Ground Water Modelling of a Semi Arid Tank Cascaded Catchment in a Hard Rock Terrain Using Remote Sensing & GIS

Submitted by Hindi on Sat, 05/07/2016 - 08:27
Printer Friendly, PDF & Email
Bhujal News Quarterly Journal, Jan-Dec, 2012


Groundwater is a dynamic and replenishable natural resource, but in hard rock terrain availability of groundwater is of limited extent and its occurrence is essentially confined to fractured and weathered zones. The occurrence, origin, movement and chemical constituents of groundwater are dependent on the geologic framework. i.e., fissures, degree of weathering and permeability of the rocks through which it moves. Exploration and utilization of groundwater especially in hard rock terrains, requires thoroughunderstanding of geology, geomorphology and lineaments of the area, which directly or indirectly control the terrain characteristics. With the increasing use of groundwater for agricultural, municipal and industrial needs, the annual extraction of groundwater is far in excess of net average recharge from natural resources. In a semi arid region, evapo- transpiration captures most of the water entering the soil, and recharge occurs only at extreme rainfall events. The study of ground water modeling in a semi arid hardrock region is important in order to estimate the available ground water resource for utilization without exploiting the resource.

Sindapalli-Uppodai sub basin is one of the sub basins of Vaippar river basin, Tamil Nadu is taken for the study. The sub basin is characterized as a semi arid region. There are 16 tanks connected in the form of cascades in the study area. The recharge from rainfall to the groundwater will take place during the monsoon seasons. The recharge from other sources such as the recharge from tanks and the recharge from the irrigation water is also an addition to the ground water. The ground water is used mainly for irrigationand also for meeting other demands such as domestic and industrial demands.

The GEC 1997 methodology was used to assess the various recharge and discharge components. In the present study the use of geospatial technology viz GPS, Remote Sensing and GIS to derive various parameters needed to estimate the spatially distributed recharge and discharge components was explored. These components are inputs to Visual MODFLOW to simulate the ground water dynamics in an unconfined aquifer of a semi arid region.


Groundwater resource is a renewable resource subjected to periodic replenishment. National Water Policy, 2002 of India stresses that ‘exploitation of groundwater resources should be so regulated as not to exceed the recharging possibilities, as also to ensure social equity. The dynamic groundwater resource is essentially the exploitable quantity of groundwater, which is recharged annually. It is also termed as annually replenishablegroundwater resource.

In India, dynamic groundwater resources are monitored and estimated jointly by the Central Ground Water Board (CGWB) and State governments at periodical intervals based on GEC methodology. The GEC 1997 has reported the following refinement and improvement for micro level studies.

1. Micro level studies in hard rock terrain should be based only on watershed as the type of groundwater assessment unit.
2. Groundwater assessment for each unit should be computed adopting the recommended methodology and the field tested values for different parameters such as specific yield, transmissivity, storage and porosity to those of ad-hoc norms.
3. The observation and pumping wells have to be monitored for the study period.
4. The assessment may be made separately for monsoon and non-monsoon seasons as well as for command and non-monsoon areas.
5. Remote Sensing techniques can be profitably employed for quantifying various components of the methodology. For example to demarcate cropped area under only groundwater, Remote Sensing and GIS may be advantageous.

In the present study, the above refinements are taken into account. Groundwater assessment unit is considered as watershed. The groundwater recharge is estimated by considering the different recharge components like recharge from rainfall, recharge from tanks, recharge from groundwater irrigation, separately for command and non-command areas. Pumping test was done to estimate the various parameters such as specific yield andtransmissivity. The ground water wells are identified using the high resolution Cartosat imagery. The ground water levels are monitored on weekly interval during the study period. Remote Sensing imagery and GIS is used to prepare the land use map. GIS and GPS are used to demarcate the command area and non command area and the irrigation area under surface and ground water irrigation. Visual MODFLOW is used effectively to simulate thehydraulic heads and to predict water levels.


Sindapalli Uppodai sub basin of Vaippar river basin, Tamil Nadu, India is chosen as study area. It receives drainage from its own catchment which originates from the plain terrain near by Duraiswamypuram village of Sivakasi taluk, runs for a distance of 26 km and it joins in Arjunanadhi at the downstream of Allampatti Village. The entire Sindapalli Uppodai sub basin falls under semi arid region. The base map of Sindapalli Uppodai is delineated fromthe Survey of India toposheet and presented in Figure 1. The total geographical area selected for study at Sindapalli Uppodai sub basin is 143.77 Km2. The sub basin lies between the latitude of 9° 25’00”N to 9° 30’00” Nand longitude 77° 45’00”E to 77° 55’00”E and Taluks of Sivakasi and Sattur in Virudhunagar District. The mean maximum temperature is 33.95 °C to the mean minimum temperature is 23.78°C. There are 16 tanks connected in the form of cascades. Total wells selected for the study is 71 out of which 14 wells are observation wells and the remaining are the pumping wells. Water level is monitored in each tank and in all ground water wells on weekly interval.

Index Map of Sindapalli Uppodai Sub basin


The data needed for the study were collected from the field and other agencies as given in Table 1.

Table 1 Data Collected


Weathered granite, gneiss and Schist with low clay content are present in the sub basin. While using the rainfall infiltration factor method apart from considering the rock type, the infiltration factor of the top soil has to be given due weightage. The soil type is identified by carrying out the soil texture analysis in the laboratory. The soil samples are collected at 53 locations selected in a distributed pattern. It was seen that the soil texture varies fromloamy sand soil to silty loam. Soil classes and area covered by three soil categories are shown in Figure 2. The sandy loam soils covering around 86.938 km2 of the study area was the dominant soil type and this soil is favorable for groundwater potential, Loamy sand group soil was found in an area of 3.995 km2 and the silty loam covering about an area of 51.893 km2.

Soil Map of the Sindappali Uppodai sub basinAs per the GEC norms the infiltration factor for the various soil types in the Sindappalli Uppodai sub basin is shown in table 2.



Ground water recharge may be defined as the downward flow of water reaching the water table, forming an addition to the groundwater reservoir. It is one of the key hydrological parameters for assessment, budgeting, modeling and management of ground water resources. Rainfall is the principal source of recharge. The othersources are seepage from tanks and canals and return flow from irrigation also contribute significantly to the ground water recharge.

Recharge from Rainfall

Rainfall recharge is one of the main components for groundwater recharge. The actual groundwater recharge during monsoon season and non monsoon season is calculated using the rainfall infiltration method as per GEC norm 1997. It says that the infiltration factor multiplied by the season rainfall to the corresponding area gives the recharge due to rainfall from the influencing station. There are 4 raingauge stations having their influence over the sub basin namely Sivakasi, Sathur, Vembakottai and Srivilliputhur. The infiltration factor depends only on the type of terrain in the case of hard rock terrain it again depends on the rock type. The type of rock present in the study area is verified during the well inventory.

Recharge from Tanks

The drainage map was prepared from the Survey of India toposheet and updated using high resolution Cartosat imagery. There are 16 tanks connected in the form of cascades in the sub basin. GPS survey was conducted in each tank to establish the Stage vs Water spread area and Stage vs Storage relationships The Stage vs Storage and Stage vs Water spread area for Annupankulam tank is shown in figure 3. Water level in all tanks during monsoon and nonmonsoon are being monitored on weekly interval.

Stage vs Storage and Stage vs Water spread area of Annupankulam TankAs per GEC norms (1997), recharge from each tank in hectare meters can be obtained as the product of the average water spread area for the season and the number of days of water is actually available and a recharge factor of 0.00144 meters per day per hectare. The number of days the tank water is available is known from the observed data. The tank water level is recorded on weekly basis from which the average water spread area for the season can be found. The details of the tanks available in the study area and its hydraulic particulars are presented in Table 3.

Tank Details

Recharge from Irrigation Water

Recharge from irrigation water applied from surface water irrigation and ground water irrigation has to be computed for the monsoon and non monsoon seasons. High Resolution Imagery CARTOSAT-1 was used to prepare the land use map. The land use map is shown in figure 4. The surface water and ground water irrigated area are delineated from the land use map and verified during field investigation.

Land use map of Sindapalli Uppodai sub basin

Recharge from Surface Water Irrigation

The agricultural crop land nearer to the tank area is commonly irrigated from the tank water. This is verified by field survey and using GPS. The amount of water supplied to each field through canal system has to be known for each tank command area. The water released from the outlet is obtained from the Publics Work Department database and also the number of days it is actually released. The water levels are obtained from the observationwells and interpolated for the sub basin using GIS.

The computation of recharge from irrigation water applied by surface water irrigation in hectare meters can be obtained as the product of the irrigation water applied in hectare meters and return flow factor as a fraction. The irrigation water applied is considered as the sum of the water released from all outlets in the canal system. The water released in hectare metres from each outlet can be computed as the product of the design discharge of theoutlet in hectare metres per day, number of days water is actually released from the outlet and a factor 0.6 (assuming the actual average discharge is 0.6 times the design discharge). The return flow factor depends only on the crop irrigated is paddy or non paddy, whether the range of depth to water table below ground level is less than 10 metres, between 10 and 25 metres or greater than 25 metres and whether the release from outlets is continuous or rotational.

Recharge from Ground Water Irrigation

The agricultural crop land that are irrigated from ground water are identified by the field survey and using GPS. The water levels are obtained from the observation wells and interpolated for the sub basin using GIS. The computation of recharge from irrigation water applied by ground water irrigation in hectare meters can be obtained as the product of the irrigation water applied in hectare meters and return flow factor as a fraction. Thereturn flow factor depends only on the crop irrigated is paddy or non paddy, whether the range of depth to water table below ground level is less than 10 metres, between 10 and 25 metres or greater than 25 metres.

In the study area, it was observed that the water table condition is less than 10m below ground level and the crop rose during monsoon and Non-monsoon season are paddy and non-paddy. Therefore the recharge from surface water irrigation is taken as 30% of the gross surface water applied for non-paddy crop and 50% of the gross surface water applied for paddy as recommended by GEC. The recharge from Ground water irrigation is taken as25% of the gross surface water applied for non-paddy crop and 45% of the gross surface water applied for paddy as recommended by GEC.

Groundwater Draft

Well inventory survey was done using Cartosat Imagery. Field investigation was carried out to identify the type of well. Ground water is primary made use of to meet domestic water supply and irrigation water requirements. Now it is also important to meet industrial water supply requirements. For the current study 71 wells are totally identified using Remote Sensing Imagery. There are 14 observation wells and 57 pumping wells. 51 wells werelocated inside the boundary are shown in figure 5. All other wells are located outside and nearer to the boundary. These wells are used to give general head boundary condition for ground water modeling.

Well Location map of the Sindappali Uppodai sub basinCurrent annual gross ground water draft is assessed using GEC (‘97) norms. The type of well found in the study area is the dug well with pump set. The wells found using the imagery is grouped in to the wells in the command area and non command area. As per GEC norms the average annual ground water draft per well from dug well with pump set is 0.4 to 1.00 hectare metres. The average annual ground water draft for all uses in command areaand non command area can be calculated.


The simulation of groundwater flow requires a thorough understanding of the hydro geologic characteristics of the site. The hydro geologic investigation should include a complete characterization of the following:

a. The top soil extends to a depth varying from 1 to 3 metres in the sub basin. Followed by the weathered formations up to a depth of 8 metres, below 8 m is the hard rock with the Granite- Gneiss.

b. Hydrologic boundaries considered here in this modeling is General Head Boundary condition. Flow into or out of a cell from an external source is provided in proportion to the difference between the head in the cell and the reference head assigned to the external source. The recharge boundaries (recharge from rainfall and recharge from tanks calculated from GEC norms) are also considered.

c. Hydraulic properties of the aquifers such as conductivity and storage co efficient and specific yield and the porosity are given for each layer. The model consists of two layers, first layer is soil and the second layer is weathered formations.

d. The initial head for the observation well is monitored in the field on weekly interval and the model is simulated under equilibrium condition.

e. Distribution and magnitude of groundwater recharge is calculated as per GEC norms, pumping of groundwater is from the dug wells and the pumped water is used for irrigation and for drinking water.

f. Some of the industries located in the sub basin are also depend on ground water. As per GEC norms the irrigation draft and the domestic and industrial allocation of ground water is calculated.


Model Simulation
At first the boundary of the study area map is imported into the model and the map is shown in Figure 5A. The study area has been discretized into an orthogonal grid of 60 rows, 60 columns and 2 vertical layers. The vertical cross-section of the aquifer is shown in Figure 6. This spatial discretization has been found to be adequate in view of the available data and the computational time. For the finite difference solution, a grid of 190m x 120m wasused furthermore, such a grid spacing, given the step chosen for the solution, meets the requirements for numerical stability, even in areas with intense pumping activity.

.a. The details about pumping wells and water levels in observation wells during the period of 2007 to 2009 were imported into the model.

b. The hydrologic properties like hydraulic conductivity and storage coefficient based on lithology were imported for each layer is shown in Figure 7 & Ground Head Boundary condition is shown in fig-8.


Model Calibration

The purpose of model calibration is to establish that the model can reproduce field measured heads and flows. Calibration is carried out by trial and error adjustment of parameters or by using an automated parameter estimation code. The flow directions and the dry cells in the study area are shown in Figure 9 and 10. The water table elevation in the model is shown in figure 11. The recharges from the study area i.e. recharge from rainfall and recharge from tanks is shown in figure 12 and figure 13.

Steady state conditions are usually taken to be historic conditions that existed in the aquifer before significant development has occurred (i.e. inflows are equal to outflows and there is no change in aquifer storage). In this model, Steady state calibration comprised the matching of observed heads in the aquifer with hydraulic headssimulated by MODFLOW. The calibration was made using 8 observation wells monitored during 2007-2009. Hydraulic conductivities estimated from Pedo Transfer Function (PTF) were used as initial values for the steady state simulation. By trial and error calibration, the conductivity values were increased during many sequential runsuntil the match between the observed and calculated water level values were obtained (Fig. 14,15, 16 &17). The computed water level accuracy was judged by comparing the mean error, mean absolute and root mean squared error calculated. Mean error is -0.096 m. Root mean square (RMS) error is the square root of the sum of the squareof the differences between calculated and observed heads, divided by the number of observation wells, which in the present simulation is 1.297 m. The absolute residual mean is 0.956 m.

The model was calibrated to transient state from 2007 to 2009. Since data on various flows in transient state were available for a period of 2007 to 2009, simulation was performed for that period taking the 2007 water levels as initial condition and then the simulation was carried out under transient conditions 2007 onward to 2009. Thetime steps in transient simulations run from 2007 to 2009 were divided into 20 time steps. Each year was also divided into two stress periods of 152 days (monsoon period) and 213 days (non- monsoon period) respectively. Recharge boundary were initially set using 30-days stress period, which was gradually increased to 152 and 213days. The actual amount of recharge was calculated for each year using GEC’97 methodology. Visual MODFLOW uses boundary condition imposed by the user to determine the length of each stress period.


Zone Budget

Zone Budget calculates sub-regional water budgets using results from steady-state or transient MODFLOW simulations. Zone Budget calculates budgets by tabulating the budget data that MODFLOW produces using the cell-by-cell flow option. At a period of 365 days, storage in to the aquifer is 3.43 Mm3/day and flow out of the aquifer from storage is 7.97 Mm3/day. The water pumping out from the wells is 3.4 Mm3/day. Recharge into the aquifer is 7.99 Mm3/day. The total inflow into the aquifer is 11.443 Mm3/day and total outflow into the aquifer is 11.441 Mm3/day. At the period of 365days, the outflow is more than the inflow by an amount of 0.0016 Mm3/day. This has been shown in the Figure 18.


The dynamic groundwater resource is the exploitable quantity of groundwater, which is recharged annually. In the present study groundwater recharge in the tank cascaded catchment, Sindapalli Uppodai sub-basin in Vaippar river basin has been estimated by using Groundwater Estimation Committee (GEC-1997) norms. The Groundwater assessment unit is considered as watershed and the groundwater recharge is estimated by considering the different recharge components like recharge from rainfall, recharge from tanks, recharge from groundwater irrigation, separately for command and non-command areas. GIS is a powerful tool successfully used to prepare the soil map, land use map and drainage maps from the high resolution satellite imagery and SOI toposheet and field survey. Visual MODFLOW is used effectively to simulate the hydraulic heads of an unconfined aquifer. The hydraulic heads have been simulated and it shows a good relation with observed values. The predicted ground water levels are used to study the ground water balance for proper ground water estimation and management. Various scenarios can be built by changing the pumping rates, by varying the recharge from irrigated area and tanks.


This research work is part of the Research project “Rainfall-runoff modelling and Groundwater Dynamics of Irrigation Tank Clustered Catchment of Semiarid Region” funded by Ministry of Water Resources through INCID. We are very much grateful to Indian National Committee on Irrigation and Drainage and Ministry of WaterResources, Government of India for funding the project.

Chand.R., 2004, Estimation of natural recharge and its dependency on sub-surface geologic parameters”, Journal of Hydrology Vol.299.pp.67-83.
Detailed Guidelines for Implementating the Groundwater Estimation Methodology – 1997, Central Ground Water Board, Ministry of Water Resources, Government of India.Haris H. Khan, 2008, “Remote Sensing and GIS in estimating Groundwater Balance in semi-arid regions”, Indo- French Centre for Groundwater Research, National Geophysical Research Institute, Hyderabad, AP, India.
Pradeep K. Naik, 2003, Groundwater resources assessment of the Koyna River Basin”, Hydrogeology Journal,Vol.11,pp:582-594.
Sahu. B.K., 2002, ‘Groundwater modeling in hard rock terrain, Proceedings of the International Conference on Sustainable development and management of groundwater resources in Semi-arid region with Special Reference to Hard rock. IGC-2002, Dindugul, Tamil Nadu.
Sekhar. M., 2000, ”Modeling the Groundwater Dynamics in a Hard Rock Aquifer Influenced By Boundary Fluxes, Spatial and Temporal Variability in Pumping/Recharge in a Semi-arid, Intensely Irrigated Agricultural Basin”, Department of Civil Engineering, Indian Institute of Science, Bangalore.
Thambi.D.S.C and Rani.V.R, 2010, Review of Groundwater Resource Estimation Methodology 1997 (GEC 97) in Hard Rock Terrain, Proceedings of Workshop, Ground Water Resources Estimation, New Delhi.
Welsh W.D, 2002, “Spatial and Temporal water balance estimates using GIS”, Engineers Australia, 29th Hydrology and Water Resources Symposium, pp: 21-23.

M.Krishnaveni, Center for Water Resources, Anna University, Chennai

Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

9 + 11 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.

Related Articles (Topic wise)

Related Articles (District wise)

About the author