Environmental Flow Assessment in a Lotic Ecosystem of Central Western Ghats, India

Environmental/Ecological flow refers to the minimum flow of water to be maintained in a water body (river, lake, etc.) to sustain ecosystem services. Understanding environmental flow is important to ensure the local ecological and social (people, agriculture and horticulture, etc.) needs in a sustained and balanced way, while designing large scale projects (such as hydro-electric, river diversion, etc.). Western Ghats are the mountain ranges extending from southern tip of India (Tamil Nadu-Kanyakumari) to Gujarat. These mountain ranges are rich in biodiversity with diverse and endemic flora and fauna, and is birth place to numerous perennial rivers namely Netravathi, Sita, Sharavathi, Aghanashini, Krishna, Cauvery, etc. Western Ghats is often referred as water tower of peninsular India, due to the water and food security provided by the ecosystem through array of services. The region is also one among 35 global biodiversity hotspots. However, deforestation due to large scale land cover changes has affected the water sustenance in the region evident from the quantity and duration of water availability during post monsoon period. Forests in the Western Ghats along with the soil characteristics and precipitation plays a major role in storing water in sub-surface (vadoze and groundwater) zones during monsoon, and releases to the streams during post monsoon periods catering to the needs of the dependent biota including humans. Some of these undisturbed/ unaltered natural flow conditions in rivers and streams have proved their worth with the presence of rich and diverse species and array of ecosystem services, which also has helped in sustaining the livelihood of dependent populations. The undisturbed flow conditions guarantees the natural flow as well as minimum flow in streams to sustain the ecosystem services, which helps in meeting the social and ecological needs. Growing demand to cater the demands of burgeoning human population coupled with accelerated pace of deforestation due to unplanned and senseless developmental projects in the ecologically fragile regions have led the water scarcity even in regions receiving high amount of rainfall. In the current communication an attempt is made to understand the linkages between the hydrological dynamics across varied landscape with the anthropogenic and ecological water needs. If the available water resource meets the societal and environmental demands across seasons, the catchment is said to achieve the minimum flow requirements. The federal government has plans to divert the water from rivers in Western Ghats region to the dry arid regions in Karnataka. In this regard, environmental flow assessment of Yettinaholé river in Central Western Ghats is carried out to understand the feasibility of river diversion through the assessment of hydrologic regime with the analysis of land use dynamics (using remote sensing data), meteorological data (rainfall, temperature, etc. from IMD, Pune), hydrological data (from gauged streams) apart from field investigations in the catchment. The catchments receive annual rainfall of 3000-5000 mm (Department of Statistics, Government of Karnataka). Land use analyses reveal that Major portion of the catchment is covered with evergreen forest (45.08%) followed by agriculture plantations (29.05%) and grass lands (24.06%). Water yield in the catchment computed for each of sub-catchments based on the current land use and other related hydrological parameters using empirical method. The total runoff yield from the catchments is estimated to be 9.55 TMC. About 5.84 TMC is required for domestic purposes including agriculture, horticulture and livestock rearing. The quantum of water required to sustain fish life in the streams is about 2 TMC, computed based on hydrological discharge monitoring and fish diversity in streams during 18 months (covering all seasons) in select streams in Western Ghats. Considering the available water is sufficient only to meet the anthropogenic and ecological needs in the region, the sustainable option to meet the water requirements in dry arid regions would be through (i) decentralized water harvesting (through tanks, ponds, lakes, etc.), (ii) rejuvenation or restoration of existing lakes/ponds, (iii) reuse of waste water, (iv) recharging groundwater resources, (v) planting native species of grasses and tree species in the catchment (to enhance percolation of water in the catchment), (vi) implementation of soil and water conservation through micro-watershed approaches. Implementation of these location specific approaches in arid regions would cost much less compared to the river diversion projects, which if implemented would help the section of the society involved in decision making, construction and implementation of the project.


Introduction
The Western Ghats is a series of hills located in the western part of peninsular India stretching over a distance of 1,600 km from north to south and covering an area of about 1,60,000 sq.km and one among the 35 global hotspots of biodiversity [1][2][3]. It harbors very rich and rare flora and fauna and there are records of over 4,500 species of flowering plants with 38% endemics, 330 butterflies with 11% endemics, 156 reptiles with 62% endemics, 508 birds with 4% endemics, 120 mammals with 12% endemics, 289 fishes with 41% endemics and 135 amphibians with 75% endemics [4][5][6][7].
Western Ghats has numerous watersheds that feed perennial rivers of peninsular India [2]. It encompasses series of west and east flowing rivers that originates from the Western Ghats, supporting as source of sustenance for existing life forms in the environment. One such source of perennial waters is Yettinaholé originating at an altitude of 950 m in Sakaleshpura taluk of Hassan district, and tributary of river Gundia, which joins Kumaradhara and finally drains to Netravathi River The region with a repository of endemic and rare biodiversity is ecologically sensitive and large scale degradation of catchment landscape have influenced the availability of water and has also affected the sustenance of biodiversity. Changes in landscape structure and the regional climate [8,9] have altered the hydrologic regime [10,11] in many lotic ecosystems in the tropical regions, affecting the potential of the catchment to retain water in the surface and sub surfaces. Various studies carried out in Western Ghats [12][13][14] and across the globe show the relevance of landscape on surface and subsurface hydrological regime [12][13][14][15][16][17][18][19]. Few studies carried out in Western Ghats also emphasize on the role of hydrological regime on the habitats, ecology, biodiversity, quality of water, soil and ecosystem etc. [20][21][22][23]. In the current communication an attempt has been made to understand the linkages between the hydrological dynamics across varied landscape of Yettinaholé catchment with the societal and environmental water needs. In this regard, the study investigates land use dynamics, hydrological yield, fish diversity in select streams and linkages with the flow during lean season and drivers of hydrological regime impairment.

Study area
Yettinaholé catchment has a pristine ecosystem with rich biodiversity ( Figure 1 and Table 1), extend from 12044'N to 12058'N Latitude and 75037'E to 75047'E longitude encompassing total area of 179.68 km 2 . The terrain ( Figure 2) is undulating with altitude varying from 666 m above MSL to 1292 m above MSL leading to higher density of stream network ( Figure 3). Geologically, rock types consist of Gneiss, the soils are loamy ranging from sandy loamy to clay loamy. Soils (Figure 4) in the region are fertile and highly permeable, hence allowing the precipitated water to percolate easily into the subsurface recharging ground water and storing water in the sub surfaces and hence keeping the water source perennial to the catchment and the downstream users during and post monsoon.
Decadal population in Sakleshpura Taluk (spatial extent 1034 sq. km) of Hassan district is given in Figure 5 and Table 2 shows a declining trend due to migration to cities during post 2001. Population dynamics of the catchments also follows the dynamics of Sakleshpura taluk. Total Population of all the catchments with respect to census data [24,25] was estimated as 17005 in 2001, has declined to 16345 in 2011 at a decadal rate of 3.88%. Population for the year 2014 was calculated as 16156 based on the temporal data. Population density for each of the sub catchments are as depicted in Figure 6 and Table 3.

Sub basin id
Stream Name Area (Ha)             The region receives an annual rainfall of 3500 to 5000 mm across the catchment. Precipitation in the catchment during June to September is due to the southwest monsoons, with July having maximum rainfall over 1300 mm. Monthly variation in rainfall is depicted in Figure 7. Spatial variation of rainfall across the catchments was assessed based on 110 years data [26] (1901 to 2010) from the rain gauge stations in and around the catchment (Figure 8). Figure 9 depicts monthly temperatures [27] variations, which ranges from 14.7°C (January) to 31.6°C (in March).

Data
Data required for hydrological and land use analyses were (i) social and demographic data from the government agencies, (ii) temporal remote sensing data from public archive and (iii) primary data through field investigations. Latest remote sensing data used is of Landsat 8 series (2014). Rainfall data was acquired from the Directorate of Economics and Statistics, Government of Karnataka [26], Temperature data was sourced from World Clim-Global Climate Data [27] of 1 km resolution. Census data collected from government of India, state and district census departments [24,25]. These data was supplemented with secondary data compiled from various sources as tabulated in Table 4. Primary data is compiled through field investigations and through structured questionnaire (household survey).

Method
The method for the evaluation of the environmental flow and hydrological status is given in Figure 10. Hydrologic assessment in the catchment involved 1) delineation of catchment boundary 2) land use analysis, 3) assessment of the hydro meteorological data, 4) analysis of population census data, 5) compilation of data through public interactions for assessing the water needs for livestock, agriculture/ horticulture and cropping pattern, and 6) evaluation of hydrologic regime.
Delineation of catchment boundary: Catchment boundaries ( Figure  1) and the stream networks ( Figure 3) were delineated considering the topography of the terrain based on CartoSat DEM using the QSWAT module-Quantum GIS 2.10 32 bit. These catchment boundaries were overlaid on the extracted boundaries from the Survey of India topographic maps for validations. Corrected catchment boundaries were further overlaid on Google earth in order to visualize the terrain variations ( Figure 2).

Data Description Source
Remote sensing data-spatial data Remote sensing data of 30 m spatial resolution and 16 bit radiometric resolution were used to analyse land uses at catchment levels. [28] Rainfall Daily rainfall data of 110 years , to assess the trends in rainfall distribution and variability across basins. [26,29] Crop Calendar To estimate the crop water requirements based on the growth phases [30][31][32][33][34][35] Crop Coefficient Evaporative coefficients used to estimate the actual evapotranspiration. [33,36] Temperature (max, min, mean), Extraterrestrial solar radiation Monthly temperature data (1 km spatial resolution) and monthly extra-terrestrial solar radiation (Every 1° North latitude) available [27,[36][37][38]    Land use assessment: Large scale land-use land-cover (LULC) changes leading to deforestation is one of the drivers of global climate changes and alteration of biogeochemical cycles. This has given momentum to investigate the causes and consequences of LULC by mapping and modelling landscape patterns and dynamics and evaluating these in the context of human-environment interactions in the riverine landscapes. Human induced environmental changes and consequences are not uniformly distributed over the earth. However their impacts threaten the sustenance of human-environmental relationships. Land cover refers to physical cover and biophysical state of the earth's surface and immediate subsurface and is confined to describe vegetation and manmade features. Thus, land cover reflects the visible evidence of land cover of vegetation and non-vegetation. Land use refers to use of the land surface through modifications by humans and natural phenomena. Heterogeneous terrain in the landscape with the interacting ecosystems is characterized by its dynamics. Human induced land use and land cover (LULC) changes have been the major driver of the landscape dynamics at local levels. Land use assessment was carried using the maximum likelihood classification technique [48,49]. Understanding of landscape dynamics helps in the sustainable management of natural resources.
Land use analysis involved i) generation of FCC-False Colour Composite ( Figure 11) of remote sensing data (bands-green, red and NIR). This helped in locating heterogeneous patches in the landscape ii) selection of training polygons (these correspond to heterogeneous patches in FCC) covering 15% of the study area and uniformly distributed over the entire study area, iii) loading these training polygons co-ordinates into pre-calibrated GPS, vi) collection of the corresponding attribute data (land use types) for these polygons from the field. GPS helped in locating respective training polygons in the field, iv) supplementing this information with Google Earth v) 65% of the training data has been used for classification, while the balance is used for validation or accuracy assessment.
Land uses were categorized into 8 classes namely. i) water bodies (lakes/tanks, rivers, streams, ii) built up (buildings, roads or any paved surface, iii) open spaces iv) evergreen forest (evergreen and semi evergreen), v) deciduous forest (Moist deciduous and dry deciduous) vi) scrub land and grass lands, vii) agriculture, (viii) private plantations (coconut, arecanut, rubber) and forest plantations (Acacia, Teak, etc.) Assessment of the hydro meteorological data: This involved assessment of the spatial and temporal variations in rainfall [26,29,50] in and around the study region. Long term precipitation data helped in understanding the rainfall variability over decades. Along with rainfall,  Rainfall: Point data of daily rainfall from rain gauge stations for the period 1901-2010 [26,29,50] were used for the analysis. Some rain gauge stations had incomplete records with missing data for few months. The average monthly and annual rainfall data were used to derive rainfall map throughout the study area and was used to derive the gross yield (RG) in the basin (equation 1). Net yield (RN) was quantified (equation 2) as the difference between gross rainfall and interception (In).
RG=A × P ……. (1) RN=RG-In …… (2) Where, RG: Gross rainfall yield volume; A: Area in Hectares; P: Precipitation in mm, RN: Net rainfall yield volume; and In: Interception volume InterceptionDuring monsoons, portion of rainfall does not reach the surface of the earth; it remains on the canopy of trees, roof tops, etc. and gets evaporated. Field studies in Western Ghats show that, losses due to interception is about 15% to 30%, based on the canopy cover. Table 5 shows the interception loss across various rainy months and land uses.

Vegetation types Period Interception
Evergreen  Runoff: Portion of rainfall that flows in the streams after precipitation [2,8,10,11] are (i) surface runoff or direct runoff and (ii) sub surface runoff.
Surface runoff: Portion of water that directly enters into the streams during rainfall, which is estimated based on the empirical [9,10,11] relationships given in equation 3.

Q=Σ(Ci × PR × Ai)/1000……(3)
Where, Q: Runoff in cubic meters per month; C: Catchment/Runoff coefficient, depends on land uses as given in  Infiltration: The portion of water enters the subsurface (vadoze and groundwater zones) during precipitation depending on land cover in the catchment. During field monitoring of streams in the forested catchment, overland flow is noticed in streams only after couple of days rainfall. This means that overland flow in the catchment with vegetation cover happens after the saturation of sub surfaces. The water stored in sub-surfaces will flow laterally towards streams and contributes to stream flow during non-monsoon periods, which are referred as pipe flow (during post monsoon) and base flow (during summer). Inf=RN-Q…..…(4) Ground water recharge: This is the portion of water that is percolated below the soil stratum (vadoze) after soil gets saturated.
Recharge is considered the fraction of infiltrated water that recharges the aquifer after satisfying available water capacity and pipe flow. Where, GWR: Ground water recharge; RC: Ground water recharge coefficient (Table 7); C: Rainfall Coefficient (Table 7); A: Area of the catchment. The recharge coefficient and the constant vary depending land uses with the annual rainfall.   Groundwater discharge: Groundwater discharge or base flow is estimated by multiplying the average specific yield of aquifer under each land use with the recharged water. Specific yield represents the water yielded from water bearing material. In other words, it is the ratio of the volume of water that the material, after being saturated, will yield by gravity to its own volume. Base flow appears after monsoon and receding of pipe-flow. This water generally sustains flow in the rivers during dry seasons. A portion of recharged water flows to the streams as ground water discharge which is dependent on the topography, geology and the land use conditions. Equation 7 defines Ground water discharge as product of specific yield and the portion of ground water recharged. GWD=GWR × YS…..…(7) Where, GWD: Ground water discharge; GWR: Ground water recharge; YS: Specific yield [2].

Estimation of water demand evapotranspiration
Evaporation is a process where in water is transferred as vapour to the atmosphere. Transpiration is the process by which water is released to the atmosphere from plants through leaves and other parts above ground. Evapotranspiration is the total water lost from different land use due to evaporation from soil, water and transpiration by plants. Some of the important factors that affect the rate of evapotranspiration are: (i) temperature, (ii) wind, (iii) light intensity, (iv) Sun light hours, (v) humidity, (vi) plant characteristics, (vii) land use type and (viii) soil moisture. If sufficient moisture is available to completely meet the needs of vegetation in the catchment, the resulting evapotranspiration is termed as potential evapotranspiration (PET). The real evapotranspiration occurring in specific situation is called as actual evapotranspiration (AET). These evapotranspiration rates from forests are more difficult to describe and estimate than for other vegetation types.
Potential evapotranspiration (PET) was determined using Hargreaves method (Hargreaves [2,36]) an empirical based radiation based equation, which is shown to perform well in humid climates. PET is estimated as mm using the Hargreaves equation is given by equation 8. Where, RA: Extra-terrestrial radiation (MJ/m 2 /day) [36]; Tmax: Maximum temperature [42]; Tmin: Minimum temperature [42]; λ: latent heat of vapourisation of water (2.501 MJ/kg) Actual evapotranspiration is estimated as a product of Potential evapotranspiration (PET) and Evapotranspiration coefficient (KC) ( Table 8), given in equation 9. The evapotranspiration coefficient is a function of land use varies with respect to different land use.  Note: the crop water requirement was estimated for different crops and different seasons based on land use, assumption is individual crop water requirement and different growth phases (need different quantum of water for their development inclusive of evaporation).
Domestic water demand: Understanding the population dynamics in a region is necessary to quantify and also to predict the domestic water demand. Population census for villages during 2001 and 2011 [24] were considered in order to compute the population of the basin level. Based on the rate of change of population (equation 10), the population for the year 2014 was predicted as given in equation 11. r=(P2011/P2001-1)/n …… (10) Where, P2001 and P2011 are population for the year 2001 and 2011 respectively; n is the number of decades which is equal to 1; r is the rate of change P2014=P2011 × (1+n × r) …… (11) Where, P2014 is the population for the year 2014; n is the number of decades which is equal to 0.3.
Domestic water demand is assessed as the function of water requirement per person per day, population and season. Water required per person includes water required for bathing, washing, drinking and other basic needs. Water requirements across various seasons are as depicted in Table 9.

Season
Water lpcd
Livestock water requirement: Household surveys were conducted with the structured questionnaires to understand the agricultural and horticulture cropping pattern and water needed for various crops in the catchment. Livestock population details were obtained from the district statistics office and water requirement for different animals were quantified based on the household interviews. Table 10 gives the water requirement for various animals.  Crop water requirement: The crop water requirement for various crops was estimated considering their growth phase and details of the cropping pattern in the catchment (based on the data compiled from household surveys and publications such as the district at a glance, department of agriculture). Land use information was used in order to estimate the cropping area under various crops. Figure 12 provides the information of various crop water requirements based on their growth phase as cubic meter per hectare. Evaluating hydrological status: The hydrological status in the catchment is analysed for each month based on the water balance which take into account the water available to that of the demand. The water available in the catchment is function of water in the soil, run off (streams and river) and water available in the water bodies (Lentic water bodies such as lakes, etc.). Water demand in the catchment is estimated as the function of societal demand and terrestrial ecosystem (AET from forested landscape) crop water demand, domestic and livestock demand and the evapotranspiration. The catchment is considered hydrological sufficient, if the water available caters the water demand completely else the deficit catchment, if the water demand is more than the water available in the system.

Water Requirement in lpcd (Liters per animal per day)
Quantification of the environmental flow: Ecological investigations include the investigations of fish diversity across seasons. Habitat simulation method [51][52][53][54][55][56] was adopted to assess flows on basis of quantity and suitability of physical habitat available to target species under different flow regimes. In order to evaluate the natural flow regime [53,54], 18-24 months field monitoring of select streams in Sharavathi river basin and at Hongadahalla and Kadumanehalla (of Yettinaholé catchment) was carried out. This field data was compared with the long term flow measurements data at Hongadahalla and Kadumanehalla [45]. The natural flow that sustains native biota during lean season is accounted as the ecological or environmental flow [57][58][59][60] for the respective lotic system. In the current study, hydrologic assessment and investigations on the occurrence of native fish species (with diversity) helped in ascertaining the minimum flow required to sustain the native fish biota.

Land use analysis
Land use analysis was carried out using remote sensing data of 2014, for Yettinaholé catchment (a tributary of Gundia River) and results are given in Figure 13 and Table 11. Major portion of the catchment is covered with evergreen forest (45.08%) followed by agriculture plantations (29.05%) and grass lands (24.06%). The valleys along the stream are dominated by agriculture lands and horticulture plantations, the hill tops dominated by grass lands, slopes covered with forest cover. The accuracy of the land use classification is 87% with kappa of 0.82. Temporal land use in the Gundia river catchment during 2000, 2006, 2010 and 2014 are depicted in Figure 14 and details are provided in Table 12  The region receives annual rainfall ranging from 3000 mm to 4500 mm. Variability of rainfall was assessed based on 11 rain gauge stations in the catchment and is given in Figure 15.     Table 12: Land use dynamics -Gundia River basin. Figure 16 provides the annual variability of hydrological parameters for understanding the hydrological regime. Gross rainfall, estimated as product of catchment area and rainfall. The gross rainfall varies from 33232 kilo.cum (in Kadumane holé 2 and Yettinaholé 2) and over 2000000 kilo.cum (in Yettinaholé and Hongada halla catchments). Portion of the water doesn't reach the earth surface, but is intercepted by the earth features namely the tree canopy, building tops, pavements etc., which gets evaporated. Runoff in the basin is estimated as a function of catchment characteristics along with rainfall. Yettinholé, catchment is covered predominantly by evergreen forests, has aided in recharging groundwater zone and sub surfaces. Infiltration of significant amount of precipitation to underlying layers, has reduced the overland flow and thus retarded the flash floods. The infiltration of water to sub-surface takes place during monsoon, and overland flow (surface runoff) happens during the monsoon (rainfall>50 mm per month) and quantity depends on the catchment characteristics namely land use/land cover in the catchment, soil porosity, texture, presence of organic matter (leave debris, decayed matter etc.). The portion of water percolates through the sub surfaces and thus recharges ground water resources. Water stored in vadoze zone (sub-surface) and groundwater zone moves laterally to streams with cessation of rain. Forests in the catchment have played a prominent role in maintaining stream flow, water holding capacity of soil, ground water, which also plays a pivotal role in catering the ecological and environmental demand of water. Sub basin wise yields are listed in Table 13; the surface runoff during the monsoon is estimated to be 9.55 TMC.  Evapotranspiration in the catchment depends on the land use, solar radiation, variations in temperature, precipitation, etc. Potential evapotranspiration was estimated using Hargreaves method. PET indicates the maximum possible water that can evaporate, PET varies between 160 mm/month (March) to 85 mm/month (monsoon season). Considering the various land use characteristics in the catchments, actual evapotranspiration was estimated in the catchments show variation of 40 mm/month (monsoon) to 120 mm/month (March).

Sub basin
Crop water demand was calculated in each catchment based on cropping pattern, area under each crop, and water required across the growth phases of the crops, which were compiled from various literatures (local, national and international) and discussion with the public regarding cropping practices and experiences. Table 9 and Figure 12, details season-wise crop water requirements and growth phases. The agricultural water demand of 2.6 TMC in the catchments is for horticultural and paddy cultivation. Livestock water demand given in Table 10  Ecological Flow Assessment: Flow measurements during the study period at two basins namely Hongadahalla and Kadumaneholé of Yettinaholé namely and based on the long term monitoring data [45] is given in Figures 16 and 17 respectively. This illustrates that flow during the lean season is about 10% of the annual flow, which is lower than the minimum flow requirement to sustain the ecosystem services in the water body.
Monthly monitoring of select streams in Sharavathi River basin over 24 months (covering all seasons) revealed the linkages of fish diversity with the duration of water flow in the respective stream. Table 14 lists the fish diversity across monitored streams. The current assessment confirm the requirement of 24 to 30% of annual flow during lean seasons to sustain the native fish diversity of endemic species [46,47]. Based on this, the ecological flow in Yettinaholé catchment (during the lean seasons) is 2.8 TMC.     The water demand and availability are listed in Table 15 and Figure  18 depicts the spatial variability of resources. Total water demand (5.84 TMC of water) across the catchments (accounting anthropogenic and evapo-transpiration of terrestrial ecosystems) was obtained as a function of evaporation, livestock, and domestic and agriculture demands. Availability of water in the catchment was assessed as a function of runoff during all seasons. The assessment showed that most streams in the forested catchment are perennial compared to steams in the catchment predominantly covered with monoculture plantations (6-9 months) or the streams in catchment dominated by open area or barren area (4 months). The available water in Yettinaholé catchment is sufficient to cater the existing water demand (social, ecological and environmental) throughout the year.

Conclusion
Yettinaholé River is currently catering to the anthropogenic and ecological water needs in the catchment. Higher discharge of water during monsoon has helped in the transport of nutrients, silt, etc., which has helped in sustaining the riparian's vegetation and aquatic life apart from meeting the anthropogenic demand (for horticulture, agriculture etc.). Many streams of Yettinaholé are perennial, which has helped in sustaining the rich and diverse aquatic life apart from sustaining horticultural, agricultural activities (3 crops per year) and fishery.
Hydrological yield computation shows the water yield in the catchment is about 9.5 TMC, About 5.84 TMC is required for domestic purposes including agriculture, horticulture and livestock rearing and the quantum of water required to sustain fish life in the streams is about 2 TMC. This highlights that water available in the catchment is sufficient to sustain the current ecological and anthropogenic (agricultural, horticultural) demand. Alterations in the catchment integrity (land cover) or water diversions would result in the variation in the natural flow regime affecting the biodiversity of riparian's and aquatic habitats and more importantly people's livelihood who are dependent on fisheries, etc. in the downstream. In this contyext, The federal government's plan to divert Yettinaholé River water to the dry arid regions in Karnataka is neither technically feasible, economically viable nor ecologically sound apart from depriving the anthropogenic demand in the Yettinaholé River catchment. The sustainable option to meet the water requirements in arid regions is through (i) decentralized water harvesting (through tanks, ponds, lakes, etc.), (ii) rejuvenation or restoration of existing lakes/ponds, (iii) reuse of treated waste water, (iv) recharging groundwater resources, (v) planting native species of plants in the catchment, (vi) implementation of soil and water conservation through micro-watershed approaches.