Available online at www.sciencedirect.com ScienceDirect

Procedia Environmental Sciences 28 ( 2015 ) 37 – 44

The 5th Sustainable Future for Human Security (SustaiN 2014) Metal transfer from marine coastal sediment to food chain: evaluating () persicus for monitoring metal bioaccumulation

Aziza Hamed Al-Farsia, Hameed Sulaimana, Hassan Ali Al-Reasia*

aDepartment of Biology, Sultan Qaboos University, Al-Khod, P.O. Box: 36, Muscat, Sultanate of Oman

Abstract

Protection of the ecosystems ultimately contributes towards their sustainability. Presence of high levels of toxic metals in seafood is considered as an environmental warning for possible influences on the ecosystem components and public health. This research quantified the concentrations of cadmium (Cd), chromium (Cr), copper (Cu), manganese (Mn), lead (Pb), vanadium (V), and zinc (Zn) in sediments and edible soft tissue of shellfish Strombus (Conomurex) persicus from the Sea of Oman. In both matrices, metals exhibited no general trend in the distribution of metals between the sampling sites. In shellfish, metal levels were below the international maximum permissible guidelines, illustrating safe consumption of this seafood item. The calculated risk factor i (Er ) for Cr, Cu, Pb, and Zn revealed no ecological damage should be anticipated at the reported metal concentrations in the i sediment samples. However, Cd had higher Er range, suggesting moderate to considerable ecological risk. For transfer of metals from sediment toS . persicus, bioaccumulation factors (BAF) estimated for Cd, Cu, and Zn were above 1.0, indicating tendency of i these metals to accumulate in S. persicus. While the BAF and Er values for Zn were insignificantly correlated, BAF values for i Cd and Cu were inversely related to their corresponding Er values. Nevertheless, correlation was not statistically significant for i Cu. For Cd, it seems that sediments having higher Er may not be necessarily resulting in higher tissue Cd burdens of S . persicus. i In conclusion, it appears that the use of correlation between BAF and Er to examine transfer of metals from abiotic component to organisms in natural waters is limited.

© 20152015 The The Authors. Authors. Published Published by Elsevier by Elsevier B.V ThisB.V. is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of Sustain Society. Peer-review under responsibility of Sustain Society

Keywords:metals; sediment; S. persicus; bioaccumulation

* Corresponding author. Tel.: +968 24146871; Fax: + 968 24143415. E-mail address:[email protected].

1878-0296 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of Sustain Society doi: 10.1016/j.proenv.2015.07.006 38 Aziza Hamed Al-Farsi et al. / Procedia Environmental Sciences 28 ( 2015 ) 37 – 44

1. Introduction

The marine ecosystems are rich with diverse wealth of organisms, fish, minerals, oil, natural gas, and other resources. Worldwide, about 3.0 billion people get 20% of their protein requirements from seafood1. Fishing with agriculture contributed about 1.3% (equivalent to about 1.0 billion US Dollars) of gross domestic product (GDP) of Oman in 20132. The development of fishing sector is important for food security and thus the Omani government is targeting the share of the fishing sector of the GDP to be raised to 2% by 20202. This goal can be primarily achieved through sustainable harvest and production of seafood. Globally, marine coastal areas are facing threats from pollution mainly due to human activities and consequently have direct and indirect impact on seawater organisms. Several pollutants in marine environments have been recognized in the Arabian Gulf, Sea of Oman, and the Arabian Sea including hydrocarbons (essentially due to oil extraction, processing, and transportation) and heavy metals. Contamination monitoring is an essential step for environmental protection which ultimately contributes towards sustainability of the ecosystems, in particular those providing services (e.g. food) for human. In Oman, metal concentrations have been recently determined in different marine samples including sediment, crustacean, and fish3, 4, 5, 6, 7.However, few studies have been directed for investigation of related phenomena like metal biomagnification4beyond the presentation of metal levels. Impacts of heavy metals on marine ecosystems are not recognized as soon as they are introduced into the seawater. For this reason, frequent monitoring programs are needed to detect changes in metal concentrations before they induce negative impacts on marine life. Since it is difficult and expensive to assess heavy metals level in many organisms from the ecosystem, biomonitor have been explored for metal pollution to indicate how healthy the ecosystem and its components is. Biomonitor species refers to any aquatic organism that accumulate metals in its tissues from the surrounding environment given that the change in its tissue metal burden is likely to reflect varying metal concentration in the surrounding environment8.The most common biomonitors in marine environment are invertebrates including molluscs (e.g. bivalves and gastropods). For example,benthic creatures can accumulate and transfer metals from sediment to higher organisms in the food chains9, but this does not exclude that benthic organisms can accumulate metals from other sources (e.g. water and food). Sediment is recognized as a potential source of trace metals for contamination of marine biota9. In aquatic sediment, presence of higher metal i 10, 11, 12 concentration can be evaluated for possible ecological risk using metal potential ecological risk index (Er ) . Regardless of the source of metals, whether it is natural or anthropogenic, the reported levels are good indication to assess the safety of seafood consumption when compared to the guidelines of the international organizations (e.g. Food and Agriculture Organization, FAO, of the United Nations and World Health Organization, WHO). But, how some metals accumulate in biota tissues while others do not, as illustrated by the wide variability in the species sampled and the locations of sampling,3, 5 remains a poorly-understood process. Bioaccumulation is the process by which an organism concentrates metal from abiotic environment (e.g. water, sediment) and through food ingestion, resulting in body concentration that is many folds higher than the metal concentration in the environment13, 14. This is usually expressed as bioaccumulation factor (BAF), the ratio of total metal concentration in the whole body of fish and/or organism to the total metal concentration from food and water and/or sediment obtained from the field 14 i studies . The main objective of this study is to investigate how likely BAF and Er to predict the transfer of metals from marine sediment to a gastropod species,Strombus (Conomurex) persicus (Swainson, 1821). This may provide insights for the use of this benthic invertebrate, which is endemic species to northern coastal water of Sea of Oman and the Arabian Gulfand is consumed by the local community15,16, asa potential candidate for monitoring metal pollution.

2. Materials and methods

On the coastal water of the Sea of Oman, coastal sediment and S . persicus samples were collected from 8 sampling sites during January 2013 from three cities or towns (Fig. 1). As detailed below, sediments and soft tissues of S . persicus were lyophilized and then digested using manual of oceanographic observations and pollutants analyses methods, MOOPAM protocol17.

Aziza Hamed Al-Farsi et al. / Procedia Environmental Sciences 28 ( 2015 ) 37 – 44 39

Fig. 1.Map showing the towns/cities where sediment and S. persicus (shown) samples were collected. The code and coordination of the sampling sites in each town/city on the Sea of Oman are given.

2.1. Sediment

Intertidal samples of sediment were obtained using plastic spoon and placed in pre-washed plastic containers. From each sampling site, different samples were collected and then homogenized to make a sample. The sediment samples were stored at ̶ 80 ºC till analysis. For the chemical analysis, sediment samples were first dried in freeze- dryer for 5 days. Then, each sediment sample was sieved using 150-μm mesh size to reduce the effect of different grain sizes on the analysis of heavy metals18. About 0.25g of fine sediment was accurately weighed in clean Teflon tubes using a single pan analytical balance to the nearest 0.01g. After that, 6 ml hydrofluoric acid (HF, 40% w/w) was added to the tube followed by the addition of 1 ml of aqua regia (3:1 Hydrochloric acid, HCl ~ 40% and nitric acid, HNO3, ~60 %). Finally, the tubes were heated using a microwave oven between 40–80 °C for 30 min.

2.2. Strombus persicus

Strombus persicus is a benthic marine organism from phylum (Class:; Family: )19. It is commonly known as Persian conch with an average size of about 50mm. It can be distinguished from other species by its cream to white shell color with brown patterns20 (Fig. 1). They inhabit sandy or intertidal bottom at ~ 5 to 10 meter below the surface water19. The S . persicus larvae and adults are herbivorous and closely associated with the sediments, representing a good candidate to study metal transfer at the base of the food web. Whole organisms were collected from the shoreline at the lowtide. About 20 to 30 S . persicus individuals of similar sizes (7.65 ̶ 9.1cm) were gathered in acid-washed plastic containers from each sampling site. Then, the samples were transported in a 40 Aziza Hamed Al-Farsi et al. / Procedia Environmental Sciences 28 ( 2015 ) 37 – 44

cool box to the laboratory where the soft tissues of each organism were separated from the shells and stored at ̶ 80 ºC until processing. Similar to sediment samples, soft tissues of S . persicus were lyophilized to remove all water content. Then, samples were ground to fine powder using a blender to ensure homogenization of samples. About 0.10 g of the ground soft tissue samples were weighed and transferred to a 30-ml polypropylene tube to which 4 ml of concentrated nitric acid (HNO3, ~60 %) was added. After that, the tubes were left at room temperature for 1 hour and then heated at 95°C for 2½ hours. Then, the tubes were left to cool down to the room temperature. Finally, the digested tissue samples were transferred to 50-ml acid-washed polypropylene tubes and the samples were diluted by addition of deionized water to final volume of 50 ml. The samples were stored at +4°C till chemical measurement.

i 2.3. Chemical analysis and calculation of bioaccumulation factor (BAF) and ecological risk index (Er )

For both sediment and S . persicus samples, the concentrations of cadmium (Cd), chromium (Cr), copper (Cu), manganese (Mn), lead (Pb), vanadium (V), and zinc (Zn) were determined using inductively coupled plasma optical emission spectrometer (ICP-OES). All concentrations are expressed in mg kg-1 dry weight. The accumulation of metals from sediment in the soft tissue of S . persicus was simply estimated according to the following BAF equation: Metal][ BAF S. persicus (1) Metal][ Sediment When BAF exceeds the unity (i.e. > 1.0), metal is likely to concentrate in S. persicus relative to the sediment. Data for metals with BAF values greater than 1.0, as dependent variable, were correlated with the metal potential i 10 ecological risk index (Er ), as independent variable. According to Håkanson , metal concentrations of sediments i could be utilized to derive Er as follow:

i i i CT Er = r CT f (2) i i i f 0 /= CCC n (3)

i i i Where Er is the individual heavy metal potential ecological risk index, Tr is the toxicity coefficient (Tr of Cd, Cu, 11 i i and Zn are 30, 5, and 1, respectively from Hilton et al. ), Cf is a contamination factor, Cn is a reference value of i -1 11 i metal (Cn of Cd, Cu, and Zn are 0.2, 30, and 80 mg kg , respectively from Hilton et al. ), C0 is the metal concentration in sediment samples in mg kg-1 dry weight.

3. Results and discussion

Figure 2 presents the mean total heavy metal concentrations ± standard deviation (SD) in mg kg-1dry weight in coastal sediments and S . persicus from three coastal town/cities on the Sea of Oman.

3.1. Metals in the sediment samples

In general, sediments exhibited wide ranges for metal concentrations. The lowest concentrations were recorded for Cd, while Cr and Mn existed in abundance relative to other metals (Fig. 2). Similar metals concentrations were obtained by Al-Rawahi7 for the coastal sediments in Muscat intertidal coastal areas. Similarly, Peer and Safahieh21 reported comparable metal levels to the present study for sedimentary metal concentrations from intertidal zone along the Iranian coasts on the Sea of Oman. The results showed no general trend in the spatial distribution of individual metals between the sites. Aziza Hamed Al-Farsi et al. / Procedia Environmental Sciences 28 ( 2015 ) 37 – 44 41

12 750 Cd Sediment Cr 600 10 S. persicus 450 8 300 150

dry wt.) 6 -1 20

4 (mg kg 10 Metal concentration 2

0 0 100 1000 Cu Mn 800 80 600 400 60 200 dry wt.)

-1 40 40

(mg kg Metal concentration 20 20

0 0 25 400 Pb V 300 20 200 15 100 dry wt.)

-1 2 10

(mg kg 1 Metal concentration 5

0 0 Sohar Saham A'Seeb Sohar Saham A'Seeb 120 Zn

100

80

dry wt.) 60 -1

40

(mg kg Metal concentration 20

0 Sohar Saham A'Seeb

Fig. 2. Concentrations of Cd, Cr, Cu, Mn, Pb, V and Zn in sediment and S . persicussamples collected from the coastal water of the Sea of Oman. Data are mean ± SD in mg kg-1dry weight

Though lower than the levels found in the present study, de Mora et al.3 reported that Cd concentrations (0.1–0.2 mg kg-1 dry weight) were the highest in Oman compared to the other Gulf countries. Even higher Cd concentrations (3.96 mg kg-1 dry weight) than those reported here were found by Al-Rawahi7 for sediment samples from A'Seeb. There may have been an increasing trend in Cr contents of the coastal sediments of along the Sea of Oman when data from this study compared to the other studies3, 7. The increasing trend in sediment contents of Cd and Cr may not be exclusively attributed to the recent industrial development and possible associated anthropogenic emission. Since Oman has Cd- and Cr-rich ophiolites3, the increase in Cd and Cr is likely due natural geologic release. The range in the present study (11.98 ̶ 19.68 mg kg-1) and in other studies (1.75 ̶ 14.8 mg kg-1 and 10.98 ̶ 15.79 mg kg-1)7, 21 suggested that Pb concentration has shown increasing trend when compared to the low range of 0.253 ̶ 1.82 mg kg- 1reported earlier3. No human activity is known to release Pb on the coastline of the Sea of Oman and atmospheric Pb deposition may be implicated. In addition, higher Mn concentration was found in the sediment (Fig. 2) than that 42 Aziza Hamed Al-Farsi et al. / Procedia Environmental Sciences 28 ( 2015 ) 37 – 44

reported by de Mora et al.3. Similar concentration ranges for Cu and V were detected in sediment of Sohar, Saham, and A’Seeb (Fig. 2) to other coastal areas of the Sea of Oman3, 7, 21. Vanadium (V) concentration in coastal environment has been used as an indicator of oil or crude oil pollution as it is an oil-related metal. Heavy traffic of crude oil tankers is known for the Sea of Oman which the passage between the Arabian Gulf and the Arabian Sea and Indian . For Zn, the data were in agreement with the data published recently7.

3.2. Metals in the S. persicus tissues

For S. persicus, higher concentration ranges for Cu and Zn were measured in the soft tissues (Fig. 2). This observation is not surprising since both metals are essential elements needed for certain metabolic pathways of organisms. The remaining metals were in the following order: Mn > Cr > Cd > Pb > V. Average metal contents in the shellfish were low and generally comparable among the sampling sites (Fig. 2). The concentrations were compared to the different maximum acceptable standards compiled for various metals in fishery products by FAO22. Average levels of Cd, Cu, Pb, and Zn in S. persicus from all the sites were within the permissible ranges suggested for shellfish and/or fishery products (Cd: 0.1 ̶ 5.5, Cu: 30 ̶ 70, Pb:1.0 ̶ 10; Zn: 40 ̶ 1500 mg-1 kg wet weight)22. It is worthy to note that concentrations in the present study were expressed on dry weight basis, nevertheless they would be lower than the proposed guidelines or standards. The findings may illustrate the safe consumption of this seafood item. Consumption of S . persicus by coastal communities is usually restricted to the winter season, making the risk to the public health minimal based on the reported concentrations (Fig. 2).

i 3.3. Correlation between bioaccumulation factor (BAF) and potential ecological risk index (Er )

Utilizing metal concentrations in sediments and the shellfish in equation (1) revealed that only Cd, Cu, and Zn exhibited BAF of greater than 1.0 for all sites (Table 1). This observation may demonstrate higher tendency of these metals to accumulate in S . persicus compared to other metals. These metals are likely to accumulate in biota close to shore3, 5. In addition, interference of Cu and Zn with reproduction of other gastropod species has been recognized23. i According to equations (2) and (3), the calculated Er values for Cr, Cu, Pb, and Zn were below 40 which is the maximum value for the low risk category as described elsewhere10, 12, suggesting no ecological damage should be observed at the metal concentrations determined in the sediment samples from every site. On the other hand, Cd had i Er range between 67 and 155 which implies moderate and considerable ecological risk for this metal.

Table 1. City/town, sampling sites and bioaccumulation factors (BAF) calculated for S. persicus from the coastal water of Sea of Oman. Metal BAF in S. persicus City/Town Code of the sampling sites Cd Cr Cu Mn Pb V Zn S-1: 8.43 0.01 2.96 0.06 0.55 0.02 2.26 Sohar S-2: 4.13 0.02 5.33 0.05 0.50 0.01 1.46 S-3: 6.84 0.03 3.76 0.08 0.89 0.01 2.87 S-4: 4.95 0.03 1.99 0.05 0.12 0.02 1.26 Saham S-5: 5.32 0.03 4.20 0.08 0.61 0.02 3.00 S-6: 4.49 0.02 4.47 0.06 0.10 0.00 2.39 S-7: 2.78 0.01 2.25 0.03 0.14 0.00 1.04 A’Seeb S-8: 5.07 0.07 2.67 0.08 0.14 0.03 1.69

i While the correlation between BAF and Er for Zn was not significant, BAF values for Cd and Cu were inversely i related to their corresponding Er values, though not significant for Cu (Fig. 3). Particularly for Cd, the relationship i i between BAF and Er is contrary to the prediction (i.e. the higher Er values were, the higher BAF values obtained) (Fig. 3). Similarly, BAF for several metals was reported to exhibit inverse relationships to metal exposure Aziza Hamed Al-Farsi et al. / Procedia Environmental Sciences 28 ( 2015 ) 37 – 44 43 concentrations (i.e. metal levels in the surrounding environment), a trend likely to result due to violation of the 14 i assumption that BAF is independent of the exposure concentration . It seems that sediments having higher Er (i.e. anticipating more ecological damage) may not be necessarily resulting in higher tissue metal burdens of S . persicus. i Based on the discussion above, the use of correlation between BAF and Er to examine transfer of metals from abiotic component (i.e. sediment) to organisms is refuted. In natural waters, the use of BAF as a regulatory tool was shown to be limited based on fish studies24. The evidence provided by the present study based on shellfish, invertebrate, emphasized BAF limitation in the field studies.

6 r = 0.45 5 p = 0.25 12 Cd 4 Cu Zn 3 10 BAF 2 1 r = 0.08, p < 0.85 8 0 01234 i 6 E r

4

Bioaccumulation factor (BAF ) 2 r = 0.77, p < 0.05

0 40 60 80 100 120 140 160

i Individual metal potential ecological risk index (E r)

i Fig. 3. Correlation between BAF and Er calculated based on concentrations of Cd, Cu and Zn of sediment and S. persicus from the coastal water of Sea of Oman.

Conclusion

Concentrations of heavy metals (Cd, Cr, Cu, Mn, Pb, V, and Zn) were determined in sediment samples and soft tissues of shellfish S. persicus from the Sea of Oman. The concentration were utilized to estimate bioaccumulation i factor (BAF) and ecological risk index (Er ). According to BAF, evaluation of S . persicus as a biomonitor species showed that the organism accumulated Cd, Cu, and Zn, but not the other metals (Cr, Mn, Pb, and V). Subsequent i correlation between BAF and Er may not be appropriate to predict accumulation of Cd, Cu, and Zn in soft tissue of the shellfish.

Acknowledgements

Special thanks to Ms. Jamila Al-Balushiand Mr. Khamis Al-Dhafri, technicians at the department of Biology, Sultan Qaboos University, for their technical help. Thanks to the people in Food and Water Laboratories Center, Ministry of Regional Municipalities and Water Resources, Oman. Special thanks to Marwa Al-Rawahi for her support, guidance and encouragement.

44 Aziza Hamed Al-Farsi et al. / Procedia Environmental Sciences 28 ( 2015 ) 37 – 44

References

1. Anonymous. Annual Report 2013. Muscat; Central Bank of Oman; 2014. 2. Anonymous. The state of world fisheries and aquaculture 2012. Rome; FAO fishieries and aquaculture department, Food and agriculture organization (FAO) of the United Nations; 2012. 3. de Mora S, Fowler SW, Wyse E, Azemard S. Distribution of heavy metals in marine bivalves, fish and coastal sediments in the Gulf and Gulf of Oman. Mar PollutBull 2004;49, 410 ̶24. 4. Al-Reasi HA, Ababneh FA, Lean DR. Evaluating mercury biomagnification in fish from a tropical marine environment using stable isotopes (δ13C and δ 15N). Environ Chem Toxicol2007;26: 1572 ̶81. 5. Fowler S, Villeneuve J-P, Wyse E, Jupp B, de Mora S. Temporal survey of petroleum hydrocarbons, organochlorinated compounds and heavy metals in benthic marine organisms from Dhofar, southern Oman. Mar Pollut Bull2007;54, 339 ̶67. 6. Al-Shuely W, Ibrahim ZZ, Al-Kindi A, Al-Saidi S, Khan T, Marikar FA, Al-Busaidi M. 2009. Heavy metals contents on beach sediments north and south of Sohar industrial area, Oman. J Environ Sci Tech 2009;2, 73 ̶9. 7. Al-Rawahi MM. An assessment of trace metals and polycyclic aromatic hydrocarbons concentrations in intertidal sediment and rock oyster from selected areas at Muscat coast, Sea of Oman. MSc thesis; 2012, Sultan Qaboos University, Muscat, Oman. 8. Rainbow PS. Biomonitoring of heavy metal availability in the marine environment. Mar Pollut Bull 1995;31, 183 ̶92. 9. Griscom, SB, Fisher NS. Bioavailability of sediment-bound metals to marine bivalve molluscs: an overview. Estuaries 2004; 27, 826 ̶38. 10. Håkanson L. An ecological risk index for aquatic pollution control of sediment ecological approach. Water Res1980;14, 975 ̶1000. 11. Hilton J, Davison W, Ochsenbein U. A mathematical model for analysis of sediment coke data. Chem Geol 1985;48, 281 ̶91. 12. Yi YJ,Wang ZY, Zhang K. Ecological risk assessment of heavy metals in sediment and human health risk assessment of heavy metals in fishes in the middle and lower reaches of the Yangtze River basin. Environ Pollut 2011;9, 257515 ̶85. 13. Leblanc GA. Basics of environmental toxicity. In: Hodgson E, editor. A Textbook of Modern Toxicology,New Jersey: John Wiley & Sons, Inc.; 2004, p. 463–78. 14. Wood CM. An introduction to metals in fish physiology and toxicology: basic principles. In: Wood CM, Farrell AP, Brauner CJ, editors. Homeostasis and Toxicology of Essential Metals, Oxford: Academic Press, Elsevier Inc.; 2012, p.1 ̶51. 15. Al-Khayat JA. Molluscs of the state of Qatar. Qatar Biodiversity Newsletter 2008;2: 1 ̶5. 16. Al-Farsi AHS. An assessment of heavy metal levels in benthic marine organism (Strombus persicus) and coastal sediment from Sohar, Saham and A’Seeb coastal areas in Oman. MSc thesis; 2013, Sultan Qaboos University, Muscat, Oman. 17. Anonymous. Manual of oceanographic observations and pollutant analyses methods ( MOOPAM). Al-Safat; Regional Organization for the Protection of the Marine Environment (ROPME); 1999. 18. Duarte CA, Giarratano E, Gil MN. Trace metal content in sediments and autochthonous intertidal organisms from two adjacent bays near Ushuaia, Beagle Channel (Argentina). Mar environ Res 2012;79, 55 ̶62. 19. Bosch DT, Dance SP, Moolenbeek RG, Oliver PG. Seashells of eastern Arabia. Dubai; Motivate Publishing; 1995. 20. Smythe K. Seashells of the Sultan Qaboos nature reserve at Qurm. Singapore; International Press Co, Ltd.; 1983. 21. Peer FE, Safahieh A, Sohrab AD, Tochaii SP. Heavy metal concentrations in rock oysterSocostrea cucullatafrom Iranian coasts of the Oman Sea. Trakia J Sci 2010;8, 79–86. 22. Nauen CE.Compilation of legal limits for hazardous substance in fish and fishery products. Food and Agricultural Organization (FAO) fishery circular No. 464. Rome; Food and agriculture organization (FAO) of the United Nations; 1983, p. 5 ̶100. 23. Spade DJ, Griffitt RJ, Liu L, Brown-Peterson NJ, Kroll KJ, Feswick A, Glazer RA, Barber DS, Denslow, ND. Queen conch (Strombus gigas) testis regress during the reproductive season at nearshore sites in the Florida Keys. PLoS ONE 2010;5, e12737(1 ̶14). 24. Couture P, Pyle G. Field studies on metal accumulation and effects in fish. In: Wood CM, Farrell AP, Brauner CJ, editors. Homeostasis and Toxicology of Essential Metals, Oxford: Academic Press, Elsevier Inc.; 2012, p.417 ̶73.