Climatization of environmental degradation: a widespread challenge to the integrity of earth science

ABSTRACT

To guide scientists and society regarding the hydrologic consequences of anthropogenic climate change, earth scientists increasingly develop qualitative predictions and quantitative ensembles of models, some of which have important economic or (geo)political implications. However, with unprecedented human population, environmental degradation, and water scarcity, climatic factors are increasingly invoked falsely to explain failures of environmental governance, a phenomenon termed climatization. We propose a first typology of climatization in hydrology. Scientific climatization occurs when – during the normal course of a scientific investigation – a hydrologic state is falsely attributed to climatic factors, often due to a conceptual model that excludes human impacts or a simplified methodology that fails to quantify uncertainty. In contrast, securitization-aligned climatization occurs when a securitizing state actor requires that scientists attribute observed hydrologic states to climatic factors. Maintaining the credibility of earth science requires that earth scientists vigorously contest both scientific climatization and securitization of global change hydrology.

KEYWORDS:

• global change
• climate change
• Anthropocene
• strong inference
• securitization
• hydrological processes

Editor S. Archfield ; Guest editor M. Sivapalan

1 Growing evidence of Earth’s finiteness

As Earth’s human population approaches eight billion people, signs of its limited carrying capacity proliferate (Warren 2015, 2016, Gotmark et al. 2018, Anderson 2019). With respect to water resources, four billion people face severe water scarcity (Mekonnen and Hoekstra 2016). This water scarcity in turn threatens human and consequently national security (El-Sayed and Mansour 2017) as well as peaceable relations amongst nations (Petersen-Perlman et al. 2017). When human or agricultural water requirements clash with those of natural systems, the latter often suffer (Dery and Salomon 1997, Vorosmarty et al. 2010), as demonstrated by increasingly widespread evidence of Tragedy of the Commons (Hardin 1968) in the water resources of Earth’s water-limited regions – manifested by decreasing terrestrial water storage (Müller et al. 2017, Rodell et al. 2018, Wang et al. 2018),¹ river flows (Hillel et al. 2015, 2019, Tal 2017, Best 2019), and lake water storage (Micklin 1988, 2007, Aghakouchak et al. 2015, Moore 2016, Fazel et al. 2017, Wurtsbaugh et al. 2017, Rodell et al. 2018, Morin et al. 2018, Chaudhari et al. 2018, Khazaei et al. 2019, Wine et al. 2019b, Wine 2019e). Similar to water, land has also become increasingly scarce (Lambin and Meyfroidt 2011), which may in part explain the increasing population living in low-lying hazard-prone coastal areas (Grant et al. 2015). Protecting the global environment while providing humanity with needed water and land requirements presents a “monumental challenge” (Goodland 1995).

Despite widespread perceptions to the contrary (Abbott et al. 2019a, 2019b), to meet needs of a growing human population for food and water, Earth systems have been extensively altered (Vörösmarty et al. 1997, 2004, 2013, Vörösmarty and Sahagian 2000, Falkenmark et al. 2003, Wilcox et al. 2011, Rockström et al. 2014). Of the water consumed by humanity, evapotranspiration from agriculture accounts for over 90% (Döll 2009, Siebert et al. 2010, Hoekstra and Mekonnen 2012, Scanlon et al. 2017). While most of humanity’s global water footprint is satisfied by green water (Hoekstra and Wiedmann 2014), increasing rates of blue water consumption by irrigated agriculture have demanded corresponding increases in groundwater pumping and reservoir storage (Graf 1999). Transboundary watersheds present a particularly acute challenge if water requirements increase amongst all riparian states, while renewable water supplies remain limiting (Bakker and Duncan 2017, Wine 2019b, 2019c, 2019e). In affected domains, failure of governance can also serve as an impetus to increase local water consumption, thereby further stressing water resources within already water-limited regions (Dalby and Moussavi 2017).

In coastal regions rising populations are driving urban expansion, even in areas vulnerable to flooding (Hallegatte et al. 2013, Hinkel et al. 2014, Güneralp et al. 2015, Mcgranahan et al. 2016). The most important driver of increasing global exposure to coastal flooding is rising population in low-lying coastal areas (Hanson et al. 2010, Bouwer 2011, Cazenave and Cozannet 2014). Furthermore, coastal areas may be managed to fulfill multiple conflicting objectives; for example, saline water for shrimp farming may threaten freshwater or soil quality (Ali 2006, Ahmed 2011, Chowdhury et al. 2011, Grant et al. 2015). Moreover, upstream diversions of water may increase coastal salinity, thereby further impacting agriculture (Mirza 1998, Gain and Giupponi 2014). Hence, even without considering impacts of climate change, rising coastal populations, upstream flow diversions, and competing interests set the stage for vulnerability to flooding and environmental degradation.

2 Enter climatization

In the Anthropocene Epoch (Lewis and Maslin 2015), governments contend with convergence of resource scarcity, increasing marginal costs to acquire scarce resources,² scientific support of anthropogenic climate change theory, and alignment between climate change projections and securitized issues (Fig. 1). In coastal areas a leading contender to mitigate water scarcity is seawater desalination, though it is expensive compared to exploitation of existing renewable freshwater resources, thereby presenting a disincentive to implement supply-side solutions. At the same time, the scientific community has increasingly supported anthropogenic climate change theory (Manabe and Wetherald 1967, Keeling et al. 1976, Oreskes 2004, Doran and Zimmerman 2009, Anderegg et al. 2010, Gleick et al. 2010, Cook et al. 2013, 2016, Maibach et al. 2014) while investing greatly to predict its hydrologic impacts (Arnell 1999, Held and Soden 2006, Maxwell and Kollet 2008, Kundzewicz et al. 2009b, Brown et al. 2019). In certain cases, impacts of climate change – relative to some control period – projected to transpire in future decades are already observed. For example, in certain parts of the world where water scarcity is predicted to result from climate change, heightened water scarcity is already observed.³ This convergence of factors has led to widespread contentious attribution to climatic factors of lake shrinkage (Madani 2014, Madani et al. 2016, Wurtsbaugh et al. 2017, Wine and Davison 2019, Zeitoun et al. 2019, Wine et al. 2019, Wine 2019a, 2019b, 2019c, 2019e), environmental degradation in Bangladesh (Grant et al. 2015), and the Syrian civil war (De Châtel 2014, Hendrix 2017, Selby et al. 2017a, 2017b), though in the Anthropocene Epoch climate change is not necessarily the most important factor (Grimm et al. 1997, Ferguson and Maxwell 2012, Feitelson and Tubi 2017, Wine et al. 2018, Wijngaard et al. 2018, Momblanch et al. 2019b) affecting hydrological processes.

Figure 1. Human domination of Earth systems presents global land and water managers with numerous challenges, which include phenomena that are inconvenient truths vis-a-vis the local politics, social norms, or perceived national security interests. Where pursuit of sustainable environmental governance has deteriorated, responsibility for environmental degradation may be mistakenly assigned to climatic drivers to fulfill expedient objectives by exploiting any number of vehicles. Several exemplars of climatization have become apparent within different domains and contexts. Each exemplar may implicate a particular subset of climatic processes with numerous consequences from both sociopolitical and scientific perspectives. The credibility of Earth sciences demands that scientists vigorously confront climatization.

False attribution to climatic factors has been termed climatization (Grant et al. 2015). Grant et al. (2015) define climatization as “framing a disastrous event or degraded environmental condition as caused by climate change, in order to reach an intended goal or to distract the discussion from the real problem which might have a different root course than caused by the climate change effects.” According to this definition, climatization is inherently driven by securitization. In the interest of framing a physical hydrology discussion without treating its sociopolitical context, Wine and Davison (2019) proposed an alternative definition of climatization: “overemphasis of climatic factors relative to the demonstrated quantitative impacts of other anthropogenic drivers in eliciting an observed state”. This second definition is broader in that it could encompass climatization that may or may not be driven by securitization. On this basis, we propose a first typology of climatization.

In scientific climatization, climatic factors are overemphasized in certain studies, as part of the normal business of science, in which multiple hypotheses are considered and, in due course, certain hypotheses are falsified, whereas others are supported (Platt 1964). In this paradigm scientists’ only objective is to seek the truth and their professional liberties are uninfluenced by any securitizing state actor in terms of funding, carrying out their study, or disseminating results. In contrast, in securitization-aligned climatization, water resources are deemed by a state actor to be a national security issue – securitized (Zeitoun and Warner 2006, Stetter et al. 2011). As part of this securitization, freedom of expression of scientists domiciled within the state⁴ is limited to a set of ideas perceived by the government to be beneficial (or non-detrimental) to national security, and therefore acceptable – sanctioned discourse (Feitelson 1999, 2002, Wine 2019d) – a central component of which may involve scapegoating climatic factors. The goal of this paper is to synthesize past research into a typology that improves understanding of the causes, characteristics, and consequences of climatization. In contrast to past research that focused solely on physical hydrology (Wine and Davison 2019), here we note the role of water resource securitization in climatization. This explicit acknowledgement of securitization is consistent with growing recognition of its relevance to hydrological process understanding (Goulden et al. 2010, Brooks and Trottier 2010, Yu et al. 2015, Schmeier and Shubber 2018, Farnum 2018, Grech-Madin et al. 2018, Wine 2019e).

3 Scientific versus securitization-aligned climatization

Scientific climatization⁵ may occur when a conceptual model excludes a key process (Wine et al. 2019a, Abbott et al. 2019a, 2019b), when uncertainty is not fully accounted for, or when an interdisciplinary and complex process is interrogated from the perspective of a single discipline. Evidence of scientific climatization has arisen in cases where researchers domiciled in countries that protect freedom of expression (Fig. 2) or in countries where the state has no apparent interest in the outcome of research on a particular topic, implicate climatic factors despite evidence to the contrary. That climatic factors (including climate change) may influence hydrologic states is a valid alternative hypothesis. Even when this hypothesis is subsequently questioned, the end result is often a better understanding of hydrologic science once all perspectives have been presented.

Figure 2. The majority of Earth’s human population resides in states classified by Freedom in the World as either not free or only partially free. Even among certain states categorized as free, water resource securitizations have been reported. Degree of freedom is from Freedom in the World 2018 compiled by Freedom House.

Securitization (Table 1) aligned climatization has been identified to fulfill a number of expedient (geo)political objectives including absolving government of responsibility for environmental degradation (Dalby and Moussavi 2017), influencing water sharing in a transboundary watershed (Wine 2019a), or obtaining funding from those countries chiefly responsible for emissions of greenhouse gases (Grant et al. 2015). In contrast to scientific climatization, which transpires in an environment in which the scientists’ freedom of expression is protected, securitization-aligned climatization may occur in states that encourage or coerce scientists to limit their publications to topics that either align with a securitization or at a minimum do not contest that securitization. These limits on freedom of expression in securitized domains contrast with the expression of a wide range of views in the international scientific literature. While elements within the scientific community have recommended that certain views be censored (Huard 2011), the international scientific community as a whole has not adopted this recommendation.

Table 1. Reported securitizations.

Domain	Securitization
Lake Urmia	According to Dalby and Moussavi (2017),“One of the most interesting features of discussions on Iranian policy about the lake is the suggestion by the Ahmadinejad regime, and specifically by the president himself, that climate change (rather than dam building, wells and agricultural overuse of water) is responsible for the shrinkage of the lake: ‘Based on estimations, Urmia Lake encounters similar conditions every 500 years so that it is predicted low precipitation and drought will govern in this area.’ The Iranian Minister of Energy, Majid Namjoo was more specific in blaming climate for the lake’s fate: ‘Eighty five percent of the problem in Urmia Lake is due to drought. It is out of our control and we can do nothing regarding climate change’.” These stated government positions are consistent with water resource securitization in the Lake Urmia basin, with the aim of decoupling government water management from hydrologic state.
Lake Kinneret	“In the presence of economic motives requiring consumption of water these drivers [competition for water in a water-scarce transboundary watershed] promote securitization manifested as secrecy regarding water management and consumption by the upstream riparian.” (Wine 2019e)
Bangladesh	“Recent examples of climatization related to Cyclone Aila (2009) and saltwater intrusion in Bangladesh. In most cases these disasters were climatized in order to create a sense of urgency in order to push for an increase in financial aid to Bangladesh and to deflect responsibility for inaction that led up to the disaster.” (Grant et al. 2015)
Lake Hulun	Amesheva (2017) notes that “environment imperatives need to become an urgent priority for the Chinese government in order to avert the cascading implications that will arise in terms of social unrest.”
India	Water resources securitization has been reported in India’s transboundary river basins (Hanasz 2014, Johnston and Smakhtin 2014).

4 Exemplars of climatization

4.1 Shrinkage of the Caspian Sea and Great Salt Lake

In inland waters, apparent instances of climatized processes include the shrinkage of the Caspian Sea and Great Salt Lake (Fig. 3). In the case of the Caspian Sea, Chen et al. (2017) describe the role of rising temperatures in the sea’s shrinkage as “dominant”, whereas Van Dijk et al. (2014) implicate inflow variability. Yet, after considering both climatic factors and human impacts, Rodell et al. (2018) implicate the Aral Sea Syndrome: “diversions and direct withdrawals of water from the rivers” in the Caspian Sea’s basin. This choice by Rodell et al. (2018) of a conceptual model that includes both climatic and direct human impacts may be the most important factor in reliable attribution of present-day hydrologic processes. In contrast, Van Dijk et al. (2014) do not consider the possibility that decreasing Caspian Sea stage could be driven by increased basin water consumption. Similarly, Chen et al. (2017)’s attribution of climate change induced increase in evaporation rates did not quantify the relative roles of increasing evaporation versus contemporary evaporative water consumption by irrigated agriculture.

Figure 3. As Earth has warmed, the mean annual water level in the Great Salt Lake has decreased (1880–2018). However, while warming can increase open water evaporation, this inverse relationship largely reflects a spurious correlation. The primary driver of the Great Salt Lake’s shrinkage is increasing human water consumption, with the role of rising temperatures being of lesser importance by an order of magnitude (Wurtsbaugh et al. 2017, Wine et al. 2019a).

Meng (2019) concludes that “climate changes … drive the dynamics” of the Great Salt Lake on the basis of statistical methods and by assuming that inflows are only climatically controlled and are not impacted by consumptive withdrawals in the lake’s basin. In contrast, when both basin water consumption (Wurtsbaugh et al. 2017) and increases in lake evaporation due to rising temperatures (Wine et al. 2019a) are quantitatively estimated, the latter is shown to be small. Wine et al. (2019a) notes that most water resource development in the Great Salt Lake basin occurred before 1900, such that to fully capture the Great Salt Lake’s pre-development condition requires a study period that extends back 150–200 years from the present. Thus, in the case of shrinking lakes, a conceptual model that includes all relevant processes along with physically based quantification of the influence of those processes can prevent inadvertent climatization.

4.2 Shrinkage of Lake Urmia

Lake Urmia’s shrinkage (Fig. 4) has also been climatized. Shadkam et al. (2016a) suggested that Lake Urmia is shrinking principally due to climate change in contrast to other research (Madani 2014, Aghakouchak et al. 2015, Tourian et al. 2015, Taravat et al. 2016, Madani et al. 2016, Alizadeh-Choobari et al. 2016, Hamzekhani et al. 2016, Jalili et al. 2016, Fazel et al. 2017, Ashraf et al. 2017, 2018, Alborzi et al. 2018, Chaudhari et al. 2018, Ghale et al. 2018, Khazaei et al. 2019) pointing to water resource development as the primary driver. In contrast to other instances of climatization in which human impacts were absent from the conceptual model, Shadkam et al. (2016a) correctly aim to quantify the relative impacts of irrigated agriculture and climatic changes. However, the equifinality thesis (Beven 2006, Müller and Levy 2019) has led Beven (1993) to conclude that the “application of distributed hydrological models is more an exercise in prophecy than prediction”. Shadkam et al. (2016a) note that a particular set of Variable Infiltration Capacity model parameters supports the conclusion that primarily climate change is shrinking Lake Urmia. However, they provide no evidence of an attempt to determine the uncertainty interval around this attribution or to “define critical experiments that will allow models as hypotheses to be adequately differentiated” (Beven 2012). Thus, even sophisticated modeling exercises can result in scientific climatization if the uncertainty inherent in hydrological modeling is not considered.

Figure 4. As agricultural water consumption increased in Lake Urmia’s basin the lake shrank from its historical extent due to a persistent negative water balance. Irrigated agricultural areas are from Meier et al. (2018); historical lake extent, Lehner and Döll (2004); watershed area (Lehner and Grill 2013); and current surface water extent from Landsat Normalized Difference Vegetation Index (NDVI) subject to the threshold of NDVI<-0.1.

Prior to 1998 Lake Urmia’s stage remained within its historical range (Alborzi et al. 2018). However, since the 1980s water consumption by irrigated land in its basin nearly tripled in response to a doubling of irrigated area, leading to desiccation of 86% of Lake Urmia’s former area by 2016 (Chaudhari et al. 2018). By the early 2000s, the increase in water consumption in its basin pushed the lake to a tipping point such that it has not since rebounded to its “ecological threshold” (Alborzi et al. 2018). This desiccation was associated with a near quadrupling of the Iranian population since 1960, which increased water stress, along with competition amongst provinces to secure scarce water resources for development (Madani 2014).

Observing that rising temperatures correlated with the decline of inflows to Lake Urmia, Fathian et al. (2014) and Fathian et al. (2016) suggest that rising temperatures drove an increase in basin evapotranspiration. Though they allow that decreasing streamflows may be caused by both rising temperatures and over-exploitation of water resources, the possibility that the importance of these two processes is essentially co-equal is left open. Similarly, Jalili et al. (2012), Fathian (2019), Arkian et al. (2018), Nazeri Tahroudi et al. (2019), and Nourani et al. (2018) acknowledge human impacts or over-exploitation, but focus on the role of climatic conditions. This literature states clearly that over-exploitation is non-zero but does not suggest that remedying over-exploitation alone would restore Lake Urmia.

Another approach resulting in ambiguity regarding the relative roles of climate and human impacts involves quantifying the impacts of different processes but reporting them together. On the basis of system dynamics modeling Hassanzadeh et al. (2012) report “results show that … changes in inflows due to the climate change and overuse of surface water resources is the main factor for 65% of the effect”. Even when authors, using similar data and methods, achieve similar results – detecting significant long-term trends in streamflow near Lake Urmia while significant trends are not detected in the lake’s headwaters – some resist implicating water management activities. For example, Fazel et al. (2017) conclude that “the results showed that irrigation was by far the main driving force for river flow regime changes in the lake basin.” In contrast, Fathian et al. (2014) conclude (indefinitely) that “it can be primarily attributed to climate change and over-exploitation”.

4.3 The feasibility of restoring Lake Urmia

The aforementioned hydrologic model developed by Shadkam et al. (2016a) was forced by a subset of bias-adjusted general circulation model predictions for mid-century (Fig. 5), resulting in the impactful conclusion that “future water management plans are not robust under climate change” (Shadkam et al. 2016b). This pessimistic conclusion exacerbates existing concerns that “there is a serious risk of implementation failure due to contradictory national policy agendas of lake restoration and agricultural development, insufficient budget allocation for realizing the restoration plan, lack of provincial accountability for the spending of resources made available for the implementation of restoration measures, and potential future political instability which may lead to less attention to the restoration process” (Salimi et al. 2019). Taken together with research on the present climate suggesting that restoring environmental flows to Lake Urmia is (technically) feasible (Alborzi et al. 2018), we would conclude that the primary challenge in restoring Lake Urmia is political, not climate change. Similarly, Beven (2011) states “I believe in climate change (and the potential impacts of other catchment changes) but I would suggest that we need better ways of deciding how precautionary to be in planning for the future.” This sentiment is substantiated by questions raised about the credibility of climate projections given that they are poorly correlated with observations in the historical record (Nohara et al. 2006, Douglass et al. 2008, Koutsoyiannis et al. 2009, 2010, Kundzewicz et al. 2009a, 2009b, Anagnostopoulos et al. 2010, Schwartz et al. 2010, Christy et al. 2010, Stockwell 2010, Kundzewicz and Stakhiv 2010, Scafetta 2012, 2013, Legates 2014, Frigg et al. 2015, Pindyck 2017, Connolly et al. 2019, Nissan et al. 2019). Despite understanding that parameter uncertainty can be as large as global climate model (GCM) uncertainty (Wilby 2005), Shadkam et al. (2016b) do not quantify how hydrological model parameter uncertainty propagates with GCM uncertainty. The result is a conclusion that climatizes the nature of the Lake Urmia restoration challenge, while disregarding the admonition of Kundzewicz and Stakhiv (2010) that “climate models are not … ready for ‘prime time’ yet, at least for direct application to water management problems. Much more research needs to be done, and models need to be improved considerably before they can be used effectively for adaptation planning and design.”

Figure 5. Change in precipitation (%) in the future (2010–2099) relative to a baseline period (1971–2000) under RCP8.5 (top left) and RCP2.6 (top right). The red bars represent the projections of the five GCMs selected by Shadkam et al. (2016b) and their ensemble mean, and the blue bars represent the projections of five other GCMs and their ensemble mean. Similar figures are provided for different projection periods (2010–2039, 2040–2069, and 2070–2099) consistent with Shadkam et al. (2016b). The figure highlights that robust projections require consideration of a wide range of possible future climatic conditions.

4.4 The Syrian civil war

Climate change has also been implicated in contributing to the initiation of the recent civil war in Syria (Kelley et al. 2015, 2017, Werrell et al. 2015). This determination has been based in part on assessment that “anthropogenic greenhouse gas and aerosol forcing are key attributable factors for this increased drying” that may have brought about elevated Mediterranean drought frequency (Hoerling et al. 2012). In contrast to the climatic attribution, Selby et al. (2017b) state “there is thus no good evidence to conclude that global climate change-related drought in Syria was a contributory causal factor in the country’s civil war.” Moreover, while acknowledging the drought in Syria, De Châtel (2014) points also to “a host of political, economic and social grievances.” While accepting evidence of a climate change signal in Syria’s pre-conflict drought, Hendrix (2017) observes: “ … the drought that affected Syria also affected neighboring Jordan, Lebanon and Cyprus, yet widespread violence did not occur there.” More recently, Mach et al. (2019) conclude: “climate has affected organized armed conflict within countries. However, other drivers, such as low socioeconomic development and low capabilities of the state, are judged to be substantially more influential, and the mechanisms of climate–conflict linkages remain akey uncertainty.” The complexity and interdisciplinarity of many of today’s critical challenges recalls the recommendation of Wesselink et al. (2014): “ …research should be done by teams of people with arange of backgrounds.” Adhering to such guidance, including on topics with important policy implications, might prevent perceptions of climatization, in favor of interdisciplinary collaboration.

4.5 Lake Kinneret

In the transboundary context, climatization can be used, together with projections of future increases in aridity, to justify reduced water-sharing obligations. Water-sharing agreements involve proportional sharing of renewable water resources. Thus, a reduction in renewable water would reduce the amount of water subject to water-sharing agreements. However, under non-stationarity, the amount of renewable water in transboundary basins is expected to change (Ward 2013, Jafroudi 2018), though how to quantify this change credibly and equitably incorporate non-stationarity into water-sharing agreements remains unresolved (Cooley and Gleick 2011). In the case of Lake Kinneret’s watershed, part of the transboundary Jordan River basin (Fig. 6), both Israel (Wine 2019b, 2019c, 2019e) and Syria (Haddadin 2011) have been implicated in climatizing decreasing flows to downstream riparian states – the Kingdom of Jordan and the Palestinian Authority – potentially allowing the upstream riparians to increase their own water consumption (Shentsis et al. 2019, Wine et al. 2019b). In the case of Israel, which accesses the Mediterranean Sea, climatization reduces its obligation to bear the expense of desalinizing water to meet needs of the downstream riparian states.

Figure 6. In the transboundary Jordan River basin, hydrologic science determines how water availability and quality will change under climatic non-stationarity. However, securitization of these transboundary water resources has manifested when a securitizing state actor enforces data secrecy (Wine 2019e) that obscures the relative roles of endogenous versus exogenous drivers (Wine 2019b). By obfuscating hydrologic process understanding in a transboundary basin, an upstream riparian may increase water consumption or impact water quality while falsely assigning to climatic factors responsibility for water scarcity or degraded water quality (Wine 2019a).

Lake Kinneret has religious significance and is the only regional freshwater lake. Consequently, it has been studied extensively. Historically, it was an exorheic lake fed primarily by the Upper Jordan River and emptying into the Lower Jordan River, which terminates at the Dead Sea. However, as Lake Kinneret’s water balance has become more negative the formerly exorheic lake has effectively become an endorheic lake, with natural outflows to the Lower Jordan River having effectively ceased decades ago. The consequence of this reduced natural flow, as well as decreased pumping rates, has been water quality impairment, such that lake water quality no longer meets US EPA drinking water standards of 250 mg L⁻¹ of chloride. The Israel Jordan Peace Treaty requires that neither country take actions that harm the quality of the water supplied to the other, such that if Israel were substantially responsible for the observed water quality degradation in Lake Kinneret, from whence the Kingdom of Jordan is supplied with water, that would be a violation of the Treaty. Rising populations in the Jordan River basin, a renewed focus on a peace agreement between Israel and the Palestinians, and the deadline for cancelling or renewing a 25-year land access agreement serve as the socio-political backdrop of present heightened tensions over water. Though many studies do not explicitly consider Lake Kinneret’s location in the transboundary Jordan River basin, any hydrological science results from Lake Kinneret are directly relevant to water sharing amongst Israel, the Kingdom of Jordan, and the Palestinian Authority.

Early on, Inbar and Bruins (2004) implicated “human interference, water pumping and flow diversion” in impacting Lake Kinneret. This attribution is similar to conclusions in the neighboring Yarmouk basin (Muller et al. 2016, Shentsis et al. 2019) and consistent with Morin et al. (2018): “From 1964 [when Israel’s National Water Carrier was completed], substantial human intervention in the lake⁶ watershed caused a steep decline of lake level.⁷” However, Givati and Rosenfeld (2007, 2009) suggested that decreasing net inflows into Lake Kinneret were driven primarily by suppression of orographic precipitation. This mechanism was disputed by Alpert et al. (2008) and Halfon et al. (2009) for most of central and northern Israel, excluding the Golan Heights – the most important recharge zone of Lake Kinneret’s securitized transboundary watershed. Numerous studies have claimed that available water in Lake Kinneret’s headwaters is decreasing and attributed this decrease to anthropogenic climate change (Tal 2017, 2018, 2019a, 2019b, 2019c, Givati et al. 2019, Gophen 2019), though in most cases failing to provide a detailed methodology.⁸ Tal (2018) suggests that available water has already decreased by 40% since the mid-1970s, which contrasts with climate change projections suggesting only a possible 17.5% decline by the end of the century (Smiatek and Kunstmann 2016). These deterministic attributions and projections, respectively, also contrast with those that consider variability and include the possibility of little or no impact of climate change on precipitation (Rimmer et al. 2011). Assertions of a 40% decline in available water also disagree with Samuels et al. (2010) who state “changes in streamflow of the Jordan River and its tributaries [as a consequence of climate change] are not significant in the near future.”

Investigations of Lake Kinneret’s water balance have been limited through access restrictions imposed on key datasets, including water consumption (Klein 1998, Avni et al. 2015, Wine et al. 2019b, Wine 2019e). However, recent studies that investigate Lake Kinneret’s shrinkage have found less evidence for the climate change hypothesis and stronger evidence of increasing water consumption associated with rising human population (Wine and Davison 2019, Wine 2019a, 2019b, 2019c, 2019e, Wine et al. 2019b). This conclusion is based on headwater streamflows and long-term precipitation where no significant trends were detected (Wine et al. 2019b). Moreover, inability to reconcile the relative magnitudes of observed (satellite derived) increases in water consumption by irrigated agriculture with streamflow decreases is most consistent with increasing diversion of water (Wine et al. 2019b). Wine (2019a), (2019c), and (2019b) specifically critique certain aspects of the science that assigns Lake Kinneret’s desiccation to climatic factors.

4.6 Bangladesh

Historically, Bangladesh has been impacted by natural disasters and environmental degradation – including cyclones (Islam and Peterson 2009) and saltwater intrusion (Chowdhury 2010). The impacts of the cyclones depend not only on their characteristics but also on the number of people and amount of infrastructure affected. These impacts, in turn, are controlled in part by the efficacy of flood control measures such as embankments. Flooding induced by cyclones, diversion of the Ganges (Mirza 1998), and farming shrimp with saline water (Chowdhury et al. 2011) contribute to saltwater intrusion. As a consequence of climate change these historic phenomena – cyclones and saltwater intrusion – could become more severe, with an increase in storm magnitude and frequency (Mirza 2003) and sea-level rise (Chen and Mueller 2018). However, despite the reality of existing human impacts on Bangladesh, government officials have focused on projected impacts of anthropogenic climate change, aiming to obtain development assistance funding (Grant et al. 2015). Grant et al. (2015) deemed this exclusive focus on climate change projections, even in the presence of known human impacts, climatization.

4.7 Lake Hulun

Studies of regional lake loss explicitly note that agricultural water consumption is a key driver of lake loss in certain regions of China, though without quantifying the extent to which irrigated agriculture contributed to the shrinkage of individual lakes (Liu et al. 2013, Tao et al. 2015). Nonetheless, shrinkage of Lake Hulun, China, has been attributed to decreasing river discharge due to decreasing precipitation, with no reference to any impacts of irrigated agriculture (Cai et al. 2016, Zheng et al. 2016), despite the apparent influence of irrigated agriculture on the basin’s water balance (Fig. 7).

Figure 7. Despite the presence of irrigated agriculture in Lake Hulun’s watershed, its contribution to the lake’s negative water balance has neither been included in conceptual models of the lake’s water balance nor quantified.

4.8 Himalayas

A recent review of global change hydrology literature in the Himalayas by Momblanch et al. (2019a) found that:

[N]one of the reviewed studies include other associated consequences of socio-economic change on water abstraction, reservoir storage capacity, hydropower and irrigation demand, which may be more important determinants of the future water security than climate change. This omission may limit the usefulness of modelling outcomes as it obscures correlations between drivers and impacts. For instance, the impact of human water consumption change triggered by non-climate factors (such as population increase, improvements in standard of living) may be misattributed to temperature increase if models do not account for relevant socio-economic component of global change.

This focus on climate may be driven by the transboundary character of the rivers draining the Himalayas combined with securitization triggered by “fragile political stability between co-riparian countries due to distrust and power asymmetry.” Securitization has led to “data secrecy around water management and consumption that prevents incorporation of these components” in hydrologic modeling. They suggest that “the particular geo-political situation of the region, which can prevent ‘sensitive’ hydro-meteorological data from being shared or even collected due to conflicts related to water resources sharing in the Indus, Ganges, and Brahmaputra transboundary basins, may hinder the identification of the main drivers of hydrological change and, thereby, the design and implementation of adequate adaptation measures.” Advancing knowledge on securitized hydrologic science topics may then require that “transboundary research projects should be promoted by independent international institutions.”

4.9 India

Water has been securitized in India’s transboundary basins (Hanasz 2014, Grech-Madin et al. 2018). “The Indian government considers hydrological data for the Ganges to be a matter of national security and has not released it publicly since 1973” (Johnston and Smakhtin 2014). Similarly, with respect to transboundary basins to which India is a riparian, Thenkabail et al. (2009) note “reluctance on part of the states to furnish irrigated area data in view of their vested interests in sharing of water.” Unavailability of reliable water consumption data “is clearly a problem for calibration and validation of models in a system where the infrastructure is evolving rapidly” (Johnston and Smakhtin 2014) in that it increases the uncertainty in hydrologic modeling. This enhanced uncertainty may then set the stage for erroneous understanding regarding the relative role of climatic versus human drivers in bringing about observed hydrologic states.

5 Mitigating climatization

Increasing recognition that human health depends on planetary health or GeoHealth (Almada et al. 2017) is apropos, for example, relative to Earth’s besieged inland waters given the dependence of humanity on freshwater and fisheries therein, as well as the negative health impacts of lake desiccation (Micklin 1988, 2007, 2016, Brooks et al. 2016). While science has the potential to rationally inform public policy for the betterment of humanity (Kirchner 2017), this relies on science remaining an objective search for the truth and scorning vested interests. In contrast, climatization undermines the credibility of environmental science literature.

What then can be done to mitigate climatization? What can editors and reviewers of hydrological sciences journals do to ensure that climatization does not co-opt environmental science literature to strengthen securitization? Publications that climatize environmental degradation may:

• Choose a conceptual model of the water cycle that inappropriately neglects instances of human domination (Abbott et al. 2019a, 2019b).

• Rely heavily on empirical correlations, falsely assuming that statistical correlations imply causation. (Müller and Levy (2019) and Ferraro et al. (2019) suggest strategies and challenges for mitigating equifinality in coupled human-natural systems.)

• Rely on qualitative lines of reasoning to justify quantitative conclusions.

• Fail to quantify volumes or fluxes attributed to different global change drivers using methods of physical hydrology.

• Address only a single aspect of a multi-disciplinary question.

• Fail to provide the methodology and data required to repeat calculations.

• Rely entirely on a single line of reasoning while excluding numerous lines of reasoning or datasets that would readily falsify the climatic thesis.

• Fail to validate selected models or otherwise demonstrate that the selected model is objectively optimal.

• Fail to quantify model uncertainty and its propagation.

• Rely heavily on consequential datasets with important implications, but that have not been verified or replicated by an independent third party.

By increasing awareness of these characteristics of publications, gatekeepers of the scientific literature may screen out dubious science, including on securitized issues, prior to publication.

6 Causes, characteristics, and consequences of climatization

Scientific climatization occurs when the role of climatic drivers in eliciting an observed state is overstated as part of the normal business of science. This overstatement can occur readily due to the often-high correlation between climatic and non-climatic drivers, growing awareness of anthropogenic climate change, and a persistent propensity for disciplinary investigation of inter-disciplinary problems. Instances of scientific climatization may occur when an idealized conceptual model is adopted (Abbott et al. 2019b), the effects of model uncertainty are neglected, a disciplinary team of scientists studies an interdisciplinary problem, or when incisive attempts to falsify the climatic thesis are overlooked. As a consequence of climatization, hydrologic process understanding is impaired until the misattribution has been corrected. The consequence of this misattribution then is a weakened scientific basis for water management. However, once scientific climatization is corrected, it often results in a more robust scientific literature.

Securitization-aligned climatization bears many similarities to scientific climatization, with the exception of impairment of freedom of expression of affected scientists through encouragement and coercion (Wine 2019d) by a securitizing state actor.⁹ This coercion arises from the great expense involved in solving hydrologic problems – often water scarcity or water quality degradation – in comparison to the substantially lower expense of attributing those problems to climatic factors outside of the government’s control. As a result, instances of securitization-aligned climatization may be characterized by secrecy regarding key hydrologic data (Momblanch et al. 2019a, Wine 2019e), questions of reproducibility (Abbott et al. 2019a), or conclusions by scientists domiciled in securitized states that differ from those of scientists whose freedom of expression is protected (Wine 2019d), even on the basis of similar results. In contrast to scientific climatization, which following correction tends to lead to improved scientific understanding, this is not the intent of securitization-aligned climatization. By persistently funding, encouraging, and coercing a particular view, this perspective tends to persist in the scientific literature. The result of this persistent securitizing state-sponsored climatization is ambiguity regarding the state of the hydrologic science, with all that that implies for water policy.

7 Implications for socio-hydrology

In traditional hydrology, relevant state variables driving hydrologic dynamics include hydraulic head, temperatures, and solute concentration. In socio-hydrology, these traditional state variables remain relevant, but an additional suite of state variables has been acknowledged (Table 2). These variables are considered in the seminal socio-hydrology literature, which has often pointed to case studies in which feedbacks between co-evolving human and hydrologic systems drive the dynamics of both (Sivapalan et al. 2012, Kandasamy et al. 2014, Sivapalan et al. 2014). In such case studies, a typical pattern may emerge in which humans modify the hydrologic system, perceive the modification to be excessively detrimental to the environment, then take action aimed at undoing that component of the environmental exploitation deemed excessive. However, when this paradigm is considered in the context of the present work, a cross-cutting variable critical to socio-hydrology emerges – the degree of water resource securitization. This securitization has the potential to influence the nature of feedbacks between human and water systems, for example, as a consequence of securitization-aligned climatization, which may persistently alter hydrologic process understanding. Securitization is related to suppression of freedom of expression and suppression of freedoms more broadly. This suppression of freedom in turn is known to strongly influence fertility rates (Fig. 8), a factor determining global population, which forces the global hydrologic system through changes in agricultural areas. Hence, in socio-hydrology, consideration of degree of water resource securitization and freedom, where appropriate, may ensure that conceptual and mathematical models attain optimal structure.

Table 2. State variables in socio-hydrology (Sivapalan et al. 2012, 2014, Sivapalan 2015, Di Baldassarre et al. 2015, Loucks 2015, Gober and Wheater 2015, Troy et al. 2015a, 2015b, Tian et al. 2019).

Variable
Total human population
Financial motivations
Environmental quality
Land cover
Famine
Human population distribution
Technology
Armed conflict
Policy
Subsidies
Economy
Fuel cost
Historical context
Market factors
Societal stability
Water governance
Socioeconomics
Norms
Values
(Geo)politics
Vulnerability
Legal framework
Environmental awareness
Culture
Public perceptions
News media content

Figure 8. With increasing freedom progress toward a sustainable human population is achieved (Campbell et al. 2013, Warren 2015, 2016). Degree of freedom is from the 2018 Freedom in the World Index compiled by Freedom House. Fertility Rate is from the World Fact Book, estimated for 2018.

Acknowledgements

Constructive critiques from three named reviewers—Drs. Demetris Koutsoyiannis, Pete Loucks, and Marc Mueller—and one anonymous reviewer are gratefully acknowledged. The support of the journal editors and contributions of several other unnamed colleagues in strengthening this manuscript were very much appreciated.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

We wish to acknowledge funding from the United States–Israel Educational Foundation—the Fulbright Commission in Israel. The lead author was supported by a post-doctoral fellowship from the Kreitman School of Advanced Graduate Studies, Ben Gurion University of the Negev.

Notes

¹ The rate of progression toward Tragedy of the Commons may be governed by the degree of mutual coercion, mutually agreed upon. For example, Müller et al. (2017) suggest that an impending Tragedy of the Commons can be postponed substantially by increasing the distance between groundwater pumping centers.

² This refers to, for example, the relatively large expense of building and operating a large-scale desalination plant as opposed to using local renewable high-quality surface water or groundwater sources.

³ For example, some climate change scenarios predict decreasing renewable water availability in the Jordan River and Lake Urmia basins. It is important to note though that climate change is only one candidate cause of water scarcity with another being increased human water consumption to support growing populations.

⁴ Evidence of securitization has arisen in both authoritarian states as well as those with a democratic system of governance.

⁵ The author currently has a paper in review arguing that one of the author’s own previous works climatized the shrinkage of Lake Kinneret by overestimating the role drought.

⁶ Dead Sea.

⁷ Lake Kinneret is an exorheic lake upstream of the Dead Sea from which water is drawn to Israel’s National Water Carrier (NWC). Hence, negative water balance in the Dead Sea due to the NWC necessarily implies negative water balance in Lake Kinneret as well.

⁸ The absence of a detailed methodology challenges falsification or verification of these results.

⁹ See the Academic Freedom Monitoring Project compiled by Scholars at Risk or the Committee of Concerned Scientists for means by which securitizing state actors seek to influence scholarly and scientific activities.