Addressing groundwater over-extraction in India: assessments, monitoring methods and interventions

Abstract

Groundwater is a vast distributed source of water that is critical for meeting the demands of various socio-environmental systems globally. However, the management of groundwater resources has proven to be challenging with groundwater depletion being observed in many regions globally. As the country with the highest groundwater extraction (and depletion) rates in the world, India is currently at the forefront of this problem where national food security and the livelihoods of millions of households have grown to become dependent on the over-exploitation of groundwater resources. This dissertation consists of three studies to support the broad goal of addressing groundwater overexploitation in India. Specifically, these studies aim to improve understanding on: (1) assessments of stress on regional groundwater resources, (2) identification of groundwater depletion hotspots using monitoring data, and (3) the potential of rain-water harvesting systems as interventions to increase groundwater supplies.

The goal of the first study was to understand the effects of incorporating environmental considerations into large-scale groundwater assessments. Assessments of regional groundwater stress (measured here as the ratio of annual groundwater usage to renewable groundwater supply) are important for setting policy targets and guiding interventions. However, the threshold of yearly groundwater supply that is considered available for human use (especially in relation to environmental water demands) remains poorly defined at the regional-scale. In this study, groundwater extraction thresholds were estimated by scaling yearly groundwater recharge volumes based on different local and global environmental considerations. Focusing on India, district-scale groundwater use thresholds were developed based on: (a) no environmental considerations ('baseline'), (b) water requirements of 'local' groundwater-dependent ecosystems, (c) 'global' considerations using the current planetary boundary framework, and (d) a 'mixed' approach that is informed by both local and global considerations, but where a national groundwater use budget is disaggregated (top-down) to estimate thresholds based on current district-level extraction rates. This was followed by an assessment of how hotspots related to groundwater stress (i.e. regions where groundwater extraction rates exceed estimated thresholds) change in each scenario. Compared to the baseline (where 26% of the districts were considered over-stressed in India), it was found that accounting for local environmental flow requirements results in 36% districts being classified as over-stressed with a groundwater stress hotspot emerging in Southern India. Under the global and mixed scenarios, results showed that nearly 70% of districts (where currently >801 million people live) are classified as over-stressed given current groundwater extraction rates. However, the effort required from over-stressed districts to stay within derived groundwater use thresholds in the mixed scenario (median groundwater stress = 143%) was found to be lower than the global scenario (median groundwater stress = 203%). Overall, the results from this analysis suggest that incorporating environmental considerations would significantly decrease the volume of groundwater resources available for human use in India (173-312 km3/year; compared to 399 km3/year in the baseline).

The aim of the second study was to improve how groundwater depletion hotspots are identified using monitoring data. Numerous recent studies have highlighted groundwater recovery in Southern India due to increasing rainfall rates and political interventions. However, these estimates of increasing groundwater storage trends obtained using hydrological data sources (monitoring wells, GRACE satellite) were found to be incongruent with reports of well failures from non-hydrological data sources (like census data and news articles). Results from this study revealed that previous trend estimates relying on monitoring well data were skewed by the presence of a survivor bias, where dry or defunct wells were excluded from trend analyses due to missing data. Upon further investigation, the timing of missing data and the location of wells with missing data were found to be strongly correlated with metrics of climate stress (i.e. dry periods) and groundwater irrigation intensity, which was indicative of a systemic exclusion. Two alternative metrics that better accounted for information from dry and defunct wells were developed to help augment analysis relying on water level measurement from monitoring wells. An assessment based on these metrics revealed increasing groundwater depletion rates in Southern India between 1996-2016.

In the third study, the potential of rain-water harvesting systems (RWH or tanks) as an intervention to increase groundwater supplies and provide farmers with an alternative source of water was assessed in Southern India. Agricultural rain-water harvesting (RWH) structures remain a promising intervention for improving water availability for small-holder farmers in arid and semi-arid regions of the world. However, the feedback between RWH systems and the surrounding aquifer remains poorly understood in regions like Southern India where these structures are nestled within a landscape of intense groundwater development. In this study, a conceptual hydrological model was developed to answer fundamental questions about how RWH structures impact groundwater availability for irrigation, and in turn how groundwater irrigation impacts the outflow fluxes from RWH structures. Model simulations highlighted that agricultural RWH structures were able to increase groundwater availability in the surrounding area. However, these impacts were meaningful (in meeting agricultural water demands) under only a narrow spectrum of landscape and climate conditions. Specifically, the impact of tanks was found to decline significantly during drought spells or when the beneficiaries of tank-induced groundwater recharge were poorly regulated. Alternatively, results showed that groundwater irrigation in the surrounding aquifer positively impacted the efficiency of output fluxes from the RWH structures by reducing the percentage of evapotranspiration losses and increasing groundwater recharge, however, this came at the cost of reduced water available for surface irrigation. This study provides crucial information to understand the potential of RWH structures in contemporary small-holder dominated agricultural systems.

Overall, the results from this dissertation provide critical insights to support science-based decision-making to minimize environmental impacts of anthropogenic groundwater use, improve monitoring of regional groundwater resources, and better evaluate interventions aimed at increasing (ground)water availability. These insights can aid current efforts to improve groundwater management in India.