Getting the groundwater resource estimates right

In this post, PLOS Water‘s Section Editor for Water Resources and Hydrology, M. Dinesh Kumar (Executive Director of Institute for Resource Analysis and Policy in Hyderabad, India) discusses the importance of groundwater resource estimates in India, and recommendations to improve them in future.

It is often stated that India is the largest user of groundwater in the world. The maximum use of this resource in the country is for irrigation, accounting for nearly 65% of the net irrigated area or 65 million ha in absolute terms. The official estimates by the Central Ground Water Board (CGWB), the premier technical agency responsible for groundwater resource evaluation and planning in the country, puts the annual renewable groundwater resources at 432 billion cubic metres. The use of the resource for agriculture is almost entirely in the private sector with several millions of farmers owing wells of different capacities.

In India, groundwater occurs in unconsolidated (alluvial), semi-consolidated (sedimentary) and consolidated formations. The spatial variations in precipitation, climate, surface morphology, aquifer storage properties, and topography cause major regional variations in annual replenishment of groundwater from precipitation. Against this, the estimated withdrawal of groundwater for all uses stands at only 240 BCM. Hence at the aggregate level, as per official data, the situation is not bad as the groundwater balance (recharge –abstraction) is still positive.

However, groundwater resources have been depleting for several decades in many semi-arid and arid regions, manifested by long-term drops in water levels. Hence the groundwater balance estimates being prepared by CGWB periodically becomes a very serious exercise. All government measures for management of groundwater, including regulatory measures, artificial recharge schemes and institutional financing for well development are dependent on these groundwater balance estimates. Hence, they are quite crucial. Sadly though, the groundwater balance estimates offered by the CGWB for different administrative units (districts) often do not reflect the situation on the ground in terms of the resource condition, especially the water level trends.

CGWB’s first attempt at developing a scientific methodology for groundwater resource assessment was in 1984, with the setting up of a groundwater estimation Committee (GEC-1984). Broadly, the CGWB methodology is based on groundwater balance, which considers the following: i) net groundwater recharge during monsoon; ii) groundwater recharge from irrigation during the non-monsoon season; iii) groundwater recharge from canal and recharge structures; iv) groundwater draft during the monsoon season; v) natural discharge of groundwater during the non-monsoon season.

Over the years, the agency had also been refining its methodology. The current methodology follows a set of norms by Groundwater Estimation Committee (GEC) to estimate the groundwater resources for monsoon and non-monsoon seasons. The CGWB methodology estimates monsoon groundwater recharge by introducing a coefficient to reconcile with the estimates obtained from two different analytical procedures. Known as the ‘percentage difference’ (PD%), it is the variation of groundwater recharge value, obtained using Water Level Fluctuation (W.L.F) method from the recharge value obtained using ‘rainfall infiltration factor method’ for the monsoon season. These norms lack proper scientific backing. One major weakness of the approach is that the specific method used for the estimates for a specific region is not revealed. This makes the assessment non-falsifiable.

The GEC and CGWB reports state that these norms were set to avoid unreasonably high or low estimates of groundwater recharge values. But these norms normalise the effects of extreme rainfall events. For an instance, it was observed by Vaishnavi and Kumar (2023) that the rainfall received by Ahmednagar district, Maharashtra during non-monsoon season is very little and thus the water available for groundwater recharge after the subtraction of effective rainfall is almost zero. But the CGWB report, 2013 showed a groundwater recharge of 254.12 million cubic meters (MCM) from rainfall during non-monsoon period for Ahmednagar district.

New analytical procedure has been proposed by CGWB to estimate the groundwater discharge into the streams during the lean (non-monsoon) season, a major factor influencing the utilizable recharge in an area.

However, in practice, the methodology being used is still the outdated one. To estimate the monsoon recharge, the monsoon rainfall is multiplied by a coefficient, which is arbitrarily chosen on the basis of the properties of the surface strata, instead of using the WLF approach which is more scientific. This is in spite of having a large number of observation wells maintained by the agency (around 26,000 across the country) for monitoring water levels four times a year. One major reason for non-use of this approach is the absence of reliable data of specific yield of the aquifer. The result is that for the high rainfall mountainous regions (like the North-East), the methodology yields very high values of recharge, whereas in reality too little groundwater is available in this region, confined to the thin valley. For estimating the irrigation return flows, a coefficient (value, 0.30) is chosen arbitrarily for wetland paddy. This again is unsuitable, as the fraction of the water that goes as return flow is a function of the soil hydraulic conductivity, depth to water table, rainfall and irrigation depth.

The lean season (natural) discharge from aquifers is neither measured nor estimated scientifically, and instead arbitrary values are used.

Importantly, there are many millions of irrigation wells spread across the country with their discharges varying drastically according to geohydrological settings–shallow alluvium, deep alluvium, hard rocks, semi-consolidated formations etc., and characteristics of the wells, and the hours of pumping varying according to the water demands and seasonal availability of groundwater. However, the current methodology doesn’t capture these nuances. No measurement of groundwater draft is made. Instead, it uses certain simplistic norms with regard to the discharge and hours of pumping for each type of well and based on the total number of wells under each category, estimates the total draft. This can cause big errors in the estimates of ‘draft’. 

Finally, the groundwater balance estimates are not corroborated with the data on annual water level fluctuations. If in an area, the groundwater outflow (abstraction + natural discharge) during a hydrological year exceeds the inflow (total recharge), the water levels should ideally drop and vice versa. However, it is quite common to find that even in areas where official estimates show positive groundwater balance, water level data show declining trends over the year. This can be because of over-estimation of recharge, or under-estimation of draft or both.

It is time that we take the groundwater assessment more seriously and put in more scientific efforts to get reliable estimates of recharge and abstraction for each district in the country, as the stakes are very high. The following actions are urgently needed: i) getting realistic estimates of specific yield of different types of aquifers in different regions; ii) quantifying the lean season flow of streams in different regions so as to arrive at the groundwater discharge rates; iii) quantifying the groundwater abstraction rates for sample wells from different settings; and iv) quantifying the return flows from irrigation using detailed water balance studies in different regions using water accounting framework.

Interested in submitting your research on groundwater and water resources to PLOS Water? Find our step-by-step guide to submitting to the journal here.