Evaluation of sustainability of groundwater resources in a semi-arid region of the Maharashtra State of India

ion and recharge. Figure 5. Stages of groundwater development in seven sub-watershed in Dhubdhubhi according to GEC (1997) norms. groundwater development, and groundwater abstraction use does not exceed the recharge. This shows that there is still scope for groundwater development in these subwatersheds. However, the stage of groundwater development is semi-critical in BM 138-1 and over exploited in BM 139-2 sub-watersheds, where groundwater abstraction exceeds the recharge, calling the sustainability of the resource into question. Water management demands a certain degree of action for example: minimising on-farm water loss, adopting water-saving irrigation scheduling, soil-water conservation through mulching, and rainwater harvesting in farm ponds. It is therefore necessary to develop alternative surface water sources such as dams (where feasible), and surface water should be used when and where available to avoid groundwater mining and promote the sustainable use of groundwater resources in these sub-watersheds.


INTRODUCTION
A large part of the population in the Maharashtra State of India suffers severe and chronic water scarcity due to its unfavorable climatic conditions such as low rainfall, frequent dry spells and high evaporation.Of the total water use in the State, irrigation is about 80% while domestic and industrial uses are 12 and 4% respectively, with the rest used for other purposes (World Bank, 2005).Widespread and progressive depletion of groundwater tables in districts of the Maharashtra State has become a major environmental concern over the past 20 years.
In the State, 19 out of 35 districts show a decline in average groundwater levels of greater than 4 m in 1981 to 2000 (MoWR, 2003).In addition, many districts in the State are concerned about the groundwater pollution problems such as salinity intrusion in the Amravati, Akola and Buldhana districts; fluoride contamination in Bhandara, Chandrapur, Nanded, Yavatmal and Satara; nitrate in the Satara, Sangli and Nagpur districts (GoM, 2003).These problems are more critical in the drought prone areas of Maharashtra State which receives an average annual rainfall of 400 to 700 mm, limiting the availability of surface water and forcing the majority of the population to depend heavily on groundwater.
Most of the population in the State depends on agriculture and related activities for their livelihoods.Construction of groundwater wells has increased the pressure on groundwater resources.Excessive groundwater abstraction followed by low rainfall in the watershed is leading to declining groundwater levels.The population of most of the villages and towns depends on water supply bodies for drinking water in the dry season and the situation is worse in drought years.Assessment of the sustainability of groundwater resources; including the stages of development, recharge, extraction, exploitable resources, and quality is therefore extremely essential for the efficient management and development of groundwater resources in the watershed.'Groundwater sustainability' refers to the development and use of the resource in a manner that can be maintained for an indefinite time without causing unacceptable environmental, economic, or social consequences (Alley, 2004).
Sustainability represents an optimal state; however, this is neither fixed nor constant but rather time and space dependent (De Carvalho et al., 2009) and therefore needs to be quantified in order to evaluate the progress in achieving sustainability over time and space.
As many researchers have stated, sustainability of groundwater bodies is a necessary goal for the long-term welfare of both humans and the environment (Henriksen et al., 2008;Sophocleous, 2000Sophocleous, , 2012)).
In this context, many findings have been reported on the development of indicators to quantify the degree of exploitation to which a groundwater body is subjected.In fact, groundwater sustainability indicators, based on monitoring programmes help to maintain the sustainable management of groundwater resources, and analyse the extent of natural processes and human impact on groundwater systems in space and time to facilitate communication and public participation in resource planning and policy making (Vrba and Lipponen, 2007).
The objective of such indicators is to define by means of diverse variables and synthetic expressions, specific aspects of the quantitative and qualitative state of groundwater and to permit comparison with different aquifers, as well as proper planning and management of the available water resources.Such indicators can be calculated from data that is relatively easy to gather, providing information about the state of an aquifer and the possible tendencies and/or impacts taking place in it.Some earlier studies (Lavapuro, 2008;Girman, 2007;Hirata et al., 2007;Pernia and Lamban, 2007) have also used such groundwater indicators to generate concise information on the present state and trends in groundwater systems, analyzing the extent of natural processes and human impact on groundwater systems, and facilitating communication and public participation in resource planning.
This study aims to assess the sustainability of groundwater resources in the Dhubdhubhi watershed using selected indicators proposed by UNESCO.These indicators can be easily derived from readily available groundwater related data.

Study area
The Dhubdhubhi watershed is located in the Southern Maharashtra State of India.The watershed area is 484 km 2 and home to a population of 94,727.The watershed further consists of seven subwatersheds (BM-138-1, BM-138-2, BM-138-3, BM-138-4, BM-139-1, BM-139-2 and BM-139-3), Figure 1.The size of sub-watershed varies between 10 km 2 to 100 km 2 .The watershed is a typical drought prone area in India.It is underlain by hard basalt rock restricting the groundwater recharge to 10 to 20% of annual rainfall (Campbell, 2013).The characteristics of seven sub-watersheds in relation to groundwater are summarised in Table 1.
According to the Groundwater Surveys and Development Agency (GSDA) of the Government of Maharashtra, groundwater is the main source of irrigation, contributing to more than 60% of the State's net irrigated area.More than 80% of the drinking water supply is also groundwater dependent.Widespread and progressive depletion of the groundwater table at various locations in Maharashtra State districts has become a major concern over the past 20 years, with only a partial (temporary) recovery during exceptionally heavy monsoon rainfall years.
According to GSDA, groundwater development is progressing slowly: 55 and 62% for Akkalkot and South Solapur sub-districts respectively.The groundwater development stage here is defined by the ratio of existing gross groundwater draft for all uses to net annual groundwater availability.For domestic water uses, most of the villages and towns depend on tanker (vehicle carrying water tanks) water supply in the dry season with the situation deteriorating in years of drought.

Data collection
Groundwater level data from 35 observation wells for the period of 2002 to 2006 and specific yield (estimated by pumping tests) from 22 locations in the watershed were collected from Shivaji University, Solapur (Sabale, 2008).Records of precipitation, human and cattle population, areas of various crops, and net irrigation were obtained from various government departments.Thematic maps of watershed boundaries, reservoirs, and water bodies were obtained from the Maharashtra Remote Sensing and Application Centre (MRSAC), Nagpur.

Calculation of indicators
Development of groundwater sustainability indicators (GWSIs) has been taken up by UNESCO under the Sixth Phase of the International Hydrological Programme (IHP), Theme 2: Integrated Watershed and Aquifer Dynamics.The indicators proposed for this group, although simple, are both scientifically based and policyrelevant (Vrba and Lipponen, 2007).In this work, we selected seven main GWSIs based on their importance in the field of groundwater sustainability and because they proved to be the most reliable, based on the use of data collection and methodology.The GWSIs selected are described below:

Renewable groundwater resource per capita (I1)
The renewal groundwater resources consist of recharge from precipitation (natural recharge), seepage from surface water Where GR is a renewable groundwater resource in m3 and N is the total population.
The renewable groundwater resource is obtained by a simple groundwater budget equation considering the negligible recharge during the non-monsoon season (December-March) (Shirahatti, 2012;MoWR, 1997).According to the norms of the Groundwater Estimation Committee (GEC) (1997) and MoWR (1997), the recharge during the non-monsoon season is considered only if the rainfall in the non-monsoon season is greater than 10% of the average annual rainfall.
In this study, rainfall during the non-monsoon season is less than 10% of average annual rainfall.Hence non-monsoon recharge is not considered in the analysis.The renewable groundwater resource is obtained as: Where A is the total groundwater abstraction for all uses (irrigation, domestic and livestock), FI is the irrigation return flow, and S is the change in groundwater storage volume and estimated as: Whereh is the change in groundwater level and SY is specific yield.
Groundwater abstraction for irrigation (referred later as AI) is obtained as the difference between the total water needed for irrigation (referred later as WI) and surface water available for irrigation.The WI for a particular crop is obtained by multiplying the irrigated area of that crop and the gross irrigation requirement (GIR).GIR is estimated as: Where NIR is the net irrigation requirement and EF is the irrigation application efficiency.
The CROPWAT 8.0 model developed by FAO was used to estimate the net irrigation requirement of each crop (Smith, 2012).The NIR is the difference between the crop water requirement and effective precipitation.Various inputs were needed to apply the model, such as reference evapotranspiration (ETO) obtained from the CLIMWAT database (Smith, 1993) for the Solapur station.The average monthly precipitation from the South Solapur and Akkalkot stations, crop characteristics (of all crops grown including crop coefficients for four crop development stages), and soil characteristics, were obtained from available literature (Sabale, 2008;Subramaniam, 1989;Boonstra, 1981;Brouwer and Heibloem, 1986).

Renewable freshwater availability per capita (I2)
Renewable internal freshwater resource flows refer to internal renewable resources (internal river flows and groundwater from rainfall) in the country.Watersheds with more than 1,700 m 3 of renewable fresh water per person per year will generally experience only intermittent or localised water shortages.As the renewable water supply falls below 1,000 m 3 per person, more serious "water scarcity" begins to occur.In this category, chronic water shortages can hamper food production and economic development and cause serious environmental degradation (Falkenmark and Widstrand, 1992).Renewable freshwater availability (I2) is obtained as: Where WR is renewable freshwater and Nis the total population.

Percentage of the population served by groundwater for drinking (I3)
This indicator shows the dependency of the population on groundwater as a source of drinking water.As explained under Indicator I1, according to the Total Water Resources Analysis Report (GoM, 2006), almost all of the population in the watershed, abstract groundwater for drinking purposes from dug wells, bore wells, and hand pumps.Indicator I3 is calculated as: Where NG is the population served by groundwater for drinking and Nis the total population.

Groundwater abstraction for irrigation as a percentage of total irrigation water (I4)
This indicator signifies the contribution of groundwater to irrigation.The indicator I4 is obtained as: Where AI is groundwater abstraction for irrigation, and WI is the total water available for irrigation.

Groundwater abstraction as a percentage of the total water available (I5)
This indicator looks at groundwater abstraction as a percentage of the total available water in the watershed, separated into groundwater and surface water.The indicator I5 is calculated as: Where A is the total groundwater abstraction for all uses and W is the total available water.W is obtained as the sum of the exploitable groundwater and surface water resources.

Groundwater abstraction as a percentage of recharge (I6)
Groundwater abstraction means the total withdrawal of water from a given groundwater body by wells, springs, and other methods for the purpose of water supply and other uses against the recharge.Excessive abstraction of groundwater without the understanding of recharge rates can cause problems, in particular, depletion of the resource.The indicator I6 is obtained as: Where A is the total groundwater abstraction for all uses and GR is the renewable groundwater resource in m 3 .

Stage of groundwater development (I7)
The indicator I7 stage of groundwater development is the ratio of the total abstraction of groundwater resource (A) to the exploitable groundwater resource (GE) as: Vrba et al. (2007) classified the stages of groundwater development in three scenarios: scenario 1: I7 <90% (scope for development), scenario 2: I7 =100% (fully developed) and scenario 3: I7 >100% (Overexploited).

RESULTS AND DISCUSSION
The variables needed to calculate seven indicators are estimated at watershed and sub-watersheds levels following the methodology outlined above and discussed in the section below.

Groundwater recharge
The change in groundwater storage volume (S) is obtained for each year as a product of change in groundwater level (h) and specific yield (S Y ).The difference between the pre-and post-monsoon groundwater level is h.The average groundwater level in the watershed is plotted in Figure 2, showing water  ).The average annual renewable groundwater resource in the study area is 16.7% of annual average rainfall, which agrees with the findings of Singhal et al. 1999 andLimaye et al. (1986).The renewable freshwater is estimated as the sum of the annual average renewable groundwater resource (46.44 million m 3 ) and surface water available for irrigation (4.44 million m 3 ), both of which were obtained under IndicatorI 1 .
The average depth of dug wells (13.31m) was used to determinate the exploitable saturated soil thickness (WRI, 2003).The product of the saturated soil thickness and specific yield multiplied by the aquifer area is the amount of exploitable groundwater resource (G E ) and was estimated to be 51.44 million m 3 .Accordingly, the estimated total available water (W) is 55.88 million m 3 .Similarly, the average groundwater recharge estimated in the sub-watershed is shown in Figure 3.

Groundwater for irrigation, domestic and livestock uses
Irrigation water requirements for eleven types of crops (Sorghum, Wheat, Maize, Sunflower, Vegetables, Pulses, Bajara, Groundnuts, Rice, Sugarcane, and Fruit) grown in the watershed were estimated using the CROPWAT model, Table 2.For example, to determine GIR for a crop of Sorghum, NIR was estimated by the CROPWAT model to be 387.20 mm (it generally varies at around 400 mm (Dara and Raghuvanshi, 1999).E F of 0.65 (MKVDC, 2011) was used for the Solapur district in Equation ( 4) above.GIR for Sorghum was estimated to be 595 mm.The irrigated area of Sorghum in the watershed is 601 ha.GIR volume as a product of the irrigated area of Sorghum and GIR is 0.358 million m 3 .Similarly, GIR volume was estimated for all eleven crops grown in the watershed.The summation of GIR volume for all crops over the watershed gives the annual W I as 35.58 million m 3 .Nearly all of the population in the watershed abstracts groundwater for drinking purposes; from dug wells, tube wells, or hand pumps.Surface water is not used for domestic purposes as it is scarce and not of good quality (GoM, 2006).In other words, the population is almost 100% dependent on groundwater for domestic uses.Groundwater abstraction for domestic use is obtained as a product of the population (N = 94,727) to provide an adequate water requirement of 40 L per person per day (GoI, 2001), which equates to1.38 million m 3 /year.The groundwater abstraction for livestock as a product of the livestock population (33,375) has a water requirement of 55 L per cattle per day (Datta, 2012), equating to 0.67 million m 3 /year.Since there is no industrial development, there is no groundwater withdrawal by industries.Groundwater abstractions merely during the monsoon season for domestic and livestock uses are 0.69 and 0.34 million m 3 respectively.Total abstraction of groundwater (for irrigation, domestic, and livestock uses) during the monsoon season is 10.92 million m 3 .The estimated irrigation return flow (assumed to be 30% of groundwater abstracted for irrigation) (GEC, 1997) is 2.97 million m 3 .Groundwater use for the domestic and livestock sector is presented in Table 2. Similarly, sub-watershed level groundwater abstraction vs. recharge is presented in Table 3.

Groundwater sustainability indicators
Utilising the values of all variables obtained above, seven indicators were estimated at the watershed and subwatershed levels and are summarised in Table 4.For the whole watershed level, the renewable groundwater resource per capita (I 1 ) is estimated at 490 m 3 per capita per year.It is highest (981) in sub-watersheds in BM 139-3 and lowest in sub-watersheds BM 138-1 and BM 139-2.The renewable freshwater availability per capita (I 2 ) is estimated to be 537 m 3 which shows water scarcity in the watershed, since it is far less than the so-called water scarcity limit (1000 m 3 /year/capita) (Falkenmark, 1989).This value (537 m 3 ) agrees with the estimates of the Water Research Institute (WRI) (2003).Similarly, renewable freshwater availability per capita in all subwatersheds is less than 1000 m 3 /year/capita which highlights the water scarcity.
In the entire watershed, the percentage of the population served by groundwater for drinking (I 3 ) shows that 100% of the population is dependent on groundwater for drinking purposes, depicting the importance of groundwater in people's lives.Generally, in the Maharashtra State, 90% of the rural population uses groundwater for drinking purposes (eSakal News Article, 2013).This difference is due to the comparison of State level results with local levels in a much smaller area.
The groundwater use for irrigation as a percentage of total irrigation water (I 4 ) ranges from 75 to 95%, showing a higher contribution of groundwater for irrigation purposes.The groundwater abstraction as a percentage of the total water available (I 5 ), is estimated to be the lowest (27%) in sub-watershed BM-139-1 and the highest (142%) in sub-watershed BM-139-2.These results indicate that the overall dependence of the community on groundwater resources is very high in some subwatersheds and overexploitation of groundwater is observed.
The groundwater abstraction as a percentage of recharge (I 6 ) at a watershed level is 71.46%, whereas it is more than 100% for sub-watersheds BM138-1 and BM-139-2.The stage of groundwater development (I 7 ) in the entire watershed is 64.5% which shows that groundwater resource development falls into the first scenario as reported by Vrba et al. (2007).It means that groundwater resources in the watershed as a whole are not fully utilized, and there is still scope for groundwater development.However, in the sub-watershed BM-139-2, the stage of groundwater development is greater than 100%, showing over-exploitation of groundwater resources.When compared with GEC-1997 norms, a watershed is safe if the stage of groundwater development is at less than 70% and no decreasing groundwater level over the pre-or post-monsoon season is observed.If this is not the case, it is otherwise classified as semi-critical (when I 7 is 70 to 90%), critical (when I 7 is 90 to 100%) and overexploited (when I 7 is >100%) with pre-or post-monsoon decreasing groundwater levels.
In Dhubdhubhi, the stage of watershed groundwater development (I 7 ) is 64.5% with no decreasing groundwater level.Hence, the watershed is safe in terms of groundwater resource development.However, at subwatershed level a different situation is observed.In the sub-watershed BM-139-2, I 7 is greater than 100% with a decreasing groundwater level classified as overexploited, Figure 4.This can be attributed to low recharge followed by high groundwater abstraction resulting in a declining groundwater level.Also, for the sub-watershed BM-138-1, I 7 is between 70-90% with a decreasing groundwater level classified as semi-critical, Figure 5, which can result in a state of over-exploitation if adequate management practices are not implemented.

CONCLUSIONS AND RECOMMENDATIONS
Groundwater resource status and its uses were assessed using seven widely applied indicators proposed by UNESCO in a drought prone Dhubdhubhi watershed in the Maharashtra State of India.The integral application of all these indicators in this particular watershed leads us to conclude that, at present, the groundwater resources of the whole watershed is at a stage of sustainable development.Five sub-watersheds have sustainable  groundwater development, and groundwater abstraction use does not exceed the recharge.This shows that there is still scope for groundwater development in these subwatersheds.However, the stage of groundwater development is semi-critical in BM 138-1 and over exploited in BM 139-2 sub-watersheds, where groundwater abstraction exceeds the recharge, calling the sustainability of the resource into question.
Water management demands a certain degree of action for example: minimising on-farm water loss, adopting water-saving irrigation scheduling, soil-water conservation through mulching, and rainwater harvesting in farm ponds.It is therefore necessary to develop alternative surface water sources such as dams (where feasible), and surface water should be used when and where available to avoid groundwater mining and promote the sustainable use of groundwater resources in these sub-watersheds.

Figure 1 .
Figure 1.Location map of the Dhubdhubhi watershed including its sub-watersheds.

Figure 2 .
Figure 2. Average groundwater levels in the Dhubdhubhi watershed during 2002 to 2006.

Figure 3 .
Figure 3. Groundwater recharge in seven sub-watersheds estimated by the hydrograph analysis method (2002 to 2006).

Figure 4 .
Figure 4. Sub-watershed of Dhubdhubhi watershed showing the status of groundwater abstraction and recharge.

Figure 5 .
Figure 5. Stages of groundwater development in seven sub-watershed in Dhubdhubhi according to GEC (1997) norms.

Table 1 .
Characteristics of seven sub-watersheds in Dhubdhubhi

Sub- watershed 1 Area (km 2 ) Elevation (masl) Area under agriculture 2 (km 2 ) Area under irrigation 2 (km 2 ) Groundwater extraction 3 (MCM/yr) Average saturated thickness
4 (m)bodies, and groundwater discharge to surface water (base flow), groundwater flow across the groundwater basin boundary and artificial recharge.At the initial stage of groundwater development, this preliminary estimate is based on surface observations and the use of direct and indirect estimation techniques depending on data availability.The expression of this calculated amount annually per capita in the study area gives an indication of the availability of groundwater.The indicator I1is obtained as:

Table 2 .
Surface and groundwater withdrawal for irrigation, livestock, and domestic purpose in sub-watersheds.

Table 4 .
Summary of results of estimated groundwater indicators in the seven sub-watersheds and at the watershed level.