HESS Opinions: The myth of groundwater sustainability in Asia

Across the arid regions of water-stressed countries of Asia, groundwater production for irrigated agriculture has led to water-level declines that continue to worsen. For India, China, Pakistan, Iran, and others, it is unrealistic to expect groundwater sustainability in a verifiable sense to emerge. Fragmented governance and the general inability to bring traditional socio-economic tools to bear on reducing groundwater demands have impeded progress to groundwater sustainability. For India and Pakistan, where operational management is at the level of states and provinces, there is no capacity to regulate. Also in both China and India, the tremendous numbers of groundwater users, large and small, confound regulation of groundwater. With business as usual, groundwater-related problems receive insufficient attention, a situation referred to as an “accelerating and invisible groundwater crisis” (Biswas et al., 2017). Another obstacle to sustainability comes from trying to manage something you do not understand. With sustainable management, there are significant burdens in the needed technical know-how, in collecting necessary data, and in funding advanced technologies. Thus, there are risks that Iran, India, and Pakistan will run short of groundwater from over-pumping in some places and will also be adversely affected by global climate change.


Introduction
About 20 years ago, hydrogeologists began more fully to appreciate the extent of non-sustainable withdrawals of groundwater worldwide. However, recognizing a problem and doing something about it are two different things. Our focus 20 here is Asia, where the need for sustainable groundwater management is essential, given impacts from irrigated agriculture and growing urbanization. Yet, progress towards sustainable development has been slow to non-existent. For many of these countries, groundwater sustainability is essentially just a myth. This paper makes a case that groundwater impacts in developing Asian countries are already bad and getting worse. Further, it describes impediments to sustainable groundwater management and presents suggestions for a pragmatic research agenda. 25 The concept of sustainability refers to the "development and use of groundwater in a manner that can be maintained for an indefinite time without causing unacceptable, environmental, economic or social consequences" (Alley et al., 1999). It builds on the foundational concept of safe yield as "the limit to the quantity of groundwater which can be withdrawn regularly and permanently without dangerous depletion of the storage reserve" (Lee, 1915). A modern concept of groundwater sustainability recognizes the additional complexity provided by the inherent coupling of groundwater and The Indo-Gangetic alluvial (IGA) aquifer system occurs across the top of India (Figure 2b), extending into Pakistan, Nepal, and Bangladesh. It includes flood plains along the Indus and Ganges Rivers and their tributaries as a thick sequence of 65 alluvial sediments derived from the Himalayan Mountains (MacDonald et al., 2016).
Yet, impacts from pumping are not the urgent problem that some measurements (e.g., Rodell et al., 2009) imply. The 75 relatively large quantities of groundwater stored in the upper 200 m of the IGA system coupled with 100 plus years of additional recharge from unintended canal leakage and irrigation return flows means that depletion is regionalized. Yet, what is concerning is that the greatest recent water-level declines are evident in northern Indian and Pakistan, areas essential for food production with irrigation.
What is often misunderstood about IGA aquifer system is the greater threat to groundwater sustainability associated with 80 water quality issues -salinity, urban and industrial contaminants, and geogenic arsenic in groundwater associated with sediments from Himalayan sources (MacDonald et al., 2016;Foster et al., 2018;Young et al, 2019). The origin of salinity in the shallow groundwater is complex but commonly associated with effects of irrigation. Leaking canals, over more than a century in some instances, have led to waterlogging and salt accumulation in soil, and the salinization of recharge (Foster et al., 2018). Large capacity irrigation wells are also capable of mobilizing naturally salty water occurring at depth with up-85 coning. Estimates are that 18,000 km 3 or 60% of shallow groundwater in the IGA system suffers water-quality impairment (MacDonald et al., 2016).
The largest cities of India exemplify the emerging problems of water sustainability. A useful example is Delhi whose population of ~25 million is poised to double in the next 30 to 50 years. Most of Delhi's drinking water comes from surfacewater sources; but groundwater from the IGA aquifer system is both important and problematic. Almost every sustainability 90 issue just discussed is a major problem for Delhi -rapidly declining water levels, salinity at depth, and nitrate concentrations commonly >45 mg/L and as high as 1500 mg/L. Various news outlets have been active in expressing concerns about the local impacts of these problems (Text Box 1).
Political and policy failures associated with groundwater and surface water have created a crisis for India that bears directly on food, water and health (Biswas et al., 2018). Citing centuries of mismanagement of water resources, and "institutional 95 incompetence" (Biswas et al., 2018) in the context of a large growing population, there has been no willingness for action politically in India beyond "cosmetic changes" (Biswas et al., 2018). Although issues involved with surface waters https://doi.org/10.5194/hess-2019-399 Preprint. Discussion started: 12 August 2019 c Author(s) 2019. CC BY 4.0 License.
(contamination, fights over allocation, and reliability of public supplies) are worsening; "the groundwater situation is even worse" (Biswas et al., 2018). Yet, data on groundwater is poor in quality or unavailable. Rampant growth in groundwater utilization is linked in part to the failure of government to provide surface water for irrigation (Biswas et al., 2018). 100 Pakistan is another country with groundwater issues threatening future sustainability. Its large population, ~208 million and growing, contributes to its water scarce status with a per capita availability of water in the lowest 10% of the world's population (Young et al., 2019). The Indus River and its tributaries are significant surface-water resources, used almost entirely to support irrigated agriculture. Yet, the use of water is inefficient with significant losses due to canal leakage, evaporation, and over-irrigation (Young et al., 2019). 105 The IGA aquifer system extends southward into Pakistan along the length of the Indus River. For now, levels of groundwater in the IGA aquifer system in Pakistan are stable or even increasing (MacDonald et al., 2016). The main problems are associated with water-level declines of ~10 m since the 1980s in the important food growing area of Punjab Province to the northeast (Young et al., 2019). Here, as in India, canal leakage and irrigation return flow have continued to provide an unmanaged aquifer recharge system that has banked water in the subsurface since the late 1800s to the point of waterlogging 110 in some places (MacDonald et al., 2016). The greater threat to sustainability comes from the kinds of water quality problems mentioned previously.
There is little progress in the development of a sustainability ethic for groundwater management in Pakistan. Assessments are frustrated by an absence of data and the lack of a quantitative understanding of groundwater-surface water interactions along the major rivers (Young et al., 2019). 115 Elsewhere in Asia, the non-sustainable production of groundwater has resulted in even more serious problems. In Iran, the significant loss of groundwater resources could render major parts of the country uninhabitable with the possibility of millions displaced as conditions worsen (Collins, 2017). In addition to widespread declines in water levels, there are significant problems related to land subsidence and declining water quality (Madani et al., 2016).
Various factors have contributed to groundwater insecurity. Iran has a growing population of ~80 million, which has doubled 120 over the last 40 years (Bozorgmehr, 2014). The country is dry, making groundwater a growing source for drinking and irrigation water. A continuing trend towards urbanization has resulted in an urban population of 70% with 18% in Tehran (Madani et al., 2016). Since 1999, there has been a succession of drought years. When coupled with an increase in annual temperature, the new normal is dryer and hotter weather with a likely decline in precipitation and recharge in coming decades due to climate change (Gohari et al., 2013;Nabavi, 2018). 125 This water crisis is also driven by socioeconomic decisions in the late 1970s to become self-sufficient in wheat, the country's most important crop (Collins, 2017). The expansion in wheat production through irrigation has had significant impacts on groundwater. Yet, there are few signs of movement towards a more sustainable groundwater future (Collins, 2014).
Another Asian hot spot for impacts associated with unsustainable groundwater production is Jakarta, Indonesia, on the island of Java. Approximately 25-30% of the more affluent residents of this large city receive piped-in surface water (Colbran, 130 2009). Others obtain drinking water from large numbers of groundwater wells, rainwater, vendors, bottled water, etc. The groundwater (Colbran, 2009).
Yet, this is not a drought story. Large, localized production from the shallow unconfined aquifer ~50 m thick and a deeper confined aquifer ~100 m thick is not sustainable even with significant natural recharge (Kagabu et al., 2013). The overuse of 135 groundwater has been evident for a long time. For example, in 1995, reported pumping rates were three times larger than recharge rates. By 2008, drawdowns in the deep aquifer were >40 m with hydraulic heads 25 m below sea level (Kagabu et al., 2013). Water quality in the shallow aquifer is impacted by urban contaminants, like NO3 (Kagabu et al., 2013) because there are virtually no sanitary sewer systems. There is evident seawater intrusion landward within the deep aquifer caused by over-pumping. Declining water levels have also resulted in subsidence that in several places exceeds 2 m (IRIDeS, 2017). 140 Now approximately 40% of the city's land surface is below sea level with only a seawall to protect land from inundation.
Yet, as far as groundwater utilization, it appears to be business as usual.

What are the Hurdles to Groundwater Sustainability?
Developing Asian countries have encountered significant roadblocks hindering progress towards groundwater sustainability. 145 So far, it has been relatively painless for countries and large cities to simply ignore groundwater issues, which in the case of India has been called an "invisible" crisis (Biswas et al., 2017). Of greater concern in Asian countries is a collection of more critical national issues related, for example, to growing their economies, feeding their people, maintaining national security, and improving the social conditions for growing populations.
An instructive example is Yemen, a slowly unfolding example of human tragedy. Yemen is located at the southern end of the 150 Arabian Peninsula bordering the Red Sea. After four years of civil war, half the population (~28 million) is short of water (Camacho et al., 2018). Approximately 60% of the population is food insecure with nearly 500,000 children under five suffering "severe acute malnutrition" (BBC, 2017). The public health system has trouble providing basic services in the face of the world's largest recorded epidemic of cholera in modern times. There have been more than one million cases from 2016(Comacho et al., 2018. It is easy to understand why problems of groundwater over-pumping evident even in 155 2002 in the Sana'a basin (Foster, 2003) are not an urgent national concern.
In Asian countries, much less deference is given to water security than food security. India's "Green Revolution" (GR) is a case in point. In the 1950s, government leaders in India were troubled by the deaths from the Bengal Famine of 1943 (Rahman, 2015). With their growing population, achieving food security became a top priority. In the 1960s, the GR began with an expansion in agricultural lands, new high-yielding seeds, expanded irrigation, double cropping, and vastly increased 160 fertilizer and pesticide applications (Rahman, 2015;Schmanski, 2008). India became food secure with large increases in the production of food and cereal grains.
Yet there is dark side, which includes severe social, economic, and environmental problems, particularly in the amazingly from pesticides, and especially the unsustainable utilization of groundwater (Schumanski, 2008;Singh and Park, 2018). 165 Groundwater impacts were slow to develop but are now serious with water-level declines ranging from 4.5 m to 35 m (Rahman, 2015).
China, India, and Iran have been able to aggressively ramp up agricultural production to feed their people without adequately planning for the impacts to groundwater. With recharge and the inherent capacity of large aquifers to store abundant groundwater, groundwater problems developed incrementally and have been difficult to recognize in data-poor settings. 170 Now, food production from irrigated agriculture is structurally part of the national economies of these countries, making it difficult to reduce the production of food and groundwater.
The second major impediment to progress in sustainable management is the inherent inability to manage anything that is not understood. For example, assuming that appropriate laws and regulations exist, government administrative actions, such as authorizing or charging for groundwater use, banning new wells, or capping production with existing wells (Garduno and 175 Foster, 2010), all depend on data. The solution to groundwater requires an understanding of the scope and extent of problems and specific information as to who is pumping what quantities of water. Populous developing countries quite simply may lack the capacity to administer groundwater.
Modern technical or socioeconomic interventions in support of aquifer sustainability depend on data for planning purposes and compliance monitoring to assure actual progress towards sustainability goals. This kind of data-centric approach is 180 different than, for example, traditional water-harvesting methods understood to promote groundwater recharge and storage.
For example, India is the world leader in the number of these traditional systems installed (Dillon et al., 2019). There are "several million recharge structures" (Dillon et al., 2019) in place, some quite old, and 11 million more planned. Typically, this type of managed aquifer recharge (MAR) has involved streambed recharge and percolation tanks/ponds to store water in the subsurface. Yet, there are few quantitative assessments of the efficacy of these approaches in promoting recharge (Dillon 185 et al., 2019;Dashora et al., 2018). While contributing to sustainability, these practices are unlikely to achieve that goal.
Moving beyond India to China and S.E. Asia, there are few active MAR projects in operation (Dillon et al., 2019).
A program invested in groundwater sustainability requires personnel with basic knowledge of hydrogeology along with specializations in relevant topical areas. Such specialized knowledge exists in the world, but not so much in Asian countries.
Until this expertise is developed and embellished through practical experience, progress will continue with the ad hoc 190 traditional practices. This idea of "understanding" in relation to the sustainable management of groundwater must extend to comprehensive national data collection, such hydrogeologic mapping, monitoring, and modeling. Yet, there is little discernable progress in data collection necessary to support sustainability initiatives in either India (Biswas et al., 2017) or Pakistan (Young et al., 2019). There may be some progress in China but information there is siloed and lacking in necessary transparency. In the 195 megacities, like Jakarta, Delhi, and Karachi, our reviews found the status of groundwater data to be meagre to nonexistent, totally inadequate to support technical or socioeconomic efforts towards sustainability. The kinds of technical information and infrastructure needs for sustainable groundwater management are well known. They include a robust qualitative and quantitative understanding of how the land-based portion of the hydrologic system functions, physically, chemically and biologically. Basic data collection involves metering or other approaches to establish water-200 utilization, aquifer characterizations, testing, sampling and measurements in the field, supported by various monitoring networks, data acquisition systems, laboratories and database systems. Figure 3 highlights the broad scope of data needs with an illustrative conceptual model of a complex coastal hydrologic system (CDWR, 2016).
Asian countries starting from scratch will need to anticipate costs associated with years of field operations in, for example, groundwater mapping, aquifer testing, and water quality measurements. Various monitoring networks will need to be 205 designed and emplaced, as well as equipment to be purchased, installed, and operated. Provision must be made for data compilation and storage, interpretations, modeling, laboratory measurements, etc. What adds even more difficulty is an absolute need to monitor for one to several decades to provide an average set of baseline conditions (CDWR, 2016). The creation of conceptual models, water balance calculations, and compliance assurance all require these kinds of data. A useful place to gain perspective is with a series of best practices reports of the California Department of Water Resources (e.g., 210 CDWR, 2016). They are intended to provide technical assistance for California's new state-wide initiative in sustainable groundwater management.
The third major obstacle is that technically-oriented sustainability initiatives require expensive infrastructure with continuing operating costs. Consider a problem where the key issue with sustainable management is water-level declines from excessive pumping. The operational objective is to end up with an aquifer system where water storage does not change over the long-215 term while maintaining appropriate natural discharges to rivers and springs. Reductions in storage due to unsustainable production can only be reversed in two ways -increasing the quantity of inflows to the aquifer (e.g., recharge) or decreasing the outflows (e.g., pumping with wells). The yellow box in Text Box 2 lists four recharge schemes to increase inflows to aquifers (i.e., MAR) with links to the associated issues/problems, as indicated by the red arrows.
Clogging is a significant problem reducing the quantities of water infiltrated or injected into the subsurface. Consequently, 220 MAR systems require regular maintenance to maintain performance. In an Asian context, the other issues affecting MAR (Text Box 2) also provide formidable challenges. Finding water to recharge an aquifer can be difficult. Surface water can be scarce because excess water is often only available with summer monsoons. Treated municipal sewage, another important source of water, is often not available or of appropriate quality. For example, ~50% of Delhi's population has no sewers (Sengupta, 2015) with significant quantities of wastewater dumped into the nearby Yamuna River or left to seep into the 225 ground. In addition, there tends to be declining interest in projects involving long transfers of water. Farmers are commonly not persuaded that government-supplied surface water for irrigation is a preferable alternative to groundwater sources (World Bank, 2010). Infrastructure, like reservoirs, pipelines or canals is needed to transfer water to where it is needed.
A variety of strategies exists to reduce groundwater withdrawals. Replacing groundwater (i, ii green, Textbox 2) in irrigation with imported surface water or treated wastewater is often not practicable, as was mentioned. Decreasing agricultural 230 production through acreage reductions, growing one crop per year instead of two, or changing to crops that use less water https://doi.org/10.5194/hess-2019-399 Preprint. Discussion started: 12 August 2019 c Author(s) 2019. CC BY 4.0 License.
will lead to less groundwater utilization (iii to vi, Text Box 2). Yet, on the one hand, with governments firmly committed to food security and poor farmers needing to maintain their livelihoods, such initiatives are unattractive. On the other hand, these strategies require minimal technical expertise. Governments can pass a law, check the sustainability box, and plan to spend money to import some food. Finally, more efficient irrigation technologies might lead reduced pumping (Garduno 235 and Foster, 2010) There are successful models for managing water resources sustainably. They are worth discussing here to illustrate (i) requirements for data and advanced technologies, (ii) the long-term commitment to complex and costly projects, and (iii) efforts necessary to turn urban wastewater into a valuable water source. The Orange County Water District (OCWD) in southern California near Los Angeles is a leader in sustainable groundwater management. OCWD serves a 900 km 2 area, 240 distributing water to ~2.4 million people. Two thirds of that water comes from groundwater produced from hundreds of deep, high capacity wells. Sustainable operation of the aquifer systems provides ~345 Mm 3 yr -1 of high-quality groundwater, The success of OCWD's hydrogeological operations is critically dependent on monitoring. They collect production data monthly for the high capacity production wells and less frequently for smaller wells (OCWD, 2015). Water level and water quality data coming from hundreds of wells provides evidenced-based compliance with sustainability goals. The quality of 260 water from the GWRS facility is monitored, as is the Santa Ana River and tributaries (OCWD, 2015). Performance of the seawater intrusion barrier is also monitored along with subsidence across the basin.
Such sophisticated water management systems are uncommon in Asia. Yet, the island state of Singapore is home for an innovate collection of management activities creating near self-sufficiency from water imports from Malaysia (Irvine et al., 2014). Drinking and industrial waters come from capturing and treating rainwater captured with urban catchments, the advanced purification of urban wastewater to a product called NEWater, and the addition of desalination plants (Irvine et al., 2014).

Planning for the Worst and Hoping for the Best: Groundwater Research Directions
There are compelling arguments why it is unrealistic to expect groundwater to be managed sustainability in developing 270 Asian countries. The indictment for India, "centuries of mismanagement, political and institutional incompetence; indifference at central, state, and municipal levels, and steadily increasing population" (Biswas et al., 2017) applies as well to other countries. Groundwater-related problems are largely invisible (Biswas et al., 2017) and seemingly irrelevant to a greater agenda. For China, India, and Iran, there is an undeniable focus on food production to support growing populations and changing food preferences of increasingly affluent societies (Young et al., 2019). The continuing trend towards 275 urbanization at all scales up to megacities is localizing water demands and exacerbating groundwater problems.
Despite progress with satellite-remote sensing, particularly GRACE (Feng et al., 2018;Long et al., 2017;Rodell et al., 2009;Rodell et al., 2018), actions around evidence-based groundwater sustainability is at an early stage. In the case of the IGA aquifer system, the greatest present threat to long-term sustainability is not from over-pumping but from human activities that have led to groundwater salinization and urban/agricultural contamination (MacDonald et al., 2016). This experience in 280 India and Pakistan and possibly China reveals how pervasive contamination can lead to the same unsustainable outcome as over-pumping.
Adding water-quality issues to the sustainability mix reveals even greater deficiencies in data and needs for research in modeling and arid-zone geochemistry. For example, salinity problems are complicated because impacts can occur in so many ways. In Pakistan, saline water exists at depth in addition to salinized recharge caused by waterlogging. Moreover, this 285 deep groundwater water can be remobilized by pumping (Foster et al., 2018). In China, shallow groundwater across the eastern half of the North China Plain is salinized (Foster and Garduno, 2004). This creates the possibility for eventual water quality impairment in the underlying deep freshwater aquifer as over-pumping there continues. Research is required to explore mechanisms, pathways and time scales of contaminant-related impacts on sustainability of aquifers. Another target of opportunity is the difficult field characterizations of the geochemistry of saline groundwaters in arid-zone settings. 290 A pragmatic research agenda must also account for the risk that sustainable groundwater management will never happen.
The necessary transition from a water policy of muddling along, stumbling from one crisis to the next without substantive actions to quantifiably sustainable systems, like those in Orange County or Singapore will be enormous. Further, it is doubtful whether the successful strategies in those two places with relatively small and economically advantaged populations are practically scalable to many millions of people in developing countries. In any case, logistical constraints mean that it 295 will be decades before sustainable systems are up and running. Such a delay increases the possibilities of predictable surprises -the problems (e.g., climate change) that are anticipated but ignored (Bazerman and . It is worthwhile to consider research to support those sustainability initiatives that are likely to be undertaken. For example, India appears poised to invest in traditional MAR schemes (Dillon et al, 2018). There are significant opportunities in reimagining this overall approach by adapting modern practices to the design and restoration of traditional water-harvesting 300 systems.
Another possibility is for research to facilitate adaption to the worst outcomes. This idea came from a recent conference address (Siegel, 2019), which challenged the audience to contemplate a future where researchers are consumed in dealing with problems of adapting to unmitigated impacts of climate change. A population-modeling study gaging the future incountry migration in arid countries due to climate change (Rigaud et al., 2018) is an interesting example. The results for 305 South Asia suggested that by 2050, there could be 35.7 million in-country climate migrants under a pessimistic climate scenario.
In the context of groundwater sustainability, we envision a need for research that would help with adapting to the ongoing decline in groundwater availability aggravated by climate change impacts. In other words, with groundwater sustainability unlikely to be achievable, research could help in understanding when the groundwater is likely to run out, possibilities for 310 creatively stretching the supply, or envisioning ways to ease the impacts of inevitable declines in food and health.