Resolving the Ganges pollution paradox: A policy‐centric systematic review

Despite the fact that it is one of the most sacred and holy rivers in the world, the Ganges River is paradoxically among the most polluted. Over the past decade, researchers have described various mechanisms and actions for improving the pollution problem within the Ganges watershed. The aim of this policy‐centric systematic review is to summarize these recommendations to make them more accessible for concerned citizen groups and planners while also critically appraising their findings. Using the Reporting standards for Systematic Evidence Syntheses (ROSES) framework, our findings indicate that there are a wide range of potential solutions for mitigating pollution in the river system that originate from 37 peer‐reviewed sources that encompass field studies, modeling analyses, and review articles. While we find that there are many actionable and thought‐provoking recommendations for improving water quality and pollution mitigation given by authors studying the Ganges, there are also areas for improvement. Notably, there is a heavy focus on state‐centric planning in the basin with only a few examples of policies that have been tailored toward encouraging community‐based solutions. This lack of community‐based planning may relate to the fact that there is also a missing social dimension to policy recommendations in the Ganges watershed, where most of the articles that we reviewed were published in natural science journals and were not interdisciplinary in nature. Better reporting standards for recommendations arising from reviews and a greater focus on the interrelations between different components of the Ganges system may also yield novel and more trustworthy policy findings for practitioners.


| INTRODUCTION
The Ganges River in many ways presents an incongruous paradox to onlookers. From its headwaters in the upper reaches of the Himalayan Range to its mouth in the Bay of Bengal, the Ganges is a source of life and livelihood for tens of millions of people across the Indian subcontinent. In Hinduism, the Ganges is seen as the mother of the human race and is viewed as a source of purity along every centimeter of its course (D. Kumar, 2017). Despite its spiritual purity, since the industrialization of India, the Ganges has become physically degraded and tainted throughout its watershed creating a contrasting image of divine spotlessness on one hand and environmental filth and decay on the other. In 2017 it ranked as the second most polluted river in the world (Mariya et al., 2019), and though the river has seen some minor improvements in water quality in recent years (Duttagupta et al., 2019;Muduli et al., 2021) it still is among the most polluted rivers globally. Common pollutants in the Ganges include plastic waste (Nelms et al., 2021), heavy metal concentrations (Dubey, 2021;Paul, 2017), total and fecal coliforms especially in nonsaline environments (Sengupta et al., 2014), and carcinogenic inorganic pollutants (Dwivedi et al., 2018). Many of the Ganges tributaries are also greatly imperiled by human activities. As an example, the Gandak River basin has seen metal concentrations increase as part of the sugar cane industry (Bhardwaj et al., 2010) while heavily urban streams such as the Yamuna are replete with industrial estates including paper mills and tanneries (Lokhande & Tare, 2021) which cause severe river contamination through drainage channels and outflow. Environmental concerns are not limited to the surface water of the Ganges either. Since surface waters have been extensively used in industrialization efforts, many impoverished and rural people have sought groundwater through tube wells as a drinking source partially as a result of a World Health Organization (WHO) and the United Nations International Children's Emergency Fund (UNICEF) recommendations from the middle of the 20th Century (Chakraborti et al., 2018). This water, as it were, is more likely to be contaminated with arsenic than traditional drinking water sources such as rainwater and dug wells which are once again being promoted for human consumption (Acharyya et al., 1999;Chakraborti et al., 2018;Rahman et al., 2009). Though arsenic pollution is generally a natural phenomenon, its plaguing of the population throughout the Ganges Basin can be tied to water quality issues and a lack of developed water infrastructure at the surface (Chakraborti et al., 2017). Given all of these noted challenges, it is not at all surprising that the Ganges pollution problems have been a topic of many academic reviews over the past few decades.
Traditional reviews of the Ganges have focused either broadly on pollution (see, Roy & Shamim, 2020;Sinha & Loganathan, 2015) or on particular pollutants such as plastic pollutions (see Roman et al., 2022) or heavy metal pollutions (see Paul, 2017;Sankhla et al., 2018). Other authors (i.e., Schiff, 2014) have overviewed some of the policy and community-based actions of regional and national actors to regulate and improve water quality on the Ganges and to rectify the political mechanisms and histories at play in the area. What is missing in the literature is a clear and authoritative bridge documenting how scientific studies about Ganges water pollution and quality can be used to inform policy decisions; this is a noted area for improvement across the environmental sciences (Pullin & Knight, 2012), but it is especially important in the Ganges which is a large river basin that supports significant populations that need better access to clean waters.
In this review paper, we ask the following questions: (A) What recommended policies and actions can be implemented to improve water quality and contamination in the Ganges River basin? (B) How can the development of policy-related items be improved in the region? To tackle these questions, a systematic review methodology is utilized under the Reporting standards for Systematic Evidence Syntheses (ROSES) framework. This review is unique as its main focus is on documenting and critically examining policy interventions that have been proposed to improve the Ganges pollution problem. Traditional systematic reviews normally focus on the effects of a treatment or intervention, asking "what works" types of questions (Snilstveit et al., 2012). In this paper though, the inquiry that we pose is more accurately described as a "how can science inform policy" question. Such a question then necessitates a broadening of the traditional systematic review to include qualitative literature as well as quantitative literature. The role of systematic reviews to be used in informing policy has been documented (Bilotta et al., 2014) but it is not often applied in practice. In this review, we aim to combine all potential policy recommendations given by researchers to form a more comprehensive resource for scholars, citizens, and practitioners who are involved in vital Ganges cleanup efforts while also analyzing the provided evidence.

| MATERIALS AND METHODS
The literature search for this review follows the standard ROSES checklist items (Haddaway et al., 2018). ROSES provides an alternative model to the commonly used Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Moher et al., 2009), and PRISMA 2020 (Page et al., 2021) guidance which among many problems (Haddaway et al., 2018) is heavily focused on meta-analysis at the expense of narrative synthesis. In this paper, ROSES is used to systematically assess the pollution mitigation recommendations of various authors working within the Ganges River Basin in an attempt to narratively and statistically summarize their suggested action items so that these items may be used to inform policy, practice, and future research within the tripartite countries that form the Ganges Watershed.

| Geographic focus of review
River which flows through much of Central India, and the predominantly rural Ghaghara River. In total, the Ganges Basin covers an area of 1,087,300 km 2 which makes it the 16th largest river basin in the world (Salahuddin & Nomani, 2021).
The Ganges Basin is the economic driver of Northern India. Agriculture is well developed and productive in the basin with 22.41 million hectares of irrigated land and 44.99 million hectares of sown area (Rasul, 2015). It also has a high potential as a hydroelectric resource, especially in the mountainous reaches of the basin (Rasul, 2015). Along the Ganges, the largest industry by the magnitude of facilities is the tannery industry, but the industry with the greatest wastewater generation is the pulp and paper industry, while the chemical industry consumes the most water resources besides these two (CPCB, 2013). These industries and land uses support a population of over 430,000,000 people, of which 22% is urban-based (CPCB, 2013).

| Searching strategy
The search conducted for this systematic review was performed using Scopus. Compared to other citation databases such as Web of Science and Google Scholar, Scopus is among the most consistent when it comes to content quality and verification (Adriaanse & Rensleigh, 2013). Scopus was searched for English language articles about Ganges pollution and policy using the search string documented in Figure 1. To be specific, the search terms "Ganges" or "Ganga" were used to isolate articles about the Ganges Basin, terms such as "water quality," "contamination," and "pollution" were used to further narrow the search to articles about pollution problems, and lastly policy-related terms such as 'mitigation' or "recommendation" were used to filter the articles even further. The adopted strategy was used to search article titles, abstracts, and keywords. Articles, review articles, book chapters, and conference papers were all searched for policy recommendations over the past decade from 2013 until the date the search was conducted on July 21, 2022.
As a means to validate the comprehensiveness of the literature search, three widely cited articles about the Ganges River Basin's pollution problem were pre-selected as benchmarks. The articles we chose are roughly evenly distributed within the search time range and cover three important pollution topics: (1) groundwater arsenic contamination, (2) organic surface water pollution, and (3) inorganic surface water pollution. While selecting these benchmark studies, we aimed to select research that has been consistently cited with approximately five citations on average per year, with continued documented citations since their publication date. The articles we chose are: "Groundwater arsenic contamination in Ganga-Meghna-Brahmaputra plain, its health effects and an approach for mitigation" (Chakraborti et al., 2013), "Use of Principal Component Analysis for parameter selection for development of a novel Water Quality Index: A case study of river Ganga India" (Tripathi & Singal, 2019), and "Policy Framework for Mitigating Land-based Marine Plastic Pollution in the Gangetic Delta Region of Bay of Bengal-A review" (Raha et al., 2021). Since all three of these articles were located during the Scopus search, we believe that it is reasonable to conclude that the use of Scopus was operative for conducting this systematic investigation of pollution.

| Article screening, inclusion, and exclusion
The complete search procedure, combined with various exclusion steps is documented in a ROSES-compliant flow diagram (Figure 1). In the initial screening stage, article titles, abstracts, and keywords were screened to ensure that each research item had relevance to pollution in the Ganges through the inclusion of search terms in the pieces. One hundred of 117 articles met this most basic requirement. Full article texts were then gathered, and articles that were unavailable at the authors' institutions were sought through interlibrary loan. In the end, only a single article was unretrievable and had to be excluded.
The remaining 99 articles left over after the screening and retrieval phases were individually read by the authors. To be included in the final review, the articles had to make at least one clear and specific policy recommendation for improving the pollution or water quality issues of surface water or groundwater in the Ganges Basin. Specificity was highly important for inclusion and exclusion. The authors worked hand-inhand with one another to ensure that any policy recommendations offered were specific to the Ganges Basin, actionable, broad in scope, and based on evidence. Articles that simply stated vague future recommendations, encouraged the use of their models or methods, or suggested nonspecific regular monitoring were excluded. Research that was found to be only tangentially related to improving pollution and water quality in the river basin on subjects such as freshwater dolphin conservation, or fishery development were also excluded. Historical articles documenting past policy and pollution problems, which made no currently actionable recommendations were also excluded along with articles that narrowly focused on a single sub-watershed within the Ganges and offered no policy recommendations that could be broadly applicable to the whole basin.
The authors independently read the articles as a consistency check. Research with borderline inclusion criteria were debated when disagreement arose over whether a specific policy recommendation was present or not and a consensus was reached to include or exclude them from the review. All articles were sorted into a spreadsheet and the excluded articles were placed in Supporting Information: Table S1.

| Critical appraisal and data extraction
As a policy-centric review, comprising both empirical and nonempirical research, critical appraisal was conducted as a means of determining whether the policy recommendations made by authors were based off of well-documented research and review procedures. Specifically, research was appraised on whether it documented quality control standards in modeling, laboratory procedures, and field procedures while review articles were appraised on whether they documented their search processes in a replicable manner. For lab and field work, we determined whether the authors documented procedures to keep a sterile lab environment and whether the authors listed a replicable protocol or methods that could be employed by others, and for modeling work, we documented whether an accuracy assessment or diagnostic was completed. We also checked to ensure if review articles followed a systematic or scoping/mapping protocol as a critical appraisal method. Where quality control procedures were documented to be F I G U R E 1 ROSES-compliant flow diagram documenting exclusion and inclusion process. ROSES, Reporting standards for Systematic Evidence Syntheses. absent, we noted this. The critical appraisals of all included articles can be found in Supporting Information: Table S2. Our approach is slightly different from other systematic reviews where a critical risk of bias may result in an article being altogether removed from the synthesis section of the review, and where each individual article is ranked for bias on an ordinal scale. Since this review is heavily focused on policy rather than interventions, we kept our critical appraisal process straightforward and descriptive in nature based on documentation of procedures rather than on a complete diagnostic of article methodologies and data. A detailed critical appraisal of individual methods would have also been impractical due to the wide range of techniques employed by authors to study pollution in the Ganges.
After screening and inclusion steps were undertaken, 37 articles were included in the final review. As the main object of interest in this review, policy recommendations were extracted and summarized in tabular form for the narrative synthesis portion of the results. A second author validated the inclusion of these recommendations and suggested additions, or subtractions based on their own individual readings of the research. Other data items that provide context to these recommendations were also extracted. The sub-watershed/geographic focus and methodological approach of each article was extracted from the methods or introductory sections of the papers, the overall topical focus was categorized based on discussions between each of the authors in this review, and bias was captured as reported in the previous paragraph. For field and lab studies, a data item called "sources of pollution" was used to differentiate between the main objective focus of each article. Since this paper asks a "how can science inform policy" question, our data items reflect this choice and are quite different from what might be expected in a review that asks a "what works" question. In this way, our paper also offers new perspectives on the Ganges pollution problem through its line of inquiry that deviates from interventional-focused reviews.

| Data synthesis and presentation
The final presentation of this systematic review is presented as a narrative synthesis. This narrative synthesis has two parts: a descriptive analysis and a content analysis.
In the descriptive analysis, we consider more than just the 37 articles that were kept after our article screening, inclusion, and exclusion processes. To provide a snapshot of the institutions and actors involved in studying the pollution problem in the Ganges River Basin, we opted to also include the 62 other articles on pollution that made it through the title and abstract screening stage but were excluded in the other steps. In the content analysis portion of the synthesis, we document descriptive information about the 37 reviewed articles and seek to find meaningful commonalities between each research article that may be relevant to understanding how policy recommendations on Ganges pollution and water quality are formed. Further, individual recommendations are discussed based on their topical contributions and on the documentation of their methodological underpinnings. All findings are combined to create a holistic picture of Ganges pollution policy to specifically address our two main research questions.
As none of the authors in this review article have any works featured in this synthesis, we did not appraise our own research in the course of this review. We also ensured that we provided no favoritism in our synthesis towards any party or approach.

| Descriptive analysis
A total of 37 articles made it through the screening and exclusion stages of this review out of an original 99 that were read for policy recommendations. In this section, we consider both groups of articles. Only slightly more than one-third of the total articles that we read had specific and actionable policy recommendations within their content. In every year, articles with strong policy recommendations were in a minority compared to articles that discussed Ganges pollution without making policy recommendations ( Figure 2). In these results, what becomes evident is that there is a gap between the production of knowledge of Ganges pollution and the ability of this science to act as a catalyst to promote change, in tandem with a plethora of other social, cultural, and administrative aspects. Just as is the case in broader climate and environmental science (Kirchhoff et al., 2013), science can only be widely usable when it engages with its audience and provides functional utility beyond complex modeling outputs and experimental insights. When researchers overcome the "entrenched institutional roadblocks" of knowledge production to increase interaction with stakeholders who use the science, it results in improvements in the uptake of scientific understanding (Kirchhoff et al., 2013). In the case of this review, the articles that were excluded from the final content analysis failed to make any actionable or relevant policy recommendations, and while some make vague statements about "future research" or encourage nebulous monitoring programs, they lack a bridge between science and application.
Another gap that exists in Ganges pollution research pertains to the way that information is created and disseminated about the topic. In environmental sustainability research, there have traditionally been two schools of knowledge creation: the natural science school, and the social science school (Kaufmann & Cleveland, 1995). The ability of these approaches to synergistically work in concert with one another to study issues such as the Ganges pollution problem would likely lead to interdisciplinary knowledge sharing and generation. In this review, however, we found that most research on Ganges pollution including both included and excluded articles come from a narrow subset of publications which may not be broad in scope. The journals, Environmental Monitoring and Assessment and Environmental Science and Pollution Research were tied as the most frequently used outlets for disseminating information about Ganges pollution, with each journal contributing six articles. These journals, though occasionally offering social science perspectives, are primarily geared toward the natural sciences. In fact, of the top 15 publication outlets chosen by authors writing about Ganges pollution, the vast majority are natural science-focused environmental publications. Only a few publications that are common venues for disseminating Ganges pollution research, such as Environment, Development and Sustainability which provide three pollutionrelated articles, could fairly be said to be interdisciplinary in their aims and scope. Publications from traditional human-environment bridge disciplines, such as Geography, are also underrepresented in this review. From a policy and planning perspective, this may be an indication that recommendations for improving Ganges water quality are primarily being generated within a philosophical echo chamber centered on empirical natural science, with lesser contributions from empirical and critical social science disciplines.
When only the 37 articles that met final inclusion criteria were considered, there is a relative parity between the methodological approaches that were used to create policy findings (Figure 3). Review articles and lab/field studies each contributed 14 articles to this paper, while modeling-centric research contributed nine articles. The majority of review articles were broad in geographic scope and examined the entire river system, whereas modeling and lab/field studies were mostly conducted within a sub-watershed of the Ganges as depicted in Figure 3. The methodological approaches can also be examined at finer scales, which can yield insights about the reaches where scholars more frequently conducted their work at. Geographically, stretches of river near industrial cities such as Kanpur, Varanasi, and Allahabad were popular locations for conducting research, as were locations nearer to the origin of the Ganges such as Haridwar. Many of these locations also have spiritual significance in Hinduism and are the location of rituals and water festivals.

| Field and lab studies
A chief policy official from the United States Department of Health and Human Services once opined that "good policy results from a combination of three inputs": good research, good data, and human stories (Olson & Bogard, 2014). It is in fieldwork and lab studies that this statement holds its greatest truth. Field research is the backbone of science, and researchers conducting studies in this area are generating new data, engaging communities, and introducing new research ideas to the realm of policymakers and planners. In the Ganges Basin, we identified 15 articles that had lab or fieldwork at the core of their methodologies and that used their hands-on research to suggest policy improvements to improve pollution conditions (see Table 1). Grouping these articles by the source of pollution that each one studied, the majority of researchers examined nonspecific point source pollution predominantly through water sampling and F I G U R E 2 Chart documenting the production of pollution knowledge in the Ganges since 2013. SIGDEL ET AL. statistical techniques while some researchers (e.g., Madhulekha & Arya, 2016) provided focus on a specific type of waste.
A commonly offered policy solution for point source pollution, particularly from industrial sources, is that wastewater effluent needs to be treated or curtailed before it is allowed to re-enter the watershed (A. Kumar et al., 2020;Madhulekha & Arya, 2016;Roy et al., 2014). Some authors have taken this simple policy recommendation a step further. Srinivas et al. (2018) suggest that local industries should install online effluent monitoring systems and engage in water recycling and organic chemical treatment while Bowes et al. (2020) are of the impression that wastewater treatment standards should be enhanced to meet the quality controls of developed industrial economies while also considering local seasonal variations in water quality. Besides industrial wastes, human and animal wastes are also a subject of policy recommendation from field scientists. Mass gatherings such as the Kumbh Mela, as well as other Hindu religious traditions involving "sacred waste" such as idols and floral arrangements have been the focus of multiple policy recommendations.   et al. (2015) detail a regime of hygiene education, drinking water monitoring, and crowd prediction monitoring to reduce the impact of the Mela on the river system, but they do not suggest a change in the material culture of religious practice. On the other hand, Sharma and Behera (2022) suggest that mass gatherings and cattle bathing should be restricted in the river. Between these two suggestions, Chakrabarty (2020) recommends a material culture change where idols and religious items are made with sustainable materials to limit pollution inputs. Another policy option for mass bathing is to designate certain locations for religious activities so that filtration techniques can be used in those designated areas along with enhanced education to teach swimmers how to prevent pollution (Dimri et al., 2021) while state-actors may want to track plastic pollution using global positioning system (GPS) units to see where it is most problematic (Duncan et al., 2020). Some authors engaged in fieldwork (e.g., Trombadore et al., 2020) suggest that beyond infrastructural and material changes, there also needs to be deeper political discourse in the basin that engages public stakeholders and coordinates and guides public water use rather than solely dictating its management through regulations and restrictions. These suggestions are rare, however. For surface water policies, most authors suggest state-led initiatives rather than community-led programs. However, for groundwater studies, the recommended policy options suggested by field researchers are notably more focused on community-led efforts. A simple effort that villagers in arsenic-impacted areas can undertake to limit their exposure to the toxic element is to gather water from traditional water sources such as lakes, ponds, rainwater catchments, and shallow dug wells instead of deep tube wells (Chakraborti et al., 2013). Other modest changes that can be made in communities include the cultivation of arsenic-resistant rice varieties (Bhattacharya et al., 2015), water sieving, and diet changes (Chakraborti et al., 2013). Even with arsenic, many large-scale changes such as the formulation of tube well regulations need to take place at a state policy level (Chakraborti et al., 2013). Infrastructural policy suggestions related to dam construction and maintenance (e.g., Srivastava et al., 2020) are also the domain of state-level planners. This view of planning is perhaps slightly slanted, as it ignores the role that intermediary nongovernmental groups may have in policy formation. Panigrahi and Pattnaik (2021) are one of the few field research groups to recommend the inclusion of intermediary groups such as students and ecologists in Ganges pollution planning efforts, but the majority of their recommendations also begin and end with the state.

| Modelling studies
In direct comparison to field studies, the modeling studies that were found in this review are much broader in their scope and include both quantitative and qualitative T A B L E 1 (Continued) Author (

Sharma and
Behera (2022) Haridwar to Diamond Harbour Multiple sources (nonspecific) Macro-invertebrate collection and multivariate statistical analysis Human activities in the Ganges such as mass bathing and cattle wading should be curtailed to support the river habitat and help pollution-sensitive organisms.
approaches. In terms of policy recommendations many of the proposals made by authors share an overlap with those from field studies (see Table 2). For remediating source pollution from tanneries and other industries, multiple research discussions (e.g., Srinivas & Singh, 2018b;Srinivas, Singh, Jain, et al., 2020;Srinivas, Singh, Shankar, et al., 2020) suggest a need for systematic monitoring and effluent treatment in the Ganges watershed. These suggestions approximately parrot those made by field investigators, but there are also new recommendations that are distinct to modeling approaches. Batabyal and Beladi (2019) suggest, for instance, that a strategic cost-benefit analysis approach could be used to prioritize the removal of certain harmful pollutants over others in the basin under a regime of scarce resources. This interesting proposal acknowledges that practical decision making may have to trump pragmatic notions of mass-scale pollution remediation while funding and support is limited in the basin. Singh and Gundimeda (2021) suggest a technocentric approach for limiting industrial pollution where the best available technologies for improving the local environment are adopted while also considering technical and economic limitations. While Singh and Gundimeda (2021) see the state as having a role in regulating the technological transition of river-polluting industries, they also advocate a need for market-based policy instruments in the basin to incentivize change. In these recommendations there is an acknowledgment that policies are more likely to be successful if they engage and reward polluting industries for gradually implementing pollution control measures instead of prognosticating sudden and drastic policy changes which are costly and lack the political capital to implement. From a health standpoint, Moors et al. (2013) promote a need for better medical and sewage infrastructure to combat climate-change-induced increases in diarrhea within the Ganges basin. For contending with arsenic, Chakraborty et al. (2019) recommend better trainings for tube well drillers and better selection strategies for choosing new well sites. There is nothing particularly trailblazing about either of these suggestions, but these recommendations do add some additional weight to the accumulated mass of research taking place in these areas and help to validate some of the policy items being enacted by state planners in India and Bangladesh.
One other area where modeling-based recommendations vary from those prognosticated by field researchers is that there is a slightly higher incidence of studies that examine and suggest policy items for remediating nonpoint source pollution. George et al. (2021) discuss how a policy of geoengineering and vegetation planting could help to reduce erosional pollution in the Ganges basin especially in hilly areas. Srinivas and Singh (2018a) also acknowledge the role of limiting erosion, especially during new infrastructure projects.

T A B L E 2 Recommendations from modelling studies.
Author (

| Review studies
A useful review paper does not merely summarize the literature body, but it adds value to it by generating new conceptualizations and linkages, offering outlines of future research goals, and producing policy and planning recommendations which can be applied in practice (De Vos & El-Geneidy, 2022). Within the Ganges Basin, there are no shortage of academic reviews on a number of topics and quite a many of these also offer pollution remediation recommendations based on syntheses of the literature (see Table 3). It should be noted, however, that many of these recommendations come from narrative reviews which lack documented steps for replicability and do not offer clear inclusion and exclusion criteria. This is a concern which may need to be considered by local authorities and stakeholders before any of these policies are implemented or funded within the river basin. Many reviewers offer "big picture" governance suggestions for the basin in a manner that is distinct from the more localized suggestions offered by field researchers. The notion that there is a need for increased central government control, oversight, and management of pollution along the river is offered by multiple review papers (e.g., Chaturvedi, 2019;Wang et al., 2016) and it has been suggested that coordination and planning needs to be synched between all of the countries in the Ganges Basin for mitigation to be successful (Islam et al., 2017). Other authors (e.g., Srinivas, Singh, Shankar, et al., 2020) recognize that policy change and education has to originate from all levels of governance. Among the most important policy items that the central government may have to tackle is the management of environmental flows (e-flows) especially during dry periods (Mariya et al., 2019;Srinivas, Singh, Shankar, et al., 2020). Kaushal et al. (2019) are of the opinion that improved e-flows may be difficult to implement in the basin, and they will likely require incentivization for farmers and a change in agricultural practices towards micro-irrigation techniques to be successful. Another option is for farmers to select crops using adaptive management strategies more carefully or by using indigenous or organic varieties of crops or organic fertilizers which would require less water and produce less chemical runoff (Ramesh et al., 2019;Srinivas, Singh, Shankar, et al., 2020). Agricultural changes alone may not be enough to restore and support the environment. R. Kumar and Ambastha (2018) recommend that more efforts need to be made by governments to understand that there is an "ecological continuum" between deltaic wetlands and upstream ecosystems and to program their investments with this in mind. Restoring riparian vegetation along the banks of the Ganges is one way of potentially restoring the pristine ecological nature of the river (Mariya et al., 2019). Natural ponds located within the watershed may also need to be comprehensively evaluated for biodiversity and their exchange with other systems to improve environmental conditions within the basin (Moorhouse et al., 2021).
Pollution issues going on at the surface of the Ganges may also impact groundwater (Mariya et al., 2019), but as was the case in the previously discussed fieldwork and modeling sections of this paper, most groundwater pollution research from reviews are oriented around arsenic pollution. These recommendations normally involve behavioral changes among groundwater users and better regulation and policy implementation from the government. Chakraborti et al. (2016) recommend that tube wells should be replaced with shallower dug wells and also combined with alternative sources of water such as rainwater capture. They follow this recommendation up in Chakraborti et al. (2018) by further suggesting that tube wells should be painted red or green to distinguish whether arsenic is or is not present. Das (2014) also discusses the need for alternative water sources such as rainwater harvesting, but their broad recommendations are made from a surface water perspective. The need for education and awareness campaigns are suggested by Das (2014), Khursheed et al. (2016), and Raha et al. (2021). Raha et al. (2021) in particular suggests some novel and easily implemented ideas for combating plastic pollution including building new billboards that discourage plastic use, instituting family-led initiatives to increase education about the detrimental nature of plastic and creating environmental protection education courses at local colleges and schools. Most of these recommendations, however, are still state-centric in their approach and may need to be broadened to be effective.

| DISCUSSION
The results of this policy-centric review can be encapsulated by three broad themes that generally hold true regardless of any specific pollutant: (1) most policy recommendations made by researchers studying pollution and degradation in the Ganges basin involve the state and are heavily reliant upon its enforcement and regulatory mechanisms for their implementation, (2) different types of studies (e.g., reviews, field studies, and modeling) generally produce similar policy recommendations with some divergence in the scale of their suggestions and implementation strategies, and (3) the bridge between human spiritual and physical health and the Ganges is well documented and most policy recommendations focus on improving the overall wellbeing of the average citizen as their main objective function.
The question moving forward then is where policyoriented research on Ganges pollution should go from here. This question likely has many possible interpretations, but based on this review, our suggestions for researchers are as follows: 1. There is a greater need for cross-collaboration between physical science and social science disciplines to build policies that can truly address both the environmental and social dimensions of the Ganges pollution problem. 2. While many functions of pollution remediation belong to the state, there are many community-centered solutions that do not require government interventions. Alternative frameworks need to be considered that work from a bottom-up approach rather than a top-down polity. 3. Water systems should not only be considered locally but as an interconnected structure. When a change takes place at the surface, it may also impact groundwater and vice versa. This also holds true between surface systems, and policy-centric research should consider the social, economic, and environmental implications that a change to one part of the watershed may have on the other parts. 4. Procedures for generating policy recommendations should be well documented, especially for recommendations that come from review articles. This is an essential step for replicability and for credibility.
To follow up on these recommendations in greater detail, the need for interdisciplinary work is best encapsulated by a statement made by the ecologist Garret Hardin who stated that, "It is easy to call for interdisciplinary synthesis, but will anyone respond?" (Hardin, 1998). It is relatively undemanding for physical scientists to document the presence of harmful pollutants in a natural system like a river and to suggest methods for their remediation, but it is ultimately the everyday citizenry and civic leaders who have to implement these recommendations. Having research then that not only considers how to educate the populace, but also how to appeal to them on an emotional and cognitive level is essential and this is where disciplines in the social sciences such as psychology, sociology, and geography excel. Social science perspectives should not merely be paid "lip service" in interdisciplinary work (Heberlein, 1988) but they should be treated as integral to the science and the dissemination of its findings. This same general tenant holds true for the nature of policy implementation. Everyday people living near rivers have their own beliefs, values, and opinions on the state of the environment in their communities (Vasudeva, 2016). Good policy is not only implemented by the state but also originates from community-led initiatives where citizens work hand in hand with researchers to take charge of their own futures. Researchers working in India and elsewhere have documented many effective river pollution remediation and planning techniques that can take place at smaller scales including the use of mycofiltration to remove organic pollutants from water (Davis et al., 2022), the institution of community-led resettlement efforts for disaster-prone waterways (Cronin & Guthrie, 2011), participatory action research programs to help improve awareness about water pollution (Peplow & Augustine, 2012), and communitybased water monitoring designed to involve citizens and spur advocacy (Metzger & Lendvay, 2006). Much of the arsenic-related research documented in this review (e.g., Chakraborti et al., 2016) are also predicated on community-engaged actions such as the reinstitution of traditional water filtration methods and a return to dug wells, but similar community actions have yet to find their way into mainstream discourse. The groundwater and surface water research documented in this review appears to be conducted separately from each other even though there is a bidirectional exchange between the two systems (Lu et al., 2022). Researchers need to contemplate how new policy recommendations such as pumping drinking water from shallower aquifers that interact more with the surface and less with geologic layers containing arsenic may limit exposure to the former pollutant but expose citizens to new pollutants that are generally unique to surface waters such as coliforms and microplastics. As multiple papers recorded in this review have also noted, there are clear connections between upstream and downstream sections of the Ganges. Creating recommendations that do not solely consider a single reach of the river but also its neighborhood is going to be an essential part of policy-centric research in the future, especially as collaborations increase between intraregional and interregional bodies.
It is also the case that many of the recommendations generated by authors reviewing Ganges pollution come from narrative-style reviews that lack clear search procedures and inclusion criteria (see Supporting Information: Table S2). Although there are frameworks for assessing the validity of narrative reviews such as the Scale for the Assessment of Narrative Review Articles (SANRA) (Baethge et al., 2019) and the International Narrative Systematic Assessment (INSA) (La Torre et al., 2015) we found no evidence that narrative reviews of the Ganges were edited and reviewed with these methods or other similar quality control procedures. We do not adhere to the notion that systematic or scoping review procedures are in any way superior to narrative review procedures and we recognize that each type of review plays a function in research (Greenhalgh et al., 2018) but for policy formation there should be a clearer set of evidence underpinning each suggestion made by researchers. If there is not clear evidence to back up a policy recommendation, then it may cast doubt on the suggested policy item and discourage action on them by policymakers. A simple solution to this problem regardless of review type is for researchers to engage with wellestablished review frameworks and provide documentary evidence in the form of protocols, procedures, and checklists.
This review has helped to identify a number of areas for improved research practice in relation to the Ganges pollution problem, but we offer only a first-level analysis of this topic. Further limited systematic reviews of a single pollution type (e.g., arsenic) would help to provide stakeholders with a closer analysis of how pollution data is methodologically analyzed and synthesized. To keep this review succinct and relevant to policymakers we also did not include grey literature in this search process which may be a limiting factor of our results. Stakeholders who are interested in pollution policy may want to consider these sources as an enhancement to the articles identified in this systematic review. As a first-order study, however, we believe that this review offers interesting findings that may help to ameliorate the pollution issues in the Ganges and inform new policies.

| CONCLUSION
Governments and policymakers in developing countries such as those in the Ganges Basin consistently set and then extend their deadlines for providing clean drinking water to their populations even though most of them have the capabilities to implement such systems relatively quickly (Biswas, 2022). This policy-centric review has documented that there are many actions that governments can take throughout the Ganges Basin to improve water quality and reduce pollution both at the surface and under the ground. Specific pollutants may require specific policy actions, but there are also broad suggestions given by researchers that may help to broadly reduce the harmful effects of pollution on the environment and the people living therewithin. Education, awareness, enhancing sewage treatment using economic incentivization and government regulation, and cleaning up religious practice are all suggested items that can be worked on by policymakers almost immediately. This work also acknowledges, however, that these steps are likely only a starting point for the real work that needs to be done in the basin across all levels of planning, and in the future, it will be vital that policymakers also consider actions that can be taken at the community-level based on sound interdisciplinary science that considers not only local concerns but also those of the wider basin.

ACKNOWLEDGMENTS
We would like to thank the associate editor and the reviewer for their valuable comments and suggestions which helped to improve the quality of this manuscript. This study did not receive any funding from any organizations or funding agencies.

DATA AVAILABILITY STATEMENT
Data will be available upon request, and are also partially available through our attached supplemental materials.

ETHICS STATEMENT
None declared.