Water scarcity

People suffer from “chronic water shortage” if they have access to less than 1,000 m3 fresh water per person per year (31). With this definition in mind, the impact of climate change on water scarcity has been assessed on a global scale (30).

At present (reference year 2000), 1.6 - 2.4 billion people (25% - 39% of global population) are estimated to be living in watersheds exposed to water scarcity. The greatest proportions of populations living in water-scarce watersheds are located in East Asia (660 million) and South Asia (491 million).

Population growth will increase pressures on future water resources (30): it is estimated that by 2050 3.1 - 4.3 billion people (37% - 53% of global population) will be living in watersheds exposed to water scarcity, globally. This is solely due to population growth, the effect of climate change is not included. Again, the greatest absolute exposure to water scarcity is in South Asia (1.5 - 1.7 billion) and East Asia (0.7 - 1.2 billion).  

Climate change will probably be an additional pressure on future water resources. The projected additional increase in the number of people that suffer from “chronic water shortage” depends on the scenario of climate change that is used in the model calculations, and the models that are used. Under a moderate scenario of climate change (the so-called A1B scenario), results of 21 global climate models indicate that by 2050 0.5 to 3.9 billion people are exposed to an increase in water scarcity solely due to climate change, when compared with the reference period of 1961 – 1990. On the other hand, water scarcity will decrease for 0.1 - 2.7 billion people, according to these projections. The large width of this range is largely due to uncertainty in the estimates for South Asia and East Asia. For other scenarios of climate change the results are quite similar. In fact, the differences in projections of water scarcity across four climate change scenarios that more or less span the bandwidth of estimated climate change (the B1, B2, A1B and A2 scenarios) are relatively small when compared with the differences across the results of the 21 models (30).

Fresh water resources: observations at the European scale

Climate change is affecting Europe's renewable freshwater resources

Runoff of rivers and streams in in the Mediterranean region shows drying trends over the last half century (1956–2005). For Northern Europe, on the other hand, weak wetting trends have been observed. These trends show that on the continental-scale Europe’s renewable freshwater resources are changing (53). These changes may be due to several contributing factors: human management including land cover change (54), large-scale irrigation (55,56) or the construction of reservoirs and dams (55,57), and changes in the hydrological cycle. A study has shown that the latter, observed changes in precipitation and evapotranspiration, largely explain the observed continental changes in runoff. What’s more, these changes can be related to anthropogenic climate change (53).

There is a high probability (≥90%) that anthropogenic climate change is amplifying the north-south contrast of runoff at the continental scale in Europe: state-of-the-art climate model simulations suggest that it is highly probable that the observed pattern is captured only if the models are forced with human emissions (53). Most remarkable are the increasingly dry conditions in southern Europe, whereas the change in the north is less pronounced. In central Europe, the transition region from a decrease to an increase, there is little change.

These results highlight the issue of increased water scarcity in the Mediterranean region under climate change. The authors, therefore, stress the need of raising awareness of the possible consequences of anthropogenic climate change for regional water resources in the already water-scarce south of Europe (53).

Climate change is not always the main driver of river discharge changes 

Climate change and climate variability affect the discharge of river basins. One would expect river discharge in, for instance, southern Europe to decrease in direct response to less precipitation, more evaporation from soils, streams, reservoirs, and lakes, and more transpiration by plants. The relationship between changes in climate and changes in river discharge is not so straightforward, however. Several other developments in a river’s catchment, including human water use, irrigated agriculture, and land use change can obscure or even reverse the impact of climate change on river discharge. Human factors are sometimes more important than climate in understanding changes in river discharge over time.

For the Atlantic part of Europe, from the south of Spain to the north of the UK, a study was carried out to disentangle the contributions of climate change versus human impacts on river discharge. Long‐term trends in annual river discharge from a very dense network of 1874 gauging stations spanning the period 1961-2012 were analysed for this (71).

Discharge has increased in the North, and decreased in the South: Across much of the British and Irish Isles, trends of increasing river discharge dominate. In contrast, most stations across the Iberian Peninsula and southern France show decreasing trends. Both the increasing trends in the North and the decreasing trends in the South are statistically significant. Northern France and central and southern Great Britain are transitional, with few significant trends. These trends agree with trends in annual precipitation totals that also show a north‐south gradient, characterized by significant increasing trends in the north of the British and Irish Isles, and predominantly significant decreasing trends in northern parts of the Iberian Peninsula and southern France. The spatial coherence between changes in river discharge and evaporation is much less (71).

Link with climate change in the North, mismatch in the South: For the British and Irish Isles most observed trends in annual river discharge are associated with climate change. In Spain, however, there is a clear mismatch between observed trends in river discharge and climate. Especially for the river basins in the south of Spain, river discharge has decreased more than expected, given annual precipitation and evaporation. This may be due to a large surface cover of irrigated crops: irrigation may have declined groundwater resources, producing feedbacks that finally contribute to the strong observed decline of river discharge (71). Also, depopulation has led to the abandonment of agriculture and grazing lands in these regions (72), with a subsequent and rapid natural afforestation (73) that increased water consumption by vegetation and reduced runoff to the river (74).

In southern France, trends in river discharge seem to be mostly driven by climate. Compared with the Iberian Peninsula, vegetation cover has not increased much here since land abandonment and natural afforestation have been modest (75). Also, crop irrigated areas represent a small and stationary percentage of basin land cover in southern France (76).

Effect wildfires: increase river discharge: In the north and west of Spain and Portugal the pattern is opposite to the one in the south: river discharge has increased more than expected from climate alone. This region has been frequently impacted by wildfires in past decades. River discharge can increase after fire events when a substantial part of the river basin is burned (77).

Fresh water resources: future projections at the global scale

Snow droughts

Snow drought, the lack of sufficient snow accumulation in the mountains in the winter to feed the rivers in spring and summer, is an emerging global threat to water resources, according to climate model projections under a moderate (SSP2-4.5) and high-end (SSP5-8.5) scenario of climate change. In major snowy basins around the world, both the frequency and intensity of snow droughts are projected to increase (79). 

Impacts on global water scarcity (drought risk) at 1.5 and 2 degrees global warming

A study has been carried out on changes in water scarcity (drought risk) at the global scale at 1.5 °C relative to 2 °C global warming. Water scarcity was defined as the magnitude of fresh water availability (= precipitation minus actual evapotranspiration) that is exceeded 80% of the time (the so-called 20th percentile Q20). A large number of climate simulations based on five general circulation models was used for this. No future population growth was taken into account. Future projections were compared with the historical period 2006-2015 (67).

The model results indicate that water scarcity (Q20) would intensify globally when global warming approaches the 2 °C (instead of 1.5 °C) above the preindustrial levels. In other words: the magnitude of droughts would be less severe in most parts of the world should we pursue climate change mitigation efforts to hold temperature increase at 1.5 °C instead of 2 °C above the preindustrial levels. A common definition of the threshold of fresh water availabilitythat indicates water scarcity in a region is an amount of 1,700 m3 per person per year. For this threshold, the number of people adversely affected by water shortage in +1.5 and +2 °C warmer world would increase by 3.7% (271 million) and 5.3% (388 million) of global population relative to the historical period, respectively. From a water scarcity point of view, the benefit of holding global warming at 1.5 °C instead of 2 °C above the preindustrial levels is apparent in East Asia, South Asia, Central Europe, West Africa, and East Africa (67).

Fresh water resources: future projections at the European scale

Impacts on European hydrology at 1.5, 2 and 3 degrees global warming

In a warmer world, the hydrological impacts of climate change are more intense. Heavy rainfall becomes more heavy, highest river flows further increase and lowest flows decrease. In addition, these changes affect wider areas. Impacts of climate change at 1.5, 2 and 3 °C mean global warming above preindustrial level have been assessed and compared for precipitation, snowpack, runoff, and discharge in Europe. This was done by using the results of a number of climate models as input for hydrological models. The hydrological impacts in a warmer Europe were compared with the situation in the period 1971-2000 as a reference (37):

• Precipitation. Generally, changes in precipitation and evapotranspiration are higher for higher levels of global warming. Where precipitation is projected to decrease with increased warming, these decreases become larger, where precipitation is projected to increase, it increases more. Regionally, there are robust increases in total annual precipitation in most of central, western and northern Europe, for all levels of warming. The decreases in precipitation projected around the Iberian coast become larger and more widespread with increasing warming. Changes in precipitation are negligible or uncertain in the rest of southern Europe and in the UK. As expected, the snowpack (annual maximum of snow water equivalent) decreases in most parts of Europe.
• Evapotranspiration. At 3 °C warming mean annual evapotranspiration is clearly higher and changes are more robust than at 1.5 °C in most of Europe except the south. Evapotranspiration is projected to decrease on the Iberian Peninsula because there is simply not enough moisture available to increase actual evapotranspiration due to the decreases in precipitation.
• Runoff. Mean annual runoff is indicative of available water resources for agriculture, water supply, navigation, etc. Mean annual maximum runoff is indicative of recurring high flows, and mean annual low runoff is indicative of 
dry conditions. 
In parts of Europe where runoff is affected by climate change, there is a distinct increase in the changes to mean, low and high runoff at 2 °C compared to 1.5 °C. Between 2 and 3 °C, the changes in low and high runoff levels continue to increase, but the changes to mean runoff are less clear. For all levels of warming, the changes to runoff are strongest in winter with large increases in runoff seen in Scandinavia and the Alpine regions. Robust increases in runoff affect the Scandinavian mountains at 1.5 °C, and extend over most of Norway, Sweden and northern Poland at 3 °C. Decreases in mean annual runoff are seen only in Portugal at 1.5 °C warming, but at 3 °C warming, decreases to runoff are seen around the entire Iberian coast, the Balkan Coast and parts of the French coast.
• Discharge Europe’s largest rivers. As a result of the increase in mean runoff for large parts of northern Europe, annual mean discharge of rivers in this area increase as warming level increases. In a similar way, decreases in river discharge with increasing warming level are consistent with runoff decreases in southern Europe. In other parts of Europe changes in mean annual discharge are less clear. This was illustrated for the rivers Rhine and Danube, and points at contrasting changes in winter to summer: in winter, discharges increase due to higher winter precipitation and earlier snowmelt, while summer discharge decreases due to lower snowmelt runoff from the Alps and increased evapotranspiration.  

Groundwater

Future climate change will affect recharge rates and, in turn, the depth of groundwater levels and the amount of available groundwater (10). Groundwater recharge depends on the distribution, amount and timing of precipitation, evapotranspiration losses, snow cover thickness and snow melt characteristics, and land use/land cover. The properties of the aquifer are also essential; small, shallow unconfined aquifers respond more rapidly to climate change, whereas larger and confined systems show a slower response (9). It is expected that in snow-dominated regions, warmer winters will cause snow melt and groundwater recharge (11) and runoff to occur over longer periods and earlier in the year (12). Increased aquifer recharge will increase wintertime groundwater levels (13), whereas in spring and summer the groundwater levels may decrease with a warmer climate (14).

Southern Europe will have less recharge overall and the region may become more water stressed than at present, with any increase in winter recharge unable to compensate for the reduced autumn recharge. Southern Spain is predicted to be among the worst affected regions in Europe, with almost total disappearance of recharge (15).

Changes in groundwater recharge patterns may increase the risk of leaching of contaminants (pesticides for instance) during winter (16). Reduced groundwater level increases the risk of contamination mainly from sea water intrusion in coastal aquifers (17). Besides, changes in river flow affect connected aquifers: reduced minimum flow, for instance, can lead to higher riverine concentration in wastewater effluents as waters are less diluted posing a risk to groundwater, with possible consequences for aquifers. Also, the effect of climate change on air temperature and river temperatures may influence groundwater temperatures (18) and dissolved oxygen concentrations (19). Because many biogeochemical processes in groundwater are temperature dependent, climate-induced changes that affect groundwater temperature may negatively affect the quality of groundwater (20).

Low flows

Although models project that climate change will cause a decrease in low flows in north-western European rivers over the coming decades, it should be noted that most models have low accuracy in the simulation of low flow extremes (36,68).

Southern Europe

Owing to climate change (enhanced evapotranspiration and reduced rainfall) alone, fresh water availability in the Mediterraneanis likely to decrease substantially (by 2-15% for 2°C of warming), among the largest decreases in the world (59), with significant increases in the length of meteorological dry spells (60) and droughts (61). River flow will generally be reduced, and water levels in lakes and reservoirs will probably decline. The seasonality of stream flows is very likely to change, with earlier declines of high flows from snow melt in spring, intensification of low flows in summer and greater and more irregular discharge in winter (62). In Greece and Turkey per-capita water availability may fall below 1,000 m3 per year (the threshold generally accepted for severe water stress) for the first time in 2030 (63). The critically low current water availabilities per capita in southeastern Spain may drop to below 500 m3 per year in the future.

The general increase in water scarcity as a consequence of climate change is enhanced by the increasing demand for irrigated agriculture to stabilize production and to maintain food security (64). Irrigation demands in the Mediterranean region are projected to increase between 4 and 18% by the end of the century due to climate change alone (for 2°C and 5°C warming, respectively). Population growth, and increased demand, may escalate these numbers (65) to between 22 and 74%. Water demand for manufacturing is also projected to increase between 50 and 100% until the 2050s in the Balkans and Southern France (66).  

Vulnerabilities - water temperature

Observed warming lakes

Lake summer surface water temperatures are warming significantly, with a mean trend of 0.34°C per decade, across 235 globally distributed lakes. This was concluded from a dataset over the period 1985 – 2009 (22). This warming rate is consistent with the rapid annual average increase in air temperatures (0.25°C per decade) and ocean surface temperatures (0.12°C per decade) over a similar time period (1979–2012) (23). Warm-water and cool-water lakes showed similar ranges of warming rates. In Northern Europe, lakes were warming significantly faster than the global average.

Similar results were found for annually averaged lake surface temperature of 20 Central European lakes during the past 50 to 100 years; a statistically significant increase was observed since the early 1960s while the majority of this warming has occurred in the last three decades, i.e. since the 1980s (52). In these lakes, spring was found to be the most rapidly warming season (52). 

Key drivers of surface water temperature include absorbed solar irradiance and heat exchange with the atmosphere, which is controlled by air temperature, solar radiation, humidity, ice cover, and wind (24), but is also mediated by local factors such as lake surface area and depth (25). For the observed lake warming, summer air temperature is the single most important and consistent predictor (22).

Consequences of this extensive warming are numerous and diverse. The global average lake summer surface water warming rate found here implies a 20% increase in algal blooms and a 5% increase in toxic blooms over the next century (26), as well as a 4% increase in methane emissions from lakes during the next decade. Increased evaporation associated with warming can lead to declines in lake water level, with implications for water security (27), substantial economic consequences (28), and in some cases, complete ecosystem loss (29). The observed widespread warming suggests that large changes in Earth’s freshwater resources and their processes are not only imminent but already under way (22,58).

Deep subalpine lakes in Europe are experiencing a decrease in the frequency of winter full turnover and an intensification of stability. As a result, hypolimnetic oxygen concentrations are decreasing and nutrients are accumulating in bottom water, with effects on the whole ecosystem functioning (58).  A reversion in the increasing thermal stability would be possible only if global GHG emissions started to be reduced by about 2020, allowing an equilibrium mixing regime to be restored by the end of the twenty-first century. Otherwise, persistent lack of complete mixing, severe water warming and extensive effects on water quality are to be expected for the centuries to come (58). 

Observed consequences warming lakes on algae blooms

Blooms of phytoplankton in freshwater lakes harm aquatic food production, recreation and tourism, and drinking-water supplies. In the United States alone, this leads to economic losses of more than US $4 billion annually (70). Two main drivers of these blooms are nutrient loads and lake warming due to climate change. Long-term trends in intense summertime near-surface phytoplankton blooms have been studied for 71 large lakes globally, all with surface areas of more than 100 km2. Landsat 5 satellite imagery was studied for this, covering the period from 1984 to 2012. The results reveal a global exacerbation of phytoplankton blooms. Peak summertime bloom intensity has increased in 68% of these lakes; the increase is statistically significant for close to a third of all lakes. A significant decrease was found for only 8% of these lakes (69). The authors do not have a clear explanation of the observed changes. There is no consistent link with water temperature, trends in fertilizer-use, or other relevant drivers. They did find, however, that lakes with a decrease in bloom intensity warmed less compared to other lakes. The latter suggests that lake warming may already be counteracting management efforts to reduce phytoplankton blooms by reducing nutrient loads (69).

Warmer water may lead to more or less algae blooms. This depends on local characteristics of water systems. This was concluded from a global study of 188 large lakes (39). Estimates of lake surface temperature and lake surface chlorophyll-a concentration (as a proxy for phytoplankton biomass) were derived from satellite observations. This was done for 188 of the world’s largest lakes over the period 2002-2016. 62% of the lakes showed positive correlations between chlorophyll-a concentration and lake surface temperature, of which 74% were statistically significant. Correlations were negative for 38% of the lakes, of which 68% were statistically significant (39).

Lakes with relatively low chlorophyll-a concentration were found to have more negative correlations between chlorophyll-a concentration and lake surface temperature whereas the opposite was found for lakes with relatively high chlorophyll-a concentration. Thus, lake warming tended to amplify lake-to-lake variability in phytoplankton biomass whereby phytoplankton-poor lakes were poorer in warm years and phytoplankton-rich lakes were richer in warm years (39).

• Why algae-poor lakes get poorer in warm years. According to the authors of this study it would make sense that lake warming leads to less algae, at first glance. After all, warming increases phytoplankton’s demand for resources to support higher metabolic rates at higher temperature (40). These resources will become scarcer relative to their demand causing a metabolic deficit at higher temperatures, thus leave lakes capable of supporting less phytoplankton biomass (41).
• Why algae-rich lakes get richer in warm years. The opposite may be the case, however, for a number of reasons. In phytoplankton-rich lakes, lake warming strongly favours phytoplankton species such as cyanobacteria. Grazers less efficiently consume these species (42) and phytoplankton biomass may accumulate in warm years as a result (43). Also, phytoplankton-rich lakes tend to have strong populations of fishes that feed on zooplankton and their consumption rates increase with lake warming. As a result there is less zooplankton to “graze” on phytoplankton (44). A third effect is thermal stratification of lakes due to surface warming, thus trapping nutrients below the photic zone where they are unavailable to surface phytoplankton (45). This effect is most pronounced in phytoplankton-poor lakes because internal nutrient loading via vertical mixing is often the primary source of nutrients to phytoplankton there (46). In contrast, external nutrient inputs and their subsequent recycling tend to dominate nutrient budgets in phytoplankton-rich lakes (47). In these lakes, external nutrient inputs may increase with warming driven by climate-mediated shift in rainfall and land use (48).

Changes in phytoplankton biomass in some large lakes have been shown to have substantial effects on species, economies, and livelihoods (45,49). Higher phytoplankton biomass in phytoplankton-rich lakes may exacerbate problems associated with anthropogenic lake nutrient enrichment, such as the expansion of anoxic zones, harmful algal blooms, fish die-offs, and reduced water clarity. Managers may need to reduce anthropogenic nutrient loads that were acceptable in the past to maintain ecosystem functions in phytoplankton-rich lakes as they warm (50). In contrast, the reduction of phytoplankton biomass in phytoplankton-poor lakes with warming presents its own potential management challenges such as reduced fisheries productivity (45,51). In some cases, managers may want to fertilize lakes to sustain fish production (51).

Projected warming rivers

Projections of daily river discharge and water temperature under future climate have been made, based on a global physically based hydrological-water temperature modelling framework forced with the output of three global climate models (GCM’s) for the SRES A2 and B1 emissions scenarios. Impacts of anthropogenic heat effluents from thermoelectric power plants on water temperature were incorporated (1).

The results show an increase in the seasonality of river discharge (both increase in high flow and decrease in low flow) for about one-third of the global land surface area for 2071-2100 relative to 1971-2000. Global mean river water temperatures are projected to increase on average by 0.8-1.6°C for 2071-2100 relative to 1971-2000; projected high (95th percentile) river water temperature increase is 1.0-2.2°C. The largest water temperature increases are projected for the United States, Europe, eastern China, and parts of southern Africa and Australia (1).

For the high northern latitude, results show an increase in mean annual river discharge and high flows (the 95th percentile of the daily distribution (Q95)); for southern and central Europe, a decrease in mean annual river discharge and high flows is projected.In Europe (except northern part) a reduction of low flows (10th percentile of the daily distribution (Q10)) is projected. River discharge seasonality is projected to increase (1).

The basin average increases in mean water temperature are 2.1°C for the Danube, 1.9°C for the Rhine, and 2.1°C for the Rhone. Projected increase of high water temperatures (95th percentile in daily distribution; Tw95) is 2.6-2.8°C; mostly due to atmospheric warming, partly due to declines in summer low flow (for the Danube, for instance: 74% due to atmospheric warming, 26% increase  due to declines in summer low flow) (1).

For the Rhine, the number of days with water temperatures above 25 °C, which is the threshold for significant stress to river fauna and flora, would increase at least five-fold (e.g. from 2-15 days 
in the reference period 2001-2010 to 32-75 days in 2071-2100) (38). 

Projected impact on water resources

Worldwide, climate change is projected to reduce raw water quality, posing risks to drinking water quality even with conventional treatment (according to IPCC: high agreement, medium evidence). The sources of the risks are increased temperature, increases in sediment, nutrient and pollutant loadings due to heavy rainfall, reduced dilution of pollutants during droughts, and disruption of treatment facilities during floods (21).

Regions characterized by a combination of large reductions in (low) river flows and increases in water temperatures under future climate, such as southern and central Europe, could potentially experience increased deterioration of water quality, freshwater habitats and reduced potentials for human water use in the future compared to the current situation (2).

Declining river flows decrease their dilution capacity, resulting in increased concentrations of effluents from point sources (2). In addition, rising water temperatures decrease oxygen solubility and concentrations (3) and increase the toxicity of pollutants (e.g. heavy metals and organophosphates) to fish and other freshwater species (4). Freshwater organisms might also experience increased stress due to lower summer flows that decrease available habitats (5) and the exceedances of critical water temperature thresholds (6).

Besides, thermoelectric power production may be reduced (7) and concentrations of microbiological pollutants may increase. This could result in increased cost of water treatment to produce potable drinking water (8).

Adaptation strategies

Forest management

Forested headwater basins are an important source of water for human use and ecosystem function. It is important, therefore, to maintain streamflow from these headwaters in light of expected climate change.

Forest and land management may alter freshwater availability and streamflow through changes in vegetation type, stand age, and associated evapotranspiration rates (34). Forest harvest can provide short-term increases in streamflow, while species conversion can either increase or decrease streamflow for long periods, depending on the structure and water use demands of the new forest (35).

Long-term data for watersheds in southeastern U.S. shows that management activities that alter species composition and structure can have lasting effects on streamflow, especially at the lowest and highest flows (32). We need to understand how forest management or disturbance counteract or exacerbate the climate effects on streamflow in order to safeguard the ecosystem service of water supply for society (33).

Forest management can mitigate the effects of climate change. Climate and forest management interact and affect streamflow differentially across the flow regime, where increasing or decreasing forest cover, altering dominant species, or converting deciduous to conifer forests can enhance or lessen the effects of changes in precipitation patterns on low and high streamflows (32).

Offshore fresh groundwater reserves

There are potentially large offshore fresh groundwater volumes of up to 10% (about 1 million km3) of total fresh terrestrial groundwater deposited in unconsolidated sediment systems in many regions worldwide (78). Most of these groundwater reserves have been reported to lie within 55 km distance from the current coastline. They reach depths of up to several kilometers. The offshore fresh groundwater volumes could provide multiple centuries (even up to two millennia) worth of fresh groundwater under current and under future water demand rates. Exploring these offshore resources could provide additional fresh water sources for agricultural, industrial or domestic use, and prove to be a vital water stress mitigating factor especially in South-east and East Asia, West Africa and several regions in South America. However, salinization of inland aquifers might occur when extracting these offshore fresh groundwater volumes and this process should be further studied before extracting these water volumes.  

References

The references below are cited in full in a separate map 'References'. Please click here if you are looking for the full references for Europe.

1. Van Vliet et al. (2013)
2. Caruso (2002); Moore et al. (1997); Van Vliet and Zwolsman (2008), all in: Van Vliet et al. (2013)
3. Kundzewicz and Krysanova (2010); Murdoch et al. (2000), both in: Van Vliet et al. (2013)
4. Ficke et al. (2007), in: Van Vliet et al. (2013)
5. Isaak et al. (2011); Sabo and Post (2008), both in: Van Vliet et al. (2013)
6. Eaton and Scheller (1996); Mantua et al. (2010); Mohseni et al. (2003), all in: Van Vliet et al. (2013)
7. Forster and Lilliestam (2011); Koch and Vögele (2009); Van Vliet et al. (2012b), all in: Van Vliet et al. (2013)
8. Delpla et al. (2009); Schindler (2001); WHO (2011), all in: Van Vliet et al. (2013)
9. Kløve et al. (2014)
10. Ludwig and Moench (2009), in: Kløve et al. (2014)
11. Jyrkama and Sykes (2007); Sutinen et al. (2007), both in: Kløve et al. (2014)
12. Veijalainen et al. (2010), in: Kløve et al. (2014)
13. Mäkinen et al. (2008); Okkonen and Kløve (2010), both in: Kløve et al. (2014)
14. Okkonen and Kløve (2010), in: Kløve et al. (2014)
15. Hiscock et al. (2012), in: Kløve et al. (2014)
16. Okkonen et al. (2010), in: Kløve et al. (2014)
17. Werner et al. (2013), in: Kløve et al. (2014)
18. Figura et al. (2011); Kløve et al. (2012); Haldorsen et al. (2012), all in: Kløve et al. (2014)
19. Taylor and Stefan (2009), in: Kløve et al. (2014)
20. Figura et al. (2011), in: Kløve et al. (2014)
21. IPCC (2014)
22. O’Reilly, C.M. et al. (2015)
23. Hartmann et al. (2013a), in: O’Reilly, C.M. et al. (2015)
24. Edinger et al. (1968), in: O’Reilly, C.M. et al. (2015)
25. Schmid et al. (2014), in: O’Reilly, C.M. et al. (2015)
26. Brookes and Carey (2011); Rigosi et al. (2015), both in: O’Reilly, C.M. et al. (2015)
27. Vorosmarty (2000); Hanrahan et al. (2010), both in: O’Reilly, C.M. et al. (2015)
28. Gronewold and Stow (2014), in: O’Reilly, C.M. et al. (2015)
29. Smol and Douglas (2007), in: O’Reilly, C.M. et al. (2015)
30. Gosling and Arnell (2016)
31. Rockström et al. (2009), in: Gosling and Arnell (2016)
32. Kelly et al. (2016)
33. Jones (2011), in: Kelly et al. (2016)
34. National Research Council (2008); McGuire and Likens (2011); Jones et al. (2012), all in: Kelly et al. (2016)
35. Bosch and Hewlett (1982); Buttle (2011), both in: Kelly et al. (2016)
36. Gudmundsson et al. (2012b), in: Willems and Lloyd-Hughes (2016)
37. Donnelly et al. (2017)
38. ICPR (2014), in: European Environment Agency (2017)
39. Kraemer et al. (2017)
40. Allen et al. (2005), in: Kraemer et al. (2017)
41. Yvon-Durocher et al. (2011); Daufresne et al. (2009), both in: Kraemer et al. (2017)
42. Jeppesen and Jensen (2000); Rigosi et al. (2014); Kosten et al. (2012), all in: Kraemer et al. (2017)
43. Jeppesen et al. (2009), in: Kraemer et al. (2017)
44. Persson et al. (1992); Jeppesen et al. (2003); Hansson et al. (2012), all in: Kraemer et al. (2017)
45. Cohen et al. (2016), in: Kraemer et al. (2017)
46. Smith et al. (1999); Langenberg et al. (2003); Sarmento et al. (2013), all in: Kraemer et al. (2017)
47. Jeppesen et al. (2009); Smith et al. (1999); Burger et al. (2008), all in: Kraemer et al. (2017)
48. Jeppesen et al. (2009); Domis et al. (2013), both in: Kraemer et al. (2017)
49. Allan et al. (2015), in: Kraemer et al. (2017)
50. Scheffer et al. (2015); Urrutia-Cordero et al. (2016), both in: Kraemer et al. (2017)
51. Stockner et al. (2000), in: Kraemer et al. (2017)
52. Woolway et al. (2017)
53. Gudmundsson et al. (2017)
54. Sterling et al. (2013), in: Gudmundsson et al. (2017)
55. Jaramillo and Destouni (2015), in: Gudmundsson et al. (2017)
56. Thiery et al. (2017), in: Gudmundsson et al. (2017)
57. Lehner  et al. (2011), in: Gudmundsson et al. (2017)
58. Fenocchi et al. (2018)
59. Jiménez Cisneros et al. 
(2014), in: Cramer et al. (2018)
60. Schleussner et al. (2016); Kovats et al. (2014), both in: Cramer et al. (2018)
61. Tsanis et al. (2011), in: Cramer et al. (2018)
62. García-Ruiz et al. (2011), in: Cramer et al. (2018)
63. Ludwig et al. (2010), in: Cramer et al. (2018)
64. Albiac et al. (2013); Iglesias et al. (2012), both in: Cramer et al. (2018)
65. Fader et al. (2016), in: Cramer et al. (2018)
66. Forzieri et al. (2014), in: Cramer et al. (2018)
67. Liu et al. (2018)
68. de Wit et al. (2007); Smakhtin (2001); Smakhtin et al. (1998); Stahl et al. (2011), all in: Liu et al. (2018)
69. Ho et al. (2019)
70. Kudela et al. (2015), in: Ho et al. (2019)
71. Vicente‐Serrano et al. (2019)
72. Lasanta‐Martínez et al. (2005), in: Vicente‐Serrano et al. (2019)
73. Lasanta and Vicente‐Serrano (2007), in: Vicente‐Serrano et al. (2019)
74. Robinson et al. (2003), in: Vicente‐Serrano et al. (2019)
75. Mottet et al. (2006), in: Vicente‐Serrano et al. (2019)
76. Campardon et al. (2012), in: Vicente‐Serrano et al. (2019)
77. Hallema et al. (2018), in: Vicente‐Serrano et al. (2019)
78. Zamrsky et al. (2022)
79. Cowherd et al. (2023)