., Wang Q.S. and Li X. (2009) Studies on CdSe/L-cyste- QDs under UV light was observed to show strong inhibi- ine quantum dots synthesized in aqueous solution for biologi- tory effect on E. coli growth. The growth rate of E. coli cal labeling. J. Phys. Chem. C 113(18): 7670-7676. decreased significantly and generation time prolonged ac- [5] Kikkeri R., Lepenies B., Adibekian A., Laurino A. and cordingly. The heat output power of E. coli cells also de- Seeberger P.H. (2009) In vitro imaging and in vivo liver tar- creased significantly. geting with carbohydrate capped quantum dots. J. Am. Chem. Soc. 131(6):2110-2112.

[6] Fu T., Qin H.Y., Hu H.J., Hong Z. and He S., (2010) Aqueous synthesis and fluorescence-imaging application of CdTe/ZnSe 4. CONCLUSIONS core/shell quantum dots with high stability and low cytotoxicity. J. Nanosci. Nanotechnol. 10(3):1741-1746. The free radicals produced by QDs under irradiation. [7] Chen L.Y., Siemiarczuk A., Hai H., Chen Y., Huang G.B. and The radicals were thought to be the cause of the damages Zhang J. (2014) Development of biocompatible and proton- on phospholipids bilayer due to preoxidation. So the free resistant quantum dots assembled on gelatin nanospheres. radicals lead to the rupture of LPS patches. Ca2+ and Mg2+ Langmuir. 30 (7):1893-1899. also released to the solution with the broken LPS patches. [8] Lovric J., Cho S.J., Winnik F.M., Maysinger D., Lovric J., Cho The cell membrane couldn’t serve as the cellular envelope S.J., Winnik F.M. and Maysinger D. (2005) Unmodified cadmium telluride quantum dots induce reactive oxygen species steadily. The permeability and functionality of the cell formation leading to multiple organelle damage and cell death. membrane were altered consequently. An increase in the Chem. Biol. 12(11):1227-1234. fluorescence polarization of DPH was observed, reflecting [9] Su Y., Hu M., Fan C., He Y., Li Q., Li W., Wang L., Shen P. a significant decrease in membrane fluidity of E. coli. The and Huang Q. (2010) The cytotoxicity of CdTe quantum dots alteration of functional group on the E. coli surface deter- and the relative contributions from released cadmium ions and mined by ATR-FTIR has confirmed the ICP and fluores- nanoparticle properties. Biomaterials 31(18):4829-4834. cence polarization results. By microcalorimetry, the ther- [10] Huang P.C., Jiang Q., Yu P., Yang L.F. and Mao L.Q. (2013) mal curves show the growth of E. coli was inhibited signif- Alkaline post-treatment of Cd(II)–Glutathione coordination polymers: toward green synthesis of water-soluble and cyto- icantly in the presence of QDs under UV light. This phe- compatible CdS quantum dots with tunable optical properties. nomenon could also be caused by the free radical produced Applied Materials & Interfaces 5 (11):5239-5246. by QDs, which would inhibit the synthesis of bio-mole- [11] Kirchner C., Liedl T., Kudera S., Pellegrino T., Javier A.M., cules in cells. Gaub H.E., 1lzle S.S., Fertig N. and Parak W.J. (2005) Cyto- toxicity of colloidal CdSe and CdSe/ZnS Nanoparticles. Nano. Lett. 5(2):331-338. ACKNOWLEDGMENTS [12] Cho S.J., Maysinger D., Jain M., Roder B., Hackbarth S. and Winnik F.M. (2007) Long-term exposure to CdTe quantum dots causes functional impairments in live cells. Langmuir We gratefully acknowledge the financial support of the 23(4):1974-1980. Youth Talent project of Science and Technology Research Program of Hubei Provincial Department of Education of [13] Tan L.J., Wan A. and Li H.L. (2014) Synthesis of near-infrared quantum dots in cultured cancer cells. Appl Mater Inter- China (Q20131105) and Coal Conversion and New Carbon faces 6 (1):18-23. Materials Hubei Key Laboratory(Wuhan University of Sci- [14] Ding L., Li X., Liu P. and Zhao X.H. (2011) Microcalorimet- ence and Technology (WKDM201301) and Wuhan Uni- rical Study of the Effect of Selenium and Cadmium on the versity of Science and Technology Foundation of China Growth Metabolism of Escherichia coli. Asian J. Chem., 23, (z00980, 2014XG006) (10) 4285-4288. [15] Kiwi J. and Nadtochenko V. (2005) Evidence for the mecha- The authors have declared no conflict of interest. nism of photocatalytic degradation of the bacterial wall membrane at the TiO2 interface by ATR-FTIR and laser kinetic spectroscopy. Langmuir. 21(10):4631-4641.

[16] Nadtochenko V., Denisov N., Sarkisov O., Gumy D. Pulgarin C. and Kiwi J. (2006) Laser kinetic spectroscopy of the inter- facial charge transfer between membrane cell walls of E. coli and TiO2. Photochem Photobio A: Chem 181(2-3):401-407.

1280 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

[17] Nadtochenko V.A., Rincon A.G., Stanca S.E. and Kiwi J. (2005) Dynamics of E. coli membrane cell peroxidation during TiO2 photocatalysis studied by ATR-FTIR spectroscopyand AFM microscopy. Photochemistry and Photobiology A: Chemistry 169 (2):131-137. [18] Liu P., Wang Q.S. and Li X. (2010) The damage of outer membrane of Escherichia coli in the presence of TiO2 combined with UV light. Colloids and Surfaces B: Biointerfaces 78(2):171-176. [19] Amro N.A., Kotra L.P., Mesthrige K.W., Bulychev A., Mobashery S. and Liu G.Y. (2000)High-Resolution Atomic Force Microscopy Studies of the Escherichia-coli Outer Mem- brane: Structural Basis for Permeability. Langmuir 16 (6):2789-2796.

[20] Kuhry J.G., Fonteneau P., Duportail G., Maechling C., Laustriat G. (1983) TMA-DPH: a suitable fluorescence polarization probe for specific plasma membrane fluidity studies in intact living cells. Cell Biophys 5(2):129-140. [21] Swan T.M. and Watson K. (1997) Membrane fatty acid composition and membrane fluidity as parameters of stress tolerance in yeast. Can J Microbiol 43(1):70-77. [22] Hauser H., Hinckley C.C., Krebs J., Levine B.A., Phillips M.C. and Williams R.J. (1977) The interaction of ions with phosphatidylcholine.Biochim. Biophys Acta 468(3):364-377. [23] Conti J., Halladay H.N., Petersheim M. (1987) An ionotropic phase transition in phosphatidylcholine: cation and anion co- operativity. Biochim Biophys Acta 902(1):53-64.

Received: August 12, 2014 Accepted: September 22, 2014

CORRESPONDING AUTHOR

Jinhui Zhou and Ling Ding

947 Heping Road

College of Chemical Engineering and Technology

Wuhan University of Science and Technology

Wuhan 430081

P. R. China

Phone: + 86 027 68862780

E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1275 - 1281

1281 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

THE MECHANISM OF ALUMINIUM TOXICITY AND RESISTANCE ON SOME BIOCHEMICAL AND HORMONAL CONTENTS OF HORDEUM VULGARE SEEDLINGS

Mona M. Abdalla

Department of Botany, Faculty of Science, Ain Shams University, Cairo, Egypt

ABSTRACT sistant to aluminium than others, and variations occur among genotypes of the same species [4]. The uptake and accumu- To investigate the threshold of aluminium toxicity and lation of Al into the root apoplasm and symplasm is rapid, tolerance in Hordeum vulgare plants, the grains were pre- and accordingly, various inter-/intra-cellular sites may be soaked in Al2(SO4)3 (0, 10, 2, 0.4 and 0.08 µM) for 6 h, sown affected causing the disruption of the normal function of in plastic pots, and samples were taken when plants were plasma membrane and plasma membrane transport system 10 and 30 days old. Aluminium, either steeply (0.08 µM [5, 6]. Extensive work has shown that Al causes an inhibi- )or slightly (0.4 µM), raised all measured growth criteria tion in root growth [7, 8], root elongation [9, 10], morpho- (mean length of shoots and roots, mean number of leaves logical disorganization in the root apex [3], as well as root and lateral roots, and mean fresh and dry weights of plants), bending which arose from unequal root cell elongation and the contents of each of total chlorophyll, total pigments, an alteration in root anatomy [9, 10]. It also inhibits the photosynthetic rates, reducing sugars, starch, total sugars, the number and length of lateral roots [11]. Decrease in both levels of each of potassium, magnesium, phosphorus, cal- shoot growth and shoot/root ratio was observed after Al cium and growth promoters (auxins, gibberellins, cytokin- treatment [12]. Moreover, this element reduces both the ins), and the activity of invertase, whereas higher concen- fresh weight and dry weight of shoots and roots [9, 13, 14]. trations (2 and 10 µM) obviously reduced them. Reversi- Al was reported to induce a reduction in the quantity of bly, Al at lower concentrations decreased the carotenoid chlorophyll pigment and in the ratio between chlorophyll a and ABA contents and the activity of IAA-oxidase while and b which was accompanied by marked decline in pho- higher concentrations elevated them. Al, at all rates, posi- tosynthetic rate [14, 15]. It also suppressed photosystem I tively accumulated ascorbate, sodium and iron contents, and and II-mediated electron transport, and O2 evolution was raised the activities of α- and β-amylases. suppressed [15]. The total respiratory rate decreased with

increased supply of Al; these circumstances were accompanied by a reduction of soluble carbohydrates, including KEYWORDS: Aluminium, growth, metabolites, photosynthetic pigments, enzymes, phytohormones reducing sugars which formed the substrate for respiration [16]. Total soluble sugars increased in blueberry when

treated with Al up to 200 µM, and then, remained nearly 1. INTRODUCTION constant [17]. Graham [12] indicated that 1 Mm Al increased the content of sucrose and starch (in stems), but decreased Aluminium is one of the most abundant elements on the level of each of glucose, sorbitol, fructose, total soluble the earth, constituting about 7% of soil minerals. The acid- carbohydrates, starch and total carbohydrates (in roots and ification of the ground has increased the level of free Al in leaves). Ascorbic acid (AA) can be used either directly or soils as well as in lakes, and there is a positive correlation indirectly by plants as an antioxidant for scavenging reac- between the decrease in pH of the lakes and the increasing tive oxygen species (ROS) produced during oxidative stress level of Al in the water [1]. About 40% of the world arable induced by Al [18-20]. Al was found to interfere with cer- soils are acidic which favors the dissolution of microscopic tain enzymes governing the deposition of polysaccharides quantities of Al+3 from metal oxides, and therefore, pre- of the cell wall, and alters the activity of hydrolytic en- sents Al toxicity hazards [1, 2]. Aluminium toxicity is a ma- zymes contained in the Golgi apparatus after being dam- jor factor in limiting plant growth in most strongly acid aged by Al [3], but no data was available concerning its soils. Toxic effects on plant growth have been attributed to role on IAA-oxidase enzyme activity. Al was found to re- several physiological and biochemical pathways [3]. The duce Ca uptake in different plants, thus reducing Ca reten- mechanism of toxicity and resistance to aluminium have tion in roots and shoots [13, 21]; it also reduces rice, maize been studied and recognized for 70 years. Although the and sugar maple contents of Ca, Mg, P and K [22, 23]. reasons are still unknown, some plant species are more re- Cells treated with Al showed lower levels of Na, K , Ca, P,

1282 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

Mg, Fe and Mn [24, 25]. Al has been reported to cause The total reducing sugars (TRS) were determined after su- damage to the endoplasmic reticulum within the root meri- crose hydrolysis, and sucrose was calculated from the dif- stem, thus altering its hormone-binding site [26]. Unilateral ference between TRS and DRS. Starch was determined in application of Al to the root cap influenced the polarity of terms of glucose using the glucose oxidase method after auxin transport along roots [27]. Moreover, Bennet et al. digestion with amyloglucosidase [37], and the resulting [28] developed a model for Al-toxicity in which Al could glucose content was then multiplied by 0.9. Ascorbic acid, indirectly inhibit (or stimulate) root growth by altering the a scavenger of oxyradicals, was assayed by the method de- production and distribution of growth hormones. Recent scribed by Roe and Keuther [38] where ascorbate is con- investigations supported the above view [29-31]. The pre- verted to dehydroascorbate by the treatment with activated sent work was carried out: (1) to investigate both the long- charcoal. Dehydroascorbate then reacts with 2,4 dinitro- term (10-30 days) effect of Al toxicity and the concentra- phenyl hydrazines to form osazones; dissolved in sulfuric tions of Al2(SO4)3 applied being toxic, and to which extent acid, they give an orange-coloured solution whose absorb- both the growth and chemical composition of Hordeum ance can be measured spectrophotometrically at 540 nm. vulgare seedlings are altered; (2) as an approach for under- Enzymes were extracted from plant tissues as adopted by standing the physiological mechanism of Al toxicity in Guerrier and Strullu [39], with some modifications. The in- plants so as to induce the production of genetically tolerant vertase activity was assayed following the method adopted traits that can overcome its deleterious effect, and to be by Russel and Jimmy [40]. The activity of α-amylase was with good performance, especially on acid soils. assayed according to the procedure adopted by Davis [41], and it was represented as the decrease in optical density/min/1 g fresh weight, while the activity of β-amylase 2. MATERIALS AND METHODS was determined following the method described by Malik and Singh [42]. IAA-oxidase enzyme was assayed follow- Grains of Hordeum vulgare Lcv. Giza 108 were ob- ing the method described by Darbyshire [43]. The method tained from the Agricultural Research Center, Giza, Egypt. of extraction of minerals from plant tissues was essentially The grains were sterilized with sodium hypochlorite (5%) similar to that of Chapman and Pratt [44]. Phosphorus was for 5 min, and washed thoroughly with distilled water. determined following the method described by Humphries They were then soaked for 6 h at 22 °C in various concen- [45]. Sodium and potassium were estimated photometri- trations of Al2(SO4)3 (0, 10, 2, 0.4, 0.08 µM). Forty plastic cally according to Williams and Twine [46]. Calcium, mag- pots (15 cm in diameter and 11cm in depth) were arranged nesium and iron were determined by atomic absorption spec- into 5 groups, (8 pots for each group) and 15 seeds were trophotometry according to A.O.A.C. [47]. For estimation of planted in each pot. Then, they were germinated on What- growth hormones, fresh samples were collected and kept in man filter paper no. 45 at relative humidity of 60-65%, day cold re-distilled 95% ethanol in which they were after- length of 12 h, day/night air temperature 22/18 °C, and light wards extracted. Then, they were fractionated into aqueous intensity of 3040 Lux. Hoagland nutrient solution (15 ml) and acidic fractions according to the method described by [32] at pH 5.5± 0.2 were added to each pot and renewed Shindy and Smith [48]; the acidic fraction contains IAA every 3 days. After 10 and 30 days of growth, the plants (indoleacetic acid), GA3 (gibberellic acid) and ABA (abscisic were harvested, and samples were taken so as to be used acid) while the aqueous one contains the cytokinin. Both frac- either immediately for both morphological, photosynthetic tions were finally quantified by HPLC analysis according to and pigment measurements, rapidly dried in an oven at 108 °C the method adopted by Muller and Hilgenbery [49]. for carbohydrates, AA and mineral determinations, or frozen for enzyme and hormonal analysis. 2.3 Statistical analysis Morphologic parameters and photosynthetic rate val- 2.1 Measurement of photosynthetic rates ues were means ± standard errors (SE) of 10 replicates The photosynthetic rate was measured using an open while those of chemical and hormonal analysis were means gas portable photosynthesis system (LI 6400, LICOR, Bi- ± standard error (SE) of five replicates. Significant differ- osciences, USA). Measurements were performed on sunny ences were calculated using student's (t) test. SPSS version days under light conditions, and between 9.00 and 12.00 h 15 was performed for multiple comparisons. on the upper, most fully expanded leaves of 10 plants randomly chosen per treatment, and expressed on a leaf area basis [33]. 3. RESULTS AND DISCUSSION

2.2 Chemical analysis 3.1 Growth parameters Photosynthetic pigments (total chlorophyll, carotenoids It is evident from Table 1 that the two lower concen- and total pigments) were determined spectrophotometrically trations of Al, either highly (0.08 µM) or moderately and [34]. Carbohydrate fractions were extracted and clarified slightly (0.4 µM) raised all growth criteria which were rep- similar to those described by Said and Naguib [35]. The resented by mean length of roots and shoots, mean number direct reducing sugars (DRS) were determined following of leaves and lateral roots, and mean fresh and dry weights the anthrone method suggested by Umbrient et al. [36]. of 15 plants above those of untreated ones, at the two ages

1283 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

of growth. Conversely, increasing Al concentration obvi- in shoot/root ratio, and in shoot, root and hypocotyl fresh ously decreased all these criteria. This inhibitory effect of and dry weights [9, 10, 14, 54], in several examined plant the higher doses of Al was reported by several authors us- species. This trivalent element also resulted in the formation ing various plants. Aluminium was found to induce abnor- of smaller young leaves that are curled along the margin malities in the root system which included dwarfing of with yellow tips and having necrotic spots while the grow- roots [14], reduction or inhibition of the growth of main ing point collapsed; older leaves show a marginal chlorosis axis of roots with consequent thickening and mottling [8, with subsequent lethality [50]. 9], forming initiation of numerous lateral roots followed by reduction in their growth accompanied by their thickening On the contrary, numerous works have indicated that and browning [50], and finally, root bending which arose exposure of plants to Al for either a short period (30 min to from unequal cell elongation that results from unequal in- 2h) or low concentrations, surprisingly, is beneficial for hibition of mitotic activity and cell enlargement at both plant growth as it accelerates root formation, root growth sides of root axis [11, 51]. Al can interact with multiple and elongation, shoot growth, and an overall plant growth sites in the apoplasm and symplasm of root cells, and dis- stimulation (hormesis) which is consistent with the present rupt the normal function of plasma membrane and plasma results [10, 11]. Such stimulatory effect of Al at the lower membrane transport system [5, 6]. The binding of Al to concentrations can possibly be due to either its ability to these sites is probably an important factor for its toxicity reduce cell surface which arose from H+ activity at the [52]. It is proved that Al could indirectly inhibit or stimu- membrane surface, thereby promoting Fe and P uptake [11, late root growth (depending on concentration) by altering 55], or by altering the distribution of growth regulators in the production and distribution of growth hormones [31] roots [11, 29-31]. and/or affecting water and nutrient uptake [15]. Al treatments caused a reduction in each shoot growth [12, 14, 53],

TABLE 1 - Changes in the growth criteria of Hordeum vulgare plants in response to aluminium toxicity (each value is a mean of 10 replicates).

Mean Mean Mean length of Mean length of Mean no. of Mean no. of Age/ Treatment fresh wt. of dry w t. of root shoot lateral roots leaves day (µM Al) 15 plants 15 plants (cm) (cm) (cm) (cm) (g) (g) 10 0 5.4±0.1 8.3±0.3 4.5±0.6 1.6±0.4 2.58 ±0.2 0.36± 0.03 10 2.1± 0.3 4.5±0.2 2.9±0.4 1.0±0.3 2 2.15±0.1 0.20±0.02 2 3.6± 0.2 5.9±0.6 4.1±0.2 1.3±0.4 2 2.35±0.2 0.25±0.05 0.4 6.2± 0.5 7.8±0.3 5.7±0.6 1.4±0.2 2.66±0.03 0.42± 0.05 0.08 7.3± 0.2 9.5±0.5 6.2±0.1 1.9±0.3 2.77±0.06 0.51± 0.06 30 0 8.3±0.4 14.0±0.4 7.6±0.3 2.8±0.3 3.11±0.02 0.66±0.04 10 5.0± 0.2 10.9±0.2 6.0± 0.5 1.2±0.1 2.8 ±0.05 0.44± 0.04 2 5.7± 0.3 12.2±0.6 7.9±0.6 1.4±0.4 3.0±0.04 0.56±0.07 0.4 8.1± 0.4 13.4±0.5 8.2±0.6 2.2±0.3 3.25±0.02 0.73 ±0.03 0.08 9.8± 0.7 16.6±0.6 9.4± 0.7 3.3±0.5 3.4±0.06 0.79±0.06 *The mean difference is significant at the 0.05 level.

TABLE 2 - Changes in the photosynthetic rates and photosynthetic pigment contents of Hordeum vulgare plants in response to aluminium treatment (each value is a mean of 5 replicates an d expressed as mg/g FW. ± SD while photosynthetic values are means of 10 replicates).

Treatment Photosynthetic rate Total Age/day 2 -1 Chl. (a+ b ) Carotenoids (µM Al) (µmol CO2 m s ) Pigments 10 0 7.63±0.1 5.2 ± 0.4 1.48 ± 0.01 6.68 ± 0.4 10 4.86±0.6 3.87 ± 0.2 1.81 ± 0.01 5.68 ± 0.2 2 5.93±0.4 5.05 ± 0.3 1.74 ± 0.02 6.79 ± 0.3 0.4 7.03±0.7 5.47 ± 0.2 1.68 ± 0.13 7.15 ± 0.3 0.08 8.24±0.4 7.34 ± 0.2 1.27 ± 0.10 8.61 ± 0.4 30 0 8.94±0.2 9.97 ± 0.50 1.52 ± 0.14 11.49 ± 0.4 10 5.32±0.5 5.02±0.16 3.31±0.03 8.33±0.01 2 6.22±0.13 6.53±0.18 2.83±0.04 9.36±0.02 0.4 8.98±0.8 8.85±0.20 2.63±0.02 11.48±0.13 0.08 10.77±0.7 11.58±0.17 1.95±0.06 13.43±0.2 *The mean difference is significant at the 0.05 level.

1284 © by PSP Volume 24 – No 4. 2015 Fresenius Environmental Bulletin

3.2 Photosynthetic pigments and photosynthetic rates ter 7 days, and then remained constant. Additionally, an- Grain pre-soaking in the higher concentrations of Al other work explored that high accumulation of Al in barley (10 and 2 µM) greatly reduced each of chl (a+b), total pig- plants was positively correlated with an increase in soluble ments and net photosynthetic rates below those of un- sugar contents [57]. The following authors suggested the treated controls (Table 2). Conversely, treatment with the increase to be due to either reduction in photorespiration, lower concentrations either produced comparable levels to hexose phosphorylation, cell wall polysaccharide synthe- those of control (0.4 µM) or elevated markedly (0.08 µM) sis, or osmoregulation [3, 17]. the pigment amounts and the net photosynthetic rates at Al at (1 mM) caused severe reductions in reducing sug- both ages of growth. Concerning the carotenoid contents, ars, total soluble carbohydrates and total carbohydrates in they increased progressively with the increase in Al con- roots, stems and leaves, increased starch in the roots and centration applied. The present data were confirmed by dif- shoots as well as sucrose amounts in leaves [12]. Such an ferent workers who realized remarkable Al-induced reduc- increase in the content of soluble sugars, which is associ- tions in the quantity of chlorophyll pigments including ated by a decline in starch and total sugars, could be at- chlorophyll a and chlorophyll a and b ratio [14, 15, 53] and tributed to the increased activity of hydrolytic enzymes (α- high accumulations of carotenoids [17], which was accom- and β-amylase, invertase) which were estimated in the pre- panied by the degradation of thylakoids in the chloroplasts sent work, and a concomitant decline in total pigment [17], and with consequent suppression in photosystems I amounts as well as an alteration of the chloroplast ultra- and II-me diated electron transport and repression of O2 structure which eventually resulted in a decline of the pho- evolution [15, 17]. Accordingly, photosynthetic rate was tosynthetic rate and the amount of photosynthates. Beside declined [14, 15]. Moreover, Pereira et al. [10] demonstrated the antioxidative enzymes, metabolites (AA and reduced that Al affects chlorophyll synthesis by inhibiting the activ- glutathione (GSH)) can be used either directly (direct reac- ity of aminolevulinic acid dehydratase enzyme (ALA-D) re- tion with ROS) or indirectly (ascorbate glutathione cycle) sponsible for the formation of monopyrrole porphobilonogen by plants for the elimination of excess ROS produced dur- (PBG) which is a part of the chlorophyll molecule as well as ing oxidative stress induced by Al exposure [18-20, 58]. the cytochromes, and also greatly impairs plant growth. Several authors consider that the regeneration mechanism The decrease in chlorophyll might be mediated of AA is fundamental for cell homeostasis allowing for re- through the reduced uptake of magnesium which is the in- use of AA in the removal of ROS, and indicating that AA tegral part of the chlorophyll molecule [14]. Carotenoids is involved in the tolerance mechanism of Al detoxification are among non-enzymatic antioxidant compounds released in barley seedlings [19, 20, 58]. by plant cells as a defensive system that helps to detoxify ROS-induced under Al stress, and also acts as a protection 3.4 Enzyme activity against chlorophyll molecule oxidation imposed by Al [2, 56]. The activities of α- and β-amylase were directly propor- tional to the concentrations of Al used (Table 4), and in the 3.3 Carbohydrate contents meantime, above those of the control activities at both stages Higher concentrations of Al obviously decreased re- of growth. Lower concentrations of Al (0.4 and 0.08 µM) ducing sugars, starch and total sugar levels below those of raised the invertase activity while higher concentrations re- untreated plants, while these fractions were raised at the duced it above and below the control activities, respec- lower concentrations (Table 3). Sucrose and ascorbic acid tively. Al, on the other hand, induced a reverse effect on (AA) contents notably increased in response to the differ- IAA-oxidase activity (i.e., its activity is increased by the ent concentrations applied. These results are in agreement two higher doses while declined by the two lower ones). In with those of Reyes Diaz et al. [17] who showed that Al up accordance, Simon et al. [25] found that the invertase ac- to 200 µM increased total soluble sugars in blueberries af- tivity in tomato roots responded consistently to Al concen-

TABLE 3 - Changes in the carbohydrate and ascorbic acid contents of Hordeum vulgare plants in response to aluminium treatment (each value is a mean of 5 replicates and expressed as mg/g DW ± SD).

Treatment Age/ day Reducing sugars Sucrose Starch Total sugars Ascorbic acid (µM Al) 0 84 ± 3 50.4 ± 2.8 441 ± 6.3 709 ± 4.9 0.32±0.01 10 62.2 ± 2 53.2 ± 3.1 339 ± 5.6 586 ± 5.2 0.64±0.03 10 2 76.4 ± 2 96.4 ± 5.4 441 ± 6.2 693 ± 4.4 0.59±0.03 0.4 88.6 ± 3 100.8 ± 5.7 543 ± 5.8 784 ± 6.1 0.48±0.01 0.08 91.0 ± 5 118.0 ± 6.1 573 ± 5.6 956 ± 6.4 0.36±0.04 0 111.2 ± 2 69.2 ± 3.2 592 ± 2.3 862 ± 4.4 0.49±0.02 10 68.4 ± 1 82.4 ± 1.9 294 ± 3.4 488 ± 2.7 0.78±0.05 30 2 83.2 ± 3 108.8 ± 2.8 403 ± 4.3 509 ± 3.1 0.64±0.05 0.4 109.2 ± 3 152.4 ± 2.9 443 ± 3.8 810 ± 5.2 0.53±0.01 0.08 128.3 ± 5 170.1 ± 3.4 490 ± 4.6 997 ± 5.6 0.50±0.03 *The mean difference is significant at the 0.05 level.

TABLE 4 - Changes in the activities of certain hydrolytic and oxidative enzymes of Hordeum vulgare plants in response to aluminium treatment (each value is a mean of 5 replicates and expressed as enzyme activity/g fresh weight/hour ± SD.

Treatment α-amylase (decrease β-amylase (ug maltose Invertase (mg reducing IAA –oxidase (ug of Age/day (µM Al) in OD/unit time) released/g.f.wt/h) sugar released/g.f.wt./h) IAA oxidised/ g.f.wt./h) 0 0.46 ± 0.11 32 ± 5.2 425.7 ± 2.2 468.2 ± 1.9 10 0.29 ± 0.03 70 ± 3.1 354.9 ± 2.4 571.8 ± 1.6 10 2 0.34 ± 0.06 56 ± 4.4 388.3 ± 2.8 526.9 ± 1.0 0.4 0.39 ± 0.09 45 ± 3.4 446.1 ± 2.7 431.8 ± 1.3 0.08 0.44 ± 0.10 40 ± 3.6 521.4 ± 1.3 382.6 ± 0.2 0 0.41 ± 0 .1 2 46 ± 3.2 595.3 ± 2.0 576.3 ± 2.3 10 0.23 ± 0 .0 3 80 ± 4.7 380.9 ± 0.9 689.2 ± 0.9 30 2 0.28 ± 0 .0 5 75 ± 2.9 419.6 ± 1.7 624.7 ± 1.8 0.4 0.34 ± 0 .1 1 62 ± 2.7 615.2 ± 1.1 517.6 ± 1.1 0.08 0.38 ± 0 .0 6 55 ± 3.6 666.7 ± 1.4 479.4 ± 1.4 *The mean difference is significant at the 0.05 level

TABLE 5 - Changes in the content of certain mineral elements of Hordeum vulgare plants in response to aluminium treatment (each value is a mean of 5 replicates and expressed as mg/ g DW ± SD).

Age/ Treatment Sodium Potassium Phosphorus Magnesium Calcium Iron day (µ M Al) 10 0 3.01 ± 0.2 4.23 ±0.1 21.26± 0.3 2.54 ± 0.5 4.13 ± 0.3 0.35± 0.03 10 5.87 ±0.5 3.08 ± 0.2 14.97 ± 0.4 1.47 ± 1.9 2.84 ± 0.4 0.80± 0.03 2 5.12 ± 0.3 3.56 ± 0.1 15.23 ± 0.3 1.87 ±0.3 3.0.7 ± 1.2 0.77 ± 0.02 0.4 4.79 ± 0.2 4.14 ± 0.2 17.67 ± 0.2 2.19 ± 0.1 4.12 ± 0.7 0.66 ± 0.01 0.08 4.06±0.2 4.85 ±0.1 29.34 ± 0.6 2.89± 0.4 4.88 ± 0.6 0.47 ± 0.02 30 0 3.97 ±0.3 5.96 ±0.2 31.56 ± 0.2 3.27 ± 0.4 5.09 ± 0.3 0.58± 0.03 10 7.33 ±0.6 3.88 ± 0.2 20.03 ±0.4 1.57 ± 0.3 3.67± 0.2 0.87 ± 0.04 2 6.27 ±0.5 4.39 ± 0.1 26.39 ± 0.6 2.09 ± 0.1 4.18 ± 0.4 0.80 ± 0.05 0.4 5.66 ± 0.5 5.55 ± 0.3 27.19 ± 0.9 2.39 ± 0.2 4.93 ± 0.4 0.73 ± 0.02 0.08 5.03 ± 0.4 6.02 ± 0.4 34.26± 0.4 2.74 ± 0.5 5.67 ± 0.3 0.69 ± 0.04 *The mean difference is significant at the 0.05 level

trations applied. Furthermore, Roy et al. [3] proved that Al amounts at all concentrations of Al used, comparable to causes the damage of Golgi apparatus, and then, alters the those of control amounts. Several reports were obtained activity of included hydrolytic enzymes which govern the that ascertain these results. K uptake was reduced in many deposition of cell wall polysaccharides. Additionally, tested plants in response to Al treatment [15, 22, 23]. Al is heavy metals were reported to affect α- and β-amylase ac- proposed to compete with K for root absorption, thus re- tivities [59, 60]. Unfortunately, no data is available to throw ducing its uptake by roots and its content in roots and tops light on the effects of Al on IAA-oxidase activity. However, [22]. Increasing Al concentration caused accumulation of Chaoui et al. [61] found that pea seedlings treated with Cd+ P either on the root surface, within the cells or in the free or Cu+ showed an elevated IAA-oxidase activity in their root space of roots, thus reducing its translocation and, there- cells. It was proposed that the stimulating effect of Al to hy- fore, its amounts in tops of various plant species [9, 11, 13]. drolytic enzymes is concurrent with the ability of this ele- The disturbance in P metabolism by Al resulted in a ment to reduce the membrane permeation to water [15, 62], marked decrease in sugar phosphorylation due to the in- thus inducing cell water stress. Such conditions, in turn, fa- creased affinity of Al to combine with ATP (40 times that vor the secretion of osmolytic substances as soluble sugars of Mg), thus forming a highly stable Al-ATP complex and which increase the cell osmotic potential thus forcing more preventing the transfer of the terminal phosphoryl group to water uptake by the cell [63, 64]. glucose by hexokinase (Mg-dependent enzyme). This case alters the respiration rate, the energy production, and vital- 3.5 Mineral contents ity of treated cells. Moreover, soil-P availability during Depending on the concentration of Al applied, it either seedling stage is an important determinant of growth, N2 increased the accumulation of each of potassium, phospho- fixation and grain yield [9, 13]. rus, magnesium and calcium (0.08 µM) or reduced them Extensive results, argued with the present work, are (10, 2 and 0.4 µM) above and below the untreated plants, concerning the reduction or accumulation of Mg+2 and Ca+2 respectively, throughout the experimental period (Table 5). due to, respectively, high and low Al concentrations [15]. Iron and sodium, on the other hand, registered higher Al inhibited the efflux of H+ from barley roots and de-

creased the activities of K+, Mg+2 and ATPase of plasma µM) or decreased (at higher rates) above and below the membrane [15]. Al reduced the uptake and transport of control levels, respectively, at the two ages of plant growth, these elements, thus causing their deficiency symptoms to while the abscisic acid (ABA) contents were reversibly in- appear in shoots [15, 22]. Al competed with Mg at the bind- creased, progressively with the increase in the concentra- ing sites of δ-aminolevulinic acid dehydratase enzyme tion of Al used. Similar results were reported by Raven and (ALA-D) responsible for the formation of phosphobi- Rubery [26], Bennet et al. [28] and Barcelo and Poshen- longen (PBG), a part of chlorophyll molecule as well as the rieder [11]; they speculated that Al, at certain concentra- cytochrome molecule, thus affecting the synthesis of pig- tions, could indirectly inhibit or stimulate root growth by ments which reduced photosynthesis with concomitant re- altering the production and distribution of growth hor- duction in the amount of organic matter and eventually mones. Additionally, during Al toxicity stress, there is a plant growth [10]. Al was absorbed by cells and competed limited cytokinin biosynthesis that limits its supply to in an exchangeable manner at almost all calcium binding shoots, inhibits meristematic development of lateral shoots sites on the cell surface causing the disruption of cytoplas- and causes an alteration in Ca transport and redistribution mic Ca+2 homeostasis [15], and the accumulation of hemi- [69]. Recently, Al-treated tomato roots caused either the cellulosic polysaccharides in walls of root tips; these, in induction of proteins involved in the biosynthesis and sig- turn, led to cell stiffening and thickening that eventually naling a pathway which includes ethylene (1-aminocyclo- caused inhibition of root elongation [2, 65, 66]. Further- propane-1-carboxylate oxidase and multiprotein bridging more, Al interaction with the plasma membrane could lead factor) and ABA (ABA/WDS-induced protein) or the re- to depolarization of the transmembrane potential and/or repression of another protein involved in the biosynthesis of duction of H+/ATPase which, in turn, can alter the activi- gibberellins (gibberellin-3-ß-hydroxylase 2-3 activity and ties of ions near the plasma membrane surface and impede gibberellins-2-oxidase) [70]. Hasenstein and Evans [27] + +2 +2 + the uptake of several cations (K , Mg , Ca and NH4 ) demonstrated that unilateral application of Al to the root thus causing nutritional unbalances for plants [2]. The en- cap could influence the polarity of auxin transport along hanced accumulation of Fe+3 and Na+ in Hordeum, in re- roots. It may also inhibit the basipetal auxin transport from sponse to Al treatment, was supported by Berger et al. [22]. root meristem to elongation zone resulting in decreased Such an increase in Na values can be considered as one of root cell elongation. Ethylene is suggested to be involved the tools that Al-treated plants may lead to increase the in fast signal transduction of Al-induced enhancement of negative osmotic potential of tissues that arose from the re- cytokinin levels in roots. These suggestions were sup- duction in membrane water permeability thus increasing ported by the finding that Al-induced transient rise in eth- the ability of cells and tissues for water and solute uptake ylene production in roots after 5min of Al exposure, which from soil [15]. Iron accumulation can be attributed to the was followed after 15 min by a substantial increase of root corresponding increases in peroxidase activities in cytokinin levels of beans [29]. Alternatively, other workers Hordeum plants [67], or it may be due to the stimulation of showed that Al induced ethylene production which acts as Al to the radical chain reactions mediated by iron ions so a signal to alter auxin distribution and accumulation in Ar- as to enhance lipid peroxidation [68]. abidopsis roots by disrupting the auxin carriers (AUX1 and PIN2), thus disrupting the auxin polar transport which ar- 3.6 Phytohormones rests root elongation [30, 31]. Another view postulated that The changes in the phytohormonal levels of untreated Al causes root inhibition through alteration of hormone and Al-treated plants are presented in Table 6. Depending gradients within the root meristems, as a consequence of on the dose of Al applied, the contents of each of auxins, damage to the endoplasmic reticulum, which is a hormone- gibberellins and cytokinins were either increased (0.08 binding site [26].

TABLE 6 - Changes in the phytohormonal contents of Hordeum vulgare plants in response to aluminium treatment (each value is a mean of 5 replicates and expressed as mg/kg f.wt. ± SD).

Treatment Age/day Auxins Gibberellins Cytokinins Abscisic acid (µM Al) 0 9.8±0.4 7.9±0.6 26.7±0.8 7.4±0.3 10 2.3±0.1 2.9±0.5 9.6±0.6 16.6±0.6 10 2 4.4±0.4 4.1±0.3 15.2±0.7 14.4±0.6 0.4 7.9±0.3 6.3±0.2 19.3±0.6 10.3±0.2 0.08 10.0±0.6 8.9±0.2 27.8±0.6 6.8±0.4 0 14.9±0.7 13.8±0.6 24.9±0.4 10.6±0.5 10 6.4±0.5 4.1±0.1 8.1±0.7 21.3±0.7 30 2 10.6±0.5 8.7±0.2 14.2±0.5 19.6±0.2 0.4 15.3±0.6 14.6±0.4 18.3±0.5 14.7±0.6 0.08 16.8±0.4 17.8±0.6 26.8±0.4 9.1±0.3 *The mean difference is significant at the 0.05 level

4. CONCLUSION [7] Blamey, F.P.C., Nishizawa, N.K. and Yoshimura, E. (2004). Timing, magnitude, and location of initial soluble aluminum injuries to mung bean roots. Soil Science Plant Nutrition 50, Thus, with progressive increasing environmental metal 67-76. load and, consequently, acid rain, soil acidification is en- [8] Jorge, R.A. and Menossi, M. (2005). Effect of anion channel hanced and Al plays a major role in the loss of specific tree antagonists and La³ on citrate release, Al content and Al re- species as well as loss of total vegetational cover at specific sistance in maize roots. Journal of Inorganic Biochemistry 99, sites. Consequently, extensive work has been done during 2039-2045. the last decade to elucidate the threshold of Al toxicity as a [9] Jemo, M., Aboidoo, R.C.; Nolte, C. and Horst, W.J. (2006). function of either its exposure time or doses applied, its Aluminium resistance of cowpea as affected by phosphorus- mechanism of action on plants and the mechanism of plant deficiency stress. Journal of Plant Physiology 1, 1-10. tolerance to it. Accordingly, in the present work, Al treat- [10] Pereira, L.B.; Tabaldi, J.F.; Goncalves, G.O.; Juckeoski, M ment either shows hormetic or toxic effects at, respectively, .M.; Pauletto, S.N.; Weis, F.T.; Nicoloso, D.; Bocher, J.B.T.; low and high doses. It is hypothesized that toxic concentra- Rocha, M. and Schetinger, R.C. (2006). Effect of aluminium tions of Al formerly, induced alterations in hormonal levels on δ-amino levulinic acid dehydratase (ALA-D ) and the development of cucumber (Cucumis sativus). Environmental and in roots (by either decreasing the biosynthesis of growth Experimental Botany, 57, 106-115. promoters or hampering their translocation from the root meristem to the elongation zone, beside increasing the lev- [11] Barcelo, J. and Poshenrieder, C. (2002). Fast growth responses, root exudates and internal detoxification as clues to els of ethylene and ABA, thus causing cell wall stiffening the mechanisms of aluminium toxicity and resistance: a re- and thickening leading to inhibition of cell elongation and view. Enviromental and Experimental Botany 48, 75-92. growth [29-31, 70]. In addition to the inhibition of root [12] Graham, C.J., (2002). Nonstructural carbohydrate and growth, Al treatment also affects plant growth by impairing prunasin composition of peach seedlings fertilized with differ- metabolic activity, reducing chlorophyll synthesis, photo- ent nitrogen sources and aluminium. Scientia Horticulturae 94: synthesis, respiration, and carbohydrate contents [12, 15], 21-32. enhancing the production of ABA and antioxidant sub- [13] Sierra, J.H.; Ozier-Lafontaine, L.; Dufour, A.; Meunier, R.; stances, such as AA and carotenoids [15], altering nutrient Bonhomme, A. and Welcker, C. (2006). Nutrient and assimi- availability in the rhizosphere, nutrient uptake and translo- late partitioning in two tropical maize cultivars in relation to their tolerance to soil acidity. Field Crops Research. 95: 234- cation by plants [2] and water uptake [15]. It eventually 249. causes extensive plasma membrane damage, peroxidation of membrane lipids, and loss of cell compartmentation [11, [14] Aftab, T.; Khan, M.M.A.; Idrees, M. and Moinuddin, M.N. (2010). Effects of aluminium exposures on growth, photosyn- 67]. Furthermore, these results indicate that the tolerance thetic efficiency, lipid peroxidation, antioxidant enzymes and indices of H. vulgare were 0.4 µM of Al or lower, which Artemisinin content of Artemisia annua L. Journal of Phytol- affirms that H. vulgare possesses a certain potential for Al ogy and Phytophysiology . 28: 23-37. tolerance and can be used as an indicator species for this [15] Panda, S. K.; Baluska, F. and Matsumoto H. (2009).Alumi- element. num stress signaling in plants. Plant Signal Behavior. 4: 592- 597. PMID:18666947; http:// dx:doi.org/10.1111/j.1744- The author has declared no conflict of interest. 7909.2007.00599x. [16] Sarkunan, V.; Biddappa, C.C. and Nayak, S.K. (1984). Physi- ology of Al toxicity in rice. Current Science. 53(15): 822-824. REFERENCES [17] Reyes-Diaz M.; Blancheteau, C.I.; Millaleo, R.; Edgardo C.; Zottele, C.W.; Alberdi, M. and Mora, M.L. (2010). Long–term [1] Von Uexkűll, H.R, Mutert, E. (1995). Global extent, develop- Aluminum Exposure Effects on Physiological and Biochemi- ment and economic impact of acid soils. In Plant-Soil Interac- cal Features of Highbush Blueberry Cultivars. Journal of tions at low pH: Principles and Management; Date, R. A., American Social and Horticultural Science. 135, 3: 212- 222. Grundon, N. J., Rayment, G.E., Probert, M.E., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, pp. 5-19. [18] Sharma, P. and Dubey, R.S. (2007). Involvement of oxidative stress and role of antioxidative defense system in growing rice [2] Silva, S. (2012 ). Aluminium Toxicity Targets in Plants. Re- seedlings exposed to toxic concentrations of aluminium. Plant view Article. Journal of Botany. 8 pages; doi number: Cell Reports 26: 2027-2038. 10.1155/2012/219462. [19] Ma, B.; Gao, L.; Zhang, H.; Cui, J. and Shen, Z. (2012). Alu- [3] Roy, A.K., A. Sharma and G. Talukder (1988). Some aspects minium induced oxidative stress and changes in antioxidant of aluminium toxicity. The Botanical Review, 54, 145-178. defenses in the roots of rice varieties differing in Al tolerance. [4] Kochian, L.V.; HoeKenga,O.A. and Pineros, M.A. (2004). Plant Cell Reports 31: 687 – 696. Annual Review of Plant Biology, 55,459-493. [20] Xu, F.J.; Li, G.; Jin, C.W.; Liu,W.J.; Zhang, S.S. and Zhang, [5] Kopittke, P.M., Blamey, F.P.C. and Menzies, N.W. (2008). Y.S.. (2012). Aluminium induced changes in reactive oxygen Toxicities of soluble Al, Cu and La include ruptures to rhizo- species accumulation, lipid peroxidation and antioxidant ca- dermal and root cortical cells of cowpea. Plant and Soil. 303, pacity in wheat root tips. Biologia Plantarum 56: 89-96. 217-227. [21] Olivares, E.; Pena, E.; Marcano, E.; Mostacero, J.; Aguiar, G.; [6] Horst, W.J., Wang, Y. and Eticha, D. (2010). The role of root Benitez, M. and Rengifo, E. (2009). Aluminium accumulation apoplast in aluminium-induced inhibition of root elongation and its relationship with mineral plant nutrients in 12 pterido- and in aluminium resistance of plants. A review. Annals of phytes from Venezuela. Environmental and Experimental Botany, 106, 185-197. Botany 65 (1):132-141.

[22] Berger, T.W.; Eagar, C.; Likens, G.E .and Stingeder, G. [38] Roe, J.H. and Keuther, C.A. (1953). The determination of (2001). Effect of calcium and aluminium chloride additions on ascorbic acid in whole blood urine through 2,4-nitrophenyl hy- foliar and throughfall chemistry in sugar maples. Forest Ecol- drazine derivative of dehydro-ascorbic acid. Journal of biolog- ogy and Management 149: 75-90. ical Chemistry 147 ,399-407.

[23] de Mendonca, R.J.; Cambraia, J.; de Oliviera, and Oliva, M.A. [39] Guerrier, G. and Strullu, D.G. (1990). Development laxes em- (2003). Aluminium effects on the uptake and the utilization of bryonnaires de pois pourvus des reserves. Canadian Journal of macronutrients in two rice cultivars. Pesquisa Agropecuaria Botany 68: 742-746. Brasileira 38 (7): 843-848. [40] Russel, P. and Jimmy, K.A. (1980). Invertase in oat seedlings, [24] Lindberg, S. and Griffiths, G. (1993). Aluminium effects on separation, properties and changes in activities in segments. ATPase activity and lipid composition of plasma membranes Plant Physiology 65: 136-142. in sugar beet roots. Journal of Experimental Botany 44, 267, [41] Davis, B.D. (1977). Occurrence of amylase in the axis of ger- 1543-1550. minating peas. Plant Physiology 60: 513. [25] Simon, L.; Smalley, T.J.; Jones. J.B.Jr. and Lasseigne, F.T. [42] Malik, C.P. and Singh, M.P. (1980). Plant Enzymology and (1994). Alminium toxicity in tomato. Part 2. Leaf gas ex- Histo-Enzymology - A Text Manual. Kalyani-Publishers. New change, chlorophyll content and invertase activity. Journal of Delhi - Ludhiana, 69. Plant Nutrition 17: 307- 317. [43] Darbyshire, B. (1997). The Effect of Water stress on indole [26] Raven, J.A. and Rubery, P.H. (1982). Co-ordination of devel- acetic acid oxidase in pea plants. Plant Physiology 47: 65. opment hormone receptors, hormone action and hormone transport. In Smith, H. and Crierson (eds.), The molecular bi- [44] Chapman, H.D. and Pratt, F.P. (1961). Method of analysis for ology of plant development. Bot. Monogr., 18: 28-48, Black- soils, plants and waters. Univ. of California. Division of Agri- well, Oxford. cultural Science. [27] Hastenstein, K.H. and Evan, M.L. (1988). Effects of cations [45] Humphries, E.C. (1956). Mineral component and ash analysis. on hormone transport in roots of Zea mays. Plant Physiology In: Peach, K. and Tracey, M.V. (Ed.) Modern Methods of Plant 86: 890-894. Analysis, Vol. I, p. 148, Springer-Verlag, Berlin.

[28] Bennet, R.J., Breen, C.M. and Bandu, V.H. (1990). A role for [46] Williams, S. and Twine, M. (1960). Flame photometric Ca+² in the cellular differentiation of root cells: a re-examina- method for sodium, potassium and calcium. In: Peach, K. and tion of root growth central mechanisms. Environmental and Tracey, M.V. (Ed.) Modern Methods of Plant Analysis, Vol. Experimental Botany 30: 515-523. V, pp 3-5. Springer-Verlag, Berlin.

[29] Massot, N.; Nicander, B.; Barceló, J.; Poschenrieder, C. and [47] A.O.A.C. (1984). Official Method of Analysis of the Associa- Tillberg, E. ( 2002). A rapid increase in cytokinin levels and tion of Official Analytical Chemists, 14th Ed., Published by enhanced ethylene evolution precede. Al-induced inhibition of the Association of Official Analytical Chemists, P.O. Box 540, root growth in bean seedlings (Phaseolus vulgaris L.). Plant Benjamin Franklin Station, Washington, DC., 2004. Growth Reulation 37 (2): 105-112. [48] Shindy, W.W. and Smith, O. (1975). Identification of plant [30] Sun, P.; Tian, Q.; Chen, J. and Zhang, W,H. (2010). Alumin- hormones from cotton ovules. Plant Physiology 55: 550-554. ium–induced inhibition of root elongation in Arabidopsis is [49] Muller, P. and Hilgenberg, W.(1986). Isomers of zeatin and mediated by ethylene and auxin. Journal of Experimental zeatin riboside in club root tissue: evidence for trans-zeatin bi- Botany 61 (2): 347-356. osynthesis by plasma Diophora brassicae. Physiologia Planta- [31] Yuan, H. M.; Lui, W.C.; Jin, Y. and Lu, Y.T. (2013). Role of rum 66: 245-50. ROS and auxin in plant response to metal-induced stress. Plant [50] Foy, C.D., 1984. Physiological effect of hydrogen, aluminium Signaling and Behaviour 8, 7; e24671-3. and manganese toxicities in acid soil. In: Adams, F. (Ed.) Soil [32] Hoagland, D.R. and Arnon, D.I. (1950). The water culture Acidity and Liming. Agronomy Monograph no. 12, second ed. method for growing plants without soil. California Agricul- ASA-CSSA-SSSA Publisher. Madison, WI, pp. 57-97. tural Experiment Station Circular 347: 1-32. [51] Potters, G.; Pasternak, T.P.; Guisez, Y.; Palme, K.J. and Jan- sen M.A. (2007). Stress-induced morphogenic responses: [33] Renault, S.; Croser C.; Franklin, J.A.; Zwiazek, J.J. and growing out of trouble? Trends Plant Science 12, 98-105. Mackinnon, M. (2001). Effect of Consolidated tailings water PMID: 17287141; http://dx.doi.org/10.1016/j.tplants. on red–osier dogwood (Conus stolonifera michx) seedlings. 2007.01.004. Environmental Pollution 113: 27-33. [52] Delhaize, E.; Ryan, R.P. and Ma, F.J. (2001). Aluminium tol- [34] Metzner, H., Rau, H. and Senger, H. (1965). Untersuchungen erance in plants and the complexing role of organic acids. zur Synchronisierbarkeit einzelner Pigmentmangelmutanten Trend Plant Science 6: 273-278. von Chlorella. Plants 65: 186-142. [53] Ali, B.; Hasan S.A.; Hayat, S.; Hayat, Q.; Yadav, S.; Fariduddin, [35] Said, A. and Naguib, M.I. (1964). Sucrose determination as a Q. and Ahmad A. (2008). A role of brassinosteroids in the ame- means of estimation of the draw back tax of Halawa Tehinia. lioration of aluminum stress through antioxidant system in mung Bulletin Faculty of Science, Cairo University, 39, 207. bean. Environment of Experimental Botany 62: 153-159. [36] Umbrient, W.W.; Burris, R.H.; Stuffer, J.F.; Cohen, P.P.; [54] Ou-yang, C ; Gao, S.; Mei, L.; Chung, T.W.; Tang, L.; Wang, Johnse, W.I.; Leepxgi, G.A; Patter, V.R. and Schneider, S. and Chen, F. (2013). Effects of aluminum toxicity on the W.C.I. (1959). Monometrie Techniques, a manual describing growth and antioxidant status in Jatropha curcas seedlings. methods applicable to the study of tissue metabolism. Burgess Journal of Medicinal Plants Research 7 (6): 234-242; Publishing Company, pp: 239. [55] Barcelo, J. and Poschenrieder, C. (2004). Heavy metal stress [37] Haissig, B.E. and Dickson, R.E. (1979). Starch measurement in plants. From biomolecules to ecosystems In: M.N.V. Prasad in plant tissue using enzymatic hydrolysis. Physiologia Planta- (Ed.), Structural and Ultrastructural Changes in Heavy Metal rum 47: 151-157. Exposed Plants. Springer Berlin, Germany, pp. 223-248.

[56] Nhan, P.P. and Hai, N.T. (2013). Amelioration of aluminium toxicity on OM4900 rice seedlings by sodium silicate. African Journal of Plant Science 7 (6): 208-212; oi;10.5897/AJPS 11.306. [57] Guo, T.R.; Guo, P.Z. and Zhang, Y.H. (2007). Physiological changes in barley plants under combined toxicity of aluminium, copper and cadmium. Colloides and Surface B: Biointer- faces 57, 2, 182-188; doi: 10.1016/j.colsurfb.2007.01.013.

[58] Ribeiro, C.; Cambraia, J.; Peixoto, P.H.P. and Junior, E.M.F. (2012). Antioxidant system response induced by aluminium in two rice seedlings. Brazilian Journal of Plant Physiology 24, 2, 107-116.

[59] Hong, R.; Yang, G. and Liu, D. et al. (1991). Effects of Cd on the growth and physiological, biochemical reactions of wheat seedlings. Acta Agriculture Boreti – Sinica 6 (3): 70-75. [60] Sethy, S.K. and Ghosh, S. (2013). Effect of heavy metals on germination of seeds. Journal of Natural Science, Biology and Medicine 4 (2): 272-275. http//www.jnsbm.org. [61] Chaoui, A.; Jarror, B. and El Fergani, E. (2004). Effects of cadmium and copper on peroxidase, NADH oxidase and IAA oxidase activities in cell wall, soluble and microsomal membrane fractions of pea roots. Journal of Plant Physiolgy 161 (11): 1225- 1234. [62] Blamey, F.P.C.; Asher, C.J.; Edwards, D.C. and Kerven, G.L. (1993). In vitro evidence of aluminium effects on solution movement through root cell walls. Journal of Plant Nutrition 16: 555-562. [63] Lutfor Rahman, S.M.; Nawata, E.; Damae, Y. and Sakaura- tani, T. (2000). Effects of water stress and temperature on activity, growth and yield of tomato. Acta Horticulture 516: 41- 45.

[64] Xu, S.; An, L.; Feng, H.; Wang, X. and Li, X. (2002). The seasonal effects of water stress on Ammopiptanthus mongol- icus in a desert environment. Journal of Arid Environment 51: 437-447. [65] Tabuchi, A. and Matsumoto, H. (2001). Changes in cell wall properties of wheat (Triticum aestivum ) roots during aluminium-induced growth inhibition. Physiologia Plantarum 112: 353-358. [66] Silva, S.; Matos, S.M. and Pinto-Carnide, O. (2011). Al toxicity mechanisms in tolerant and sensitive rye genotypes. En- vironmental and Experimental Botany 75: 89-97.

[67] Abdalla, M.M., (2010): Role of Al in altering the biochemical contents, protein profile and genomic DNA in Hordeum vulgare seedlings. Egyptian Journal of Botany 50; 1-20. [68] Yamamoto, Y.; Kobayashi, Y. and Matsumoto, H. (2001). Li- pid peroxidation is an early symptom triggered by aluminium, Received: August 17, 2014 but not the primary cause of elongation inhibition in pea roots. Accepted: November 12, 2014 Plant Physiol. 125: 199-208.

[69] Pan, W.L.; Hopkins, A.G. and Jackson, W.A. (1988). Alumin- ium- inhibited shoot development in soybean. A possible con- CORRESPONDING AUTHOR sequence of impaired cytokinin supply. Communications in Soil Science and Plant Analysis. (Commun SOIL SCI Plant Mona M. Abdalla Anal) 19: 1143-1153; doi:10.1080/00103628809368001. Department of Botany [70] Okekeogbu, I.; Ye, Z.; Sangireddy, S.R.; Li, H.; Bhatti, S.; Faculty of Science Hui, D.; Zhou, S.; Howe, K.J.; Fish, T.; Yang, Y. and Ain Shams University Thannhauser, T.W. (2014). Effect of aluminum treatment on proteomes of radicals of seeds derived from Al-treated plants. Cairo Proteomes. 2, 169-190. doi: 10.3390 / Proteomes 2020169. EGYPT

E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1282 - 1290

EFFECTS OF Mn-LOADING ON ACTIVATED

CARBON TREATED BY HNO3 FOR SO2 REMOVAL

Xuejiao Wang1, Mengdan Gong1, Jiaxiu Guo1-3,*, Yifan Qu1, Huaqiang Yin1-3, Yongjun Liu1-3 and Jianjun Li1-3

1College of Architecture and Environment, Sichuan University, Sichuan Chengdu 610065, China 2National Engineering Technology Research Center for Flue Gas Desulfurization, Sichuan University, Sichuan Chengdu 610065, China 3Sichuan Provincial Environmental Protection, Environmental Catalytic Materials Engineering Technology Center, Chengdu 610064, PR China

ABSTRACT Many studies have shown that the metals or oxides loaded on activated carbon can increase catalytic activity Using Mn(NO3)2 as precursor, a series of Mn-based ac- effectively [6-9], including Mn, Cu, Fe, Ce, Co, and V, es- tivated carbons treated by HNO3 were prepared by the ex- pecially Mn [10]. Huang [11] found that manganese oxides cessive impregnation method, and characterized by scan- (MnOx) have the advantages of high sulfur capture capacity ning electron microscopy (SEM), nitrogen adsorption-de- and good regenerative capacity. In our previous study, it sorption, X-ray diffraction (XRD), X-ray photoelectron was also found that different chemical states of Mn species (XPS) and Fourier-transform infrared spectroscopy (FTIR). exhibit different SO2 removal abilities [12]. It is generally The desulfurization showed that different Mn loadings on believed that the metal proﬁle is controlled by the impreg- activated carbons performed best SO2 removal ability, with nation step where the metal precursor contacts the solid the optimal dosage of 7 wt.% of Mn, which had the best support for the ﬁrst time, and different loadings affect both breakthrough sulfur capacity of 211 mg/g and break- textural and chemical properties of the catalysts and, con- through time of 391 min. MnO and Mn3O4 coexist in cata- sequently, their catalytic activity [13]. Hu et al. [14] has lysts and exhibit best SO2 removal ability, but after desul- found that different Mn loadings can influence catalytic furization, MnO2 is detected, indicating that it may be a combustion of methane. In our studies, it was found that reason of the deactivation of catalysts. Different Mn load- Ni, loaded on activated carbon with different contents, can ings could influence the surface functional groups, when change oxygen-containing functional groups and influence loading 7 wt. % Mn on the sample, the highest SO2 adsorp- SO2 removal ability [15]. However, desulfurization activi- tion capacity is mainly attributed to the highest relative ties of different Mn loadings on activated carbons are rarely content of C=O in the catalyst. After desulfurization, sur- studied. face oxygen-containing functional groups are changed, which indicates that they could react with SO2 and be restructured In this paper, a series of activated carbon-based cata- by generated H2SO4. lysts with different Mn loadings are prepared by the excessive impregnation method, and they have been character-

ized by a wide number of techniques, in order to correlate KEYWORDS: Manganese; activated carbon; surface functional catalytic properties and study the influence of the texture groups; desulfurization; manganese oxides; catalyst and surface chemical properties as well as Mn oxide spe-

cies. The catalytic activities for the desulfurization under the simulated flue gas, the possible changes of surface 1. INTRODUCTION metal species and functional groups on the activated carbons during the desulfurization process are also studied. More energy-saving and deeper desulfurization has better development for the environment. Activated carbons possess large surface areas, abundant micropore structures 2. MATERIALS AND METHODS and surface oxygen species; they have been extensively studied as suitable materials or adsorbent of many sub- 2.1. Catalyst preparation stances both in liquid and gaseous phase as well as supports A commercial activated carbon (Xingtong Chemical of catalysts [1-3]. In recent years, the carbon-based catalyst Ltd., Henan, China) was crushed into 10-20 mesh particles. is a new type of oxidative desulfurization catalysts and can The obtained activated carbon was washed with distilled remove SO2 from flue gas with the requires of high effi- water until the washing liquid became neutral, and then ciency, resource conservation and integration [4, 5]. dried at 105 °C for 12 h. The obtained activated carbon was denoted as AC. Subsequently, the AC was completely im- * Corresponding author mersed into 39 wt.% HNO3 solution for 2 h at 60 °C. After

that, the sample was filtrated and washed with distilled wa- The N2 adsorption isotherms of the samples at the tem- ter until the washing fluid became neutral. Finally, the ob- perature of liquid nitrogen were obtained on a Micromerit- tained sample was dried in air at 105 °C for 12 h and des- ics ASAP M 2020 apparatus. Before each analysis, the ignated as NAC. samples were outgassed at 250 °C for about 3 h to reduce Catalysts were prepared by the excessive impregnation the residual pressure to 30 mm Hg. The adsorption isotherm data were used to calculate the surface area (S ) of method. The precursor was Mn(NO3)2 (AR grade; Kelong BET Chemical Reagent Factory, Chengdu, China). The NAC each sample via the Brunauer–Emmett–Teller (BET) equa- was completely immersed into an appropriate concentration at relative pressures between 0.05 and 0.35. The total tion precursor solution to achieve different Mn loadings (0- pore volume was directly calculated from the volume of 10 wt.%). The mixtures were allowed to stand overnight. nitrogen held at the relative pressure (P/Po = 0.97), and the All samples were heated in a water-bath at 60 °C until the volume of micropores was estimated using the Dubinin– liquid was eliminated completely, and dried overnight at Radushkevich (D–R) equation. The mesopore volume was 105 °C; then, the samples were calcined in a muffle furnace determined from the subtraction of micropore volume from from room temperature to 650 °C with a heating rate of 5 °C/ total pore volume. The microporous pore size distribution was calculated using Horvath–Kawazoe (H–K) model. min under a N2 (99.99%) atmosphere and maintained for 2 h. Finally, the samples were cooled down to room tem- Scanning electron microscopy (SEM) was performed perature in N2 atmosphere. The real contents of Mn in the using a JMF–7500F scanning electron microscope (JEOL prepared catalysts were 1.04, 3.0, 4.85, 6.85 and 9.70 wt.% Co., Japan), with an acceleration voltage of 5 kV. Powder using atomic absorption spectrometry (AAS; SpectrAA X–ray diffraction (XRD) was carried out on a DX–2007 220FS, USA). The samples were labelled as 1% Mn/NAC, diffractometer (Dandong Haoyuan Instrument Co., Ltd, 3% Mn/NAC, 5% Mn/NAC, 7% Mn/NAC and 10% Mn/ China) using Cu Kα radiation (λ = 0.1542 nm) at 40 kV NAC, respectively. and 30 mA. The data were recorded for 2θ values between 10° and 70°, with an interval of 0.03. X–ray photoelectron 2.2. Activity evaluation spectroscopy (XPS) spectra were performed using a XSAM–

SO2 removal was carried out at 80 °C in a continuous 800 spectrometer (KRATOS Co., UK) with an Al Kα radia- flow multiple fixed-bed micro-reactor. The simulated flue tion under ultra–high vacuum (UHV) at 12 kV and 15 mA. gas was controlled by a rotor flow-meter before entering the Energy calibration was done by recording the core level blender. The water vapor was introduced by the gas mixture spectra of Au 4f7/2 (84.0 eV) and Ag 3d5/2 (368.3 eV). Peak passing through a humidifier before entering the reactor. areas including satellites were computed by a program The simulated flue gas contained 0.28% SO2, 10% O2, which assumed Gaussian–line shapes and flat background 10.2% water vapor, and N2 as the balance. The gas space subtraction. FTIR spectra were recorded on a Nicolet 6700 velocity (SV) was 1809 h-1. The flue gas before and after spectrometer (Thermo Electron Co., USA) using KBr pel- -1 the reactor passed through a H2O2 solution (3%), and lets in the wave-number region of 4000–400 cm with a −1 formed H2SO4 which was determined by titrating with resolution of 4 cm . NaOH (0.01 mol/L) solution, using a bromcresol green and methyl red mixture as an indicator to determine the ending point [16]. When SO2 concentration in the outlet reached 3. RESULTS AND DISCUSSION 200 mg/m3, it could be considered that the catalyst bed had been penetrated. The corresponding cumulative working 3.1. SO2 removal performance of catalysts time is regarded as the breakthrough time, and the cumula- The curves of the relation between SO2 removal effi- tive amount of SO2 removal per unit mass of catalyst is the ciency and working time are presented in Fig. 1. SO2 break- breakthrough sulfur capacity. SO2 removal efficiency (%) through sulfur capacity (expressed as mg SO2/g of sample) was obtained by analyzing the SO2 inlet and outlet concen- and the corresponding breakthrough time are listed in Ta- trations: ble 1. It is found that the SO2 removal ability of the catalysts is significantly improved when Mn species are loaded SO2(inlet ) SO 2( oulet ) SO2 removal efficency (%) 100% on the NAC, indicating that Mn species play a key role in SO2(inlet ) the SO2 removal process. 1% Mn/NAC has a breakthrough sulfur capacity of 152 mg/g and breakthrough time of 2.3. Catalyst characterization 280 min. With the increase of the loading, SO2 removal The real content of Mn was AAS-detected (SpectrAA ability of the samples gradually increases, which may be 220FS, USA). Before the testing, the catalyst was digested caused by the different texture and surface chemical prop- by dry method as follows: (1) 1.0 g sample was accurately erties of samples as well as Mn oxides. 7% Mn/NAC ex- weighed and heated for 6 h at 800 °C in air; (2) 10 ml HCl hibits the best SO2 removal ability, and it has the break- (12 mol/L) was added to the cooled residue and then heated through sulfur capacity of 211 mg/g and breakthrough time gently to dissolve residue; (3) the mixture was diluted with of 391 min. The SO2 removal ability of 10% Mn/NAC is HCl (3 mol/L) to 50 ml, and the insoluble material was fil- slightly less than that of 7% Mn/NAC, and corresponds to trated; and (4) the filtrate was diluted with distilled water the breakthrough sulfur capacity of 197 mg/g and break- to 100 ml. through time of 367 min, suggesting that both 7% Mn/NAC

and 10% Mn/NAC may exist similar, with regard to active 3.2. Texture properties components and functional groups. Based on the results in The surface micrographs of selected samples before Fig. 1 and Table 1, the SO2 removal ability of the catalysts and after desulfurization are illustrated in Fig. 2. For 5% from poor to excellent is in the order: NAC < 1% Mn/NAC Mn/NAC, the activated carbons have abundant pores, and < 3% Mn/NAC < 5% Mn/NAC< 10% Mn/NAC < 7% some fine crystals on the carbon surface are observed, NAC. The improvement of desulfurization activity could which may be active components such as Mn oxide species be related to the catalytic oxidation activity of Mn oxide and some inorganic compounds such as SiO2. For 7% species, as well as the textural and chemical properties of Mn/NAC, more fine crystals are observed, and they are rel- the activated carbons because of the different loadings. atively uniformly covered on the surface, leading to a good desulfurization activity. For 10% Mn/NAC, fine crystals lessen and aggregated on the surface. The tiny particles 100 may block the channels in activated carbons, which results in the decrease of desulfurization activity because of diffu- 95 sion resistance. After desulfurization, the shapes of all samples are changed from granular to sheet and more smooth, 90 and the pores of the used catalysts are covered by the by- products, increasing the mass transfer resistance and pre- 85 venting the diffusion of oxygen and SO2 to the active sites NAC resulting in the decrease of SO2 removal efficiency. There- 80 1%Mn/NAC Removal efficency (%) 2 3%Mn/NAC fore, the blockage of pores is considered as a reason for the % SO 5 Mn/NAC % catalytic deactivation, which is consistent with [10, 17]. 75 7 Mn/NAC 10%Mn/NAC N2 adsorption–desorption isotherms at liquid nitrogen

70 temperature and micropore size distribution curves of all 0 100 200 300 400 500 samples are shown in Fig. 3, and the structure parameters Time (min) calculated are listed in Table 2. In Fig. 3 (1), the shapes of FIGURE 1 - The curves of the relation between SO2 removal effi- all isotherms are similar and might be considered as type I, ciency and working time of all samples. which is typical of microporous materials. As shown in Fig. 3 (2), the peaks of all samples are predominantly be- TABLE 1 - Breakthrough sulfur capacities and breakthrough time of all samples. low 0.65 nm. In Table 2, the NAC has a BET surface area of 936 m2/g and total pore volume of 0.480 cm3/g including Breakthrough sulfur capac- Breakthrough time 3 Samples a micropore volume of 0.318 cm /g. After Mn modifica- ity (mg/g) (min) tion, the BET surface area and total volume of Mn/NAC NAC 30 67 are lower than those of NAC, which is probably due to the 1%Mn/NAC 152 280 presence of Mn species that can block the access of N mol- 3%Mn/NAC 175 325 2 ecules into the pores. It indicates that the loadings can in- 5%Mn/NAC 186 339 fluence the texture of samples, resulting in different desul- 7%Mn/NAC 211 391 furization activity because of diffusion of reactive molecu- 10%Mn/NAC 198 369 lar to active sites. Furthermore, 7% Mn/NAC has a surface

FIGURE 2 - SEM of 5% Mn/NAC (a), 7% Mn/NAC (b) and 10% Mn/NAC (c) before SO2 removal, and 5% Mn/NAC (d), 7% Mn/NAC (e) and 10% Mn/NAC (f) after SO2 removal.

(1) Adsorption 300 Desorption

) a

/g 250 b c 3 cm ( 200

150

100 Quantity Adsorbed Adsorbed Quantity 50

0 300

) e 250 d f /g 3 cm ( 200

150

100 Quantity Adsorbed Adsorbed Quantity 50

0 0.0 0.2 0.4 0.6 0.80.0 1.0 0.2 0.4 0.6 0.80.0 1.0 0.2 0.4 0.6 0.8 1.0 Relative Pressure (P/Po) Relative Pressure (P/Po) Relative Pressure (P/Po)

0.3 (2) 0.07 0.15 0.15 c a /g) b 0.06 3 /g) 3 /g)

0.2 3 0.10

0.10 /g)

3 0.05 dV/dW (cm

dV/dW (cm 0.05 dV/dW (cm

0.1 0.05 0.04 0.45 0.50 0.55 0.60 0.65 0.45 0.50 0.55 0.60 0.65 0.45 0.50 0.55 dV/dW (cm Pore Width (nm) Pore Width (nm) Pore Width (nm)

0.0 0.3 0.15 0.15 0.052

d e f /g) /g) 0.10

3 0.050 3 /g) 0.10 0.2 3

/g) 3 0.048 dV/dW (cm

dV/dW (cm 0.05 0.05 dV/dW (cm

0.046 0.1 0.45 0.50 0.55 0.60 0.65 0.45 0.50 0.55 0.60 0.65 0.45 0.50 0.55 0.60 0.65 dV/dW (cmdV/dW Pore Width (nm) Pore Width (nm) Pore Width (nm)

0.0

0.5 1.0 1.5 2.0 0.5 1.0 1.5 2.0 0.5 1.0 1.5 2.0 Pore Width (nm) Pore Width (nm) Pore Width (nm)

FIGURE 3 - N2 adsorption-desorption isotherm curves (1) and H-K pore size distributions (2) of NAC (a), 1% Mn/NAC (b), 3% Mn/NAC (c), 5% Mn/NAC (d), 7% Mn/NAC (e) and 10% Mn/NAC (f).

TABLE 2 - Structure parameters calculated from nitrogen adsorption isotherms.

Samples SBET VMicro VMeso VTotal Average pore size (nm) (m2/g) (cm3/g) (cm3/g) (cm3/g) NAC 936 0.318 0.162 0.480 2.051 1%Mn/NAC 837 0.337 0.071 0.408 1.951 3%Mn/NAC 849 0.306 0.123 0.429 2.020 5%Mn/NAC 884 0.353 0.071 0.424 1.918 7%Mn/NAC 865 0.339 0.097 0.436 2.061 10%Mn/NAC 717 0.286 0.079 0.365 2.037

2 3 area of 865 m /g and a total pore volume of 0.436 cm /g the increase of calcination temperature (>600 °C), Mn2O3 3 including a micropore volume of 0.339 cm /g, and corre- further losses oxygen to form Mn3O4 or MnO [12]. Diffrac- sponds to a good desulfurization performance. The in- tion peaks at 2θ = 20.80°, 26.65° and 50.10° belong to SiO2 creased micropore volume is attributed to the interaction (JCPDs No.77-1060), and some small unknown peaks are between manganese nitrate and activated carbon and de- detected in the samples, which may be due to the impurities composition of manganese nitrate. It may have an ability in the supports. For 7% Mn/NAC, characteristic peaks of to create pores. This is in favor of active component dis- MnO at 2θ = 35.05°, 36.65° and 59.30° (JCPDs No.07- persion, which can enhance the catalytic activities of the 0230) are also observed, and no Mn3O4 or Mn2O3 are de- catalysts. The average pore diameters of the catalysts are tected. For 10% Mn/NAC, the diffraction peaks of MnO at 1.918-2.061 nm, but with the increase of the loading, the 2θ = 35.05°, 36.65°, 59.30°, 74.41°, 73.91° (JCPDs No.07- pore volumes decrease. Especially for the 10% Mn/NAC, 0230) and Mn3O4 at 2θ = 50.34°, 64.43° (JCPDs No.24- 2 the specific surface area decreases to 717 m /g, and the to- 0734) are observed, indicating that MnO and Mn3O4 coex- tal pore volume and micropore volume decrease to 0.365 ist in the catalysts. Comparing with the peak intensity of and 0.286 cm3/g, respectively, which indicates that exces- 10% Mn/NAC, 5% Mn/NAC and 7% Mn/NAC show sive loading can block the pores and is not conducive to the lower peak intensity of MnO and no Mn3O4 phase, which textural properties. This is proven by SEM. Combined with may be a good dispersion of metal oxides, because there is Fig. 1 and Table 1, 10% Mn/NAC exhibits higher desulfu- a mono-layer dispersion capacity when metal oxides dis- rization ability than NAC, indicating that the texture of perse on a support [20]. When loading is lower than the modified activated carbons is not the key factor to affect capacity, the oxides will be in a mono-layer state, but its SO2 removal ability. loading exceeds the mono-layer dispersion capacity, and the surplus oxides will remain as crystalline phase in the 3.3. Surface chemical properties system together with its mono-layer phase, and may be de- The XRD patterns of selected samples before desulfu- tected by XRD. 10% Mn/NAC has high peak intensity and rization are shown in Fig. 4a. All selected samples exhibit narrow band width, revealing crystal agglomeration or a two broad peaks around 20~30° and 40~50° in 2θ range, poor dispersion of Mn oxide particles on the carbon sur- and correspond to the characteristic peaks of carbon struc- face, which is resultant in the decrease of desulfurization ture, showing that different Mn loadings do not destroy the activity. structure of carbons in catalyst preparation. According to [12], more Mn3O4 is formed on the carbon when the calci- The Mn 2p3/2 XPS spectra of selected samples before nation temperature is at 650 °C. For 5% Mn/NAC, small desulfurization are illustrated in Figs. 5a–c. It is reported characteristic peaks of MnO at 2θ = 35.05°and 59.30° that the binding energy of MnO is 640.1~641.4eV [12, 20]. (JCPDs No.07-0230) are detected, indicating that Mn(NO3)2 The Mn 2p3/2 binding energy at about 642.0 eV is observed can be decomposed into MnO in pure N2 atmosphere at for all selected samples, which is a typical value of oxides 3+ 650 °C. It is reported that Mn(NO3)2 is decomposed into containing Mn , such as Mn3O4 or Mn2O3 [21, 22]. Com- NO2 and MnO2 between 200–350 °C, and MnO2 losses ox- bination with the results of XRD, the Mn oxides should be ygen to form Mn2O3 between 500–600 °C [18, 19]. With attributed to Mn3O4. It indicates that Mn oxides with dif-

a b

5%Mn/NAC

7%Mn/NAC 7%Mn/NAC

10%Mn/NAC 10%Mn/NAC

10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 o o 2 ( ) 2 ( )

FIGURE 4 - XRD patterns of 5% Mn/NAC, 7% Mn/NAC and 10% Mn/NAC before (a) and after (b) SO2 removal.

ferent chemical valences coexist in the catalysts calcined at inorganic substances. The characteristic diffraction peaks 650 °C. In Fig. 1, the SO2 removal abilities of 7%Mn/NAC of MnO (2θ = 40.2°, 40.53° and 58.96°; JCPDs No.07- and 10%Mn/NAC are better than that of the other samples, 0230), Mn3O4 (2θ = 18.44°, 18.26°, 35.11°, 34.93°, 35.29°, indicating that the coexistence of oxides containing Mn2+ 43.35°, 43.53° and 43.17°; JCPDs No.24-0734) are still de- 3+ and Mn is very important for SO2 removal. In the SO2 tected, and MnO2 (2θ = 28.61° and 28.79°; JCPDs No.44- oxidation catalytic reaction, it firstly utilizes lattice oxygen 0141) are observed, indicating that the Mn oxide species 2- (O ) which is active oxygen. Mn3O4 is used as an active on the catalyst surface could gradually transform into a species because it can provide lattice oxygen and promote new Mn oxide species. The characteristic peaks of MnSO4 the oxidation of SO2 to SO3, and oxygen in the gas phase in sulfated samples are also not detected by XRD, which is absorbed to the reductive MnO. The adsorbed O2 can be indicates that no Mn oxide species react with generated dissociated and transformed into lattice oxygen, which al- H2SO4 in SO2 removal. In other words, deactivation of cat- lows the catalytic reaction to proceed. Thus, a cyclic redox alysts is not due to the reaction between Mn oxide species mechanism is achieved. Based on the above analysis, it can and generated H2SO4. In Fig. 5d, the Mn 2p3/2 peaks of 5% be concluded that Mn species play a crucial role in SO2 re- Mn/NAC after desulfurization are at 640.6 and 642.9 eV. moval, and different Mn loadings influence the dispersion The former is due to MnO, and the latter belongs to the of Mn species, which leads to the differences of desulfuri- presence of Mn at +4 oxidation state [23]. For 7% Mn/NAC zation activity. after SO2 removal, the peaks at 640.4 and 642.8 eV are observed, indicating that MnO and Mn4+ still coexist in sam- 5%Mn/NAC, 7%Mn/NAC and 10%Mn/NAC after SO2 ples. For 10% Mn/NAC after SO2 removal, a single peak 4+ removal are characterized by XRD and XPS to clarify the of Mn 2p3/2 appears at 642.5 eV, which is assigned to Mn . possible changes of Mn species in SO2 removal. In Fig. 4b, This suggests that Mn3O4, Mn2O3 and MnO are gradually compared to the fresh samples, many diffraction peaks of transferred into MnO2. It is possible that Mn oxide species selected samples after SO2 removal disappear, indicating are more readily oxidized in acidic medium to generate that most of the impurities are eliminated in SO2 removal MnO2, which results in the deactivation of catalysts. because of the generated H2SO4 which can dissolve some

a 640.3 b c 640.0 640.11

642.0 641.9 641.9

670 660 650 640 630 670 660 650 640 630 670 660 650 640 630 642.5 f d 640.6 e 640.4

642.9 642.8

670 660 650 640 630 670 660 650 640 630 670 660 650 640 630

B.E. (eV) B.E (eV) B.E (eV)

FIGURE 5 - Mn 2p3/2 XPS spectra of 5% Mn/NAC, 7% Mn/NAC and 10% Mn/NAC before (a) and after (b) SO2 removal.

Cls and O1s XPS spectra of selected samples before are assigned to the C–C, C–O, C=O and O=C–OH carbon, * SO2 removal are demonstrated in Fig. 6a to elaborate the and π–π , respectively, but their relative contents are effects of surface oxygen-containing functional groups. changed, which indicates that the types of functional The Cls peaks of 5% Mn/NAC, 7% Mn/NAC and 10% groups are not destroyed in SO2 removal. The relative con- Mn/NAC are deconvoluted into 5 peaks, including C–C tent of C–C carbon in the selected samples after SO2 re- (BE = 284.6~284.7 eV) [24], C–O (BE = 285.4~286.3 eV) moval decreases, implying that graphite carbon may be [24], C=O (BE = 287.2~287.9 eV) [25, 26], O=C–OH (BE partially destroyed by generated H2SO4. The relative con- * = 288.7~289.3 eV) [24], and π–π (290.20~290.80 eV) [27, tent of C–O and C=O carbons in 7%Mn/NAC after SO2 28]. The relative content of O=C–OH carbon in 5% removal decrease but in 5%Mn/NAC and 10%Mn/NAC in- Mn/NAC is 5.83%, but with the increase of the loading, the crease compared to the fresh samples, which suggests that relative content of O=C–OH carbon decreases to 4.58% in C–O and C=O are restructured because of electron transfer 10% Mn/NAC. It may be that the decomposition of Mn(NO3)2 in the catalytic oxidation of SO2, and the generated H2SO4 can consume O=C–OH. For C=O carbon, the relative con- has a role to create the functional groups [41]. The O1s tent of 6.30% in 7% Mn/NAC is the highest, which leads XPS spectra of all selected samples after SO2 removal also to a good desulfurization performance, because C=O has show the changes of the oxygen-containing functional Brønsted basic property and is the active center for SO2 ox- groups, proving that the functional groups can participate idation [29]. The C–O carbon in samples is increasing with in the oxidation reaction of SO2. the increase of the loading, and the C–O carbon in FTIR spectra of all samples before and after SO2 re- 10%Mn/NAC increases to 21.65%, indicating that the moval are shown in Fig. 7, to further analyze the functional loadings significantly influence the relative content of ox- groups on the carbon surface. In Fig. 7a, all spectra present ygen-containing functional groups, because the generated a broad absorption band at 3135~3635 cm-1, with a maxi- Mn oxides can react with the functional groups, and also mum near 3400 cm-1, which is assignable to the O–H stretch- influence the relative content of different oxides, which ing of carboxylic groups, hydroxyl groups and adsorbed wa- leads to different desulfurization activity. On the other ter [42]. The broad band at 1300~1000 cm-1 is attributed to hand, the surface oxygen functional groups can affect the the C–O stretching in ethers, alcohols, carboxylic acids and interaction of the metal precursor and carbonaceous sup- phenols [43]. The intensity of the band increases with the in- ports, promoting the metal dispersion (Fig. 2). crease of the loading, but for 10%Mn/NAC, the intensity of As shown in Fig. 6a, the O1s peaks at 530.1~530.6, C–O is lowered, indicating that excessive Mn(NO3)2 is de- 531.7~532.4, 533.2~533.8 and 534.8~535.6 eV are at- composed and the generated MnO is oxidized by C–O be- tributed to oxygen in oxides [30, 31], C=O [32, 33], C–O cause of electron transfer, and then, forms more Mn3O4. The [34] and absorbed water [35, 36], respectively, which could point has been explained by XPS and XRD. Furthermore, be caused by the surface interaction of Mn(NO3)2 and acti- the spectra of all samples have several other bands, which vated carbon during the calcination process. The decompo- may be ascribed to aromatic ring C=C stretching vibration -1 sition of Mn(NO3)2 can release NO2 and MnO or Mn3O4. (1400~1700 cm ) [43], and C=O stretching vibration at −1 NO2 is an acid gas, which may etch the surface of activated 1558 cm of carbonyl groups [44]. The intensity of C=O carbon and leave defects, dislocations or discontinuities for is the highest in 7%Mn/NAC, which is in agreement with chemisorbing oxygen when molecular oxygen is intro- the result obtained from XPS analysis, corresponding to a duced into the surface of activated carbon. Mn oxides also good desulfurization activity. Based on the above analysis, play a role in oxygen transfer and influence the content of one can conclude that different Mn loadings could influ- oxygen-containing functional groups. 10%Mn/NAC has ence the relative content of functional groups because of more oxygen in oxides, which is due to the high loading. For strong interaction between the generated Mn oxides and the C=O oxygen, 7%Mn/NAC is higher than 5%Mn/NAC functional groups. and 10%Mn/NAC, corresponding to better desulfurization, As shown in Fig. 7b, C=O, C=C, C–O and –OH are because C=O with Brønsted basic property is the active also observed after SO2 removal, and the intensities of all center of catalytic oxidation of SO2 [37]. And the C=O can functional groups are changed, which indicates that the strengthen interaction between O and Mn during the cata- functional groups could participate in the oxidation reac- lyst preparation, and form a Mn oxide phase, which leads -1 tion of SO2. A new peak appears at about 590 cm , which to achieve higher dispersion of Mn species [38]. C=O and is attributed to the S=O and S–O group [33], indicating that C–O–oxygen in catalysts may provide an electron-rich sur- 2- -1 SO4 or HSO4 exists in the catalysts after desulfurization. face, which can transfer electrons to the oxygen chemically In Fig. 8, the S 2p3/2 spectra of 5%Mn/NAC show two bind- bonded to the carbon surface [39], and Mn oxide species ing energy peaks at 168.3 and 169.9 eV, and they are at- can also transfer electrons to adsorbed molecular oxygen, − tributed to H2SO4 and sulfate, respectively [45]. Three resulting in the formation of O2 [40]. This “ionic” oxygen peaks in 7%Mn/NAC are observed at 163.4, 168.3 and can improve SO2 oxidation. 170.4 eV corresponding to elemental sulfur, H2SO4 and 2- Cls and O1s XPS spectra of 5%Mn/AC, 7%Mn/AC and SO4 , respectively [33, 45]. For 10%Mn/NAC, three peaks 10%Mn/AC after SO2 removal are illustrated in Fig. 6b. The are also observed at 164.7, 167.8 and 169.6 eV, which still 2- peaks at around 284.6, 285.8, 287.3, 288.9 and 290.7 eV belong to elemental sulfur, H2SO4 and SO4 , respectively.

a C1s 5%Mn/NAC 284.6;C-C;72.11% C1s 7%Mn/NAC 284.6;C-C;67.22% C1s 10%Mn/NAC 284.6;C-C;64.94%

286.0;C-O;12.48% 285.7;C-O;18.25% 285.7;C-O;21.65%

287.2;C=O;5.78% 287.2;C=O;6.30% 287.2;C=O;4.99%

288.9;-COOH;5.83% 289.0;-COOH;4.90% 288.8;-COOH;4.58%

290.8;-;3.80% 290.7;-;3.33% 290.6;-;3.84%

300 295 290 285 280 275 300 295 290 285 280 275 300 295 290 285 280 275

O1s 5%Mn/NAC O1s 7%Mn/NAC O1s 10%Mn/NAC

533.3;C-O;38.01% 532.0;C=O;27.80% 533.2;C-O;36.23% 533.2;C-O;33.54% 531.9;C=O;29.37% 532.0;C=O;34.20% 530.4;Oxides/-OH; 534.7; 31.89% 534.6; absorbed water; 530.2;Oxides/-OH; 534.4; 530.3;Oxides/-OH; absorbed water; 17.33% 15.30% absorbed water; 14.44%

15.15% 6.77%

540 535 530 525 540 535 530 525 540 535 530 525 B.E (eV) B.E (eV) B.E (eV) b C1s 5%Mn/NAC 284.6 ;C-C; 61.07% C1s 7%Mn/NAC 284.6; C-C; 67.17%) C1s 10%Mn/NAC 284.6; C-C; 60.66%

285.7;C-O; 24.63% 285.9; C-O; 24.92% 285.8; C-O; 17.76% 287.2; C=O; 6.65% 287.4; C=O; 6.12% 287.3; C=O; 5.91% 288.9; -COOH;3.93% 288.9; -COOH; 5.02% 288.9; -COOH; 4.45% 290.7; -; 3.82% 290.5; -;3.52% 290.8;-; 4.14%

300 295 290 285 280 275 300 295 290 285 280 275 300 295 290 285 280 275 O1s 5%Mn/NAC O1s 7%Mn/NAC O1s 10%Mn/NAC

532.9; C-O; 52.65% 533.4; C-O; 43.78% 531.8; C=O; 27.98% 531.8; C=O; 25.55% 532.0; C=O; 34.73% 532.8; C-O; 38.91%

530.5; Oxides/-OH; 534.3; 24.61% absorbed H O; 530.9; Oxides/-OH; 534.2; 2 11.40% 10.40% 534.9; absorbed H O; absorbed H O; 530.5; Oxides/-OH; 2 2 8.50% 13.21% 8.28%

540 535 530 525 540 535 530 525 540 535 530 525 B.E (eV) B.E. (eV) B.E. (eV)

FIGURE 6 - Cls and O1s XPS spectra of 5% Mn/NAC, 7% Mn/NAC and 10% Mn/NAC before (a) and after (b) SO2 removal.

a b NAC C=O NAC O-H C=O C=C S=O, S-O O-H C-O C=C C-O 1%Mn/NAC 1%Mn/NAC

3%Mn/NAC

5%Mn/NAC 5%Mn/NAC

7%Mn/NAC 7%Mn/NAC

10%Mn/NAC 10%Mn/NAC

3500 1500 1000 4000 3500 1500 1000 -1 -1 Wave number (cm ) Wave number (cm )

FIGURE 7 - FTIR spectra of NAC, 1% Mn/NAC, 3% Mn/NAC, 5% Mn/NAC, 7% Mn/NAC, 10% Mn/NAC before (a) and after (b) SO2 removal.

a 168.3 b 168.3 c 169.6

169.9 167.8 170.4 164.7

163.4

176 172 168 164 160 176 172 168 164 160 176 172 168 164 160 BE(eV) BE(eV) BE(eV)

FIGURE 8 - S2p XPS spectra of 5% Mn/NAC (a), 7% Mn/NAC (b), 10% Mn/NAC (c) after SO2 removal.

When combining the Mn 2p3/2 XPS spectra and XRD 7%Mn/NAC. Different loadings could influence the tex- of samples after SO2 removal, MnSO4 is not observed. The ture of samples, leading to different desulfurization activity 2- SO4 is derived from the oxidation reaction of SO2. In this because of diffusion resistance. Mn3O4 is an active species system, SO2 is firstly adsorbed on active sites and reacts because it can produce active oxygen required for the cata- with the oxygen species to form SO3. The formed SO3 can lytic oxidation of SO2, and the coexistence of MnO and be hydrated with H2O and generate H2SO4. Some H2SO4 is Mn3O4 plays a crucial role during the SO2 removal, but both washed by excess amount of water to recover the SO2 ad- are gradually transferred into MnO2 in the SO2 removal, sorption sites, and another part is retained in the pores of which results in the deactivation of catalysts. The loadings activated carbon, resulting in the deactivation of catalysts can influence the relative content of surface oxygen-containing functional groups, because generated Mn oxide species can be oxidized by C=O and C–O. Oxygen-containing 4. CONCLUSIONS functional groups can participate in the oxidation reaction of SO2 but could be restructured by generated H2SO4. Some of SO2 removal capacity of activated carbon can be im- it is washed out by excess amount of water to recover the proved strikingly by different Mn loadings, and that of sam- SO2 adsorption sites, but some H2SO4 is still retained in the ples from poor to excellent is as follows: NAC< 1% pores of activated carbon, which is another reason for the Mn/NAC < 3%Mn/NAC < 5%Mn/NAC < 10%Mn/NAC < deactivation of Mn supported on activated carbons.

ACKNOWLEDGEMENTS [14] Hu, Y., Chu, W. and Shi, L.M. (2008) Effects of carrier and Mn loading on supported manganese oxide catalysts for catalytic combustion of methane, Journal of Natural Gas Chemis- This work is financially supported by the National Na- try 17, 159-164. ture Science Youth Fund of China (No.5110828) and the [15] Lü, G., Wu, H., Song, C.L. and Bin, F. (2011) Mn/ZSM-5 Cat- Sichuan Provincial Science and Technology Agency Pub- alysts and the Properties of Sel- ective Catalytic Reduction, lic Research Projects (No.2012GZX0028). We would like Journal of Engineering Thermophysics 32, 1597-1600. to thank the Analytical & Testing Center of Sichuan Uni- [16] Davini, P. (2003) Flue gas desulphurization by activated car- versity for the characterization of the catalysts. bon fibers obtained from polyacrylonitrile by-product, Carbon 41, 277-284. The authors have declared no conflict of interest. [17] Fang, H.B., Zhao, J.T., Fang, Y.T., Huang, J.J. and Wang, Y. (2013) Selective oxidation of hydrogen sulfide to sulfur over activated carbon-supported metal oxides, Fuel 108, 143-148. REFERENCES [18] Gallagher, P.K., Schrey, F. and Prescott, B. (1971) The thermal decomposition of aqueous manganese (II) nitrate solution, [1] Tseng, H. H. and Wey, M.Y. (2006) Effects of acid treatments Thermochimica Acta 2, 405-412. of activated carbon on its physiochemical structure as a sup- [19] Zhou, F., Zhao, X.M., Yuan, C.G. and Xu, H. (2007) Synthesis port for copper oxide in DeSO2 reaction catalysts, Chemo- of γ-MnOOH nanorods and their isomorphous transformation sphere 62, 756-766. into β-MnO2 and α-Mn2O3 nanorods, Journal of Materials Sci- ence 42, 9978-9982. [2] Grzybek, T. (1992) The influence of the addition of cobalt nickel, manganese and vanadium to active carbons on their ef- [20] Gao, X., Liu, J., Zhang, Y., Luo, Z.Y. and Cen, K.F. (2011) ficiency in SO2 removal from stack gases, Fuel 71, 1303-1308. Physicochemical properties of metal-doped activated carbons and relationship with their performance in the removal of SO [3] Xiao, Y., Liu, Q.Y., Liu, Z.Y., Huang, Z.G., Guo, Y.X., and 2 and NO, Journal of hazardous materials 188, 58-66. Yang, J.L. (2008) Roles of lattice oxygen in V2O5 and activated coke in SO2 removal over coke-supported V2O5 cata- [21] Fernández, J., Caballero, A., Villegas, M., Khatib, S., Banares, lysts, Applied Catalysis B: Environmental 82, 114-119. M., Fierro, J.L., Costa-Kramer, J., Lopez-Ponce, E., Martín- González, M.S. and Briones, F. (2006) Structure and mag- [4] Mochida, I., Miyamoto, S., Kuroda, K., Kawano, S., Sakani- netism in the Zn–Mn–O system: A candidate for room temper- shi, K., Korai, Y., Yasutake, A. and Yoshikawa, M. (1999) ature ferromagnetic semiconductor, Journal of the European Oxidative fixation of SO2 into aqueous H2SO4 over a pitch- Ceramic Society 26, 3017-3025. based active carbon fiber above room temperature, Energy & fuels 13, 374-378. [22] Moulder, J.F., Chastain, J. and King, R.C. (1995) Handbook of X-ray photoelectron spectroscopy: a reference book of [5] Yan, Z., Wang, J.P., Zou, R.Q., Liu, L.L., Zhang, Z.T. and standard spectra for identification and interpretation of XPS Wang, X.D. (2012) Hydrothermal Synthesis of CeO2 Nano- data, Physical Electronics Eden Prairie, MN. particles on Activated Carbon with Enhanced Desulfurization Activity, Energy & Fuels 26, 5879-5886. [23] Nesbitt, H. and Banerjee, D. (1998) Interpretation of XPS Mn (2p) spectra of Mn oxyhydroxides and constraints on the [6] Davini, P. (2001) The effect of certain metallic derivatives on mechanism of MnO2 precipitation, American Mineralogist 83, the adsorption of sulphur dioxide on active carbon, Carbon 39, 305-315. 419-424. [24] Ling, L., Li, K., Liu, L., Miyamoto, S., Korai, Y., Kawano, S. [7] Ma, J.R., Liu, Z.Y, Liu, S.J. and Zhu, Z.P. (2003) A regenera- and Mochida, I. (1999) Removal of SO2 over ethylene tar pitch ble Fe/AC desulfurizer for SO2 adsorption at low tempera- and cellulose based activated carbon fibers, Carbon 37, 499-504. tures, Applied Catalysis B: Environmental 45, 301-309. [25] Li, K.X., Ling, L.C., Lu, C.X., Qiao, W.M., Liu, Z.Y., Liu, L. [8] Tseng, H.H. and Wey, M.Y. (2004) Study of SO2 adsorption and Mochida, I. (2001) Catalytic removal of SO2 over ammo- and thermal regeneration over activated carbon-supported nia-activated carbon fibers, Carbon 39, 1803-1808. copper oxide catalysts, Carbon 42, 2269-2278. [26] Polovina, M., Babić, B., Kaluderović, B. and Dekanski, A. [9] Atakül, H., Wakker, J.P., Gerritsen, A.W. and Van den Berg, (1997) Surface characterization of oxidized activated carbon P.J. (1995) Removal of H2S from fuel gases at high tempera- cloth, Carbon 35, 1047-1052. tures using MnO/γ-Al2O3, Fuel 74, 187-191. [27] Rey, A., Faraldos, M., Bahamonde, A., Casas, J.A., Zazo, J.A. [10] Bashkova, S., Baker, F.S., Wu, X.X., Armstrong, T.R. and and Rodríguez, J.J. (2008) Role of the activated carbon surface Schwartz, V. (2007) Activated carbon catalyst for selective ox- on catalytic wet peroxide oxidation, Industrial & Engineering idation of hydrogen sulphide: on the influence of pore struc- Chemistry Research 47, 8166-8174. ture, surface characteristics, and catalytically-active nitrogen, Carbon 45, 1354-1363. [28] Zhang, G.X., Sun, S.H., Yang, D.Q., Dodelet, J.P. and Sacher, E. (2008) The surface analytical characterization of carbon fi- [11] Tang, Q.H., Huang, X.N., Chen, Y.T., Liu, T. and Yang, Y.H. bers functionalized by H2SO4/HNO3 treatment, Carbon 46, 196- (2009) Characterization and catalytic application of highly dis- 205. persed manganese oxides supported on activated carbon, Jour- [29] Zhao, X.S., Cai, G.Y., Wang, Z.Z., Wang, Q.X., Yang, Y.H. nal of Molecular Catalysis A: Chemical 301, 24-30. and Luo, J.S. (1994) Influences of surface functional groups [12] Qu, Y.F., Guo, J.X., Chu, Y.H., Sun, M.C. and Yin, H.Q. on catalytic activity over activated carbon catalysts for sulfur (2013) The Influence of Mn Species on the SO2 Removal of dioxide removal from flue gases, Applied Catalysis B: Envi- Mn-Based Activated Carbon Catalysts, Applied Surface Sci- ronmental 3, 229-238. ence 28, 425-431. [30] Smirnov, M.Y., Kalinkin, A., Pashis, A., Sorokin, A., Noskov, [13] Elkhalifa, E.A. and Friedrich, H.B. (2010) Oxidative dehydro- A., Bukhtiyarov, V., Kharas, K. and Rodkin, M. (2003) Com- genation of n-octane using vanadium-magnesium oxide cata- parative XPS Study of Al2O3 and CeO2 Sulfation in Reactions lysts with different vanadium loadings, Applied Catalysis A: with SO2, SO2 + O2, SO2 + H2O, and SO2 + O2 + H2O, Kinetics General 373, 122-131. and catalysis 44, 575-583.

[31] Ismail, H.M., Cadenhead, D.A. and Zaki, M.I. (1996) Surface reactivity of iron oxide pigmentary powders toward atmospheric components: XPS and gravimetry of oxygen and water vapor adsorption, Journal of colloid and interface science 183, 320-328.

[32] Hontoria-Lucas, C., Lopez-Peinado, A., López-González, J.D.D., Rojas-Cervantes, M. and Martin-Aranda, R. (1995) Study of oxygen-containing groups in a series of graphite oxides: physical and chemical characterization, Carbon 33, 1585- 1592. [33] Terzyk, A.P. (2001) The influence of activated carbon surface chemical composition on the adsorption of acetaminophen (paracetamol) in vitro: Part II. TG, FTIR, and XPS analysis of carbons and the temperature dependence of adsorption kinetics

at the neutral pH, Colloids and Surfaces A: Physicochemical and Engineering Aspects 177, 23-45. [34] De la Puente, G., Pis, J., Menendez, J. and Grange, P. (1997) Thermal stability of oxygenated functions in activated carbons, Journal of analytical and applied pyrolysis 43, 125-138. [35] Burg, P., Fydrych, P., Cagniant, D., Nanse, G., Bimer, J. and Jankowska, A. (2002) The characterization of nitrogen-en-

riched activated carbons by IR, XPS and LSER methods, Car- bon 40, 1521-1531. [36] Biniak, S., Szymański, G., Siedlewski, J. and Światkowski, A. (1997) The characterization of activated carbons with oxygen and nitrogen surface groups, Carbon 35, 1799-1810. [37] Zhao, X.S., Cai, G.Y., Wang, Z.Z., Yang, Y.H., Wang, Q.X., Luo, J.S., Sheng, S.S. and Chen, H.R. (1993) Studies on the

active center of carbon catalysts for SO2 conversion, Environ- mental Chemistry 3, 194-199. [38] Calafat, A., Laine, J., López-Agudo, A. and Palacios, J. (1996) Effect of surface oxidation of the support on the thiophene hy- drodesulfurization activity of Mo, Ni, and NiMo catalysts supported on activated carbon, Journal of Catalysis 162, 20-30.

[39] Brown, T.C., Lear, A.E. and Haynes, B.S. (1992) Oxygen

chemisorption on carbon, in: Symposium (International) on Combustion, Elsevier 24, 1199-1206. [40] Stöhr, B., Boehm, H. and Schlögl, R. (1991) Enhancement of the catalytic activity of activated carbons in oxidation reactions by thermal treatment with ammonia or hydrogen cyanide and observation of a superoxide species as a possible intermediate, Carbon 29, 707-720.

[41] Guo, J.X., Liang, J., Chu, Y.H., Sun, M.C., Yin, H.Q. and Li, J.J. (2012) Desulfurization activity of nickel supported on acid-treated activated carbons, Applied Catalysis A: General 421, 142-147. [42] Tazibet, S., Boucheffa, Y., and Lodewyckx, P. (2013) Heat Received: August 29, 2014 treatment effect on the textural, hydrophobic and adsorptive Accepted: November 27, 2014 properties of activated carbons obtained from olive waste, Mi- croporous and Mesoporous Materials 170, 293-298.

[43] Gomez-Serrano, V., Pastor-Villegas, J., Perez-Florindo, A., Duran-Valle, C. and Valenzuela-Calahorro, C. (1996) FT-IR CORRESPONDING AUTHOR study of rockrose and of char and activated carbon, Journal of analytical and applied pyrolysis 36, 71-80. Jiaxiu Guo [44] Przepiórski, J. (2006) Enhanced adsorption of phenol from College of Architecture and Environment water by ammonia-treated activated carbon, Journal of hazard- Sichuan University ous materials 135, 453-456. Sichuan Chengdu 610065 [45] Cellier, C.M., Vromman, V., Ruaux, V., Gaigneaux, E.M. and P.R. CHINA Grange, P. (2004) Sulfation mechanism and catalytic behavior of manganese oxide in the oxidation of methanethiol, The Phone: +86-28-85403016 Journal of Physical Chemistry B 108, 9989-10001. E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1291 - 1301

HEAVY METAL AND MINERAL LEVELS OF SOME FRUIT SPECIES GROWN AT THE ROADSIDE IN THE EAST PART OF TURKEY

Mücahit Pehluvan1,*, Metin Turan2, Tuncay Kaya1 and Uğur Şimsek1

1 Igdir University, Faculty of Agriculture, Department of Horticulture, 76000 Igdir, Turkey 2 Yeditepe University, Faculty of Engineering and Architecture, Department of Genetics and Bioengineering, 34755 Kayisdagi, Istanbul-Turkey

ABSTRACT The consumption of fruit is well known to have beneficial effect on human health. Fruits have become a part of The aim of this study was to determine mineral and people’s daily diets more and more with the improvement heavy metal levels of some temperate fruit species such as of living conditions in many parts of the world. However, sweet cherry, black mulberry, white mulberry, apricot, ap- consumers are not aware of the content of the fruits they ple, plum, peach, pear, hawthorn and rosehip grown in eat, because harmful substances in fruit normally cannot be Aras Valley, located in the east part of Turkey. The con- perceived by tasting or smelling them. Exceeding permis- centration of mineral and heavy metal levels in the leaves sible limits of heavy metal level in fruits do great damage and fruits of fruit species were detected by ICP-OES. The to the human health [6]. Therefore, increasing the con- content of heavy metals in fruit samples were determined sumption of fruits may bring about not health benefits but in the range of 50.16-90.11, 9.45-82.15, 12.69-65.24, risks [7-9]. Recently, it has been reported that fruits con- 10.24-30.24, 1.12-5.89, 1.62-3.42, 0.36-1.36 and 0.01-0.09 taminated with heavy metals in a few regions of Turkey [8, mg/kg for Fe, Cu, Mn, Zn, Cd, Pb, Ni and Cr, respectively. 10, 11] and in different parts of the word [9, 12-14]. The highest content of Fe, Cu, Mn, Zn, Cd, Pb, Ni and Cr The study area (Aras Valley), takes an important place in fruit samples were detected in Plum (yellow), Sweet in the production of temperate fruit species such as apricot, cherry, Plum (red), Black mulberry, Plum (sarali), Plum apple, pear, plum, sweet cherry, mulberry and peach in the (yellow), White mulberry and Plum (red), respectively. All eastern Anatolia region of Turkey, has not yet been inves- of fruit samples were found to be contaminated with high tigated in terms of heavy metal contaminations in fruit. It levels of Cu, Cd, Pb. Moreover, the contents of Zn in fruits has been estimated that fruit species may be polluted by were higher levels than the permissible limits of chemical fertilizers, pesticides, vehicle emissions and pos- FAO/WHO. However, the contents of Ni, Cr, Fe and Mn sible sources of the heavy metal pollution in the study area. did not appear to reach pollution levels in the fruit samples. Therefore, the aim of the present investigation was to The results illustrated that a strong relationship exists be- determine the level of trace elements and heavy metal con- tween leaf and fruit samples with regards to all minerals in tamination in the leaves as well as edible part of some fruit the fruit species and cultivars. species in Aras Valley of Turkey.

KEYWORDS: Temperate fruit species, heavy metal pollution, minerals 2. MATERIAL AND METHODS

1. INTRODUCTION 2.1. Study Area The study was carried out in Aras Valley of Turkey in Unorganized urbanization and industrialization has 2012. The sampling points vary from 858 to 1216 m altitudes. constantly introduced large quantities of pollutants into Four sampling sites (Kağızman, İncesu, Iğdır, Melekli) with ecosystems and has brought about heavy metals contami- 18 sampling points situated along the D-080 road from Kağız- nation in the environment [1, 2]. Heavy metals have a sig- man to Iğdır were selected for this study (Table 1 and Fig- nificant toxicity for human, animals, microorganisms and ure 1). The road, carrying about 4000 motor vehicles per plants [1]. The contamination of fruit with heavy metals day, is the main road connecting Kağızman to Iğdır and the poses a serious threat to its quality and safety [3-5]. road is also used international connecting Turkey to Nakhi- chevan and Iran [15]. It was reported that some soil prop- * Corresponding author erties such as salinity (EC), organic matter and pH varied

from 3.0 to 4.0 µmhos/cm, 0.8% to 1.6% and 7.5 to 8.5, ature, average humidity, total rainfall, average wind speed respectively in the study area [16]. Some climatic factors and wind direction were 12.1 oC, 55.6%, 256 mm, 1.2 m/ of the study area, the year, 2012 and data obtained from sec and WNW (west-north-west), respectively in the study long-terms 1975-2005 are given Table 2. Average temper- year, 2012 [17, 18].

FIGURE 1 - The map of study area with sampling points and road.

TABLE 1 - Some descriptive information on fruit species and study area

Samp. Botanical name General name Cultivar name Harvest Dates Site name Longitude Altitude No Latitude 1 P.avium Sweet cherry 900-Ziraat 06.17.2012 Iğdır N 39o 56.295’ 873 E 44o 00.786’ 2 M.nigra Black mulberry Wild 07.01.2012 Iğdır N 39o 56.185’ 873 E 44 o 00.689’ 3 M.alba White mulberry Wild 07.01.2012 Iğdır N 39 o 56.214’ 873 E 44 o 00.711’ 4 P.armeniaca Apricot Şalak 06.30.2012 Iğdır N 39 o 56 410’ 872 E 44 o 00 595’ 5 P.armeniaca Apricot Teberze 07.02.2012 Iğdır N 39 o 56.209’ 872 E 44 o 00.723’ 6 M.domestica B. Apple Grany Smith 09.28.2012 Iğdır N 39 o 56.256’ 871 E 44 o 00.644’ 7 P.domestica Plum wild (red) 07.27.2012 Melekli N 39 o 57.178’ 855 E 44 o 07.088’ 8 P.domestica Plum wild (yellow) 07.27.2012 Melekli N 39 o 57.087’ 856 E 44 o 06.931’ 9 P.persica Peach Zeferan 07.27.2012 Melekli N 39 o 56.818’ 863 E 44 o 06.625’ 10 P.domestica Plum Sarali 07.27.2012 Melekli N 39 o 57.117’ 858 E 44 o.07.003’ 11 P.domestica Plum Karali 07.27.2012 Melekli N 39 o 57.117’ 858 E 44 o 07.003’ 12 P.domestica Plum wild (yellow) 08.10.2012 Kağızman N 40 o 07.718’ 1416 E 43 o 07.280’ 13 P.domestica Plum wild (red) 08.10.2012 İncesu N 40 o 06.613’ 1042 E 43 o 26.338’ 14 P. communis Pear Nenezil 09.08.2012 Kağızman N 40 o 10.609’ 1180 E 43 o 08.157’ 15 P.armeniaca Apricot Şalak 08.10.2012 Kağızman N 40 o 10.496’ 1178 E 43 o 08.345’ 16 P. armeniaca Apricot wild (Zerdali) 08.10.2012 Kağızman N 40 o 10.622’ 1216 E 43 o 08.580’ 17 C.oxyacantha Hawthorn Wild 09.08.2012 Kağızman N 40 o 10.592’ 1200 E 43 o 09.303’ 18 R.canina Rosehip Wild 09.08.2012 Kağızman N 40 o 10.463’ 1180 E 43 o 09.550’

TABLE 2 - Some climatic data of study area in 2012 and in long-term (1975-2005)

2012 long-term climatic data (1975-2005) Temp. Hum. Rain. Wind S. Wind D. Temp. Hum. Rain. Wind S. Wind D. Months OC % mm m/sec OC % mm m/sec January -2,6 66,3 12,8 0,9 WNW 0,3 59,5 0,0 1,1 WNW February -0,1 59,9 15,9 1,2 WNW -3,1 63,0 12,5 1,1 WNW March 6,2 51,8 24,3 1,6 WNW 3,6 47,1 13,5 1,9 WNW April 13,3 49,4 34,5 1,6 WNW 16,1 43,1 16,2 1,4 WNW May 17,3 51,2 47 1,4 WNW 19,7 51,3 57,4 1,5 WNW June 21,9 47,3 33,7 1,5 WNW 25,0 37,6 26,7 1,8 WNW July 25,8 44,7 13 1,4 WNW 26,1 43,8 23,0 1,6 WNW August 25,1 46,7 8,6 1,3 WNW 27,6 36,0 0,6 1,5 WNW September 19,8 51 9,6 1,1 WSW 21,5 47,7 29,3 1,1 WSW October 12,6 62,2 24,5 0,9 WNW 15,4 62,5 14,5 0,7 WNW November 5,5 69,6 19,4 0,9 WNW 8,4 75,0 20,7 0,6 WNW December 0,5 67,2 12,7 1,1 WNW 1,4 74,5 25,8 0,9 WNW Average 12.1 55.6 -. 1.2 WNW 13.5 53.4 - 1.27 WNW Total - - 256 - - - - 240 - -

2.2. Samples collection and preparation for anaylses 0, 5, 10, 20, and 40 for P, K, Ca, Mg, Na, S; 0, 1, 3, 5 and In each sampling point (18 point), three fruit trees were 10 for Fe, Cu, Mn, Zn, B and Mo; 0.1, 0.3, 0.5, and 1.0 for selected in terms of the age (plum, apricot, peach and sweet Cd, Pb, Ni, Cr. Blank solutions were prepared in the same cherry, apple and pear were about ten years old) and grow- media. Calibration ranges were adjusted for each element ing status homogenously (black mulberry, white mulberry, according to the expected concentration range. The preci- hawthorn and rosehip). Fruit samples which were at fully sion of the method was evaluated using the relative stand- maturity stage, were taken about 1 to 4 kg depending on ard deviation of repeated determinations of the analytes. fruit species and leaf samples were collected a sufficient The sensitivity of the method with respect to each metal amount between 6 June 2012 and 28 September 2012 (Ta- was evaluated using the resulted slope of the calibration ble 1). The fresh fruit and leaf samples were put into clean curves. The correlation coefﬁcients for all curves (digest polyethylene bags and transported to the laboratory. The solutions and diluted samples) were more than 0.990. Data collected samples were cleaned by using dry air to remove were presented as mg/g for P, K, Ca and Mg and as mg/kg the air borne pollutants, and fruit samples fragmented with for Fe, Cu, Mn, Zn, Cd, Pb, Ni and Cr. the clean plastic spoon and knife and dried at ambient temperature in air conditions. After drying, the seeds were re- 2.4. Data analysis moved from dried fruits, but mulberry. Fruit and leaf sam- Descriptive statistics for each trait were expressed as ples were oven-dried at 68 oC for 48 and ground to pass Mean ± SE (standard errors). Data analyzed using one-way through 1 mm sieve. Fifty-four fruit and leaf samples were ANOVA. The research was performed by three replicates stored in 4 oC until the analyses. for each sampling point [21].

2.3. Analyses of plant tissue heavy metal levels Plant tissues weighed about 0.5 g for fruit and 0.5 g for 3. RESULTS AND DISCUSSION leaf samples, digested with concentrated HNO3-H2O2 acid mixture (2:3 v/v) in three step (first step; 145ºC, 75% ra- The concentration of macro-elements, micro-elements dio-frequency power (RF), 5 min; second step; 180ºC, and heavy metals in leaf and fruit samples were determined 90%RF, 10 min and third step; 100ºC, 40%RF, 10 min) in on the basis of dry weight and are summarized in Tables 3- microwave (Bergof Speedwave Microwave Digestion 5. Leaf minerals of samples were not discussed but these re- Equipment MWS-2) [19]. Contents of P, K, Ca, Mg, Fe, sults were used for correlation analyses between leaf and Cu, Mn, Zn, Cd, Pb, Ni and Cr in plant tissue were deter- fruit samples. The literatures discussed below were on the mined by using an Inductively Couple Plasma, Optical basis of dry weight. However, only Grembecka and Szefer Emission Spectrophotometer (Perkin-Elmer, Optima 2100 [22] and Krejpcio et al. [23] gave the results on the basis of DV, ICP/OES, Shelton, CT 06484-4794, USA) [20]. the fresh weight for minerals concentration of sweet cherry, pear, apple and plum in their study. The water content of Standard stock solution (1000 mg L−1 ) of each element fruits was also given by Grembecka and Szefer [22] in their was used for the preparation of standard solution in 2% study. Therefore, we recalculated these data for turning on HNO3. The standard stock solution was diluted with 2% the basis of fresh weight into on the basis of dry weight and HNO3 and prepared the following concentrations (mg L−1): compared to our results.

TABLE 3 - Macro-element contents of fruit species in leaves and edible part (mg/g dry weight)

Sample P K Ca Mg Numbers Sample name Leaf Fruit Leaf Fruit Leaf Fruit Leaf Fruit 1 Sweet cherry 2.23±0.05* 1.81±0.03 24.31±0.39 15.78±0.27 13.99±0.41 11.20±0.20 3.41±0.10 2.42±0.04 2 Black mulberry 2.34±0.07 1.84±0.02 20.02±0.27 14.70±0.18 12.64±0.40 10.36±0.15 2.40±0.06 1.65±0.03 3 White mulberry 2.02±0.05 1.75±0.03 17.85±0.36 13.25±0.23 13.69±0.42 11.69±0.21 2.21±0.06 1.85±0.03 4 Apricot 2.99±0.03 2.01±0.04 5.81±0.05 7.85±0.15 11.23±0.40 8.75±0.15 0.51±0.04 0.62±0.01 5 Apricot 2.72±0.04 2.27±0.03 7.21±0.12 9.24±0.10 13.22±0.42 10.23±0.12 0.78±0.02 0.71±0.01 6 Apple 2.23±0.07 1.62±0.03 22.11±0.36 10.23±0.25 12.76±0.40 12.07±0.23 4.20±0.08 2.81±0.09 7 Plum 1.51±0.03 1.45±0.03 25.44±0.54 15.42±0.24 28.91±0.46 22.14±0.39 9.22±0.18 6.87±0.13 8 Plum 1.81±0.04 1.23±0.03 22.15±0.39 12.36±0.28 27.11±0.45 20.14±0.48 8.32±0.18 5.84±0.13 9 Peach 2.10±0.04 1.45±0.03 23.91±0.42 14.58±0.26 26.57±0.45 21.97±0.35 7.21±0.14 5.70±0.09 10 Plum 1.92±0.04 1.51±0.03 23.12±0.40 14.62±0.25 26.13±0.43 18.66±0.35 8.02±0.13 5.21±0.09 11 Plum 2.16±0.06 1.52±0.01 25.70±0.42 15.50±0.16 27.15±0.47 20.16±0.13 7.81±0.13 5.41±0.04 12 Plum 2.00±0.06 1.48±0.02 22.15±0.37 16.12±0.27 25.42±0.46 21.70±0.36 6.86±0.14 5.37±0.09 13 Plum 1.72±0.07 1.25±0.01 24.12±0.85 17.42±0.18 26.15±0.43 21.30±0.18 8.71±0.15 6.49±0.05 14 Pear 1.43±0.03 1.27±0.03 15.78±0.28 12.60±0.22 22.12±0.46 17.56±0.29 5.41±0.09 4.51±0.08 15 Apricot 3.01±0.23 2.01±0.02 4.87±0.40 6.41±0.05 11.96±0.39 11.26±0.08 0.57±0.05 0.72±0.01 16 Apricot 2.41±0.04 1.75±0.03 3.69±0.07 4.02±0.09 10.32±0.56 10.36±0.20 0.43±0.01 0.82±0.02 17 Hawthorn 2.41±0.04 1.65±0.01 16.14±0.44 9.48±0.08 24.16±0.39 18.55±0.16 2.82±0.06 1.87±0.01 18 Rosehip 2.81±0.05 1.99±0.03 18.15±0.29 10.25±0.17 32.15±0.43 21.36±0.39 3.22±0.10 2.27±0.03 Mean 2.22±0.07 1.66±0.04 17.94±1.00 12.21±0.50 20.35±.1.00 16.08±0.68 4.56±0.42 3.40±0.30 Maximum 3.01±0.23 2.27±0.03 25.70±0.42 17.42±0.18 32.15±0.43 22.14±0.39 9.22±0.18 6.87±0.13 Minimum 1.43±0.03 1.23±0.03 3.69±0.07 4.02±0.09 10.32±0.56 8.75±0.15 0.43±0.01 0.62±0.01 *Data are presented as means±SE (standard errors), n=3 replicates

TABLE 4 - Micro-element contents of fruit species in leaves and edible part (mg/kg dry weight)

Sample Fe Cu Mn Zn Numbers Sample name Leaf Fruit Leaf Fruit Leaf Fruit Leaf Fruit 1 Sweet cherry 102.32±2.06* 55.42±0.94 212.65±4.63 82.15±1.38 20.14±1.37 13.67±0.22 42.32±2.77 25.12±0.45 2 Black mulberry 125.45±3.21 60.13±0.76 36.45±1.48 30.45±0.39 15.28±0.37 12.69±0.18 37.46±1.17 30.24±0.44 3 White mulberry 130.18±3.23 55.68±0.98 25.18±0.55 22.34±0.39 14.16±1.53 15.47±0.28 30.26±0.94 23.44±0.37 4 Apricot 143.12±6.37 70.12±2.07 24.34±1.34 16.47±0.68 18.13±1.29 16.98±0.27 38.45±1.01 18.65±0.35 5 Apricot 162.13±2.71 75.69±0.83 28.34±1.08 15.14±0.22 26.15±0.99 17.25±0.19 42.58±2.25 22.54±0.25 6 Apple 123.11±3.15 64.14±1.52 15.43±0.74 11.34±0.59 62.34±2.48 30.24±0.97 20.15±1.47 18.63±0.48 7 Plum 187.32±5.00 90.11±1.35 19.24±1.99 12.20±0.23 102.33±3.16 65.24±0.98 30.99±1.35 23.67±0.62 8 Plum 212.30±4.30 85.86±2.44 24.52±0.67 18.63±0.56 80.15±1.24 25.15±0.63 36.41±1.10 20.15±0.85 9 Peach 140.34±3.26 70.41±1.15 14.15±0.22 12.14±0.25 50.13±1.90 24.12±0.98 23.41±0.36 18.63±0.29 10 Plum 185.36±2.94 85.16±2.80 20.11±0.30 16.58±0.29 75.69±1.17 40.36±0.94 30.41±0.58 17.69±0.27 11 Plum 178.63±2.94 80.32±0.97 15.26±0.62 11.34±0.16 62.67±2.36 57.62±0.62 26.58±1.24 15.24±0.16 12 Plum 138.69±2.99 86.65±1.66 22.34±0.65 12.49±0.24 45.11±0.81 47.52±0.73 25.34±0.43 16.98±0.27 13 Plum 165.23±2.68 75.14±2.10 17.10±0.72 9.45±0.12 85.62±1.48 50.13±0.57 24.13±0.96 10.24±0.08 14 Pear 102.34±3.50 53.32±2.30 12.66±0.41 11.63±0.18 56.41±0.93 45.69±0.77 13.09±0.66 15.67±0.34 15 Apricot 151.29±4.62 72.13±1.70 32.04±1.11 18.65±0.36 21.65±0.86 36.45±0.44 44.14±1.74 24.37±0.39 16 Apricot 138.99±3.75 65.42±1.04 27.92±1.27 14.52±0.60 17.89±0.79 42.15±1.11 40.56±0.41 20.41±0.45 17 Hawthorn 112.36±3.17 50.16±0.35 16.58±0.46 12.69±0.10 75.68±1.26 40.12±0.71 22.24±0.43 16.75±0.38 18 Rosehip 134.69±1.69 65.42±1.55 20.13±0.96 14.15±0.22 124.13±2.46 36.28±1.01 30.45±1.07 17.55±0.57 Mean 146.55±4.13 70.07±1.68 32.55±6.07 19.02±2.21 53.05±4.43 34.29±2.11 31.12±1.22 19.78±0.62 Maxi- 212.30±4.30 90.11±1.35 212.65±4.63 82.15±1.38 124.13±2.46 65.24±0.98 44.14±1.74 30.24±0.44 mum Minimum 102.34±3.50 50.16±0.35 12.66±0.41 9.45±0.12 14.16±1.53 12.69±0.18 13.09±0.66 10.24±0.08 *Data are presented as means±SE (standard errors), n=3 replicates

TABLE 5 - Heavy metal levels of fruit species in leaves and edible part (mg/kg dry weight)

Sample Cd Pb Ni Cr Sample name Numbers Leaf Fruit Leaf Fruit Leaf Fruit Leaf Fruit 1 Sweet cherry 12.07±0.19* 3.41±0.07 8.74±0.27 1.75±0.03 4.56±0.09 1.27±0.04 0.18±0.01 0.06±0.01 2 Black mulberry 5.46±0.59 2.10±0.05 3.48±0.15 1.62±0.06 4.11±0.08 1.03±0.03 0.14±0.01 0.04±0.00 3 White mulberry 4.15±0.07 2.34±0.05 4.15±0.06 2.15±0.04 3.85±0.09 1.36±0.02 0.18±0.01 0.05±0.00 4 Apricot 9.32±0.27 3.40±0.06 6.55±0.07 2.20±0.05 3.11±0.10 1.24±0.03 0.10±0.01 0.04±0.00 5 Apricot 11.12±1.01 4.11±0.06 10.23±0.57 3.15±0.03 2.53±0.09 0.45±0.02 0.08±0.01 0.03±0.00 6 Apple 6.75±0.16 1.12±0.04 5.13±0.12 2.06±0.06 2.34±0.09 0.70±0.03 0.21±0.01 0.02±0.00 7 Plum 15.41±0.61 5.87±0.14 12.09±1.05 3.41±0.08 4.43±0.14 0.86±0.02 0.20±0.01 0.07±0.00 8 Plum 13.10±0.46 4.36±0.11 14.25±0.36 2.86±0.07 6.51±0.14 0.76±0.02 0.22±0.01 0.07±0.01 9 Peach 12.76±0.19 5.11±0.08 7.69±0.12 3.15±0.09 5.10±0.15 1.15±0.03 0.21±0.01 0.07±0.01 10 Plum 15.27±0.30 5.89±0.12 10.39±0.27 3.24±0.06 4.52±0.12 0.78±0.03 0.19±0.01 0.06±0.00 11 Plum 13.68±0.36 5.11±0.05 15.47±0.26 3.15±0.03 7.51±0.27 1.03±0.07 0.18±0.01 0.06±0.00 12 Plum 12.41±0.20 2.34±0.19 10.69±0.24 3.42±0.21 6.51±0.14 1.21±0.04 0.22±0.00 0.07±0.01 13 Plum 17.12±0.76 1.26±0.03 11.03±0.20 2.86±0.02 4.59±0.07 0.42±0.05 0.30±0.01 0.09±0.00 14 Pear 8.15±0.37 1.12±0.04 10.43±0.23 2.15±0.03 3.89±0.08 0.36±0.03 0.18±0.01 0.06±0.01 15 Apricot 18.81±3.43 3.45±0.11 6.93±1.23 3.02±0.12 4.47±0.78 0.58±0.02 0.04±0.01 0.01±0.00 16 Apricot 8.60±0.21 1.52±0.08 3.19±0.09 1.86±0.03 2.08±0.07 0.69±0.03 ND ND 17 Hawthorn 8.59±0.15 1.78±0.03 8.67±0.13 3.10±0.08 4.11±0.08 0.85±0.02 0.12±0.01 0.02±0.00 18 Rosehip 11.25±0.11 1.32±0.04 10.36±0.06 2.85±0.07 5.45±0.07 0.91±0.04 0.08±0.00 0.01±0.00 Mean 11.52±0.60 3.09±0.22 8.92±0.47 2.67±0.08 4.47±0.20 0.87±0.04 0.16±0.01 0.05±0.00 Maxi- 18.81±3.43 5.89±0.12 15.47±0.26 3.42±0.21 7.51±0.27 1.36±0.02 0.30±0.01 0.09±0.01 mum Mini- 4.15±0.07 1.12±0.04 3.19±0.09 1.62±0.06 2.08±0.07 0.36±0.03 0.04±0.01 0.01±0.00 mum *Data are presented as means±SE (standard errors), n=3 replicates, ND: non-detectable

In fruits, the amounts of P varied from 2.27 mg/g had the highest concentration in fruit samples followed by (Apricot/Teberze-Sample 5) to 1.23 mg/g (Plum-Sample Mn, Zn and Cu. Stone fruits had more Fe concentration 8); K contents were found between 17.42 mg/g (Plum- than the other fruits as reported by Grembecka and Szefer Sample 13) and 4.02 mg/g (Apricot/Zerdali-Sample 16); [22]. The same researchers [22] represented comparable Ca ranged between 22.14 mg/g (Plum-Sample 7) and 8.75 values for Fe but lower Zn, Cu, Mn in sweet cherry, plum, mg/g (Apricot/Şalak-Sample 4) and amounts of Mg varied apple and pear. Krejpcio et al. [23] reported similar values from 6.87 mg/g (Plum-Sample 7) to 0.62 mg/g in the pre- for Fe. Generally, the present study demonstrated that the sent study (Table 3). concentration of Fe and Mn were within the maximum per- The results concerning P and K in sweet cherry, pear, missible limits of international standards (425 mg/kg and apple, plum and mulberry fruits are comparable to findings 500 mg/kg, respectively) [26]. However, our study re- of Grembecka and Szefer [22], Hamurcu et al. [8]., Ercisli vealed that fruits accumulated Cu and Zn higher than the and Orhan [24]. However, Ozcan et al. [10] determined set limits of international standards (10.00 and 5.00 mg/kg, lower concentration of P in apple and pear. The concentra- respectively) [27]. tion of Mg and Ca in sweet cherry, pear, apple and plum In the current investigation, heavy metal (Cd, Pb, Ni and were higher than those detected by Grembecka and Szefer Cr) levels of leaf and fruit samples are given in Table 5. As [22]. The data obtained in the present study for Mg and Ca seen from these results, Cd content of the fruit samples var- were also higher than those detected in plum and rosehip ied form 5.89 mg/kg (Plum-Sample 10) to 1.12 mg/kg [8] and in hawthorn [25]. (Pear-Sample 6). The average Cd content of fruit samples In the present work, Fe, Cu, Mn and Zn contents of analyzed was 3.09 mg/kg. Its accumulation in stone fruits fruit samples varied from 90.11 mg/kg (plum-Sample 7) to was higher than in the other fruits. Especially, plums were 50.16 mg/kg (hawthorn-Sample 17), 82.15 mg/kg (sweet found to have high Cd content in the study area. Our results cherry-Sample 1) to 9.45 mg/kg (plum- Sample 13), 65.24 were much higher than those recorded by Krejpcio et al. mg/kg (plum-Sample 7) to 12.69 mg/kg (black mulberry- [23] in sweet cherry (0.021 mg/kg) in Poland by Pehluvan Sample 2) and 30.24 mg/kg (black mulberry-Sample 2) to et al. [11] and Duran et al. [28] in white mulberry (0.015 10.24 mg/kg (plum-Sample 13), respectively (Table 4). and 0.63 mg/kg, respectively) in Turkey, by Radwan and The obtained results of the present study showed that Fe Salama [29] in apple (0.05 mg/kg) in Egypt, and by

Hamurcu et al. [8] in plum (0.14 mg/kg) in Turkey. This probably depend on the physical and chemical nature of might be attributed to the trailer trucks which make up of a soil and absorption capacity of each metal by the plant, large proportion of traffic density. It is well known that which is altered by countless environmental and human these vehicles burns diesel fuels and emit Cd into the at- factors and nature of plant. mosphere. Therefore, plant tissues may be contaminated with high levels of Cd in the study area. However, the re- In the present work, it was found that leaf samples con- sults obtained in the present study for Cd concentration tained more minerals than fruit samples as reported by were lower than those detected in some fruit and vegetable Hamurcu et al [8]. Moreover, the correlation analysis showed with a 25 mg/kg average value [30]. Fruit material from all a strong and positive relationship between all mineral concen- sampling points were contaminated by an excessive tration in leaves and fruits. The correlation coefficient varied amount of Cd compared to the permissible limit (0.05 from r=0.267 with 0.0508 probability value (Ni) to r=0.990 mg/kg) proposed by the FAO/WHO [31] and thus might be with 0.0001 probability value (Mg) (Table 6). a great threat for the consumers because of its negative ef- TABLE 6 - Pearson correlation coefficient between minerals contents fect on liver, kidney, lung, bones, placenta, brain and cen- in leaf samples and in fruit samples (n=54) tral nervous systems [32]. Minerals r Probability The highest content of Pb was detected in Plum-Sam- P 0.853 <.0001 ple 12 with a 3.42 mg/kg value while it’s the lowest con- K 0.888 <.0001 Ca 0.969 <.0001 centration was in Black mulberry-Sample 2 (1.62 mg/kg). Mg 0.990 <.0001 The average content of Pb was 2.67 mg/kg in the current Fe 0.864 <.0001 study. As in Cd, Pb content of stone fruits was also found Cu 0.977 <.0001 higher than the other fruits. These results were in agreement Mn 0.597 <.0001 Zn 0.641 <.0001 with [33] who reported between 2.1 and 7.0 mg/kg of Pb in Cd 0.468 0.0004 apricot. The result obtained in the present study was also in Pb 0.674 <.0001 accordance with those detected in apple (2.21 mg/kg), plum Ni 0.267 0.0508 (2.82 mg/kg) and rosehip (1.54-2.86 mg/kg) [8]. Pb content of Cr 0.843 <.0001 all samples exceeded the permissible limit of FAO/WHO [31] recommended for berries and other small fruits as 0.2 mg/kg 5. CONCLUSIONS and for foods as 0.5 mg/kg. When intake of Pb excessive, it may result in brittle bones, weakness in the wrists and fingers, musculoskeletal, renal, ocular, neurological, cancer and re- The current study showed that fruit samples were pol- productive effects [32]. Increase of Pb levels in these sam- luted by Cu, Cd and Pb metals. The study also indicated ples was attributed to motor vehicles in the study area which that concentration of Zn in fruit samples was higher than lead to the accumulation of Pb emitted from cars exhaustion. set limits by FAO/WHO. Generally, stone fruit species and mulberries were found to be more contaminated with trace The content of Ni in fruit samples ranged between 1.36 elements and heavy metals than other fruits. There are mg/kg (White mulberry-Sample 3) and 0.36 mg/kg (Pear- many explanations for heavy metal accumulation in fruits Sample 14). The average content of Ni in fruits was 0.87 and one of them are motor vehicles. Increase of Pb and Cd mg/kg in the present study. According to SEPA [34], rec- levels in fruits might be attributed to motor vehicles in the ommended limits of Ni for foods are 10 mg/kg and our re- study area which lead to the accumulation of Pb and Cd sults were within the safe limits proposed by SEPA [34]. emitted from trailer trucks and cars exhaustion. Second Moreover, present study was comparable with those de- may be soil which was polluted by coal combustion. It is tected in apple, plum and rosehip [8], in white and black an important source and has high heavy metal contents. mulberries [35] and in hawthorn [25]. Low quality coal has been used for heating in the study area for many years thus soil might be contaminated with Cd Observed Cr content of fruit samples varied from 0.09 and Pb metals and accumulated in fruits. And third are ag- mg/kg (Plum-Sample 13) to 0.01 mg/kg (Apricot and ricultural practices. Pesticides are used intensively for pro- Rosehip-Sample 15 and 18) with a 0.05 mg/kg average tection fruit trees from insects and some pathogens. It value. The present results were lower than those obtained might be result in accumulation of Cu and other metals in by Duran et al. [28] who reported the levels of Cr between fruit samples. There are no any industries close to the study 0.8 and 6.17 mg/kg in some fruit species such as apricot, area within borders of Turkey. However, Yerevan, the cap- pear, white mulberry, rosehip, plum. Generally, the present ital city of Armenia, is very close to the study area with a study also illustrated that the concentration of Cr was gold extraction plants, a copper-molybdenum mining com- within the safe limits (0.5 mg/kg) [34]. plex and Metsamor Nuclear Power Plant. With a bare possibility, these may all be contributors to the contamination Data obtained in the present study had some differ- of fruits because of the opposite wind duration (WNW) in ences with some literatures [8, 22, 25] in some fruit species the study area. Moreover, some other factors may also con- in terms of some minerals discussed above. Because the tribute to metal accumulation in fruits such as fertilizers range of the metal contents observed in this fruit species and unknown atmospheric deposits. The levels of Cr, Ni,

Fe and Mn did not appear to reach pollution levels in the [16] Şimşek, U., Erdel, E., Tohumcu, F. and Sürmeli, S. (2013). fruit samples. Evaluation of Soil pH and Salinity Status of Agricultural Lands in Igdir Provience. III. Ulusal Toprak ve Su Kaynakları Kongresi, Ekim 2013, Tokat. The authors have declared no conflict of interest. [17] Karaoğlu, M. (2011). The study of Iğdır climate in terms of Agricultural Meteorology. Iğdır Univ. J. Inst. Sci & Tech. 1(1): 97-104. [18] TSMS (2012). Turkish State Meteorological Service. REFERENCES [19] Mertens, D. (2005a). AOAC Official Method 922.02. Plants Preparation of Laboratuary Sample. Official Methods of Anal- [1] O’Connell, D.W., Birkinshaw, C. and O’Dwyer, T.F. (2008). ysis, 18th edn. Horwitz, W., and G.W. Latimer, (Eds). Chapter Heavy Metal Adsorbents Prepared from the Modification of 3, pp1-2, AOAC-International Suite 500, 481. North Frederick Cellulose: a Review. Bioresource Technology 99, 6709-6724. Avenue, Gaitherburg, Maryland 20877-2417, USA. [2] Sharma, R.K, Agrawal, M. and Marshall, F.N. (2009). Heavy [20] Mertens, D. (2005b). AOAC Official Method 975.03. Metal in Metals in Vegetables Collected from Production and Market Plants and Pet Foods. Official Methods of Analysis, 18th edn. Sites of a Tropical Urban Area of India. Food and Chemical Horwitz, W., and G.W. Latimer, (Eds). Chapter 3, pp 3-4, Toxicology, 47: 583-591. AOAC-International Suite 500, 481. North Frederick Avenue, Gaitherburg, Maryland 20877-2417, USA. [3] Kuno, K. (1984). Effects of Heavy Metals on Photosynthetic Rates and Morphogenesis in Mulberry Leaves. J. Seric. Sci., [21] Duzgunes, O., Kesici, T., Kavuncu, O. and Gürbüz, F. (1987). 53 (3), pp 198-204. Statistical Analyses Methods II. A.U., Agricultural Faculty Publications p. 381, Ankara. [4] Fotakis, G. and Timbrell, J.A. (2006). Role of Trace Elements in Cadmium Chloride Uptake in Hepatoma Cell Lines. Toxi- [22] Grembecka M. and Szefer P. (2013). Comparative assessment col. Lett., 164: 97-103. of essential and heavy metals in fruits from different geographical origin. Environ Monit Assess, 185:9139-9160. [5] Sharma, R.K, Agrawal, M. and Marshall, F.N. (2008). Heavy Metals (Cu, Zn, Cd and Pb) Contamination of Vegetables in [23] Krejpcio, Z., Sionkowski, S. and Bartela, J. (2005). Safety of Urban India: A Case Study in Varanasi. Environmental Pollu- fresh fruits and juices available on the Polish market as deter- tion, 145: 254-263. mined by heavy metal residues. Polish journal of Environmen- tal Studies, 14: 877-881. [6] Ali, M.H.H. and Al-Qahtani, K.M. (2012). Assessment of some heavy metals in vegetables, cereals and fruits in Saudi [24] Ercisli, S. and Orhan, E. (2007). Chemical composition of Arabian markets. Egyptian Journal of Aquatic Reseach, 38: white (Morus alba), red (Morus rubra) and black (Morus 31-37. nigra) mulberry fruits. Food Chemistry, 103: 1380-1384.

[7] Duruibe J.O., Ogwuegbu M.O. and Egwurugwu J.N. (2007). [25] Özcan, M., Hacıseferoğulları, H., Marakoğlu, T. and Arslan, D. Heavy metal pollution and human biotoxic effects. Interna- (2005). Hawthorn (Crataegus spp.) fruits: some physical and tional Journal of Physical Sciences, 2(5): 112-118. chemical properties. Journal of Food Engineering, 69: 409- 413. [8] Hamurcu, M., Özcan, M.M., Dursun, N. and Gezgin, S. (2010). Mineral and Heavy Metal Levels of Some Fruits [26] FAO/WHO (2011). Codex Alimentarius Commission. Food Grown at the Roadsides. Food and Chemical Toxicology, 48: additives and contaminants. Joint FAO/WHO Food Standards 1767-1770. Program; ALINORM 01/12A:1-289. [9] Fang, B. and Zhu, X. (2014). High content of five heavy met- [27] WHO (1994). Quality directive of Potable Water, Geneva, sec- als in four fruits: Evidence from a case study of Pujian Cunty, ond ed., p. 197. Zhejiang Province, China. Food Control, 39: 62-67. [28] Duran, A., Tuzen, M. and Soylak, M. (2008). Trace Element [10] Ozcan M.M., Harmankaya M. and Gezgin S. (2012). Mineral Levels in Some Dried Fruit Samples from Turkey. International and heavy metal contents of the outer and inner tissues of com- Journal of Food Sciences and Nutrition, 59(7-8): 581-589. monly used fruits. Environ Monit Assess 184: 313-320. [29] Radwan, M.A. and Salama, A.K. (2006). Merket basket survey [11] Pehluvan, M., Karlidag, H. and Turan, M. (2012). Heavy metal for heavy metals in Egyptian fruits and vegetables. Food and levels of mulberry (Morus alba L.) grown at different distances Chemical Toxicology 44: 1273-1278. from the roadsides. The Journal of Animal & Plant Sciences, 22(33): 665-670. [30] Turkoglu, M.K., Kilicel, F., Kara, K., Tuncer, I. and Uygun, I. (2002). Heavy metals in soil, vegetables and fruits in the en- [12] Bukvić, E., Huremović, J., Memić, M. and Gojak-Salimović, demic upper gastrointestinal cancer region of Turkey. Environ S. (2013). Heavy metals in fruits and vegetables from markets Toxicol Pharmacol, 13: 175-179. in Sarajevo, Bosnia and Herzegovina. Technologica Acta, 6: 29-35. [31] FAO/WHO (1995). Codex General Standard for Contaminants and Toxins in Food and Feed (193-1995), pp. 31-32. [13] Orisakwe, O.E., Nduka, J.K., Amadi, C.N., Dike, D.O. and Bede, O. (2012). Heavy metals health risk assessment of pop- [32] Castro-González, M.I. and Méndez-Armenta, M. (2008). ulation via consumption of food crops and fruits in Owerri, Heavy metals: Implications associatedto fish consumption. South Eastern, Nigeria. Chemistry Central Journal, 6:77. Environmental Toxicology & Pharmacology, 26: 263-271. [14] Al-Obeed, R.S., Kassem, H.A. and Ahmed, M.A. (2011). Leaf [33] Ali, S., Masud, T. and Abbasi, K.S. (2011). Physico-chemical petiole mineral and fruit heavy metals content of different characteristics of apricot (Prunus armenica L.) grown in grape cultivars grown under arid environments and irrigated Northern Areas of Pakistan. Scientia Horticulturae, 130: 386- with treated domestic wastewater. AAB Bioflux, 3(1): 5-14. 392. [15] TMTMAC (2012). Turkish Ministry of Transport, Maritime [34] SEPA (2005). The limits of pollutants in food. State environ- Affairs and Communications, Department of Highways. mental protection administration, China. GB2762.

[35] Micić, R., Dimitrijević, D.S., Kostić, D.A., Stojanović, G.S., Mitić, S.S., Mitić, M.N., Pavlović, A.N. and Randelović, S.S. (2013). Content of Heavy Metals in Mulberry Fruits and Their Extract-Correlation Analyses. American Journal of Analytical Chemistry, 4: 674-682.

Received: March 07, 2014 Revised: July 16, 2014 Accepted: September 10, 2014

CORRESPONDING AUTHOR

Mücahit Pehluvan Igdir University Faculty of Agriculture Department of Horticulture 76000 Igdir TURKEY

Phone: +90 476 2261314-2009 Fax: +90 476 2261251 E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1302 - 1309

EFFECTS OF ROADS AND RAILWAYS ON LARGE GAME IN THE BELGRADE AREA: A CASE-STUDY OF NINE MUNICIPALITIES

Dragan P. Gacic1,*, Milorad Danilovic1, Jasmina Gacic2 and Dusan Stojnic1

1 University of Belgrade, Faculty of Forestry, KnezaViseslava 1, Belgrade, Serbia 2 University of Belgrade, Faculty of Security Studies, GospodaraVucica 50, Belgrade, Serbia

ABSTRACT Spain [5]. The primary environmental effects of transport infrastructure on wildlife and their habitats are: loss and The effects of roads and railways on major species of fragmentation of wildlife habitat, barrier effects, collisions large game (roe deer and wild boar) have been studied in between vehicle and wildlife, disturbance and pollution, and eight hunting grounds in the Belgrade area (nine munici- ecological function of verges (edges of infrastructure devel- palities), as well as the consequences of chemical accidents opment). The secondary effects include changes in landuse, in transport of hazardous materials. Road accidents involv- human settlement patterns or industrial development in- ing birds, game and other animals (WVC - wildlife vehicle duced by the construction of traffic infrastructure. In prac- collisions) during 2009-2013 in the Belgrade area result in tice, these effects usually interact and may significantly in- 10 cases (4.8%) in injuries for drivers and passengers. In crease their negative impact through synergistic effects [6]. total, 267 roe deer and 8 wild boar were killed in traffic by In recent years, wildlife vehicle collisions are becom- cars and trains. Traffic accidents with roe deer have their ing more and more frequent in many hunting grounds in maximum in April-May and then in December. The adult Serbia but this problem has never been thoroughly ana- sex ratio in roe deer killed in traffic by cars and trains was lyzed nor systematically studied. Moreover, there are no 2.1 (173 ♀ and 80 ♂), indicating that females were more accurate and comprehensive records of the number of vulnerable than males. The yearly traffic kill of roe deer killed wild animals and the total economic costs, not only represented between 0.8% and 1.1% of the spring popula- at the national level, but also in different regions, years and tion size. In the Belgrade area, the mobile eco-toxicologi- game species. cal laboratory of the City Institute of Public Health Bel- grade during 2004-2012 registered 177 chemical accidents In the Belgrade area there is very heavy traffic of re- (mean = 19.7, range = 13 to 26). The chemical accidents in gional and national importance with the tendency of further transport of hazardous materials were sporadic and their development. Together with several thermal power plants consequences were small-scale contamination of the envi- (Lazarevac and Obrenovac) it poses the major source of air ronment and economic losses. At the main transport routes and soil pollution [7], as well as the source of noise. More- of hazardous materials passing through hunting grounds in over, there are many dangerous industrial plants which use, the Belgrade area occurred 26 chemical accidents, out of store or produce huge quantities of hazardous materials, which 11 on the railways and 9 on the roads, and the remain- whose transport is carried out by road, railway, water and der in air and river transportation (4 and 2, respectively). air traffic. However, each transportation involving hazard-

ous materials is accompanied by the risk of adverse effects [8]. The greatest concentration of hazardous and harmful KEYWORDS: materials is in the municipalities of Cukarica (warehouse roe deer, traffic, chemical accident, hunting ground, Belgrade of oil and oil derivatives), Rakovica (production of technical gases), Obrenovac (industrial complex “Prva iskra” – production of most modern high explosives and civil chem- 1. INTRODUCTION ical industry), Palilula (paint and varnish industry) and Zemun (pharmaceutical and chemical industry). At the Traffic collisions involving wildlife represent a con- same time, the greatest part of the territory of the men- stant and very serious problem in many European coun- tioned municipalities represents hunting grounds with the tries, e.g. Sweden [1], Slovenia [2, 3], Belgium [4] and highest spring density of reared large game species – roe deer and wild boar [9]. Therefore, the domestic as well as * Corresponding author international transportation of hazardous materials can be

potential and serious threat to large game and its habitat for The geographical position of studied hunting grounds the reason that a large number of chemical accidents occurs and municipalities in the Belgrade area is presented in Fig- on the roads and railway lines. ure 1. Hunting statistics (spring population size, annual harvest, non-hunting mortality, trophy scoring data) for roe The aim of this paper was to determine and analyze the deer and wild boar at the municipality level were provided effects of roads and railways on major species of large by the local hunting societies (Annual and Long-term Hunt- game (roe deer and wild boar) and their habitats in the Bel- ing Plans). Roe deer and wild boar are the main and the most grade area, as well as frequency, spacial distribution and abundant large game species (Table 1). The official statisti- consequences of chemical accidents in the transport of haz- cal data on the number of traffic accidents involving birds, ardous materials. wildlife and other animals (WVC) on the roads of Serbia and the Belgrade area during 2009-2013 were obtained from the Ministry of Interior of the Republic of Serbia (Di- 2. MATERIAL AND METHODS rectorate for Analytics). The collected data include the annual number of road accidents, effects and the assessed ma- The Belgrade area covers 3.224 km2 and is divided into terial damage (in RSD), which is converted in EUR at the 17 municipalities, the largest of which covers 447 km2 official middle RSD exchange rate of the National Bank of (Palilula), and the smallest 3 km2 (Vracar). This area is in- Serbia for the purpose of more effective analysis and com- tersected or surrounded by 83 km of the Danube river flow parison. The total length of the Serbian road network ac- and 63 km of the Sava river flow. The altitude ranges from cording to the data of the Public Enterprise “Roads of Ser- 71 to 628 m. The average annual temperature is 11.6°C, bia” amounts to 40,845 km, and comprises 5,525 km of and annual precipitation ranges from 610-730 mm. From class I state roads, 11,540 km of class II state roads and the aspect of the quality of the environment it falls within 23,780 km of local roads. The analysis of the consequences the most endangered areas in Serbia (hot spots). The agri- of chemical accidents during the transportation of hazard- cultural land covers 218,055 ha and there can be distin- ous materials on the large game was conducted on the basis guished intensively and extensively cultivated areas. For- of the data provided by the City Institute of Public Health ests and forest plantations cover 35,980 ha or 11.2% of the Belgrade published in the Annual Reports on engagement total area. The municipalities with the high level of forest of the mobile eco-toxicological laboratory (2004-2012). cover are Barajevo (20%), Lazarevac (18%), Vozdovac Consequences of a chemical event include deaths, illness, (17%) and Cukarica (14%). economic losses, and contamination of environment, peo-

FIGURE 1 - The study of hunting grounds (municipalities): 1. Avala (Vozdovac); 2. Topciderskareka (Cukarica (a) and Rakovica (b)); 3. Barajevskareka (Barajevo); 4. Gavranskipotok (Grocka); 5. Varovnice (Mladenovac); 6. Posavina (Obrenovac); 7. Kosmaj (Sopot); 8. Kolu- bara (Lazarevac).

TABLE 1 - Spring population size of roe deer (RD) and wild boar (WB) in the Belgrade area

Year Hunting ground (Municipality) Species 2009 2010 2011 2012 2013 RD 300 366 300 270 280 1. Avala (Vozdovac) WB 0 0 0 16 16 RD 285 278 282 286 282 2. Topciderskareka (Cukarica and Rakovica) WB 0 0 0 0 0 RD 700 700 700 700 700 3. Barajevskareka (Barajevo) WB 18 18 18 18 18 RD 155 152 140 150 150 4. Gavranskipotok (Grocka) WB 22 36 30 32 36 RD 1.096 1.100 1.000 940 1.000 5. Varovnice (Mladenovac) WB 30 30 30 30 30 RD 1.325 1.345 1.320 1.390 1.400 6. Posavina (Obrenovac) WB 0 0 0 0 0 RD 944 860 870 880 910 7. Kosmaj (Sopot) WB 60 60 60 60 60 RD 780 780 780 780 780 8. Kolubara (Lazarevac) WB 0 0 34 34 34

ple and property. All statistical analyses were performed in 2009 and 8.0% in 2011). The yearly distribution of esti- with STATISTICA 8.0 (StatSoft). mated damage to vehicles (EUR) for both the Republic of Serbia (χ2 = 74887.68, df = 4, p < 0.001) and the Belgrade area (χ2 = 1786.85, df = 4, p < 0.001) was not uniformly 3. RESULTS distributed. During the five year survey a total number of 960 cases of WVC were recorded by the Ministry of Inte- The official number of vehicle collisions with wildlife rior (Directorate for Analytics). The ratio of recorded during 2009-2013 in the Republic of Serbia and in the Bel- WVC in total road accidents in the Republic of Serbia dur- grade area is presented in Table 2. Road accidents involving this period is shown in Table 3. The total number of ing birds, game and other animals (WVC – wildlife vehicle road accidents in the period between 2009 and 2013 has collisions) in the Belgrade area result in 10 cases (4.8%) in decreased by 42.8%, and the number of WVC ranged be- injuries for drivers and passengers (ranged between 2.2% tween 175 and 212 over the same period.

TABLE 2 - The official records of the Ministry of Interior (Directorate for Analytics) on road accidents involving birds, game and other animals (WVC – wildlife vehicle collisions) in the Republic of Serbia and in the Belgrade area

Number of Number of Number of Estimated damage Year road accidents accidents with casualties fatalities (EUR) Serbia Belgrade Serbia Belgrade Serbia Belgrade Serbia Belgrade 2009 175 46 4 1 0 0 46.917 12.790 2010 190 42 1 2 0 0 41.312 12.546 2011 212 50 12 4 0 0 61.002 18.459 2012 190 35 6 1 0 0 52.899 12.726 2013 193 35 14 2 1 0 127.357 13.770 Total 960 208 37 10 1 0 329.487 70.291

TABLE 3 - Road accidents involving birds, game and other animals (WVC – wildlife vehicle collisions) in relation to total road accidents in the Republic of Serbia

Total number Wildlife vehicle Percentage Year of road accidents collisions (WVC) (%) 2009 64.877 175 0,3 2010 47.757 190 0,4 2011 42.453 212 0,5 2012 37.617 190 0,5 2013 37.127 193 0,5 Total 229.831 960 0,4

In total, 267 roe deer and 8 wild boar were killed in killed in traffic by cars and trains between 2009 and 2013 traffic during the period 2009-2013 in the Belgrade area in the Belgrade area, 26.5% of road accidents and 50.0% (Table 4). The number of individual roe deer killed peaked of railway accidents occured in April and May. The adult in 2010 (n=63) and 2013 (n=61), while wild boar killed sex ratio in roe deer killed in traffic by cars and trains was peaked in 2012 (n=5). The official number of WVC in the 2.1 (173 ♀ and 80 ♂), indicating that females were more Belgrade area (mean = 41.6, range = 35 to 50) registered vulnerable than males, and 14 (5.2%) of the individuals by the Ministry of Interior (Directorate for Analytics) was were juveniles. significantly lower than the number of roe deer and wild boar killed in traffic by cars and trains registered by Hunt- TABLE 4 - Number of roe deer (RD) and wild boar (WB) traffic kills ing societies (mean = 55.0, range = 46 to 64; t-test: t = – per year in the Belgrade area 2.755, p = 0.025). Road-kills Railway-kills Total (n) Year RD WB RD WB RD WB The monthly distribution of collisions between roe deer 2009 40 1 5 0 45 1 and vehicles in the Belgrade area is shown in Figure 2. The 2010 60 1 3 0 63 1 monthly distribution of road accidents in case of roe deer is 2011 46 0 0 0 46 0 2 2012 50 5 2 0 52 5 not uniform (χ = 42.921, df = 11, p < 0.001, n = 253). Cas- 2013 57 1 4 0 61 1 ualties with roe deer have their maximum in April-May and Total 253 8 14 0 267 8 then in December (Figure 3). Out of 267 reported roe deer

January roads 50 December February ra ilwa ys 40 30 November March 20 10 October 0 April

September May

August June July

FIGURE 2 - Monthly distribution of roe deer killed in traffic by cars and trains in the Belgrade area

30 male female juvenile

5 Number roe traffic deer of in killed Number 0 j fmamj ja sond Month

FIGURE 3 - Number of male, female, and juvenile (12 months old and younger) roe deer killed in traffic by cars and trains in the Belgrade area

TABLE 5 - Number of roe deer killed in traffic by cars and trains in the Belgrade area according to the records of the eight hunting societies(n = annual number of roe deer kills; % = percentage of spring population)

Year Hunting ground (Municipality) 2009 2010 2011 2012 2013 n % n % n % n % n % 1. Avala (Vozdovac) 6 2,0 3 0,8 1 0,3 6 2,2 2 0,7 2. Topciderskareka (Cukarica and Rakovica) 1 0,3 3 1,1 6 2,1 0 0,0 1 0,3 3. Barajevskareka (Barajevo) 11 1,6 10 1,4 1 0,1 1 0,1 24 3,4 4. Gavranskipotok (Grocka) 2 1,3 0 0,0 0 0,0 0 0,0 0 0,0 5. Varovnice (Mladenovac) 6 0,5 9 0,8 8 0,8 12 1,3 9 0,9 6. Posavina (Obrenovac) 7 0,5 16 1,2 10 0,8 9 0,6 9 0,6 7. Kosmaj (Sopot) 2 0,2 14 1,6 9 1,0 9 1,0 3 0,3 8. Kolubara (Lazarevac) 10 1,3 8 1,0 11 1,4 15 1,9 13 1,7 Total 45 0,8 63 1,1 46 0,8 52 1,0 61 1,1

The yearly traffic kill of roe deer represented between tion); 2008 (PadinskaSkela – slipping out of six tank wag- 0.8% and 1.1% of the spring population size (Table 5). ons with Liquefied Petroleum Gas and capsize of two tank There was a statistically significant positive relationship wagons onto railway tracks). between the average total mortality of roe deer (annual harvest, poaching, predation and other deaths) and the average recorded traffic kills of roe deer (F1,6 = 16.28, r = 0.85, p < 4. DISCUSSION 0.05) within eight hunting grounds during 2009-2012. Millions of individuals of a wide range of wildlife spe- In the Belgrade area, the mobile eco-toxicological la- cies are killed on roads and railways each year, and many boratory of the City Institute of Public Health Belgrade more are seriously injured [6, 10, 11]. In the Lublin Region during 2004-2012 registered 177 chemical accidents (mean (Poland) most frequently killed animals on roads are am- = 19.7, range = 13 to 26). The most common were the phibians, medium-sized mammals (hedgehog, marten, fox, chemical accidents of low risk in communal areas due to badger, brown hare) and large mammals (roe deer, wild the deliberate poisoning of stray dogs and pets (32.8%). boar, red deer) [12, 13]. In general, mortality of animals The greatest risk to the environment was caused by chem- caused by traffic in agrarian landscapes with diminished ical accidents at industrial complexes (28.8%), transporta- biodiversity is alarmingly high, and includes protected spe- tion of hazardous materials (19.8%) and improper disposal cies and major game animals [14]. In Europe, 10 species of of waste (18.6%). The chemical accidents in transport of wild ungulates are involved in collisions with traffic [10]. hazardous materials were sporadic and their consequences Many authors have claimed that roe deer most frequently were small-scale contamination of the environment and gets killed in traffic accidents, on roads [2, 15, 16, 17 ] as economic losses. At the main transport routes of hazardous well as railways [18], which is confirmed also by the re- materials passing through hunting grounds in the Belgrade sults of our research in the Belgrade area (Table 4). area occurred 26 chemical accidents, out of which 11 on the railways and 9 on the roads, and the remainder in air Using the scant available information, some authors and river transportation (4 and 2, respectively). estimate the annual number of collisions with ungulates in traffic in Europe (excluding Russia) at 507,000 [10], which The greatest consequences were from chemical acci- results in 300 people killed and 30,000 injured, and mate- dents of high risk in road transport due to capsize and spill rial damage amounting to $ 1 billion (U.S.). In the hunting of oil derivatives from road tanks: 18 t in 2005 (into the grounds of Serbia the total number of recorded individual soil in the vicinity of the well water supply of Mladeno- large game kill in traffic by cars and the total estimated vac); 15 t in 2006 (into the soil near the road Belgrade - damage (about € 0.3 million during 2009-2013) are signif- Pancevo, municipality Palilula); 20 t in 2006 (into the soil icantly lower than in other European countries, e.g. the es- along the Ibar highway, municipality Lazarevac); 7 t in timated annual costs of damage in Czech Republic ex- 2010 (into the soil along the motorway Belgrade - Obren- ceeded € 100 million [19], and in Slovenia exceeded € 15 ovac, municipality Obrenovac). In railway transport, the million [3]. However, it can be very reliably assumed that most serious chemical accidents occurred at the railway the effects of roads and railways on large game and their stations: 2004 (Topcider – slipping out of three tank wag- habitats in the Belgrade area will increase in the future, be- ons containing H SO without contamination of the soil); 2 4 cause the development of the traffic infrastructure is de- two accidents in 2007 when there was the risk of contami- fined as one of the priorities of economic and social policy. nation of soil and subterranean waters, as well as great material damage (Mladenovac – spill out of fuel from a tank Many authors suggest that the actual number of wild- wagon in a composition transporting oil derivatives, and life killed in traffic is much higher than the official data PadinskaSkela – slipping out of three tank wagons contain- (police and hunting societies’ statistics), because injured ing Liquefied Petroleum Gas in the 12-wagon composi- wild animals, especially wild ungulates, in most cases es-

cape from the place of the traffic accident, while the over- The potential threat for large game in the hunting run wild animals frequently get poached by the driver, es- grounds in the Belgrade area pose dozens of industrial pecially small game that rarely causes damage to the vehi- plants which use, store or produce hazardous materials (ap- cle [1, 4, 11, 15]. Similarly, in the Belgrade area, the offi- proximately 1.250.000 t/year, out of which 15.000 t of haz- cial number of WVC – wildlife vehicle collisions regis- ardous waste), and some of them are located in the central tered by the Ministry of Interior (Directorate for Analytics) city zones (municipalities of Cukarica, Rakovica, Palilula was significantly lower than the number of roe deer and and Zemun). They are one of the top priority problems and wild boar killed in traffic by cars and trains registered by pose high risk to the environment and safety and human Hunting societies. Moreover, in Serbia, the data on the health [22]. The main routes of hazardous materials trans- number and the effects of large game vehicle collisions are portation are the main corridors for freight traffic, as well not recorded separately, but are shown as a category of traf- as some streets (roads) within the city as: Radnickast., fic accident caused by knocking down and overruning live- Savskast., Karadjordjevast. and many others. At the main stock or other animals. transport routes passing through hunting grounds in the Belgrade area during 2004-2012 occurred 26 chemical acci- In Slovenia and Croatia, the risk for vehicle collision dents, which is much less than in the period 1991-2001 with roe deer is highest in April and May, but the risk is (around 40). Our analyses show that most of those accidents also high during the summer and autumn. Majority of daily occurred on railway stations and roads (spills of hazardous collisions occurred in the morning (5 a.m. – 7 a.m.) and in materials from rail tanks or road trucks capsize), and sporad- the evening (6 p.m. – 10 p.m.) [2, 3, 15, 16]. Similarly, ically in the air and river transport. However, the greatest some authors [20] reported that roe deer-related accidents consequences were from chemical accidents in road in northwest Spain have their maximum in April and May transport (e.g. the accident of high risk which occurred in (coinciding with the breeding season), followed by July 2005 due to capsize and spill of road tank containing 18 t of (coinciding with the rut period), while wild boar-related ac- materials into the soil in the vicinity of the well water supply cidents have their maximum between October and January of Mladenovac). This risk has been significantly reduced af- (coinciding with the hunting season and with months with ter the construction of the bypass around Belgrade (Ba- the longest nights). In their study, peaks of roe deer acci- tajnica - Dobanovci - Bubanjpotok) which is almost com- dents occurred in the morning (5 a.m. – 8 a.m.) and in the pletely relocated from the inhabited part of the municipali- evening (7 p.m. – 10 p.m.). According to the data of our ties of Zemun, Cukarica, Rakovica and Vozdovac. study, traffic accidents with roe deer have their maximum in April-May and then in December (Figure 2). In addition, In Serbia, the transport of hazardous materials is regu- our results suggest that more than 60% of the adult roe deer lated by the Law on transport of hazardous materials [23], killed in road and railway traffic were females (Figure 3). which sets forth the authorities of state bodies and special- This result can be explained by the fact that population ized agencies in transport of hazardous materials, special structure is disturbed in some hunting grounds (females are conditions for conducting the transport of hazardous ma- more numerous), as well as the hunting of females and terials, manner of conducting transport and procedures in fawns is most common in December and January, although case of emergencies in the transport of hazardous materi- open hunting season in Serbia starts on 1 September and als. Moreover, this law provides for the transport of haz- ends on 31 January. The same result has been found in ear- ardous materials to be performed in accordance with the lier studies on roe deer in Slovenia, e.g. municipality provisions of the 6 ratified international agreements (Arti- Smartno (sex ratio in roe deer killed in traffic was 1.6 ♀ : cle 2, paragraph 2). 1 ♂) [3]. Also, in Karlovac county (Croatia), 76.4% of roe In the Belgrade area, the preventive measures provide deer individuals killed in traffic were females, due to dis- for the protection against chemical accidents and pollution turbed sex ratio in hunting grounds [15]. This indicates that of the environment by manufacturing, storage or transport areas with different landscapes, climates and population of hazardous materials in the existing and new plants. structures may differ in between sexes and monthly proba- Those measures refer to the activities necessary to be taken bility of accidents [21]. in the industrial plants and on the transport routes for hazardous materials for the purpose of reducing the potential We found that yearly traffic kill of roe deer represented accidents and consequences. The routes for transport of between 0.8% to 1.1% of the spring population size in the hazardous materials are established by special decisions in Belgrade area (Table 5). Some authors reported that traffic accordance with the new locations of hazardous plants, as losses in roe deer populations ranged between 1.6% (Nor- well as the level of construction of new motorways. In gen- way) to 6.0% (Germany) of the annual spring population eral, the transport of hazardous and dangerous materials [10]. Recent analyses in the UK suggest that traffic losses can be conducted by the sections of highway (E-70, E-75) for roe deer ranged from 3% to 7% [11]. Similar data were as well as motorways passing through lower population found in relation to the hunting grounds of Slovenia, where density areas [24]. The choice of optimal routes for hazard- yearly road kill of roe deer represented between 11% to ous materials transportation is one of the preventive 15% of the total recorded mortality of this species in the measures to minimize possible risk [8]. Also, the land use period 2000-2006 [3]. planning represents an important tool for prevention or re-

duction of damages on people and other assets due to acci- REFERENCES dents involving hazardous materials [25]. These authors made preliminary assessment of criticality and vulnerabil- [1] Seiler, A. (2004) Trends and spatial patterns in ungulate-vehi- ity of the assets within Belgrade city area in respect to cle collisions in Sweeden. Wildlife Biology 10, 301-313. chemical sites and transportation roads that can be exposed [2] Pokorny, B. (2006) Roe deer-vehicle collisions in Slovenia: to chemical accidents, or terrorist attacks. Their results in- situation, mitigation strategy and countermeasures. Veterinar- dicate that most of hazardous zones and transport routes are skiArhiv 76 (Suppl.), 177-187. in conflict with densely populated areas and protected un- [3] Pokorny, B., Jelenko, I., Policnik, H. and Marolt, J. (2009) derground water sources. Responses to the present weak- Znacilnostiprehajanjaprostozivecihparkljarjevprekdrzavnih- cestpred in poimplementacijizvocnihodvracalnihnaprav. ER- ness of spatial arrangement of sensitive and hazardous as- ICoVelenje, Koncnoporocilo, Ljubljana, 1-170. sets were set up in the new Belgrade Master Plan (e.g. planned options for sitting of new establishments, recon- [4] Morelle, K., Lehaire, F. and Lejeune, P. (2013) Spatio-temporal patterns of wildlife-vehicle collisions in a region with a struction and relocation of existing establishments, safety high-density road network. Nature Conservation 5, 53-73. barriers and corridors on transportation routes, monitoring and public participation). [5] Rodríguez-Morales, B., Diaz-Varela, E. and Marey-Pérez, M. (2013) Spatiotemporal analysis of vehicle collisions involving wild boar and roe deer in NW Spain. Accident Analysis and Prevention 60, 121-133. 5. CONCLUSION [6] Iuell, B., Bekker, G., Cuperus, R., Dufek, J., Fry, G., Hicks, C., Hlavác, V., Keller, V., Rosell, C., Sangwine, T., Tørsløv, The Belgrade area is the largest urban center and vul- N. and Wandall, B. (Eds.) (2003) Wildlife and Traffic: A Eu- nerable territory in Serbia due to the numerous industrial ropean handbook for identifying conflicts and designing solu- plants which use, store or manufacture hazardous materi- tions. Prepared by COST 341 – Habitat Fragmentation due to Transportation Infrastructure, Ministry of Transport, Public als, as well as due to the dense and well-developed network Works and Water Management, Delft, The Netherlands. of roads and railways which are frequently used for [7] Stankovic, D., Jovanic, P., Krstic, B., Sijacic Nikolic, M., transport of hazardous materials. Our results showed that Trivan, G., Ivanovic, S. and Vucinic, A. (2013) Concentration the effects of roads and railways on large game (roe deer of PAHs in forest ecosystems of the protected natural resource and wild boar) in eight hunting grounds in the Belgrade “Avala”. Fresenius Environmental Bulletin 22, 136-141. area (nine municipalities) are significantly lower than in [8] Jovanovic, D. and Zivkovic, N. (2010) Routing problems in many European countries (e.g. Slovenia and Croatia). The transportation of hazardous materials. FactaUniversitatis 7, chemical accidents in transport of hazardous materials 43-51. were sporadic and their consequences were small-scale [9] Gacic, D.P., Danilovic, M. and Mladenovic, S. (2012) Com- contamination of the environment and economic losses. In parative analysis of hunting grounds in the area of Belgrade. recent years, there has been increased number of accidents International Scientific Conference “Forests in Future - Sus- caused by improper disposal of various types of chemical tainable Use, Risks and Challenges”, Belgrade, Proceedings, 1067-1074. waste and inappropriate implementation of agro-chemical products (pesticides and insecticides). For the purpose of [10] Groot Bruinderink, G.W. and Hazebroek, E. (1996) Ungulate protection of wildlife and their habitat due to transportation traffic collisions in Europe. Conserv. Biol. 10, 1059-1067. infrastructure in the hunting grounds of Serbia there should [11] Langbein, J., Putman, R. and Pokorny, B. (2011) Traffic colli- be further research with the aim to reliably determine the sions involving deer and other ungulates in Europe and available measures for mitigation, In: Ungulate Management in Eu- locations of intersection of ecological corridors of large rope: Problems and Practices, (Eds. R. Putman, M. Apollonio, game with the roads and railways, as well as to apply mit- and R. Andersen), 215-259. Cambridge University Press. igation measures more frequently and improve traffic acci- [12] Tajchman, K., Gawryluk, A. and Drozd, L. (2010) Effects of dent database at the national and local level. roads on populations of wild game in the Lublin region. TekaKom. Ochr. Kszt. Środ. Przyr. PAN, Lublin, 7, 420-427.

[13] Kitowski, I., Letowski, J. and Kepowicz, M. (2014) Fre- quency, phenology and ecological factors in roe deer (Capre- ACKNOWLEDGEMENTS oluscapreolus L.) – vehicle collisions in the Lublin County (East Poland) in the light of the police data. AnnalesUniversi- This study was realized as a part of the project “Estab- tatisMariae Curie-sklodowska Sec. Ee. Zootechnica 32, 9-17. lishment of Wood Plantations Intended for Afforestation of [14] Hell, P., Plavý, R., Slamecka, J. and Gasparík, J. (2005) Losses Serbia” (No 31041) financed by the Ministry of Education of mammals (Mammalia) and birds (Aves) on roads in the Slo- and Science of the Republic of Serbia. vak part of the Danube Basin. Eur J Wildl Res 51, 35-40. [15] Pintur, K., Slijepcevic, V., Popovic, N. and Andrijasevic, D. The authors have declared no conflict of interest. (2012) Dynamics of wildlife-vehicle collisions on roads of Karlovac County, Croatia. Journal of Central European Agri- culture 13, 340-349. [16] Sprem, N., Dudukovic, D., Keros, T. and Konjevic, D. (2013) Wildlife-vehicle collisions in Croatia – A hazard for humans and animals. Collegium Antropologicum 37, 531-535.

[17] Balciauskas, L. (2009) Distribution of species-specific wildlife-vehicle accidents on Lithuanian roads, 2002-2007. Esto- nian Journal of Ecology 58, 157-168.

[18] Kusta, T., Jezek, M. and Keken, Z. (2011) Mortality of large mammals on railway tracks. ScientiaAgriculturaeBohemica 42, 12-18.

[19] Mrtka, J. and Borkovcová, M. (2013) Estimated mortality of mammals and the costs associated with animal-vehicle col- issions on the roads in the Czech Republic. Transportation Re- search Part D 18, 51-54.

[20] Lagos, L., Picos, J. and Valero, E. (2012) Temporal pattern of wild ungulate-related traffic accidents in northwest Spain. Eur J Wildl Res 58, 661-668. [21] Dal Compare, L., Sturaro, E., Cocca, G. and Ramanzin, M. (2007) An analysis of roe deer (Capreoluscapreolus) traffic collisions in the Belluno province, eastern Italian Alps. Ital. J. Anim. Sci. 6 (Suppl. 1), 848-850. [22] City of Belgrade Development Strategy - Draft (2008) PALGO Center, (ed. Stojkov, B.), Belgrade, 1-216. [23] The Law on Transport of Hazardous Materials, Official Ga- zette of the Republic of Serbia, No 88/2010. [24] Regional Spatial Plan of the Belgrade Administrative Area, Official Gazette of the City of Belgrade, No 38/2011. [25] Stojanovic, B. and Jovasevic-Stojanovic, M. (2006) Chemical and radiological vulnerability assessment in urban areas. Spa- tium 13-14, 21-26.

Received: August 01, 2014 Revised: November 21, 2014 Accepted: January 26, 2015

CORRESPONDING AUTHOR

Dragan P. Gacic

University of Belgrade

Faculty of Forestry

no.1KnezaViseslavast

11000 Belgrade

SERBIA

Phone./fax.: +381113053865

E-mail: [email protected]

FEB/ Vol 24/ No 4/ 2015 – pages 1310 - 1317

PCCD/Fs AND BTEXs IN THE VICINITY OF AN INDUSTRIAL WASTE INCINERATOR IN NORTHERN ALGERIA

Fetta Ait Ahsene-Aissat1, Yacine Moussaoui2,*, Yacine Kerchich3, Feriel Guenane4 and Messaoud Hachemi1

1Unité de recherche matériaux, procédés, environnement (UR-MPE), Université de M’Hamed Bougara, Boumerdes, Algeria 2Université Kasdi Merbah Ouargla, Faculté des Mathématiques et Sciences de la Matière, Ouargla, 30000, Algeria 3Laboratoire des Matériaux et l’Environnement, Faculté des Sciences et Technologie, Université de Médéa, Ain D’Heb, 26001, Médéa, Algeria 4 Laboratoire d'Analyse des rejets solides et gazeux (ECFERAL), 14, Route de Larbâ, Z.I. El Harrach, Alger, Algeria.

ABSTRACT added a stored volume of 4,500,000 tons (MATE report, 2003 [1]) as well as dangerous expired pharmaceutical prod- Waste incineration is just beginning to develop in ucts (estimated amount 15,000 tons; MATE report, 2007 Algeria, and waste pollution is caused by PCDD/Fs and [2]), and large quantities of persistent organic pollutants VOCs emissions that should be monitored and quantified. (POPs), such as pesticides and other waste operators, are This study presents a qualitative and quantitative analysis stored across the country and waiting for to be eliminated. of volatile and semi-volatile organic compounds (BTEXs, In order to resolve this problem of increasing waste PCDDs and PCDFs), and their average concentrations were amount rapidly, the Public Authority sees to equip the main evaluated by emissions stack monitoring of an industrial Algerian cities with proper industrial waste incinerators waste incinerator (ECFERAL Company). For the BTEXs (IWI) due to their low costs. At the moment, the policy- collection, the active sampling was deployed in the vicinity makers hesitate between incinerators and technical land- of the ECFERAL Company. However, the PCDD/Fs col- fills as a solution. lection was accomplished by active sampling using an isokinetic sampler in the gas release from the incinerator In order to select the right solution for the best tech- chimney. The BTEX results of samples by GC/FID analy- nical method of household and industrial waste destruction, sis showed that the average level was 77.3 µg.m-3, while we must evaluate the right way leading to the reduction of the recorded atmospheric level of PCDD/Fs varied from 21 environmental pollution. to 774 pg TEQ m-3. In this case, the continuous control of emissions from incinerator stacks or those released from the technical landfills are necessary to assess the level of air pollutants, espe-

KEYWORDS: industrial waste incinerator (IWI), GC/FID, High cially those well-known to be very dangerous, such as poly- Resolution Gas Chromatography/High Resolution Mass Spectrom- chlorodibenzodioxins (PCCDs), dibenzofurans (PCBs), etry (HRGC/HRMS), benzene, toluene, ethylbenzene, and xylene- polychlorinated biphenyls, mono- and poly-aromatic hydro- volatile aromatic compounds (BTEXs), polychlorinated dibenzodi- carbons [3, 4]. oxins (PCDDs), polychlorinated dibenzofurans (PCDFs), incineration, toxic equivalent quantity (TEQ). Persistent organic pollutants, such as PCDDs and PCDFs, are known by their toxicity and persistence properties in environmental media. They are emitted from combustion or re-emitted from reservoirs (water, soils, and sedi- 1. INTRODUCTION ments), and are transported to distant locations through atmospheric or aquatic pathways. However, they are also bi- In the recent years, waste management in Algeria has oaccumulated through the food web, and pose a risk of become one of the major scientist and government con- causing harmful effects to human health. The food concerns. The diversity and quantity of household and indus- sumption is the most important way for human exposure to trial wastes are growing up, and there is a need to act these contaminants. In addition to their presence in outdoor quickly in order to find appropriate solutions. air, PCDD/Fs have been found in different indoor environ- According to the Algerian Ministry of Territorial In- ment [5-7]. stallations and Environment report (Ministère de l’ Amé- In addition to the PCDD/Fs resulting from chimney nagement du Territoire et de l’Environnement, MATE), the emissions of IWI, inhaled volatile organic compounds total amount of industrial waste is 2.6 million tons per year, (VOCs) can cause various disorders in exposed humans [8]. with 325,000 tons of hazardous waste. To these amounts is Within VOCs, the BTEXs are known to be very dangerous and cause risks for the human health, particularly benzene, * Corresponding author which is known to be cancerous [9, 10]. However, the VOCs

transformation in the atmosphere generates secondary pollu- represents a schematic diagram of the industrial waste in- tants involved in the urban smog photochemical reactions. cinerator, model “Nar 5000” equipped with a flue gas treat- PCDD/Fs are formed by two main ways, such as the ment station and purge neutralization. novo synthesis [11, 12]. Several parameters are involved The BTEX and PCDD/F samplings were taken follow- and have impacts on PCDD/Fs level emissions, such as ing the incineration of 3 expired drugs, such as solid temperature [13, 14], oxygen and chlorine content [15, 16], Doliprane 500, liquid Maltitol and pasty Tefos. and the waste nature and composition [17]. Mainly the chlorine presence in the waste is until today controversed 2.2 Sampling methods by several researchers; there are those who published that 2.2.1 BTEXs sampling there is no relationship between dioxin emissions and con- In this case, the collection was carried out by using a tent of chlorinated waste [18, 19] whereas others have pub- high-precision portable gas pump (Buck I.H. pump, Or- lished against the addition of chlorinated compounds lando, USA). The flow-rate was set at 250 ml min-1. The which increase the PCDD/F production [20, 21]. chosen cartridges for the active VOC collection (ORBO- The goal of this study is to determine the PCDD/Fs and 32 type, Supelco) were filled with twin-bed activated co- BTEXs levels at the emissions stack of the industrial waste conut charcoal (6 mm o. d., 75 mm length, 20-40 mesh). incinerator by monitoring. For this purpose, an isokinetic The former bed (100 mg) acted as a VOC collector, while sampler for the active sampling and a portable pump for the second one (50 mg) as a breakthrough control. The col- BTEXs sampling in the vicinity of the IWI Company, lo- lection time ranged from 3-5 h, depending on incineration cated at El Harrach, Algiers, were used. This paper adds to process. In total, 11 active tubes were collected. the current literature to increase the PCDD/Fs and BTEXs atmospheric measurement data of industrial areas in north- 2.2.2 PCDD/Fs sampling ern Algeria. The particle sampling was performed by means of an isokinetic sampler (Clean Air Engineering, Marseille, France) operating at an average flow of 40 L min-1. Two 2. MATERIALS AND METHODS m3 of air were drawn by a Whatman glass micro-fiber filter (110 mm diameter) where atmospheric particles >0.1 lm in 2.1 Site sampling diameter were trapped (grade GF/F, Whatman, purchased Sampling was carried out at the stack emission of an from Fourni-Labo, Versailles, France). IWI company (IWI, ECFERAL SPA) located at El Har- The particles-loaded filters were directly maintained in rach, 10 km at the east periphery of Algiers (Fig. 1). methylene chloride to prevent any degradation, and, then The sampling of BTEXs was carried out by using an stored at 4 °C before analysis. The sampling was carried active sampler in the vicinity of the industrial waste incin- out at the stack emission of an IWI company (IWI, erator (ECFERAL Company), at an altitude of 3.5 m and a ECFERAL SPA) located at El Harrach, at 10 km distance, radius of 30 m from the IWI. For PCDD/Fs sampling, an at the east periphery of Algiers (Fig. 1). isokinetic sampler at the stack emission was used. Figure 2

El- Harrach

FIGURE 1 - Detailed position of the sampling site.

FIGURE 2 - Schematic diagram of the industrial waste incinerator, model “Nar 5000”. 1 - Charging system 17 - Trays absorption acid gas 2 - Primary Fan 18 - Inspection door 3 - Burner combustion chamber 19 - Devisiculor 4 - Combustor 20 - Sheath connecting 5 - Burner afterburner 21 - Overflow 6 - Secondary air fan 22 - Chinese heat 7 - Bedroom afterburner 23 - Fireplace 8 - Junction equipped with a safety register 24 - Control panel 9 - Quench 25 - Induced draft fan 10 - Venturi 26 - Pit rainwater drain 11 - Neutralization tank water purge 27 - Fose collection of filtered water 12 - Metering pump 28 - Submersible pump 13 - Tank saturated aqueous solution of caustic soda 29 - Decanter water purge 14 - A circulation pump of the washing waters 30 - Sand filter 15 - Washwater tank 31 - Water reservoir 16 - Plate column 32 - Heat exchanger

2.3 Chemicals and reagents and [13C]1,2,3,4,7,8,9-HpCDF was used as the recovery A certified benzene mixture, the n-heptane, toluene, standard (EDF-4145). ethylbenzene, m-xylene, p-xylene, o-xylene, isopropylben- zene and (EPA TO-1 Toxic Organic Mix 1A, 2 mg/ml of 2.4 The control quality/assurance quality each in methanol), purchased from Supelco (Sigma Al- 2.4.1 BTEXs drich, Milan, Italy), and CS2 (Fluka, reference 84713-low All the analytical procedures were monitored using in benzene, ≥99.5%) were used for experiments; 1-chlorooc- strict assurance measurements of quality and control. Labor- tane (Supelco) was adopted as internal standard compound. atory and field blanks consisted of charcoal coconut glass tubes, and Analyst-I devices were run in each field cam- Analytical purity nonane puriss., standard for GC, was paign. Three laboratory blanks and eight field blanks were purchased from Fluka (Steinheim, Germany). Anhydrous analyzed in total. sodium sulfate was analyzed by Baker (J.T. Baker, The process detection limit (DL) for each BTEX was Deventer, The Netherlands). determined as equal to three times the standard deviation of Helium (99.99%) was purchased from Air products the obtained signal from three repeated measurements of (Vilvoorde, Belgium). The internal standard solution of the blanks, divided by the calibration curve slope. DL values as 13 -3 17 2,3,7,8-chloro-substitued C12 congeners labeled low as 0.05-1 μg m were obtained for the BTEXs series. PCDD/Fs (EDF-4144), the calibration standard solution All the recorded analytical results exceeded the correspond- (EDF-4143), and the syringe (recovery) standard (EDF- ing DLs. To evaluate the analyte recoveries, three blanks of 4145) were purchased from Cambridge Isotope Laboratory both active and passive sampling devices were mixed with (Cambridge Isotope Laboratories, Andover, MA, USA). the target BTEXs from 0.04 to 35 mg L-1. Recoveries 13 For PCDDs and PCDFs, a mixture of [ C6]1,2,3,4-TeCDD, ranged from 80 ± 5% to 110 ± 3%.

2.5 Extraction and analysis Vap® II). The last clean-up was carried out with 200 mg 2.5.1 BTEX sulphuric acid (44%, v/v) of silica gel in a Pasteur pipette For the BTEX analysis, the activated carbon sample column. The purified extract was then transferred into a GC was desorbed using 2 ml and left to rest with a weak shaking flask and concentrated to10 µl in nonane. Finally, the ex- during 0.5 h. The extracted substance was analyzed by gas tract was enriched with 5 µl of recovery standard (EDF- chromatography with flame ionization detector (GC/FID), 4145) before HRGC–HRMS analysis to determine the 13 and 1-chlorooctane was used as an internal standard. achieved recovery rates for the C12-labeled internal stand- The gas chromatograph (GC-17A, Shimadzu) was ards which were 83 ± 11 for PCDFs and 77 ± 10 for PCDDs. equipped with a capillary column Supelco BP5 (length 15 m, diameter 0.53 mm, 0,5 µm film thickness). The temperature program was as follows: initial temperature 35 °C (0 min), 3. RESULTS AND DISCUSSION rising rate 3 °C min-1 up to 50 °C; then with 5 °C min-1 up 3.1 BTEX to 100 °C, and 20 °C min-1 up to 220 °C. Other operating conditions of the GC analyzer were injector temperature Table 1 presents the BTEXs identified and quantified 250 °C and detector temperature 250 °C. The BTEXs iden- at the IWI vicinity throughout 1 week (active sampling). tification was based on comparison with pure standard. The chromatogram of a sample collected at the IWI vicinity eluted through the BP5 column is shown in Fig. 3. A 5-µl sample extract was used for one injection. Each sample was injected twice and the average results were re- Benzene and xylenes (o-, m- and p-) are the most abun- ported. Tube blanks were analyzed in the same way as the dant BTEXs identified at the vicinity of industrial waste samples, and all results were corrected. incinerator; they represent about 77 % of the total BTEXs. Toluene and ethylbenzene were also quantified and their -3 2.5.2 PCDD/Fs concentrations were, respectively, 20.6 and 5.9 μg m . The The glass microfiber filters were Soxhlet-extracted total BTEXs recorded concentration at the vicinity of the -3 with toluene during 16 h. The extract was then evaporated IWI was equal to 73.3 µg m . at 250 ml with precision, using a rotary evaporator. An al- TABLE 1 - Average concentrations of BTEXs (μg m-3) at the vicinity iquot portion (one tenth) of the extract was transferred into of IWI. a bottom balloon for the analysis. Few nonane droplets were added as keeper and 10 µl of the internal standard con- Compounds Retention time (min) C (µg.m-3) ±SD 13 taining the 17 C12-labeled PCDD/Fs specified as (EDF- Benzene 4.03 31.6 ± 28.7 4144) were dropped to the aliquot. The toluene was then Toluene 6.48 10.6 ± 1.7 evaporated with a rotary evaporator. Ethtylbenzene 10.56 5.9 ± 5.2 (m/p)-Xylene 11.76 20.6 ± 13.7 The extract in hexane was then subjected to an acid/ o-Xylene 12.51 4.6 ± 3.2 base clean-up, followed by a clean-up on a silica gel column BTEXs ̶ 77.3 (modified with silver nitrate). The extract was evaporated at Toluene/Benzene ̶ 0.34 10 ml and transferred into a 50-ml tube for a Zymark addi- (m/p)-Xylene/Benzene ̶ 0.65 tional evaporation down to 500 µl, with a turbovap (Turbo o-Xylene/Benzene ̶ 0.15

1 1 : Benzene 2 : Toluene 3 : Ethylbenzene 4 : (m+p)-Xylene 5: o-Xylene 2

5 3 4

FIGURE 3 - Gas chromatographic separation of BTEXs.

The toluene, m-, p-xylenes and o-xylene vs. benzene Table 2 gives the average value of BTXs recorded with concentration ratios were calculated to get information about previous studies resulting from different industrial areas. the BTEX sources. In fact, when the 3 ratios approach the According to Table 2, the benzene, toluene and the total values of 2.7, 1.8 and 0.9, respectively, cars are clearly con- BTXs concentrations measured in this study were much firmed as the emission source [22-24]. According to Table 1, lower than those recorded at a gasoline refinery in Taiwan the 3 calculated ratios above for BTEXs, in the vicinity of [32], and a municipal waste (UK) [33]; they were much IWI, were much lower than those proposed by technical lower than those recorded at Algerian municipal waste literature; this fact clearly indicated that the BTEXs result- [34], and a petrol refinery in Taiwan [35]. In contrast, the ing from IWI emissions were only with minor contribution benzene, toluene and average BTX concentrations meas- of traffic emissions. ured in this study were very high with regard to those recorded at municipal waste and a refinery in Rome (Italy) The recorded BTEX levels in this study were higher [36], industrial area in Korea [37], an oil refinery in Greece than those recorded in the vicinity of a waste organic frac- (Kalabokas et al. [38], 2001), a gasoline refinery in Taiwan tion treatment plant, at Barcelona (Catalonia, Spain), -3 and the industrial area in Canada (Cheng et al. [39], 1997). where the BTEX levels were 9.88 and 8.49 µg m , respec- However, several sampling procedures have been used in tively in February and July [25]. More, they were higher previous studies, as well as the period of sampling as indi- than those measured in the vicinity of a waste organic frac- cated below in Table 2. tion treatment plant at Tarragona (Catalonia, Spain), where -3 17 µg m was recorded in December 2007, but in June and 3.2 PCDD/Fs December 2008, the BTEX levels were, respectively, 7,1 and 6.3 µg m-3, and in June 2009, the BTEX level was 7.7 The PCDD/Fs and dl-PCBs profile of an IWI emission µg m-3. is represented in Fig. 4. The total PCDD/Fs concentration ranged from 0.47 to 10.82 ng m-3 (Table 3). The PCDFs are The BTEXs level measured in this study was lower the main contributors to the overall PCDD/Fs concentra- than that detected recently in the industrial zone of Kocaeli tion of samples. For the sample filter 1 (Doliprane 500), the (Turkey), where a BTEXs range of 3.7–335 μg m-3 was rec- profile is dominated by OCDD (21%), followed by OCDF orded [26], while a high variability of results has been re- (20%), then 1,2,3,4,7,8,9-HpCDF (20%), and finally, ported in urban areas worldwide [27-29]. Concerning the 2,3,4,6,7,8-HxCDF (7%). For the sample filter 2 (Maltitol), benzene [30], recently, levels from non-detected to 16.6 μg the profile is dominated by 1,2,3,4,7,8,9-HpCDF, followed m-3 in the Tarragona urban and industrial areas have been firstly by 2,3,4,6,7,8-HxCDF OCDD (11%), then by detected, lower than the records of this study. Contrarily to 1,2,3,4,6,7,8-HpCDD (9%), and 1,2,3,6,7,8-HxCDF (8%), this study, these investigators have also noted that toluene and finally, OCDF (7%). For the sample filter 3 (Tefos), was the most abundant compound. This fact could be ex- the profile is dominated by 1,2,3,4,6,7,8-HpCDD (21%), plained by the toluene results coming from motorway followed by OCDD (20%), then by 1,2,3,4,6,7,8-HpCDF emissions. Besides, benzene is the only one which VOC (18%), and finally, by OCDF (13%). The TEQ contents for levels are currently legislatively restrained in the urban air. the PCDD/Fs and the doliprane-500 sample according to A threshold value of 3.5 μg m3 is set by the EU through the the NATO and OMS I-TEF were, respectively, 0.551 and Air Quality Framework Directive (2000/69/CE). The ben- 0.447 ng I-TEQ m-3. A high level was recorded for the zene concentration measured herein exceeded, by about ten Tefos sample, where the TEQ contents for the PCDD/Fs, ac- times, the guidelines established by the European Union cording to the NATO and OMS I-TEF, were, respectively, for ambient air, which recommended the limit threshold of 0.762 and 0.716 ng I-TEQ m-3. However, the TEQ content for 3.5 μg m-3 (European Directives 2008/50/CE) [31]. the PCDD/Fs for the Maltitol sample was 0.21 ng I-TEQ m-3,

TABLE 2 - Comparison of average values of BTXs between industrial areas of different countries and in this study.

This Study [A] [B](a, c) [C]e [D]c [E]e [F]f [G]g [H]h [I]i Benzene 31.6 ± 28.7 1.0 a 0.4 a 2.1 0.8 110 6.8 12.7 0.8 71 Toluene 10.6 ± 1.7 3.0 a 1.3 a 3.9 1.7 118 5.4 17.0 1.2 253 2.5 (b, c) 1(b,c) BTX 77.3 8.7 3.8 248 13.2 157 2.6 391 7.6 (c, d) 3(c, d) [A]: Refinery (Rome) [36]; [B]: Municipal waste (Rome) [36]; [C]: Industrial area (Korea) [37]; [D]: Oil refinery (Greek) [38]; [E]: Petrol refinery (Taiwan) [35]; [F]: Petrol refinery (Taiwan) [32]; [G]: Municipal waste (United King-dom) [33]; [H]: Industrial area (Canada) [39]; [I]: Municipal waste (Algeria) [34]. a Automatic analyzer; b Average diurnal concentration; c Adsorbent trap sampling method and GC-FID analysis; d Average nocturnal concentration; e Stainless steel container sampling method and GC-FID analysis; f Canisters sampling method and GC-FID or MSD analysis (average of petroleum refinery boundary (2–4 sampling points, June 2000)); g Passive sampling method and GC-FID analysis, average of SS1–SS9 sampling points pre- incinerator shutdown; h Canister sampling method and GC-FID analysis; i Adsorbent trap sampling method and thermal desorption GC-MS (TD/GC/MS) analysis.

1 TABLE 3 - Toxicity equivalent factors of PCDD/Fs in the gas stack of IWI emissions (F1: Doliprane 500; F2; Tefos; F3: Maltitol).. I-TEQ NATO; 2 I-TEQ OMS

I-TEF NATO I-TEF OMS F1 F2 F3 Dioxines 2,3,7,8-TCDD 1 1 0.012 0.075 0.001 1,2,3,7,8-PCDD 0.5 1 0.042 0.137 0.006 1,2,3,4,7,8-HxCDD 0.1 0.1 0.055 0.147 0.006 1,2,3,6,7,8-HxCDD 0.1 0.1 0.076 0.176 0.013 1,2,3,7,8,9-HxCDD 0.1 0.1 0.057 0.133 0.010 1,2,3,4,6,7,8-HpCDD 0.01 0.01 0.664 0.704 0.099 OCDD 0.001 0.001 2.257 0.510 0.095 Furanes 2,3,7,8-TCDF 0.1 0.1 0.079 0.271 0.003 1,2,3,7,8-PCDF 0.05 0.03 0.343 0.486 0.005 2,3,4,7,8-PCDF 0.5 0.3 0.399 0.527 0.010 1,2,3,4,7,8-HxCDF 0.1 0.1 0.555 0.636 0.016 1,2,3,6,7,8-HxCDF 0.1 0.1 0.602 0.689 0.016 2,3,4,6,7,8-HxCDF 0.1 0.1 0.760 0.860 0.021 1,2,3,7,8,9-HxCDF 0.1 0.1 0.111 0.102 0.004 1,2,3,4,6,7,8-HpCDF 0.01 0.01 2.110 1.754 0.083 1, 2,3,4,7,8,9- HpCDF 0.01 0.01 0.557 0.307 0.024 OCDF 0.001 0.001 2.144 0.608 0.061 Somme PCDD/Fs (ng m-3) 10.82 8.12 0.47 ng I-TEQ m-3 (PCDD/Fs)1 0.516 0.762 0.021 ng I-TEQ m-3 (PCDD/Fs)2 0.447 0.714 0.021

(A) (B)

(C)

FIGURE 4 - PCDD/Fs levels for IWI emissions (A): Doliprane-500, (B): Tefos, (C): Maltitol.

for both NATO and OMS I-TEF (Table 2). The TEQ con- than 7 times for the Tefos sample; this fact illustrates tent for PCDD/Fs measured in gas stack emissions in this clearly that the PCDD/Fs formation is linked to sample na- study exceeded those measured at the same incinerator in ture as subject for the incineration. the previous study, when the level of TEQ content of PCDD/Fs and dl-PCBs was 264 pg for I-TEQ m-3 [40], this latter value was similar to those measured for Maltitol, ACKNOWLEDGEMENTS PCDD/Fs for the doliprane-500 and Tefos samples, and exceeding the PCDD/Fs maximum limit given by the Euro- The authors would like to thank the management of the pean Union legislation and set at 100 pg I-TEQ m-3 as a company ECFERAL for the disposal of the human and ma- threshold limit. The high value of I-TEQ resulting from the terial resources in order to accomplish this work. gas stack emission during the sample burning can be explained by the presence of chlorine sources. The chemical The authors have declared no conflict of interest. structure of three samples does not content chlorine, except for 1 sample which is the doliprane-500, where the com- REFERENCES bustion of this sample took place in presence of the package made from polyvinylchloride. Another important pa- [1] MATE (2003). Report National for Algeria, Sustainable De- rameter, the responsibility of the dioxin-furan formation is velopment Commission of United Nation. http://enpi- the CuCl2, later acting as a catalyst. Another way to explain seis.ew.eea.europa.eu/south/algeria/country-report-alge- high levels of dioxin-furan in the gas stack of emission is ria.pdf/ the present source in the old ash removal in the different [2] MATE (2007). Report National for Algeria, Sustainable De- parts of chlorine NAR 5000 incinerator, and we note that velopment Commission of United Nation. http://enpi- seis.ew.eea.europa.eu/south/algeria/country-report-alge- the incineration process is continuous, and occurs without ria.pdf/ cleaning the different incinerator parts. [3] Buser, H.R., Bosshardt, H.P., Rappe, C., and Lindahl, R. The PCDD to PCDF ratios have been calculated. The (1978). Identification of polychlorinated dibenzofuran isomers average ratios for the 3 samples were found to be, respec- in fly-ash and PCB pyrolyses. Chemosphere. 7, 419-29. tively, 0.41 for Doliprane 500; 0.30 for Tefos, and 0.95 for [4] Olie, K., Vermeulen, P.L. and Hutzinger, D. (1997). Chloro- Maltitol. These average ratios are very high to PCCD ver- benzo-p-dioxins and chlorodibenzofurans are trace compo- sus PCDF ratio recorded at the stack flue gases emission of nents of fly ash and flue gas of some municipal incinerators in sinter plants in Taiwan (i.e. ratio between 0.10-0.14) [6], the Netherlands. Chemosphere. 6, 455-459. and very close to those recorded in an industrial complex [5] Loos, R., Vollmuth, S. and Niessner, R. (1997). Group separa- of chemical and oil refinery industries in South Korea (i.e. tion of ortho-PCBs, coplanar non-ortho-PCBs and PCDDs/PCDFs using an alumina column as an one-step clean- ratio of 0.31) [41]. However, ratios reported in the litera- up procedure. Fresen. J. Anal. Chem. 357, 1081-1087. ture can vary from <0.2 to >5, typically providing clues [6] Wang, L.C., Lee, W.J., Tsai, P.J., Lee, W.S., and Chang- about sources and degradation processes [42]. The ob- Chien. G.P. (2003). Emissions of polychlorinated dibenzo-p- served ratio in this study tends to support the influence and dioxins and dibenzofurans from stack flue gases of sinter impact of degradation processes, which had as origin the plants. Chemosphere. 50, 1123–1129. removal fly ash. [7] Weschler, C.J., and Nazaroff, W.W. (2008). Semi-volatile organic compounds in indoor environments. Atmos. Environ. 42, 9018-9040. 4. CONCLUSIONS [8] Nedellec, V., Lapkoff, J., and Rabl., A. (2012). Hiérarchisa- tion des polluants chimiques émis par les installations de traitement des déchets ménagers en France basée sur les im- Herein, the 17 toxic PCDD/Fs congeners emitted by pacts sanitaires non cancérigènes. Environnement, Risques et gas stack chimney and BTEXs at the vicinity of an indus- Santé.11, 120-136. trial waste incinerator (IWI), located in Northern Algeria, [9] Benedetti, M., Iavarone, I., and Comba, P. (2001). Cancer risk have been analyzed. associated with residential proximity to industrial sites: A Re- The measured benzene concentration in this study ex- view. Archives of Environmental Health. 56, 342-349. ceeded by about ten times the guidelines established by the [10] Elliott, P., Shaddick, G., Kleinschmidt, I., Jolley, D., Walls, P., Beresford, J., and Grundy, C. (1996). Cancer incidence near European Union for ambient air (recommended limit: municipal solid waste incinerators in Great Britain. British -3 threshold of 3.5 μg m ). Journal of Cancer. 73, 702-710. The TEQ content of PCDD/Fs of the three drug sam- [11] Hatanaka, T., Katajima, A., and Takeuchi, M. (2005). Role of ples was found to vary between 0.47 and 10.82 ng m-3. Re- chlorine in combustion field in formation of polychlorinated gardless of the sample drugs, the PCDFs were the major dibenzo-p-dioxins and dibenzofurans during waste incineration. Environ. Sci. Technol. 39, 9452-9456. contributors to TEQ. The observed PCDD to PCDF ratio tends to support the influence and the impact of degrada- [12] Bie, R., Li, S., and Wang, H. (2007). Characterization of PCDD/Fs and heavy metals from MSW incineration plant in tion processes as results of removal fly ash. Harbin. Waste Management. 27, 1860-1869. The TEQ content at the IWI sampling site exceeded [13] Fiedler, H. (1998). Thermal formation of PCDD/PCDF a sur- the European threshold limit of 100 pg I-TEQ m-3 by more vey. Environ. Eng. Sci. 15, 49-58.

[14] Cheng, H., and Hu, Y. (2010). Curbing of dioxins emission [32] Chang, C.C., Sree, U., Lin, Y.S., and Lo, J.G. (2005). An exami- from municipal solid waste incineration in China: rethinking nation of 7:00-9:00 PM ambient air volatile organics in different about management policies and practices. Environ. Pollut. seasons of Kaohsiung city, southern Taiwan. Atmos. Environ. 39, 158, 2809-2814. 867-884. [15] Wang, L.C., Lee, W.J., Lee, W.S., Chang-Chien, G.P., and [33] Leach, J., Blanch, A., and Bianchi, A.C. (1999). Volatile organic compounds in an urban airborne environment adjacent to a munic- Tsai, P.J. (2003). Effect of chlorine content in feeding wastes ipal incinerator, waste collection centre and sewage treatment of incineration on the emission of polychlorinated dibenzo-p- plant. Atmos. Environ. 33, 4309-4325. dioxins/dibenzofurans. Sci. Total Environ. 302, 185-198. [34] Yassaa, N., Meklati, B.Y., Brancaleoni, E., Frattoni, M., and Cic- [16] Zhang, H.J., Ni, Y.W., Chen, J.P., and Zhang, Q. (2008). In- cioli, P. (2001). Polar and non polar volatile organic compounds fluence of variation in the operating conditions on PCDD/F (VOCs) in urban Algiers and Saharan sites of Algeria. Atmos. En- distribution in a full-scale MSW incinerator. Chemosphere. viron. 35, (4), 787-801. 70, 721-730. [35] Lin, T.Y., Sree, U., Tseng, S.H., Chiu, K.H., Wu, C.H., and Lo, [17] Nakao, T., Aozasa, O., Ohta, S., and Miyata, H. (2006). For- J.G. (2004). Volatile organic compound concentrations in ambient mation of toxic chemicals including dioxin-related compounds air of Kaohsiung petroleum refinery in Taiwan. Atmos. Environ. by combustion from a small home waste incinerator. Chemo- 38, 4011-4122. sphere. 62, 459-468. [36] Gariazzo, C., Pelliccioni, A., Di Filippo, P., Sallusti, F., and Ceci- [18] Rigo, H.G., Chandler, A.J., and Lanier, W.S. (1995). The Relation- nato, A. (2005). Monitoring and analysis of volatile organic com- ship between chlorine in waste streams and dioxin emissions from pounds around an oil refinery. Water, Air, and Soil pollution. 167, waste combustor stacks, in: ASME Research Report CRTD, 17-38. American Society of Mechanical Engineers, New York, 36. [37] Na, K., Kim, Y.P., Moon, K., Moon, I., and Fung, K. (2001). Con- centrations of volatile organic compounds in an industrial area of [19] Rigo, H.G., and Chandler, A.J. (1998). Is there a strong dioxin: Korea. Atmos. Environ. 35, 2747-2756. chlorine link in commercial scale systems? Chemosphere. 37, 2031-2046. [38] Kalabokas, P.D., Hatzianestis, J., Bartzis, J.G., and Papagianna- kopoulos, P. (2001). Atmospheric concentrations of saturated and [20] Katami, T., Yasuhara, A., and Okuda, T. (2002). Formation of aromatic hydrocarbons around a Greek oil refinery. Atmos. Envi- PCDDs, PCDFs, and coplanar PCBs from polyvinylchloride dur- ron. 35, 2545-2555. ing combustion in an incinerator. Environ. Sci. Technol. 36, 1320- 1324. [39] Cheng, L., Fu, L., Angle, R.P., and Sandhu, H.S. (1997). Seasonal variations of volatile organic compounds in Edmonton, Alberta. [21] Wikstrom, E., Ryan, S., Touati, A., and Teffer, M. (2003). Im- Atmos. Environ. 31, 239-246. portance of chlorine speciation on de novo formation of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzo-furans. En- [40] Moussaoui, Y., Tuduri, L., Kerchich, Y., Meklati, B.Y., and Eppe, viron. Sci. Technol. 37, 1108-1113. G. (2012). Atmospheric concentrations of PCDD/Fs, dl-PCBs and some pesticides in northern Algeria using passive air sampling. [22] Brocco, D., Fratarcangeli, R., Lepore, L., Petricca, M. and Ven- Chemosphere. 88, 270-277. trone, I. (1997). Determination of aromatic hydrocarbons in urban air of Rome. Atmos. Environ. 31, (4), 557-566. [41] Oh, J.E., Choi, J.S., and Chang, Y.S. (2001). Gas/particle partitioning of polychlorinated dibenzo-p-dioxins and dibenzofurans in at- [23] Guicherit, R. (1997). Traffic as a source of volatile hydrocarbons mosphere; evaluation and predicting models. Atmos. Environ. 35, in ambient air. Sci. Total Environ. 205, (2-3), 201-213. 4125-4134. [24] Stevenson, K.J., Stacey, B., and Willis, P.G. (1997). Air quality at [42] Raun, L.H., Correa, O., Rifai, H., Suarez, M., and Koenig, L. (2005). Heathrow Airport Annual Report for 1996, AEA Technology. Re- Statistical investigation of polychlorinated dibenzo-p-dioxins and port No. AEA/RAMP/16402400/G3. dibenzofurans in the ambient air of Huston, Texas. Chemosphere. 60, 973-989. [25] Vilavert, L., Nadal, M., Figueras, M.J., and Domingo, J.L. (2012). Volatile organic compounds and bioaerosols in the vicinity of a municipal waste organic fraction treatment plant. Human health risks. Environ. Sci. Pollut. Res.19:96–104. [26] Pekey, B., and Yilmaz, H. (2011). The use of passive sampling to monitor spatial trends of volatile organic compounds (VOCs) at an industrial city of Turkey. Microchem. 97, 213-219. [27] Esplugues, A., Ballester, F., Estarlich, M., Llop, S., Fuentes- Leonarte, V., Mantilla, E., and Iñiguez, C. (2010). Indoor and out- Received: October 05, 2014 door air concentrations of BTEX and determinants in a cohort of Accepted: November 28, 2014 ne-year old children in Valencia, Spain. Sci. Total Environ. 409, 63-69.

[28] Giakoumi, A., Maggos, T.H., Michopoulos, J., Helmis, C., and Va- silakos, C.H. (2009). PM2.5 and volatile organic compounds CORRESPONDING AUTHOR (VOCs) in ambient air: a focus on the effect of meteorology. En- viron. Monit. Asses.152, 83-95. Yacine Moussaoui [29] Parra, MA., Elustondo, D., Bermejo, R., and Santamaría, J.M. Université Kasdi Merbah Ouargla (2009). Ambient air levels of volatile organic compounds (VOC) Faculté des Mathématiques et Sciences de la Matière and nitrogen dioxide (NO2) in a medium size city in Northern Spain. Sci. Total Environ. 407, 999-1009. Ouargla, 30000 ALGERIA [30] Ras. M.R., Marcé, R.M., and Borrull, F. (2010). Volatile organic compounds in air at urban and industrial areas in the Tarragona region by thermal desorption and gas chromatography-mass spec- Phone: 00 213 696 24 68 19 trometry. Environ. Monit. Assess. 161, 389-402. E-mail: [email protected] [31] European Directives 2008/50/CE, European Parliament and of the Council, May 21, 2008. FEB/ Vol 24/ No 4/ 2015 – pages 1318 - 1325

MEASUREMENTS OF MERCURY ASSOCIATED WITH AIRBORNE PARTICLE MATTER: USE OF TWO SAMPLING DEVICES

Zia Mahmood Siddiqi1,2* and Julia Lu1

1Department of Chemistry and Biology, Ryerson University, 350 Victoria St., Toronto, ON, M5B 2K3, Canada 2Jubail University College, P.O.Box10074, Jubail Industrial City 31961, KSA

ABSTRACT form, mainly Hg (II), in both gas and particulate phase [5]. When theses mercury species are deposited, through wet Two different sampling devices, quartz mini-sampler and dry processes, to surface of the planet, Hg (II) species with quartz fiber filter (6 mm diameter) [1] and Teflon open- can be converted to methyl mercury compounds that can face sampler with quartz fiber filter (47mm diameter) were be bio-accumulated in aquatic food chain [6]. The determi- compared for collecting airborne particulate matter for the nation of particulate mercury (p-Hg) in ambient air has analysis of total filterable mercury (TFM) in the atmos- been the subject of various studies because of the im- phere. Thermal pyrolysis was used for sample preparation and portance given to the dry and wet deposition flux of Hg and Cold Vapor Atomic Fluorescence Spectrometry (CVAFS) for its recycling in the environment [7-9]. mercury detection and quantification. The mini-sampler has a better time resolution (daily) than the Teflon open- A review of scientific and technical literature on sam- face sampler (normally one week integration time). The pling and analytical methodology for determination and chemical speciation of total particulate mercury (TPM) in latter has the advantages of better precision, convenience, time saving and cost effectiveness. The comparative meas- ambient air has been published [10]. For the first time urements of TFM in the ambient atmosphere have been eleven laboratories from North America and Europe met for international field inter-comparison of measurement done simultaneously using the two methods in downtown Toronto from June- December 2004. It is observed that, techniques for mercury species in ambient air [11]. The re- under the identical weather and analysis conditions, the sults of airborne particulate phase mercury showed significant difference between the participating four laboratories weekly concentration of TFM ranged from 22.9 10.6 to -3 -3 although two laboratories used thermal desorption fol- 56.0  52.2 pg m and 20.1  1.6 to 63.0  19.1 pg m by lowed by Cold Vapor Atomic Fluorescence Spectrometry mini-samplers and open-face sampler, respectively. Thus (CVAFS) detection. The difference could be attributed to the results showed a good agreement between the two sam- the two different sampling methods (using quartz wool pling methods. plug and Au trap preceded by Au denuders for collecting the p-Hg). The results suggested higher variability in p-Hg concentration on trap containing quartz wool plug than on disc filters. The conclusion drawn out after above study KEYWORDS: Methods inter-comparison, Mini-sampler, Teflon open-face sampler, Total filterable mercury, Atmospheric mercury was that the inter-comparability of the applied methods for p-Hg was not satisfactory.

Another study compared four different filters namely

quartz fiber filter (=6 mm), cellulose acetate filter (=47 1. INTRODUCTION mm), glass fiber filter (Gelman Type A/E), and Teflon fiber filter (=47 mm) for sampling and analysis of total par- Mercury is a toxic trace metal that can be released to ticulate mercury [12]. The results showed a large variabil- atmosphere from natural and anthropogenic sources and its ity in TPM measurements, possibly because of the differ- cycling has been significantly altered on local, regional and ences in filter types and local conditions (dust, vehicles, global scales by human activity [2, 3]. The prenatal expo- etc.) at sampling site. Further inter-comparison [12], how- sure to mercury has resulted in delayed development and ever, showed comparable results between the technique- cognitive changes in children [4]. In ambient air mercury combining mini-trap for sample collection/thermal desorp- may be found as elemental mercury (Hg0) and its oxidized tion for sample preparation and the one using Teflon filter for sample collection/acid-digestion for sample prepara- * Corresponding author tion. Another comparative study on sampling and analysis

for TPM using open-faced quartz filter (47 mm in diame- trometer) were obtained. Three parallel samples of airborne ter) showed that the thermo-reductive method (i.e., heating particulate matter and one B-2 type passive mode field the entire filter in a 0.95cm quartz tube at 800 - 900C) blank [10] were collected on a daily basis under a rain hood yielded lower values compared with microwave acid diges- using the mini-samplers described above. During sam- tion of filters [13]. The lower results might be a result of pling, ambient air was pulled through the mini-samplers the interfering species released during the pyrolysis pro- using an air pump (GAST, model DAA-V132-GB) and the cess or incomplete release of mercury from the samples un- flow rates were controlled to 1.94-2.00 L min-1. After colder the experimental conditions [13]. Also, it was observed lecting the sample for 24 hrs, the mini-samplers, along with that there was significant difference between the amounts the collected airborne particulate matter, were brought of p-Hg collected onto a filter at the down stream of a KCl- back to our laboratory for analysis. coated annular denuder and those collected without a front- In another method, two Teflon open-face holders with end KCl-coated denuder. While correction for this artifact quartz fiber filters (=47 mm) were used under another rain in the measurement of p-Hg was not given, but a complex hood side by side at the same sampling location, one of set of factors were suggested including sample duration, which served as a sample while the other as a passive mode 2+ 0 time of day, season, concentration of Hg (g) and Hg . (without air pulled through) field blank. Yet another four Metrological conditions as well as chemical composition Teflon open-face holders with quartz fiber filters (=47 of the air shed may also be determined in artifact for- mm), one of which served as passive mode field blank, mation. Another field inter-comparison study [14] showed were used under another rain hood. The quartz fiber filters the mini-sampler technique [10] is reliable, reproducible, were pre-cleaned by heating in an air oven at 600C for 4 simple, fast and cost effective as compared to others. A dif- hours. During sampling, the ambient air was pulled through ference of less than 4 folds and 1.3 folds has been observed the open-face sampler at the rate of 10.0 L min-1 by the when comparing methods operated by different laborato- similar pump as used for mini-samplers. After collection ries and the same laboratory, respectively. (for 168 hrs. and 24 hrs. respectively from two sets of sam- The determination of mercury in airborne particles is plers), the quartz fiber filters were taken out from the sam- of common interest in atmospheric chemistry. For the ac- pler, stored in petri dishes with the collected airborne par- curate determination of particulate-phase mercury species ticulate matter’s side up, and brought back to our labora- in ambient air, no internationally accepted and standard- tory for analysis. ized method currently exists. Therefore systematic research in this field is highly necessary and this manuscript 2.2 Determination of total filterable mercury deals with a very topical aspect of comparing two sampling A manual experimental system for TFM analysis was methods for collecting airborne particulate matter for anal- set up based on our published procedure [10]. Three heat- ysis of total filterable mercury (TFM), which is inter- ing steps in the procedure named as pre-cleaning of the an- changeably referred as total p-Hg or total particulate mer- alytical system, sample analysis (mercury detection and cury (TPM) by different investigators. data acquisition) and cleanup after sample analysis, were controlled by a programmable timer (ChronTrol, model XT-4F). During the sample analysis, the collected particles 2. MATERIALS AND METHODS on the filter were heated at 900°C for 5 min using nickel- chromium alloy resistance heating ribbon. A stream of argon 2.1 Sample collection at a flow rate of 100 mL min-1 was used to carry the released Two different devices, quartz mini-samplers [10] with mercury species from the sampling device to a pyrolyzer nickel screen support and Teflon open-face filter holders where all mercury compounds are converted into its ele- (Savillex, model 0-90-8) were used in this study. In each mental form, then to a gold trap where the elemental mercury device the same type of quartz filter (Whatman, pore size was absorbed. The elemental mercury was then thermally <10 m) but different size, 47 mm diameter in Teflon hold- desorbed at 500°C from the gold trap located downstream of ers and 6 mm diameter in quartz mini-sampler, was used to the pyrolyzer and detected by a CVAFS analyzer (Tekran collect airborne particulate matter on a rooftop of Ryerson 2500). The output signal from the detector was captured by University in downtown Toronto, Ontario, Canada. A brief an integrator (Agilent, 3396C). The pre-cleaning and post- description of the mini-sampler [10] is given as follows: a cleaning were done as described earlier [10]. custom- built quartz mini-sampler contains a quartz fiber For the analysis of the samples collected on to first set of filter disk. The filter disks, 6 mm in diameter, were cut two quartz filter (=47 mm) in the Teflon open-face holders, from a 47 mm quartz fiber filter. A nickel screen disk was Four (4) replicate filter disks, 6 mm in diameter each, were used to support the quartz fiber filter disk during the sam- carefully cut from different places (center, 10 mm, 15 mm and pling and determination steps. The pre-assembled samplers 20 mm from the center of filter) using a piece of nickel tubing were thoroughly cleaned by repeatedly heating at 500C (inner diameter: 6 mm, length: 12.5 cm). The tubing with filter for 5 min, with ultra-high purity (UHP) argon flushing on one end was placed one by one into a mini-sampler for through at a rate of 100 mL min-1, until a constant back- the determination of TFM following the procedure as de- ground signal (measured by cold vapor fluorescence spec- scribed above. A calibration curve for this analytical sys-

tem was constructed each day by injecting the known vol- once in a mini-sampler, and (3) after sample analysis, umes of air saturated with Hg0 vapor at a pre-selected tem- which was used for sample collection seven times in a perature. The field blank values were subtracted from the mini-sampler. analytical results (“raw data”). For the analysis of the samples collected on to second set of four quartz filter (= 47 mm) in the Teflon open-face holders, one filter disks, 3. RESULTS AND DISCUSSION 6 mm in diameter were carefully cut from the center of filters (using same nickel tuning) and analyzed as described 3.1 Effect of sample cutting position on open-face filter above). As a 6 mm diameter filter disk has to be cut from the filter of open-face samplers, the effect of cutting position 2.3 Raman Microscopy on filters was studied in this work. Practically around 35 filter disks Raman microscopic 20 images of fiber filter were can be cut from the 47 mm diameter filter. We selectively taken using a Renshaw Raman Microscope (WiRETM, ver- cut four discs, specified as the center, 10, 15 and 20 mm sion v1.3) to study the surface property of quartz fiber fil- from the center (Figure 1), to represent the coverage of the ter. The filter discs examined were the ones (1) after clean- entire filter surface and to be analyzed for TFM concentra- ing but without being used for sample collection, (2) after tions. Figure 1 reveals an unequal distribution of TFM on sample analysis, which was used for sample collection the open-face filter. The farther the distances from the cen-

FIGURE 1 - TFM concentration (pg m-1) from different cutting positions (mm) of filter disc (Weekly averages of TFM by mini-sampler is shown by stand-alone bars)

TABLE 1 - Comparison of the Quartz mini-sampler and Teflon open-face sampler for sampling and analysis of TFM in ambient air in downtown Toronto, Canada (Sampling time: 168 hrs)

Quartz mini-sampler Teflon open-face sampler na Range Weekly Av- Weekly na Range at the Weekly SDb nw Average SDw Date 2004 (pg m-3) erage (pg m- SDb center of filter Average (pg m-3) (pg m-3) (pg m-3) 3) (pg m-3) (pg m-3) (pg m-3) June 14-20 21 * 19.5-82.4 50.2 21.6 21 * 20.1-84.6 50.6 25.6 4 53.6 15.8 June 22-28 21 * 29.5-59.9 43.9 10.4 21 * 30.5-61.6 44.0 15.2 4 34.0 1.8 July 12-18 21 * 6.7-38.7 22.9 10.6 21 * 7.8-38.7 23.1 13.8 4 20.1 1.6 July 19-25 21* 19.7-71.5 39.7 18.7 21* 20.8-72.8 40.5 20.9 4 35.5 3.9 August 9-15 21 * 24.3-63.9 42.6 13.5 21 * 24.8-64.5 43.8 16.9 4 62.8 17.0 October 12-19 21 * 16.7-149.3 56.0 52.2 21 * 17.6-151.2 57.2 58.4 4 63.0 19.1 December 6-19 21 * 21.1-75.7 41.5 19.7 21 * 23.1-78.1 43.4 21.3 4 43.4 11.1 *Three samplers were operated side-by-side aNumber of determinations on daily samples wNumber of determinations on weekly sample bStandard deviation on daily samples wStandard deviation of measurements on weakly samples (4 positions)

ter of the filter, the lower the concentrations of TFM are. It sponds to a enrichment factor of 2 (=[TFM]0mm/[TFM]20mm). is reasonable when the spatial configuration of the open- Comparable results were found between the mini-sampler face filter sampler is considered. As the ambient air is and open-face sampler (ranged from 22.9 10.6 to 56.0  pulled in from the center of the open-face filter through a 52.2 pg m-3 and 23.1 13.8 to 57.2  58.4 pg m-3 respec- ¼ inch tubing by a pump, the area at the center of the open- tively). However, researchers [12] observed a higher con- face filter is expected to have more particles accumulated centration results by the open-face sampler as compared to than the outer areas. The values of weekly TFM concentra- mini-sampler probably because of (1) the acid-extraction tions at the center of filter (using open-face sampler) have analytical technique used by them is likely to contaminate been found to be 71.6, 41.5, 87.0, 84.0 and 58.7 pg m-3 the analyte (2) use of filters with different pore sizes in the (which are higher) whereas those at the cutting position samplers (quartz filter, pore size: 2.5-10 m in mini-sam- 20 mm away from the center of the filter are 38.2, 28.2, pler and Teflon filter, pore size 1-2 m in open-face sam- -3 32.1, 35.1 and 34.3 pg m (which are lower) compared to pler). For this comparison study we have used same type weekly averaged TFM concentrations by mini-sampler of filter (quartz) and identical integration time for the two -3 (50.2, 39.7, 42.6, 56.0, 41.5 pg m ). samplers, thus achieved better correlation between two However, Table 1 shows that the average values of the methods. Furthermore in our studies thermal pyrolysis tech- TFM concentrations from the open-face sampler’s four nique was used for analysis, which requires less tedious pro- cutting positions (the center, 10, 15 and 20 mm from the cedure and no chemical were used for sample pretreatment. center) are comparable with those from the mini-samplers. The different flow rates and sampling periods, in the two From these results, we recommend to cut 6-mm filter disks sampling methods (consider mini-sampler’s daily sam- from a belt area about 8-14 mm from the center of the filter. pling and open-face sampler’s weekly sampling), possible Up to 10 filter disks (=6 mm each) can be cut from this interaction of filter discs (and particulate material attached) belt area for observations. with inorganic mercury species (RGM: reactive gaseous mercury) could lead to artifact formation during sampling, 3.2 Comparison of quartz mini-sampler and Teflon open-face which is an “old” problem in the accurate determination of sampler particulate-phase mercury species in ambient air; but this Table 1 shows the comparison between quartz mini- inaccuracy problem is not prominent in the measured- sampler and Teflon open-face sampler for measurement of weekly concentration of TFM (ranged from 22.9 10.6 to TFM under the identical weather conditions between June- 56.0  52.2 pg m-3 and 20.1  1.6 to 63.0  19.1 pg m-3 by December, 2004. The results for the quartz mini-samplers mini-samplers and open-face sampler, respectively. and Teflon open-face samplers (both with 24-hr integration It is clear that the mini-sampler method has higher time sampling time) were compared. The results for the quartz resolution and less chance of contamination of sample when mini-samplers (24-hr integration sampling time, averaged compared with open-face sampler. But the Teflon open-face on weekly basis) were also compared with Teflon open- sampler technique is more precise (see Table 1) and may ef- face samplers (368-hr integration sampling time, weekly fectively replace the mini-sampler whenever data for TFM is basis) .When , ambient air is pulled at 10 L/min through required on weekly basis, thus eliminating the daily sampling, a 4mm-inner diameter (ID) tubing of an open-face filter cleaning, and analysis steps required for the mini-sampler. pack ( =40mm) for one week, the [TFM ] (pg) collected on the filter as a function of distance ( x, mm) from the center of 3.3 Reusability of mini-sampler filter the filter paper can be described by the following equation: For demonstrating the reusability of the filter, daily [TFM]= 1.7495x+69.882, R²=0.9809 (n=4), which corre- sampling using mini-samplers were conducted without

FIGURE 2 - The reusability of fiber filter using mini-sampler. Sampling dates: July 19-25, 2004, sampling time: 24 hrs, flow rate: 1.9-2.0 L/min, sampling location: downtown Toronto, Canada. Error bars represent standard deviation of three replicates (3 samplers were operated side- by-side)

FIGURE 3 - Comparison of weekly average concentration of TFM. July 19-25 2004, downtown Toronto, Canada; A and B: filters used in mini-sampler; C: filter used in open-face sampler. Error bars represent standard deviation of three replicates (3 samplers were operated side- by-side)

changing the filter disk. The results (Figure 2) show those residual materials on the filters after TFM has been ana- 5 days in a week, the re-used filter gives compatible results lyzed resulting in the reduction of the porosity of the filter. (after considering the error bars). This hypothesis has been confirmed from the microscopic images shown in Figure 4. The comparison of weekly average results obtained using the used filters, the brand new filters, and the open-face 3.4 Determination of TFM in ambient air filters (weekly sampling) show a good agreement (Figure 3). Figure 5 shows the overall temporal variations of TFM However large error bars in Fig. 3 shows that the sum collected by mini-sampler method from June to December of 5 days measurements rather average of measurement 2004 (on daily basis). A part of the data is recently reported could give better comparison with weakly measurement of for the calculation of dry deposition mercury species in To- open-face filter. The slightly higher concentrations of TFM ronto [15]. Mini-samplers were used for ambient air sam- from used filters may be a result of the accumulation of the pling because of the need for data on daily-basis as it has

FIGURE 4 - Raman Microscopic images of fiber filters at 20X (A) Filter before analysis (B and C) Filter after analysis sampled one time and seven times respectively

160 140 )

3 120 ‐ m

100 (pg

80 60 TFM 40 20 0 Jul Jul Jun Jun ‐ ‐ Oct Oct Sep Sep Dec Aug Aug Nov Nov Nov ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ 4 6 9 12 26 1 14 28 18 20 13 23 15 29 Year 2004

FIGURE 5 -Temporal distribution of TFM in downtown Toronto, Canada, Jun-Dec 2004

to be compared with other studies around the world. The 56.0 pg m-3), which is reasonably resulted from the weekly results from our study ranged between 6.7 – 149.3 pg m-3 integration characteristics of open-face filter method. The (see Table 1). These data are comparable with the values frequency distribution for the concentration of TFM shows reported [Lu et al., 1998] in metropolitan Toronto (3- 91 pg that majority concentration lies between 20- 60 pg m-3 from m-3) and in the Great Lakes region (1- 100 pg m-3) by [16]. both methods during the period of study. However 69- 90 pg m-3 range of total particulate mercury concentration was reported at Woodbridge site near To- ronto [17]. With the exception of the sampling site and 4. CONCLUSION date, meteorological parameters such as wind direction and relative humidity, are suggested to contribute to the differ- Both mini-sampler and Teflon open-face sampler offer ence between TFM concentration ranges by different re- compatible results of the total filterable mercury in the am- searchers [10]. The urban sampling environment is also a bient air. The quartz filter may be used in mini-sampler reason why our data is higher than other studies, as indi- without changing the filter on a daily basis. The open-face cated that buildings serve as sources of mercury to the ur- sampling approach is more economical for weekly analysis ban atmosphere [18]. Anthropogenic effects from down- while quartz mini-sampler offers better time resolution town Toronto resulted in the high spike (up to 149.3 pg m-3 (daily-basis). Researchers then have the options of choos- in 3 days) during our 46 days sampling period. The coning the method depending on the project objective and the centration range (20.1- 63.0 pg m-3) by open-face sampler laboratory conditions (e.g., funding, manpower, and instru- method falls in the range by mini-sampler method (22.9- ments). Total filterable mercury (TFM) in downtown To-

ronto varies from 20.1  1.6 to 63.0  19.1 pg m-3 detected [13] Lynam, M.M.; Keeler, G.J. (2002) Comparison of methods for particulate by using our two methods between June-December 2004. phase mercury analysis: sampling and analysis, Anal. Bioanal. Chem., 374, 1009-1014.

[14] Lu, J.Y.; Schroeder, W.H.; Keeler, G. (2003) Field intercom- ACKNOWLEDGEMENTS parison studies for evaluation and validation of the AES- miniSamplR™ technique for sampling and analysis of total Julia Lu is thankful to Natural Sciences and Engineer- particulate mercury in the atmosphere. Sci. Total Environ., ing Research Council of Canada (NSERC) and Ontario 304, 115-125. Premier Research Excellence Awards Program for support. [15] Zhang, X., Siddiqi, Z.; Song, X.; Mandiwana, K.L.; Yousaf, M.; (2012) Lu, J. Atmospheric dry and wet deposition of mer- The authors have declared no conflict of interest. cury in Toronto. Atmos. Environ., 60, 60-65. [16] Keeler, G., (1996) Atmospheric Mercury in the Great Lakes Region. UMAQL Newsletter issue 1. The University of Mich- igan, Ann Arbor, MI. [17] Schroeder, W.H.; Jackson, R.A. (1987) Environmental meas- REFERENCES urements with an atmospheric mercury monitor having speciation capabilities. Chemosphere., 16, 183-199. [1] Lu, J.Y.; Schroeder, W.H.; Berg, T.; Munthe, J.; [18] Cairns, E.; Tharumakulasingam, K.; Athar, M.; Yousaf, Schneeberger, D.; Schaedlich, F. (1998) A device for sampling M.; Cheng, I.; Huang, Y.; Lu, J.; Yap, D. (2011) Source, con- and determination of total particulate mercury in ambient air, centration, and distribution of elemental mercury in the atmos- Anal. Chem., 70, 2403-2408. phere in Toronto, Canada. Environ Pollut. , 159, 2003-2008. [2] Schroeder, W.H.; Munthe, J. (1998) Atmospheric mercury – An overview. Atmos. Environ., 32, 809-822.

[3] Lamborg, C.H., Fitzgerald, W.F., O’Dannell, J., Torgensen T. (2002) An examination of global-scale mercury biogeochemistry using a non-steady state compartment model which features interhemispheric gradients in the atmosphere as constraints. Geochimca et cosmochimca Acta., 66, 1-14. [4] Clarkson, T.W. (1997) The toxicology of mercury. Crit. Rev. Cl Lab Sci., 34, 369-403. [5] Feng, X.B.; Sommar, J., Lindqvist; O., Hong, Y. (2002) Oc- currence, emission and deposition of mercury during coal combustion in the Province Ghuizou, China. Water Air Soil Pollut., 139, 311-3247.

[6] Ulrich, S.M.; Tanton, T.W.; Abdrashitova, S.A. (2001) Mercury in the aquatic environment- A review of factors affecting meth- ylation. CRC Crit. Rev. Environ. Sci. Technol., 31, 241-293.

[7] Keeler, G.J.; Glinsorn, G.; Pirron , N. (1995) Particulate mercury in atmosphere: its significance, transport, transformation and sources. Water Air Soil Pollut., 80, 159-168. [8] Fitzerald, W.F.; Mason, R.P.; Vandal, G.M. (1991) Atmospheric cycling and air water exchange of mercury over mid continental lacustrine regions. Water Air Soil Pollut., 56, 745-767. [9] Lamborg, C.H., Fitzgerald, W.F., Vandal, G.M., Rolfhus, K.R. (1995) Atmospheric mercury in northern Wisconsin: sources and species. Water Air Soil Pollut., 80,189-198. Received: October 25, 2014 [10] Lu, J.Y.; Schroeder, W.H. (1998) Sampling and determination Accepted: January 09, 2015 of particulate mercury in ambient air: a review. Water Air Soil Pollut., 112, 279-295. [11 ] Ebinghous, R.; Jennings, S.G.; Schroeder, W.H.; Berg, T.; CORRESPONDING AUTHOR Doughy, T.; Guentzel, J.; Kenny, C.; Kock, H.H.; Kvietkus, K.; Landing, W.; Muhleck, T.; Munthe, J.; Prestbo, E.M.; Schneeberger, D.; Slemr, F.; Urba, A.; Wallschlager, D.; Xiao, Zia Mahmood Siddiqi Z. (1999) International field intercomparison measurements of Jubail University College atmospheric mercury species at Mace Head, Ireland. Atmos. P.O.Box10074 Environ., 33, 3063-3073. Jubail Industrial City 31961 [12] Munthe, J.; Wangberg, I.; Pirrone, N.; Iverfeldt, A.; Ferrara, KSA R.; Ebinghaus, R.; Feng, X.; Gardfeldt, K.; Keeler, G.; Lanzil- lotta, E.; Lidberg, S.E.; Lu, J.; Mamane, Y.; Prestbo, E.; Schmolke, S.; Schroeder ,W.H.; Sommar, J.; Sprovieri, F.; Phone: +966 546356165 Stevens, R.K.; Stratton, W.; Tuncel, G.; Urba, A. (2001) In- E-mail: [email protected] tercomparison of methods for sampling and analysis of atmospheric mercury species. Atmos. Environ., 35, 3007-3017. FEB/ Vol 24/ No 4/ 2015 – pages 1326 - 1332

SUBJECT INDEX

A G activated carbon 1291 growth 1282 aluminium 1282 Gulf of Aqaba 1232 Amman Zarqa basin 1176 antagonism 1250 H anthropophytes 1189 heavy metal pollution 1302 antioxidant 1204 High Resolution Gas Chromatography/ 1318 apophytes 1189 High Resolution Mass Spectrometry aquatic 1263 (HRGC/HRMS) artificial lagoons 1232 honey 1204 atmospheric mercury 1326 hunting ground 1310 azoxystrobin 1258 I B in vitro analyses 1204 Belgrade 1310 incineration 1318 benzene 1318 industrial waste incinerator (IWI) 1318 biofilter 1209 iron species 1168 biomass 1209

biosorption 1269 J C Jordan Valley 1176 castle flora 1189 catalyst 1291 K CdTe/MPA QDs 1275 Kozan 1189 chemical accident 1310 climate change 1176 L Colocasia esculenta 1250 life cycle assessment (LCA) 1215 contamination 1250 creeping jenny 1263 M

D manganese 1291 manganese oxides 1291 desertification 1195 MEDALUS 1195 desulfurization 1291 medicinal 1263 disinfection by-products (DBPs) 1209 membrane 1275 dissolved oxygen 1232 metabolites 1282 drinking water 1209 methods inter-comparison 1326 DOM 1168 minerals 1302 DON 1209 mini-sampler 1326 dumping ground 1227 Moringa oleifera seed 1269

E N ecology 1195 ecosystem 1195 nodal segments 1263 environmentally friendly 1269 nutrition 1250 environmentally sensitive areas 1195 enzymes 1282 O ethylbenzene 1318 oxidant 1204

F P flood event 1168 photosynthetic pigments 1282 fluidity 1275 phytohormones 1282 fluoride removal 1269 pilot pyrolysis plant 1215 flusilazole 1258 polychlorinated dibenzodioxins (PCDDs) 1318 polychlorinated dibenzofurans (PCDFs) 1318 G protein amount 1258 GC/FID 1318 pyrolysis 1215

AUTHOR INDEX

R A reclamation 1227 Aasim, Muhammad 1263 residence time 1232 Abdalla, Mona M. 1282 roe deer 1310 Agirman, Nur 1258 Ahsene-Aissat, Fetta Ait 1318 S Akgul, Hasan 1204 salinity 1232 Akyol, Ethem 1204 Scenedesmus acutus 1258 Al- Karablieh, Emad K. 1176 shoot tip 1263 Al-Houri, Zain M. 1176 simapro 8 1215 Al-Omari, Abbas S. 1176 soil properties 1227 Al-Weshah, Radwan A. 1176 Songhua River 1168 spectra 1275 B surface functional groups 1291 Banar, Müfide 1215 synanthropization 1189 Bedil, Burcin 1258 synergy 1250 C T Cetin, A. Kadri 1258 target plants 1227 Teflon open-face sampler 1326 D temperature 1232 Danilovic, Milorad 1310 temperate fruit species 1302 Dindaroğlu, Turgay 1195 threatened species 1189 Ding, Ling 1275 toluene 1318 Dobaradaran, Sina 1269 total dissolved iron 1168 Dogan, Hamide 1204 total filterable mercury 1326 Düzenli, Atabay 1189 toxic elements 1250 toxic equivalent quantity (TEQ) 1318 E traffic 1310 Gacic, Dragan P. 1310 Turkey 1204 Gacic, Jasmina 1310 Gajewski, Piotr 1227 U Gan, Huihui 1209 UV light 1275 Gilewska, Mirosława 1227 Gong, Mengdan 1291 W Guan, Jiunian 1168 waste tire 1215 Guenane, Feriel 1318 Water resources management 1176 Guo, Jiaxiu 1291 WEAP 1176 H X Hachemi, Messaoud 1318 xylene-volatile aromatic compounds 1318 Han, Xiujin 1275 He, Zhenyu 1275 Y yams 1250 J Jia, Xusheng 1275 Jin, Huixa 1209 Jin, Ling 1275 Jonnalagadda, Sreekanth B. 1250

K Kaczmarek, Zbigniew 1227 Kakuee, Marya 1269 Karakuş, Havva 1189 Karataş, Mehmet 1263 Kaya, Tuncay 1302

K Y Kendirlioglu, Gokce 1258 Yan, Baixing 1168 Kerchich, Yacine 1318 Yin, Huaqiang 1291 Keshtkar, Mozhgan 1269 Khorsand, Maryam 1269 Z Zazouli, Mohammad Ali 1269 L Zhang, Huining 1209 Li, Jianjun 1291 Zhang, Kefeng 1209 Liu, Maolan 1275 Zhao, Qingzhu 1275 Liu, Yongjun 1291 Zhou, Jinhui 1275 Lu, Julia 1326 Zhu, Hui 1168

M Manasrah, Riyad 1232 Mngadi, Sihle 1250 Mocek, Andrzej 1227 Moodley, Roshila 1250 Moussaoui, Yacine 1318

N Nabipour, Iraj 1269

O Otremba, Krzysztof 1227 Owczarzak, Wojciech 1227

P Pazira, Abdolrahim 1269 Pehluvan, Mücahit 1302

Q Qu, Yifan 1291

S Salman, Amer Z. 1176 Selamoglu, Zeliha 1204 Siddiqi, Zia Mahmood 1326 Şimsek, Uğur 1302 Stojnic, Dusan 1310

T Turan, Metin 1302 Türkmen, Necattin 1189

U Uma, Medine M. 1189 Unalan, Adnan 1204

W Wang, Lixia 1168 Wang, Xuejiao 1291

X Xu, Yingying 1168

1335