Groundwater chemistry is a function of recharge and the input chemistry of the rain, which gets transformed as it moves through the soil matrix. Apart from mineral transformations, anthropogenic activities are other external factors, which affect the groundwater chemistry. Stream – aquifer interactions alter the chemistry of groundwater in the regions nearer to the stream. A study is carried out to analyse the hydrogeochemical behavior under the influence of lithologic, precipitation and anthropogenic controls in the upper Cauvery basin. This is followed by the analysis of contributions made by the components of the hydrogeochemical cycle. A geochemical model is developed, which is used to study the spatiotemporal variations in groundwater chemistry of a silicatic rock group in a small experimental watershed. In order to study the effects of precipitation control on the groundwater chemistry the Upper Cauvery river basin (~ 10000 km2) is selected for the analysis, which stretches along three climatic zones – ‘semi-arid’ (500 – 800 mm/year rainfall), ‘sub-humid’ (1000 - 1200 mm/year) and ‘humid’ (1200 – 1500 mm/year) zones. The basin is mainly formed by granitic gneissic group of rocks with some traces of amphibolites and charnockites. Groundwater is observed to occur either in the saprolite or in the deeper hard rock zone based on the geomorphology even at the scale of a small watershed. Parts of this basin are under canal irrigation and are drained by Kabini and Cauvery Rivers. Groundwater – surface water interactions play an important role in these regions. Irrigation with different levels of intensities is practiced through groundwater in the upland areas. Observation wells considered in these three zones are classified into four classes based on the mean annual groundwater fluctuations. Wells in each of these four classes are further classified into ‘shallow’ and ‘deep’ categories based on the depth to groundwater. Analysis of the groundwater chemistry in the basin (widely spread with 120 wells in the three zones) shows a gradient in chemistry along the climatic gradient with sub-humid zone bridging between the semi-arid and humid zones. Ca/Na and Mg/Na ratios decrease from humid zone (unimodal rainfall) to semi-arid (bimodal rainfall) zone since both Na and Mg concentrations in groundwater increase along this gradient. These elements are mainly controlled by weathering reactions. Apart from the weathering of Ca, calcrete formations also play an important role in the semi-arid zone. Ion exchange process cycles between Cl and SO4 and between Ca and Na. Dissolution of CaCO3, silicate weathering and evaporation are the major mineralogical reactions. Variations in Na/Cl and Ca/Cl molar ratios indicate that shallow wells have higher molar ratios with higher variance than the deeper wells. Semi-arid zone is more silicaceous (higher Na/Cl value) than the humid zone, which has higher Ca/Cl ratio (~ 14). Effective seasonal patterns are identified using ‘recharge – discharge’ concept based on the rainfall intensity. Wells under normal scenario have low Na/Cl and Ca/Cl ratios in the ‘recharge period’ than in the corresponding ‘discharge period’ (dilution chemistry). Wells in the relatively higher pumping regions, which receive sufficient annual recharge exhibit dilution chemistry though groundwater level fluctuations are higher. However, wells in regions with insufficient recharge show ‘anti - dilution’ chemistry. Thus, the ‘recharge – discharge’ concept is useful in identifying the pumped wells from deeper wells and helps in characterizing the anthropogenic effects on the basin. Rainfall and its chemistry are to be analysed to understand the groundwater chemistry. Hence, data from various monitoring stations in India are analyzed for assessing the influence of several major factors such as, topographic location of the area, its distance from sea and annual rainfall. These stations are categorized as ‘urban’, ‘suburban’ and ‘rural’. pH, HCO3, NO3 and Mg concentrations have not changed much from coast to inland. On the other hand, SO4 and Ca concentrations changes are subjected to local emissions. Cl and Na (marine elements) originate solely from sea and a model is developed to quantify the variation in concentration of these elements under the influence of inland distance and annual rainfall. Non – linear regression model for the various categories shows that both rainfall amount and precipitation chemistry follow a power law reduction with distance from sea. Cl and Na decrease rapidly for the first 100 km distance from sea, then decrease marginally for the next 100 km and then later stabilize. Regression parameters estimated for different cases are found to be consistent (R2 ~ 0.8). Variation in one of the regression parameters accounts for the effect of urbanization. Model developed for precipitation chemistry is validated using stations from the southern peninsular region of the country. Model predictions are found to be in good correlation with observations with a relative error of ~ 5%. This relationship between the three parameters – rainfall amount, coastline distance, and concentration (in terms of Cl and Na) was validated with experiments conducted at Mule Hole SEW and Kalekere. Monthly variations in precipitation chemistry at these stations are predicted from a downscaled (in time) model and then compared with the observed data. Models developed at both annual and monthly scale are found to perform well with the field observations. Hence, this model is used for predicting the precipitation chemistry (in terms of Cl and Na) of different station points in the upper Cauvery basin. Comparative performance of alternate methods of recharge estimation i.e. Chloride mass balance (CMB) and water table fluctuation (WTF) approaches, is analyzed at various stations in the basin. Annual rainfall, Cl concentration in rain (predicted from precipitation model) and the concentration of Cl in the groundwater are the inputs for the CMB approach. Since main source of Na is from atmosphere, Na is taken as an alternative for Cl in the CMB approach and recharge is estimated using sodium mass balance (SMB) approach. Na concentrations contributed from weathering are quantified and eliminated in the analysis. Recharge estimated using SMB approach is found to be higher than CMB estimate in the semi-arid and the sub-humid zones. Water table fluctuation (WTF) method is used to compare the recharge obtained from both CMB and SMB approaches. Estimates using WTF approach are found to be higher than both CMB and SMB in the semi-arid and the sub-humid zones while SMB is found to be higher than CMB estimates. SMB and WTF estimates match well in the humid zone. An exponential relationship between recharge and annual rainfall is observed. Recharge coefficient estimated on an annual scale varied from 0.1 to 0.25 across the basin. Among CMB and SMB approaches, SMB is a better alternative for recharge estimation in semi-arid zones, where WTF approach performed poorly. Water – rock reactions are driven by the inequilibrium reactions of water with the mineral assemblage in the rock. These reactions evolve towards equilibrium with the primary minerals while a series of secondary minerals precipitate. Mass balance approach is adopted to quantify the rate at which the water – rock interactions occur in order to reach the equilibrium. Field experiments in the experimental watershed (Mule Hole SEW, ~ 4.5 km2) are carried to identify the minerals present in the region and their composition. Quartz, oligoclase, sericite, epidote and chlorite are the primary minerals while kaolinite and Fe-oxides are the secondary minerals present in this region. Percentages of oxides of different elements in each of these minerals are determined from the field experiments. Stoichiometric coefficients of different elements in each of these minerals are determined from these percentages. Long – term weathering rates are determined using these stoichiometric coefficients along with the mass fluxes of each element. Set of minerals present at different depths are found to vary among the thirteen observation wells of Mule Hole SEW. Hence, the mass balance calculations resulted in different weathering rates for a particular mineral based on the spatial location and the particular depth of the occurrence of the mineral. These weathering rates are tested for the sensitivity to carbonates with the inclusion of calcite in the mass balance calculations. With this sensitivity analysis it is observed that the presence of carbonates in the nodular form in the shallow wells has not changed the weathering reactions and their rates, and hence these wells are termed to be in the ‘silicate with secondary carbonate’ system. On the other hand, carbonates are not present in deeper wells, inclusion of which alters the equilibrium of the mass balance calculations. Thus, these wells are said to belong to the ‘silicate’ system. Anorthite present in some of the wells (MH2 and MH6) dissolves accompanied with the dissolution of carbonates. These wells are observed to belong to the third group the ‘amphibolites with primary carbonate’ system. Weathering rates of all the minerals present in these three different systems are also determined annually (short term rates). Mean of these short – term rates are observed to be the same as the long – term (over a period of 4 years) weathering rates with a minor difference of 3 – 10% in values. Thus, the weathering rates determined using mass balance approach is used to determine the quantities of concentrations of different elements contributed from the mineralogical reactions. Temporal variations in the concentrations of different chemical species in this small experimental watershed are simulated using a hydrogeochemical model. The model is developed based on a mixing cell approach, which considers the spatiotemporal variations in the recharge and the weathering inputs. Most of the weathering reactions are observed to take place in the saturated zone, which is termed as the ‘mixing zone’. This zone extends from few meters above the groundwater table to few meters below the water table. Mixing zone is discretized into series of ‘cells’ and concentrations in this zone are simulated. This group of cells is assumed to move along with the groundwater fluctuation. Sensitivity of the model is analysed subject to the variations in the recharge and the weathering fluxes. The developed model is used to simulate the concentrations of the groundwater in the three systems – ‘silicate’, ‘silicate with secondary carbonate’ and ‘amphibolites with primary carbonate’. Field data for chemical species is observed to vary in this mixing zone, boundaries of which are defined from the model simulations. Simulations corresponding to the cell at the mid depth of this mixing zone are observed to correlate well with the field data. Hence, the model developed is able to simulate the temporal variations in the groundwater chemistry. In summary, the study analyses the effects of lithological, climatic and anthropogenic factors on groundwater chemistry. The transformations in the rainwater chemistry as it reaches groundwater are analysed along different stages. A hydrogeochemical model is developed to simulate the groundwater concentrations in three different mineralogical settings over a period of three years.

The economic growth and the increase in population has led to an increased demand for water for various purposes such as domestic consumption, irrigation, industrial use, power generation, navigation, recreation, and ecological requirements. With the increase in population, the per-capita water availability is continuously decreasing. Due to increase in demand and accompanying scarcity of water the conflict among the potential users of the resource is on raise. Hence, the allocation of the available water resource is a big challenge as the intersect oral and inter-regional water allocation is often competing and conflicting in nature. In the above context a good model to manage the available water resources would require reliable inputs on the available water resources. In the first part of this thesis we compare different techniques that are typically used for modeling the river water flow. Time series analysis (ARIMA) is compared with machine learning techniques such as support-vector regression (SVR) and neural network models. The performance of these techniques is compared by applying them to a long-term time-series data of the inflows of three tributaries of the river Cauvery into the Krishnaraja Sagar reservoir (KRS). Flow data over a period of 30 years from three different observation points established in upper Cauvery river sub-basin is analyzed. Specifically, a multi-layer feed forward network trained with a back-propagation algorithm and support vector regression with epsilon-insensitive loss function is compared with the ARIMA models. It is found that the performance of support vector regression model is superior to those of the other techniques considered. The second part of our thesis is to develop a model for optimal water allocation to the different sectors with the aim of maximizing the total utility of available water resource in a river basin. A hydro-economic modeling framework is developed that incorporates the economic assessment of the value of water. This inter-sectoral allocation problem is studied in the context of enforcing certain minimum water rights to every person for domestic use and a certain minimum irrigation need set out by the contingency plans of the state agriculture department in Cauvery river basin. A non-linear optimization model is built to obtain an optimal inter-sectoral water allocation policy. The study evaluates the economic impact of different parameters of competing demands such as water availability, population, basic water right (quantity), ground water contribution, and crop benefit. The optimal policies that implements the water allocation priorities as set out by the National Water Policy (2012) are compared. Further, results show that the basic water right can be secured for essential needs with optimal management of available surface and ground water resources. In the third part of thesis, we study the conflict of water sharing that arises between sectors/regions. We consider the river water-sharing problem between two agents along a river. Each agent has a stated claim to the river water. The Absolute Territorial Sovereignty (ATS) and Absolute Territorial Integrity (ATI) principles are promoted by different agents along the river as a means to maximize their individual benefit. However, these principles are invariably considered to be unjust by one or more of the other agents. Hence, it is preferred to have a negotiated water treaty that is perceived to be equitable and just by all. A one way downstream stream bilateral bargaining model can be used to guide the negotiated water treaty between the agents. In this bargaining framework we introduce the issue of negative externalities imposed by the upstream agent on the downstream agent/s in the form of pollution and/or flooding. This imposes a cost on the downstream agent to mitigate losses due to the negative externalities. A bargaining model that incorporates the impact of negative externalities is developed to guide the negotiated treaties. We identify individually rational bargaining strategies for a two agents transferable utility one way downstream river water sharing problem. The results characterize the agreement and disagreement points for bilateral trading