Biodiversity

ISSN 0976-6901 ONLINE: ISSN 2456-6543 JOURNAL OF BIODIVERSITY

J Biodiversity, 9(1-2): 69-80 (2018) © Kamla-Raj 2018 DOI: 11.258359/KRE-179

Assessment of Physico-chemical Integrity of Lotic Ecosystems in Central Western Ghats through Multivariate Techniques

T. V. Ramachandra1, 2 ,3*, V. Sincy1, K. S. Asulabha1 and S. Vinay1

1Energy and Wetlands Research Group, Centre for Ecological Sciences (CES), Indian Institute of Science, Bangalore, Karnataka, India 2Centre for Sustainable Technologies (ASTRA), Indian Institute of Science, Bangalore, Karnataka, India 3Centre for Infrastructure, Sustainable Transportation and Urban Planning (CiSTUP), Indian Institute of Science, Bangalore, Karnataka, India

KEYWORDS Aghanashini River Basin. Cluster Analysis. Principal Component Analysis. Stream. Water Discharge. Water Quality

ABSTRACT The Western Ghats is the water tower sustaining perennial rivers of peninsular India. This study assesses the physico-chemical integrity of lotic ecosystems in central Western Ghats through field investigations and multivariate analyses. This helped in understanding the seasonal variation pattern of water quality in different streams of Aghanashini river basin, Karnataka. Principal Component Analysis (PCA) reveal that water quality parameters such as total dissolved solids, electrical conductivity, total hardness, calcium, magnesium, potassium, ortho-phosphate, nitrate and water discharge play an important role in the streams across seasons. T he cluster analysis grouped stations based on physico-chemical integrity considering temporal and spatial aspects. PCA and cluster analyses confirm the vital role of water discharge during the three seasons. Discharge characteristics of a stream which varies among seasons and anthropogenic activities bring in wide variations in water quality of lotic ecosystems. Regular monitoring of streams is necessary to maintain and protect these pristine ecosystems. J Biodiversity, 9(1-2): 69-80 (2018) © Kamla-Raj 2018 DOI: 11.258359/KRE-179

Assessment of Physico-chemical Integrity of Lotic Ecosystems in Central Western Ghats through Multivariate Techniques

T. V. Ramachandra1, 2 ,3*, V. Sincy1, K. S. Asulabha1 and S. Vinay1

KEYWORDS Aghanashini River Basin. Cluster Analysis. Principal Component Analysis. Stream. Water Discharge. Water Quality

ABSTRACT The Western Ghats is the water tower sustaining perennial rivers of peninsular India. This study assesses the physico-chemical integrity of lotic ecosystems in central Western Ghats through field investigations and multivariate analyses. This helped in understanding the seasonal variation pattern of water quality in different streams of Aghanashini river basin, Karnataka. Principal Component Analysis (PCA) reveal that water quality parameters such as total dissolved solids, electrical conductivity, total hardness, calcium, magnesium, potassium, ortho-phosphate, nitrate and water discharge play an important role in the streams across seasons. The cluster analysis grouped stations based on physico-chemical integrity considering temporal and spatial aspects. PCA and cluster analyses confirm the vital role of water discharge during the three seasons. Discharge characteristics of a stream which varies among seasons and anthropogenic activities bring in wide variations in water quality of lotic ecosystems. Regular monitoring of streams is necessary to maintain and protect these pristine ecosystems.

INTRODUCTION Water quality of riverine systems depends on climate, topography, geology, soil proper- The Western Ghats is a chain of mountains ties, atmospheric deposition and catchment hy- extending from north to south along western drology of a region. Variations in river flow af- peninsular India, harbors diverse and endemic fect water quality by increasing or decreasing flora and fauna (Ramachandra et al. 2015). It has the effects like dilution, residence time, mixing 37 west flowing and 3 east-flowing rivers and and erosion (Purnaini et al. 2018). The rate of their tributaries (Sreekantha et al. 2007). Lotic evapotranspiration, infiltration, interception and ecosystem permit complex interactions as well percolation changes through modifications of as mass/energy transfer between all its abiotic catchment. Health of lotic ecosystems (streams and biotic entities and have the capability to re- and rivers) reflects a picture of ecosystem func- cover itself from the minor perturbations (Shar- tioning as well as the prevailing human interac- ma and Kansal 2013). Western Ghats harbor nu- tions including disturbances (Allan 2004). An- merous sacred groves (also known as Devar- thropogenic activities in the catchment can also akadus, Kavu etc.), which are patches of forests lead to higher concentrations of suspended having spiritual and religious importance in ad- solids, salts and nutrients (Lintern et al. 2018). dition to providing hydrological services, habi- Water bodies loose the capability of self-purifi- tat, bio-geochemical cycling, etc. (Ray et al. 2015; cation due to the influence of anthropogenic Ramachandra et al. 2016a). activities near the catchment (Aishvarya et al. 2018). Human activities in the watershed alter- *Address for correspondence: Dr. T.V. Ramachandra ing the landscape structure drives variability in Energy and Wetlands Research Group, CES TE15, the duration of flow and water quality (Huang Centre for Ecological Sciences, Indian Institute of Science, et al. 2014). The alterations in natural flow re- Bangalore - 560019 gime, changes an ecosystem completely caus- Website: http://ces.iisc.ernet.in/energy Phone: 080-22933099/22933503/Extn 107 ing variations in physical, biological and chem- E-mail: [email protected]; [email protected] ical conditions of streams and its riparian zones, 70 T. V. RAMACHANDRA, V. SINCY, K. S. ASULABHA ET AL. thereby affecting ecosystem biodiversity (Rolls cal factors like temperature, light availability, tur- et al. 2012; Bunn and Arthington 2002). Water bulence, precipitation, terrain, surface runoff, discharge varies spatially as well as temporally, soil, groundwater flow, vegetation, quantity and and increases along a stream network due to composition of litter, other land use practices inputs from rainfall, tributaries and groundwa- and aquatic biota. The main objective of the cur- ter. The rate of flow recession affects the habitat rent research is to investigate the changes in preference, growth and survival of aquatic biota the physical, chemical and nutrient parameters (Rolls et al. 2012). in selected streams of Aghanashini river basin, Land cover in the catchment and riparian mainly, Chandikaholé sub catchments. forests are vital ecological elements supporting diverse flora and fauna and perform a major role MATERIAL AND METHODS in nutrient cycling and maintaining pristine ecosystem (Girardi et al. 2016; Magdaleno and Mar- Aghanashini is a west flowing river with tinez 2014). Important drivers of diversity include catchment area of 1450.9 sq.km (Fig. 1), spread factors – type of the habitat, hydrologic vari- across the districts of Uttara Kannada (Ankola, ables, disturbance and stream morphometric, Kumta, Sirsi, Honavar, Siddapur) and portions geologic and lithological variables (Tornwall et of Shimoga (Sorab). The Aghanashini river orig- al. 2015). Precipitation rate, forests and soil char- inates in Sirsi taluk, Uttara Kannada district. The acteristics in the Western Ghats decides the river has two sources – Bakurholé tributary (at amount of water stored in vadose and saturated Manjguni) and Donihalla (near Sirsi) which zones during monsoon, which later get released meets at Mutthalli (Balachandran et al. 2012; to streams in post monsoon season (Ramachan- Ramachandra et al. 2015). Aghanashini river es- dra et al. 2016a). Monitoring of hydrological vari- tuary has dense and diverse mangrove forests ables with statistical analysis helps in better in- along the coast. Rainfall across the basins var- terpretation of hydrologic regime and quality of ies from as low as 3000 mm at the plains of Sirsi an ecosystem (Oketola et al. 2013). and to as high as over 5000 mm at the Ghats. Hydrological changes in forest dominated Temperature in the basins varies from as low as watershed are due to changes in LULC – land 15.30 in January to as high as 35.40 in April. Ele- use and land cover altering the structure of the vation in the catchment varies between less than landscape (Cui et al. 2012). Water retention ca- 0 meters to 786 m of Mean sea level (Ramachan- pability of the catchment declines with alter- dra et al. 2016b). In the current study, Chandika- ations in the catchment integrity. Forest type hole’s sub catchments were considered for as- and its soil modify precipitation rate, bio- sessment of flow. Chandikaholé originated at geochemical processes, water yield, water dis- Yaana and joins Aghanashini at Uppinapattana. charge and habitats thus interfering with the About 8 macro and micro watersheds were con- quality of water flowing to streams and other sidered taking into account i) similar climatic hydrological aspects (Neary et al. 2009). Run- conditions (about 3500 mm rainfall), ii) all the off and water yield reduces due to reduction in micro watersheds connect to a macro watershed, total forests, especially, old forest (Singh and iii) variability in landscape, etc. Mishra 2012). Conversion of forest lands for Yaana, Nanalli, Beilangi, Mastihalla, Harita agriculture/other land use increases overland are the micro watersheds connecting Aanegun- flow bringing about a decline in recharge and di (AGT1), whereas Aanegundi (AGT1 and finally, discharge to streams during non-mon- AGT2), and Bialgadde (BGT) are the macro wa- soon seasons. The sub-basins receiving good tersheds. Aanegundi AGT1 and AGT2 join near rainfall, with native species of trees in the catch- Yaana cross along the Sirsi-Kumta Road. The ment (covering to an extent more than 60-65%) descriptions of various micro/macro watersheds have increased stream flow even in lean sea- sampled are given in Table 1. Yaana and Nanalli sons (Ramachandra 2014). Riparian zone regu- catchments are dominated by evergreen forests, lates stream water quality and are involved in followed by Beilangi which consists of moist exchange and transfer of elements from terres- deciduous forests and evergreen forests where- trial to aquatic ecosystems. The seasonal and as, rest of the catchment areas have mixed land spatial variations in stream water quality is gov- use dominated by horticulture and agriculture erned by various abiotic, biotic and hydrologi- activities. PHYSICO-CHEMICAL INTEGRITY OF LOTIC ECOSYSTEMS 71

Fig. 1. Study area – Aghanashini river basin, Chandikaholé sub catchments

Table 1: Description of watersheds PCSTestr 35); turbidity (Hach Turbidimeter); dissolved oxygen (Winkler’s Method); chemical S. Watershed Area Major Land use No. Ha catchment oxygen demand (Closed reflux, titrimetric method); total alkalinity (Titrimetric method); bio- 1 Yaana 122.3 AGT2 Evergreen Forest chemical oxygen demand (5 day BOD test); chlo- 2 Nanalli 84.2 AGT2 Evergreen Forest ride (Argentometric method); total hardness and 3 Beilangi 393.1 AGT2 Mixed Forest 4 Harita 12.6 AGT2 Mixed land use calcium (EDTA Titrimetric method); magnesium; 5 Mastihalla 863.3 AGT2 Mixed land use nitrate (Phenol disulphonic acid method); ortho- 6 Aanegundi 7361.6 BGT Mixed land use phosphate (Stannous chloride method); sodi- AGT2 um and potassium (Flame emission photometric 7 Aanegundi 2012.6 BGT Mixed land use AGT1 method) of water samples collected from select 8 Bialgadde 10270.04 - Mixed land use streams were done according to the standard protocol of APHA AWWA WEF (2005). Water Sample Collection Statistical Analysis Water samples were collected covering all seasons for a period of two years during April, Multivariate techniques, Cluster Analysis 2014 to March, 2016 from nine different sites (CA) and Principal Component Analysis (PCA) namely Yaana (YK), Nanalli (YNK), Beilangi (BE), were computed using PAST software consider- Harita (HA), Mastihalla (MH), Aanegundi (AG), ing the water quality data. These techniques like Aanegundi tributary 1 (AGT1), Aanegundi trib- principal component analysis (PCA) and cluster utary 2 (AGT2) and Bialgadde (BGT) and analy- analysis (CA) helps in the data dimensionality sed physico-chemical parameters. Water temper- reduction while prioritizing the factors respon- ature, discharge, EC, TDS, pH, and DO were sible for variations in water quality parameters, measured at the collection site and samples were thus helping in better interpretation of water brought to the laboratory for further analysis. quality data (Arslan 2009; Zarei and Bilondi 2013; The analysis of physico-chemical parame- Gebreyohannes et al. 2015; Ramirez et al. 2014). ters like water temperature (laboratory thermom- The assemblage of sampling locations based on eter); discharge (current meter), pH (Eutech: their similarity were done through the cluster PCSTestr 35); total dissolved solids (Eutech: analysis (Bhat et al. 2014; Han et al. 2012; Mushatq PCSTestr 35); electrical conductivity (Eutech: et al. 2013). 72 T. V. RAMACHANDRA, V. SINCY, K. S. ASULABHA ET AL. MH HA MH 104.48 45.43 412.19 5.62 45.19 Monsoon YK YNK YK BGT BE AGT2 AG AGT1 Avg.SDAvg. 25.2SDAvg. 41.4 0.91SDAvg. 25.5 25.71 90.86SDAvg. 30.97 49.58 0.60 7.4SD 12.24 61.43 27.5Avg. 0.38 7.64SD 34.88 24.07 1.56Avg. 71.38 1.01 6.77 9.31SD 7.2 26.8Avg. 51.45 0.36 23.56 12.49 2.09 30.1SD 21.86Avg. 0.78 10.59 60.14 24.76 6.77SD 3.94 19.26 7.1 15.26 27 0.62Avg. 18.23 1.80 76.57 7.69 48.43SD 7.85 29.9 96.75 8.95Avg. 1.43 30.59 6.34 17.65 35.93 17.16 13.92SD 19.09 68.57 19.9 7 71.68 27.5 1.54Avg. 0.45 44.29 39.4 3.83 10.69SD 29.17 18.68 0.93 16.03 5.02 2.36Avg. 35.65 28.15 20.91 6.25 17.07 11.56 69SD 3.05 26.57 47.5Avg. 0.46 1.86 3.35 22.30 26.9 13.36 17.99 8.17 33.33 60.42 3.76 17.82SD 6.7 6.67 13.47 9.00 1.51Avg. 5.5 22.92 2.01 5.89 30.5 1.92 0.149SD 26.77 56.33 0.54 25.7 64.57 3.99 2.92Avg. 23.49 1.29 29.63 18.87 2.33 4.36 18.49 7.03 3.13 0.13 0.184 2.20SD 6.9 22.06 3.87 17.84 11.25 8.26 44.2 6.23Avg. 0.467 27.4 6.71 0.84 0.05 28 6.22 7.2 31.89 69.67SD 21.47 2.89 17.45 1.66 8.72 0.49 35.92 1.86Avg. 0.186 2.67 4.54 3.35 26.05 0.62 13.89 2.26 6.06 0.829SD 0.12 4.60 8.08 52 1.21 7.1 6.6 45.66 23.32 0.09 5.5 75.1 17.61 0.60 26.39 17.96 0.38 8.73 0.148 29.50 33.5 11.02 2.11 0.11 0.9 2.48 18.75 68.74 1.99 3.87 3.48 21.55 0.263 25.06 117.2 32.17 8.05 7.2 0.66 6.17 0.05 16.48 1.17 13.07 0.59 15.86 119.32 18.15 7.75 0.172 7.82 15.2 1.72 47.33 2.15 34.2 2.71 2.00 0.34 0.888 3.41 6.28 18.83 111.4 0.246 7.1 25.19 1.10 17.51 100.17 9.69 2.24 54.67 0.89 1.22 0.191 0.09 2.89 9.87 18.74 21.33 12.84 0.44 0.291 1.95 57.6 0.886 1.66 0.23 1.79 27.98 28.5 0.127 5.47 0.00 22.33 0.06 0.85 3.74 9.98 2.93 1.038 0.328 0.36 1.94 9 22 1.35 2.93 2.94 94.1 0.156 0.04 0.64 4.54 0.44 0.226 1.04 1.23 9.14 1.31 8.00 2.68 0.183 1.96 0.317 0.53 43.7 4.97 0.86 0.209 0.94 3.56 2.34 0.06 10.62 1 0.25 0.63 1.21 98.2 0.12 9.17 0.34 0.69 0.85 2.44 49.1 0.24 9.7 1.01 1.18 2.84 358.4 1.00 units: 1-°C; 3- µS/cm; 5- NTU; 2, 6-17 mg/l; 18-mm/day Water Temperature Water TDS EC pH Turbidity DO BOD COD Alkalinity Chloride Hardness Total Calcium Magnesium OP Nitrate Sodium Potassium Discharge parameter Table 2a: Water quality variations among select streams across monsoon season across quality variations among select streams Water 2a: Table S.No. Parameters 8 Note: 1 2 3 4 5 6 7 9 10 11 12 13 14 15 16 17 18 PHYSICO-CHEMICAL INTEGRITY OF LOTIC ECOSYSTEMS 73 Post-monsoon Pre-monsoon YK YNK BGT BE YK AGT2 AGT1 MH YNK BGT YK AGT2 BE Avg.SD 24.2Avg.SD 68.9 1.51 24.8Avg. 134SD 14 34.8 1.75Avg. 27.5SD 28.6 8.6 65.4Avg. 32.1 26.3 2.08 8.66SD 13.2 0.37 4.28Avg. 20.8 62.4 26.5 1.6 8.3SD 4.28 0.7 10.3 7.31 0.69Avg. 29.1 37 6.2 39.9SD 2.23 1.05 11 8.2 0.9Avg. 7.47 26.7 0.42 2.65 2.06 70.9 5.65SD 21.6 2.31 22.7 2.27 3.61Avg. 7.6 0.48 12 1.09 6.98 42 24.5 2.51 107 0.62SD 21.4 1.66 5.37 3.19 5.99 20.4 4.79Avg. 0.37 6.13 3.27 0.68 8.2 105 24.9 0.51SD 20.5 7.84 40.9 1.37 18.5 1.18 10.6 80 8.3Avg. 24.1 5.69 0.64 1.1 7.26 0.85SD 4.24 29.6 1.63 194 26.1 75.3 7.8 13 5.15 19.4 1.2 12.6Avg. 56.8 28 6.28 0.58 6.62 7.19 59.9SD 14 0.06 79 1.01 1.6 21.4 28.8 38.5 7.7 108 5.42Avg. 11.3 19.5 3.31 39.8 2.4 16.8 6.64 5.98 0.73 6.47SD 0.18 0.61 19.4 12.3 4.96 5.3 10.3Avg. 34.8 25.9 66.5 19 9.32 3.28 1.71 5.46 5.5 50.6 11.7 8 12.8 4.5 76.5SD 1.04 17.7 6.84 0.7 1.98 0.63 0.61 0.18 2.26 4.51Avg. 10.2 23 76 6.98 4.42 9.8 8.52 3.72SD 18.2 4.37 8.81 41 19.2 96.2 6.91 0.79 7.41 0.25 2.6 17.3 1.86 0.19 1.07 0.53Avg. 0.18 16.1 1.13 7.9 5.58 5.11 11.5SD 6.56 54.9 3.29 18.5 39.8 3.42 24.7 5.68 0.06 5.59 70 0.68 9.06 0.22 1.75Avg. 0.2 0.53 0.73 2.83 1.13 0.13 6.08 14.9 9.72 53.1SD 18.7 18.9 7.3 2.5 6.41 5.58 2.57 4.21 11.9 29.6 4.2 0.45 0.08 5.45 0.21Avg. 0.64 0.13 0.31 2.89 1.67 1.31 0.17 4.01 20.8 146SD 22.3 0.18 33.6 17.5 0.45 7.32 6.6 3.78 7.5 3.94 45 19.8 0.8 2.11 1.2 0.05 0.27 1.95 2.56 6.5 1.13 0.73 4.81 0.21 113 3.2 14.6 0.16 16 5.47 17.8 2.16 0.52 80 8.39 1.93 4.23 0.05 1.46 0.74 31.7 3.6 0.32 8.78 0.1 2.22 7.5 0.83 30.1 0.18 32.5 24.2 2.51 2.15 2.76 22.7 46.7 32.4 11.2 0.5 10.3 3.95 0.66 0.12 0.06 11.8 84 8.76 0.73 0.21 0.16 12.4 1.4 40 8.36 10.6 37 17 1.19 0.42 0.06 3.35 2.22 2.33 0.36 7.97 0.13 2.76 0.95 7.82 0.27 10 1.6 75.2 2.25 1.24 0.04 4.22 4.89 0.1 0.51 1.59 0.91 1.15 42.4 1.21 0.17 17 9.62 0.39 5.98 2.01 4.27 44 1.7 0.06 0.48 0.73 0.83 5.55 0.98 0.28 9.55 0.2 24 0.26 4.01 1.27 1.88 0.65 4.48 0.11 1.4 0.78 1.17 13.5 0.14 1.65 0.2 0.15 2.01 0.08 0.31 3.41 1.6 0.95 0.08 8 0.42 0.18 1.12 0.2 0.1 0.39 0.88 0.6 0.54 0.22 10.8 5.35 0.61 1.04 0.1 0.8 0.54 9.1 0.6 0.1 0.09 0.1 0.0 nits of: 1-°C; 3- µS/cm; 5- NTU; 2, 6-17 mg/l; 18-mm/day Water Temperature Water TDS EC pH Turbidity DO BOD COD Alkalinity Chloride Hardness Total Calcium Magnesium OP Nitrate Sodium Potassium Discharge Parameter u Table 2b: Water quality variations among select streams across post-monsoon and pre-monsoon seasons post-monsoon and pre-monsoon across quality variations among select streams Water 2b: Table S.No. Parameters1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Note: 74 T. V. RAMACHANDRA, V. SINCY, K. S. ASULABHA ET AL.

RESULTS AND DISCUSSION ride, calcium, magnesium, aluminium, nitrate, sulphate, sodium and iron. Organic compounds Stream water is being used for domestic pur- such as oil, phenol, alcohol and sugar are bad poses, irrigation, recreation, etc. Mapping and conductors and thus, affect the electrical con- monitoring of water quality would help in main- ductivity (Gandaseca et al. 2011; Al-Badaii et al. taining the quality and sustainable management 2013). The water discharge is very high at HA1 of the ecosystem. Water sampling was done at during monsoon (358.4 mm/day). In pre-mon- stations depending on the water availability in soon season, water temperature, total alkalinity, streams. Some streams are perennial like YK, chloride, total hardness, calcium and magnesium YNK, BGT, BE1 and AGT2. AG1 and HA1 dries were very high because of very less water level up in post monsoon season whereas, AG1, AGT1, and high evaporation rate. But, ion contents like MH1 and HA1 dries up during pre-monsoon. chloride, total hardness, calcium and magnesium Variability in the physico-chemical parameters and nitrate were very low in monsoon season of streams monsoon, pre-monsoon and post- because of dilution effect. monsoon seasons are presented in Figure 2. Turbidity and ortho-phosphate were present Tables 2a and 2b lists water quality across sam- in all sampling stations due to the entry of enor- pling locations for monsoon and post-monsoon mous quantity of rain water from the catchment seasons respectively. during monsoon (Table 2a). The ion contents All ion values were very high in all seasons (TDS, EC, chloride, total hardness, calcium, mag- at Yaana in comparison to other streams due to nesium and sodium), pH, total alkalinity were anthropogenic stress. Yaana is a popular eco- less due to the dilution with precipitation. Simi- tourist place visited by large number of pilgrims lar study reports of increased suspended sol- every day, and due to mismanagement of solid ids, turbidity and COD during monsoon as the and liquid waste the stream is polluted. Turbid- rivers received silt and debris from the catch- ity ranges from 1.1 – 51.45 NTU and is very high ments, which lowered the ionic contents due to in monsoon due to silt transport during runoff. the dilution (Rani et al. 2011). Large amount of pH which measures the alkalinity and acidity of phosphates entered the stream in rainy season water and influences chemical reactions in wa- from surrounding catchment areas dominated ter bodies. pH in the sampled streams ranges by rocks (Singh et al. 2013). from 6.7 – 8.6 and values are higher in post- During post-monsoon season (Table 2b), monsoon and low during monsoon due to the high pH and DO prevailed due to increased al- presence of organic matter. gal photosynthesis. The turbidity had decreased Dissolved oxygen in streams fluctuates due due to low discharge while organic matter and to movement/mixing of water, organic pollutants, nutrient levels (COD, ortho-phosphate, nitrate, photosynthetic and respiratory activities by flora potassium and BOD) decreased due to enhanced and fauna in aquatic ecosystems. DO ranges algal uptake and microbial activity. The water from 5.45 – 16.26 mg/L and higher values during temperature in post-monsoon appeared to be monsoon due to rapid water discharge with the suitable for the phytoplankton growth and re- aeration of water. COD ranges from 50.6 -17.71 production (Nassar et al. 2014). Higher chloride mg/l and BOD ranges from 33.33 – 6.98 mg/l and levels in post-monsoon season was due to re- higher BOD and COD values reflect the pres- duction in water levels and concentration of salts ence of organic matter. For unpolluted to moder- with the enhanced evaporative losses at higher ately polluted water, BOD5 values ranges be- temperatures (Shetty et al. 2013). The DO level tween >1.0 to <10.0 mg/l (Salman et al. 2013). depended on all the abiotic and biotic factors Turbidity, orthophosphate, BOD and COD were (Hacioglu and Dulger 2009) evident from higher very high in monsoon due to inflow of suspend- DO levels with turbulence and lowered DO dur- ed matter and organic compounds into streams ing higher organic content with escalated de- with the catchment run off. TDS and EC are very composition rates. Similarly, DO vary with tem- high in pre-monsoon season, but very low in perature (Arslan 2009). monsoon. TDS ranges from 104.67 – 19.9 mg/l The sampling sites had high ionic (TDS, EC, whereas EC ranges from 194.33 – 39.4 µS. Varia- total hardness, calcium, magnesium, sodium) and tions in electrical conductivity are brought about organic content (COD and BOD) but low DO by the inorganic dissolved solids such as chlo- and water discharge during pre-monsoon peri- PHYSICO-CHEMICAL INTEGRITY OF LOTIC ECOSYSTEMS 75 Fig. 2. Variations in water quality among 9 select streams in monsoon, post-monsoon and pre-monsoon seasons in monsoon, post-monsoon and pre-monsoon quality among 9 select streams in water Variations Fig. 2. 76 T. V. RAMACHANDRA, V. SINCY, K. S. ASULABHA ET AL. od (Table 2b). Chandrabhaga River, Maharash- about 9.32 percent showed high positive load- tra had maximum BOD of 31mg/l in summer and ings on turbidity. minimum BOD recorded was 9.9 mg/l in winter During post-monsoon (Fig. 3b), PC 1 ex- (Watkar and Barbate 2015). During pre-monsoon plained seventy-five percent of the total vari- season, YK was found to have high pH and ion ance (eigenvalue of 13.50) and showed strong contents such as TDS, EC, total alkalinity, total positive loading (>0.75) on TDS, EC, total hard- hardness, calcium and magnesium. Turbidity, ness, calcium, magnesium, ortho-phosphate, ni- orthophosphate, sodium, water discharge and trate, potassium and water discharge, whereas DO were higher in YNK. COD, water tempera- high negative loadings on water temperature, ture and chloride were higher in BGT. Tempera- pH, COD, total alkalinity, chloride and DO. PC 1 ture and precipitation influences the discharge showed moderate positive loadings (>0.50) on pattern in rivers (Kriauciuniene et al. 2012). Vari- sodium, BOD and DO. PC 2 corresponding to ations in temperature alter the physico-chemical the second eigenvalue (1.66) accounted for 9.21 properties of water and accelerate various chem- percent of the total variance, showed moderate ical reactions as well as metabolic activities. BOD negative loading on turbidity. PC 3 (eigenvalue and nitrate are higher in BE1. The amount of of 1.39) accounted for 7.73 percent of the total nitrates in water increases due to heavy rainfall, variance showed moderate positive loading land drainage, oxidation of ammonia (Vankar et (>0.50) on sodium. al. 2018); and the agricultural activities which During pre-monsoon (Fig. 3c), the first PC makes use of huge amount of chemical fertilizers explained 63.45 percent of the total variance and animal farming (Dunca 2018). (eigenvalue of 11.42) and was best represented by turbidity, total hardness, calcium, magnesium, Principal Component Analysis (PCA) ortho-phosphate, nitrate, potassium and discharge. PC 1 showed strong negative loadings In order to prioritize the factors affecting on WT, COD, chloride and BOD and moderate positive loading on DO. PC 2 corresponding to water quality of the sampled streams, PCA was the second eigenvalue (4.47) accounted for 24.83 performed on data for monsoon, pre-monsoon percent of the total variance was loaded by TDS, and post-monsoon seasons. PCA of 18 physi- EC and pH. The third factor corresponding to co-chemical variables pertaining to water sam- the second eigenvalue (1.47) accounted for ap- ples from 9 different sampling points (during proximately 8.17 percent of the total variance April, 2014 to March, 2016) helped in understand- showed moderate positive loading on sodium ing the main decisive factors of stream water and turbidity. quality across seasons. PCs chosen based on Thus, PCA revealed, the parameters like EC, Eigen values (of 1.0 or greater are considered TDS, ortho-phosphate, nitrate, total hardness, significant), which explained maximum variance calcium, magnesium, potassium and water dis- of experimental dataset (Fig. 3) and are grouped charge played an important role in these sam- as “strong” (>0.75), “moderate” (0.75 - 0.50) and pling sites. These variations are due to varied “weak” (0.50 - 0.30), according to the loading land uses in the catchments of lotic ecosystems values (Liu et al. 2003). – Yaana (YK), and Nanalli (YNK) has evergreen Here, the first PC corresponds to the higher forest whereas Beilangi (BE) has mixed type of eigenvalue (10.76), accounting for 59.79 percent forest. The other sampling sites like Harita (HA), of the variance with strong positive loading Mastihalla (MH), Aanegundi (AG), Aanegundi (>0.75) on pH, calcium, magnesium, ortho-phos- tributary 1 (AGT1), Aanegundi tributary 2 phate, nitrate, potassium, DO and discharge, (AGT2) and Bialgadde (BGT) have mixed land whereas high negative loadings on WT, EC and use dominated by horticulture and agriculture total alkalinity during monsoon (Fig. 3a). PC 2 activities. Landscape alterations in catchment corresponding to the second eigenvalue (4.12) due to the anthropogenic factors changes the accounted for 22.87 percent, contributed by quantity (stream flow) and quality of streams BOD. PC 2 showed moderate positive loading (Petersen et al. 2017). Water discharge played (>0.50) on Water temperature, COD, chloride, an important role in the present study. The BOD and sodium. The third factor correspond- marked difference in seasonal variations in river ing to the third eigenvalue (1.68) accounted for discharge and concentration of pollutants is due PHYSICO-CHEMICAL INTEGRITY OF LOTIC ECOSYSTEMS 77

to variations in precipitation, surface run-off, interflow, groundwater flow etc. (Singh et al. 2004).

Cluster Analysis

Cluster analysis of physico-chemical parameters of different streams in Aghanashini river basin showed varied patterns during monsoon, post-monsoon and pre-monsoon seasons (Fig. 4). Four clusters were evident based on water quality parameters of monsoon (Fig. 4a). Harita has high water temperature, turbidity, COD, orthophosphate, nitrate, sodium, potassium, BOD and water discharge. AGT2, YK and Mastihalla had high ionic contents as evident in levels of TDS, EC, total alkalinity, total hardness, calcium and magnesium. BGT, AGT1 and YNK had high total hardness, DO and alkaline pH. Beilangi and AG had high chloride but comparatively low pH,

Fig. 3. Principal component analysis for physico- chemical parameters of different streams in Aghanashini river basin of Central Western Ghats during a) monsoon, (b) post-monsoon and (c) pre-monsoon seasons Source: Author 78 T. V. RAMACHANDRA, V. SINCY, K. S. ASULABHA ET AL.

bidity, ortho-phosphate, sodium and potassium. YK had high TDS, EC total alkalinity, pH, DO, total hardness, magnesium and calcium. Yaana is famous for two rocky outgrowth of solid black, crystalline karst limestone. This is evident from the sampled water at Yaana with high ion contents and pH due to the influence of calcium carbonate. All stations showed wide seasonal variations in all the physico-chemical parameters mainly due to variations in discharge. Cli- mate, watershed characteristics and forest cover also play a decisive role in quality and quantity of water in a forested watershed (Gokbulak et al. 2017).

CONCLUSION

Fig. 4. Cluster analysis of physico-chemical Monitoring of lotic ecosystem through the parameters of different streams in Aghanash- analyses of physico-chemical parameters in the ini river basin of Central Western Ghats dur- streams of the Aghanashini river basin in the ing a) monsoon season, (b) post-monsoon and (c) pre-monsoon season. Central Western Ghats revealed that water dis- Source: Author charge played an important role evident from TDS, EC, total hardness, calcium, magnesium, ABBREVIATIONS ortho-phosphate, nitrate, potassium, etc. High- WT: Water Temperature (°C); TDS: Total Dissolved Solid (mg/l); EC: Electrical er ion concentrations at Yaana are due to an- Conductivity (µS/cm); TB: Turbidity (NTU); DO: thropogenic influences due to influx of tourists. Dissolved Oxygen (mg/l); BOD: Biochemical Water discharge plays an important role in lotic Oxygen Demand (mg/l); COD: Chemical Oxygen ecosystems as per PCA and Cluster analysis. Demand (mg/l); TA: Total Alkalinity (mg/l); Chl: Water discharge is highest in Harita compared Chloride (mg/l); TH: Total Hardness (mg/l); Ca: Calcium (mg/l); Mg: Magnesium (mg/l); OP: to other sampling stations due to catchment run Orthophosphate (mg/l); N: Nitrate-nitrogen (mg/ off that also brought in suspended matter (tur- l); Na: Sodium (mg/l); K: Potassium (mg/l); WD: bidity), nutrients (orthophosphate and nitrate) Water Discharge (mm/day). and organic contents (COD and BOD). Climate, season, types of forest, rocks, soil, land use like turbidity, calcium, total hardness, nitrate, DO and agriculture and horticulture fields have influ- water discharge. Among all the stations, Harita enced the quality of water in streams. This ne- had highest discharge in monsoon season which cessitates an understanding of hydro-morpho- had influenced the water quality. logical with chemical characteristics for sustain- Figure 4b illustrates of the three clusters able management of lotic ecosystems. during post-monsoon. YK reveals of disturbance with high TDS, pH, EC, COD, total alkalinity, RECOMMENDATIONS total hardness, magnesium, ortho-phosphate, The following measures have to be adopted nitrate, DO and water discharge but low water to protect the lotic ecosystems in central West- temperature, chloride, sodium and potassium. ern Ghats: (i) avoid littering of wastes near wa- BE, AGT1 and Mastihalla had high water tem- ter bodies, (ii) restrictions on letting untreated perature but low water discharge, ionic, organic sewage and industrial effluents, (iii) adoption of and nutrient levels. YNK, AGT2 and BGT had constructed wetlands to prevent agricultural high turbidity, chloride, sodium, potassium and runoff with high nutrient loads, (iv) conserva- BOD. During pre-monsoon, about four clusters tion of sacred groves and native forest ecosys- are evident in Figure 4c. BE had high nitrate and tems to enhance the water retention capability BOD but low ionic contents, ortho-phosphate of the catchment and to avert scarcity of water and discharge. BGT had high water temperature, quantity and quality, (v) planting native spe- COD and chloride. YNK and AGT2 had high tur- cies in the catchment to enhance water infiltra- PHYSICO-CHEMICAL INTEGRITY OF LOTIC ECOSYSTEMS 79 tions, (vi) restrictions on large scale monocul- Cui X, Liu S, Wei X 2012. Impacts of forest changes on ture plantations in the catchment, (vi) protec- hydrology: A case study of large watersheds in the tion of the shoreline and riparian vegetation zone, upper reaches of Minjiang river watershed in China. Hydrol Earth Syst Sci, 16: 4279-4290. (vii) evolve sustainable management strategies Dunca AM 2018. Water pollution and water quality to mitigate pollution due to tourism, (viii) restric- assessment of major transboundary rivers from Ba- tion on the use of fertilizers in agriculture and nat (Romania). J Chem, Article ID #9073763, 8 horticulture fields, (ix) restriction on the diver- pages. sion of streams that alters natural flow impact- Gandaseca S, Rosli N, Ngayop J, Arianto CI 2011. Sta- tus of water quality based on the physico-chemical ing downstream biota and people’s livelihood, assessment on river water at wildlife sanctuary Sibu- etc. ti Mangrove Forest, Miri Sarawak. Am J Environ Sci, 7(3): 269-275. ACKNOWLEDGEMENTS Gebreyohannes F, Gebrekidan A, Hadera A, Estifanos S 2015. Investigations of physico-chemical parameters and its pollution implications of Elala river, The financial and infrastructure support for Mekelle, Tigray, Ethiopia. MEJSi, 7: 240-257. ecology research from (i) NRDMS Division, The Girardi R, Pinheiro A, Garbossa LHP, Torres E 2016. Ministry of Science and Technology (DST), Gov- Water quality change of rivers during rainy events ernment of India, (ii) The Ministry of Environ- in a watershed with different land uses in Southern ment, Forests and Climate Change (MoEFCC), Brazil. Braz J Water Resour, 21: 514-524. Gokbulak F, Serengil Y, Yurtseven I, Erdogan BU, Ozce- Government of India and (iii) Indian Institute of lik MS, Ozhan S, Sengonul K 2017. Relationship Science is acknowledged. between water yield and water quality in broadleaf forested watersheds. EWRA, 58: 281-286. REFERENCES Hacioglu N, Dulger B 2009. Monthly variation of some physico-chemical and microbiological parameters Aishvarya N, Malviya MK, Tambe A, Sati P, Dhakar in Biga stream (Biga, Canakkale, Turkey). Afr J K, Anita P 2018. Bacteriological assessment of riv- Biotechnol, 8: 1929-1937. er Jataganga, located in Indian Himalaya, with ref- Han YM, Cao JJ, Wu F, Zhang BC, Zhan CL, Wei C, erence to physico-chemical and seasonal variations Zhao ZZ 2012. Geochemistry and environmental under anthropogenic pressure: A case study. J Envi- assessment of major and trace elements in the sur- ron Microbiol, 1(1): 10-16. face sediments of the Wei river, China. J Environ Al-Badaii F, Shuhaimi-Othman M, Gasim MB 2013. Monit, 14: 2762-2771. Water quality assessment of the Semenyih river, Huang J, Huang Y, Zhang Z 2014. Coupled effects of Selangor, Malaysia. J Chem, 2013 (ID 871056): 1- natural and anthropogenic controls on seasonal and 10. http://dx.doi.org/10.1155/2013/871056 spatial variations of river water quality during base Allan JD 2004. Landscapes and riverscapes: The influ- flow in a coastal watershed of Southeast China. PLoS ence of land use on stream ecosystems. Annu Rev ONE, 9(3): e91528. Ecol Evol Syst, 35: 257-284. Kriauciuniene J, Meilutyte-Barauskiene D, Reihan A, APHA, AWWA, WEF 2005. Standard Methods for the Koltsova T, Lizuma L, Sarauskiene D 2012. Vari- Examination of Water and Wastewater. 21st Edition. ability in temperature, precipitation and river dis- Washington, DC: American Public Health Associa- charge in the Baltic States. Boreal Environ Res, 17: tion. 150-162. Arslan O 2009. A GIS-based spatial-multivariate statis- Lintern A, Webb JA, Ryu D, Liu S, Bende Michl U, tical analysis of water quality data in the Porsuk Waters D, Leahy P, Wilson P, Western AW 2018. River, Turkey. Water Qual Res J Can, 44: 279-293. Key factors influencing differences in stream water Balachandran C, Dinakaran S, Chandran MDS, Ram- quality across space. Wiley Interdiscip Rev: Water, achandra TV 2012. Diversity and Distribution of 5(1): 1-31. Liu WX, Li XD, Shen ZG, Wang DC, Wai OWH, Li SY Aquatic Insects in Aghanashini River of Central 2003. Multivariate statistical study of heavy metal Western Ghats, India. Paper Presented in LAKE enrichment in sediments of the Pearl River Estu- 2012: National Conference on Conservation and ary. Environ Pollut, 121: 377-388. Management of Wetland Ecosystems, 6-8 Novem- Magdaleno F, Martinez R 2014. Evaluating the quality ber, MG University, Kottayam, Kerala, pp. 1-10. of riparian forest vegetation: The Riparian Forest Bhat SA, Meraj G, Yaseen S, Pandit AK 2014. Statisti- Evaluation (RFV) index. Forest Syst, 23: 259-272. cal assessment of water quality parameters for pol- Mushatq B, Raina R, Yaseen T, Wanganeo A, Yousuf lution source identification in Sukhnag stream: An AR 2013. Variations in the physico-chemical prop- inflow stream of Lake Wular (Ramsar site), Kash- erties of Dal lake, Srinagar, Kashmir. Afr J Environ mir Himalaya. J Ecosyst, Article ID #898054, 18 Sci Technol, 7: 624-633. pages. Nassar MZ, Mohamed HR, Khiray HM, Rashedy SH Bunn SE, Arthington AH 2002. Basic principles and 2014. Seasonal fluctuations of phytoplankton com- ecological consequences of altered flow regimes for munity and physico-chemical parameters of the north aquatic biodiversity. J Environ Manage, 30: 492- western part of the Red Sea, Egypt. Egypt J Aquat 507. Res, 40: 395-403. 80 T. V. RAMACHANDRA, V. SINCY, K. S. ASULABHA ET AL.

Neary DG, Ice GG, Jackson CR 2009. Linkages between Salman JM, Jawad HJ, Nassar AJ, Hassan FM 2013. A forest soils and water quality and quantity, Forest study of phytoplankton communities and related Ecol Manag, 258: 2269-2281. environmental factors in Euphrates river (between Oketola AA, Adekolurejo SM, Osibanjo O 2013. Water two cities: Al-Musayyab and Hindiya), Iraq. J Envi- quality assessment of river Ogun using multivariate ron Prot, 4: 1071-1079. statistical techniques. J Environ Prot, 4: 466-479. Sharma D, Kansal A 2013. Assessment of river quality Petersen CR, Jovanovic NZ, LeMaitre DC, Grenfell MC models: A review. Rev Environ Sci Bio, 12: 285- 2017. Effects of land use change on stream flow and 311. stream water quality of a coastal catchment. Water Shetty S, Tharavathy NC, Lobo RO, Shafakatullah N SA, 43(1): 139-152. Purnaini R, Sudarmadji, Purwono S 2018. Tidal influ- 2013. Seasonal variation in the physico-chemical ence on water quality of Kapuas Kecil river down- characteristics along the upstream of Tungabhadra stream. In E3S Web of Conferences, 31: 1-5. River, Western Ghats, India. IJPAES, 3: 242-246. Ramachandra TV 2014. Hydrological responses at re- Singh KP, Malik A, Mohan D, Sinha S 2004. Multivari- gional scale to landscape dynamics. J Biodivers, 5: ate statistical techniques for the evaluation of spa- 11-32. tial and temporal variations in water quality of Go- Ramachandra TV, Chandran MDS, Joshi NV, Karthick mti River (India): A case study. Water Res, 38: 3980- B, Mukri VD 2015. Ecohydrology of lotic systems 3992. in Uttara Kannada, Central Western Ghats, India. Singh S, Mishra A 2012. Spatiotemporal analysis of the In: Ramkumar et al. (Eds.): Environmental Man- effects of forest covers on water yield in the West- agement of River Basin Ecosystems: Springer Earth ern Ghats of peninsular India. J Hydrol, 446-447: System Sciences, Switzerland: Springer Internation- 24-34. al Publishing, pp. 621-665. DOI 10.1007/978-3- Singh TA, Meetei NS, Meitei LB 2013. Seasonal vari- 319-13425-3_29 Ramachandra TV, Chandran MDS, Vinay S, Sudarshan ation of some physico-chemical characteristics of B, Vishnu DM, Rao GR, Shrikanth N, Bharath HA three major rivers in Imphal, Manipur: A compara- 2016a. Sacred Groves (Kan forests) of Sagara Ta- tive evaluation. Curr World Environ, 8: 93-102. luk, Shimoga District. ENVIS Technical Report 102, Sreekantha, Chandran MDS, Mesta DK, Rao GR, Guru- Bangalore, India: CES, IISc. raja KV, Ramachandra TV 2007. Fish diversity in Ramachandra TV, Vinay S, Aithal BH 2016b. Environ- relation to landscape and vegetation in central West- mental flow assessment in a lotic ecosystem of Cen- ern Ghats, India. Curr Sci, 92: 1592-1603. tral Western Ghats, India. Hydrol Curr Res, 7: 1-14. Tornwall B, Sokol E, Skelton J, Brown BL 2015. Trends Ramirez A, Rosas KG, Lugo AE, Ramos-Gonzalez OM in stream biodiversity research since the river con- 2014. Spatio-temporal variation in stream water tinuum concept. Diversity, 7: 16-35. chemistry in a tropical urban watershed. Ecol Soc, Vankar J, Tatu K, Kamboj RD 2018. Water quality of 19: 45. two important inland wetlands (Pariej and Kanew- Rani N, Sinha RK, Prasad K, Kedia DK 2011. Assess- al) in the neighbouring districts of Central Gujarat, ment of temporal variation in water quality of some India. Int Res J Environment Sci, 7(2): 24-32. important rivers in middle Gangetic plains, India. Watkar AM, Barbate MP 2015. Seasonal variations in Environ Monit Assess, 174: 401-415. physico-chemical properties of Chandrabhaga river Ray R, Chandran MDS, Ramachandra TV 2015. Hy- in Dhapewada, Dist. Kalmeshwar Maharashtra, In- drological importance of sacred forest fragments in dia. Res J Recent Sci, 4: 1-4. Central Western Ghats of India. Trop Ecol, 56: 87- Zarei H, Bilondi MP 2013. Factor analysis of chemical 99. composition in the Karoon river basin, southwest Rolls RJ, Leigh C, Sheldon F 2012. Mechanistic effects of Iran. Appl Water Sci, 3: 753-761. of low-flow hydrology on riverine ecosystems: Eco- logical principles and consequences of alteration. Paper received for publication on March 2018 Freshwater Sci, 31: 1163-1186. Paper accepted for publication on September 2018