In the Pamirs of Tajikistan, meeting food needs is an ongoing struggle. One of the challenges to local agricultural production is water scarcity. This contribution presents a case study of a small-scale, self-organized irrigation governance system in a village in the western Pamirs, applying the concept of ‘‘hydrosocial arrangements’’ to explore its physical and social components and the interactions between them. It concludes that the system has proven very robust, largely because of its strong sense of community ownership and its flexibility. Political actors and development practitioners often lack detailed knowledge about such well-functioning local solutions, but external interventions applied without that knowledge risk destroying effective systems that are already in place. It is important to design interventions that are tailored to local needs and based on a comprehensive and well-contextualized understanding of local structures, relations, and values.

The contribution deals with selected questions raised by the Workshop 2.3.A:

Question 2: Which human activities enhance the effects of climate change on water resources in mountain regions? And which human activities moderate them?;

and Workshop 3.3.A

Question 1: How do farmers deal with the new options and challenges?

Question 2: What collective strategies are emerging on different spatial levels?

(Findings from recent research activities)

In the Pamirs of Tajikistan, meeting food needs is an ongoing struggle. One of the challenges to local agricultural production is water scarcity. This contribution presents a case study of a small-scale, self-organized irrigation governance system in a village in the western Pamirs, applying the concept of ‘‘hydrosocial arrangements’’ to explore its physical and social components and the interactions between them. It concludes that the system has proven very robust, largely because of its strong sense of community ownership and its flexibility. Political actors and development practitioners often lack detailed knowledge about such well-functioning local solutions, but external interventions applied without that knowledge risk destroying effective systems that are already in place. It is important to design interventions that are tailored to local needs and based on a comprehensive and well-contextualized understanding of local structures, relations, and values.

Dams and hydropower plant operation affect Alpine rivers and streams since they generate highly variable river stage fluctuations (hydropeaking), threatening the ecosystem integrity. In this work, we propose to model hydropeaking using support vector machines trained with the output of a SWAT (soil water assessment tool) hydrological model, the day of the week and the energy price. The model does not require therefore the knowledge of actual management operation of the reservoirs present in the catchment which is particularly useful as these data are often not available since they are sensitive information for hydropower production companies. We test the model for the Upper Adige river basin, in North-East Italy and we show its performance using the wavelet coherence spectrum computed considering model results and streamflow measurements. The newly proposed model is able to capture the fluctuations in streamflow caused by hydropeaking also when both the energy price and the river discharge display clear non-stationary behaviors.

Dams and hydropower plant operation affect Alpine rivers and streams since they generate highly variable river stage fluctuations (hydropeaking), threatening the ecosystem integrity. In this work, we propose to model hydropeaking using support vector machines trained with the output of a SWAT (soil water assessment tool) hydrological model, the day of the week and the energy price. The model does not require therefore the knowledge of actual management operation of the reservoirs present in the catchment which is particularly useful as these data are often not available since they are sensitive information for hydropower production companies. We test the model for the Upper Adige river basin, in North-East Italy and we show its performance using the wavelet coherence spectrum computed considering model results and streamflow measurements. The newly proposed model is able to capture the fluctuations in streamflow caused by hydropeaking also when both the energy price and the river discharge display clear non-stationary behaviors.

In alpine catchments, fluctuations of river discharge occur at multiple temporal scales due to natural and anthropogenic factors. In the last century, several modifications of the natural flow regime have been observed, caused by a complex interplay between construction of dams for hydropower production, changes in water management policies and climate. In this work, we focus on changes associated to hydropower production. The alteration of natural flow conditions has negative impacts on the freshwater biodiversity and threatens the ecosystem integrity of a catchment. Therefore, understanding how river discharge changed over time in different locations of a catchment is relevant for environmental protection and contributes to achieve sustainable water resources management. For this purpose, we conduct time series analysis for selected gauging stations of the Inn and the Adige catchments, which cover a large part of the central and eastern Alps. We use multiresolution wavelet analysis and continuous wavelet transform to identify both old breakpoints caused by dam and hydropower plant construction as well as more recent breakpoints caused by changes in hydropower plant operation. This work proves that changes in hydropower plant management have a major impact on streamflow variability and should be rigorously considered in hydrological models.

Mountainous basins are the strategic area of both water resources and flood control in Japan. Therefore, the accurate reproduction and reliable prediction of runoff from mountainous basins are significant issue. However, hydrological models still have uncertain parameters on which the accuracy of the model performance depends. In order to develop reliable hydrological model, this study investigate the property of the parameters in the storage–discharge relationship for low flows and floods. In addition, the hydrological analysis is carried out to estimate the runoff variation in the future using GCM data.

We selected four snowy mountainous basins with area from 206 to 331 km², which located in different regions of topographical, geological and climatological conditions. The hydrological model used in this study consists of the infiltration model presented by Diskin and Nazimov (1995) and the storage–discharge relations originally described by Horton (1936). The storage–discharge equation comprises two parameters, the exponent and the constant, which are different in each basin and each flood events.

The hydrological analysis is carried out for the term of 14 years and 15 years by hourly time step for each basin, while changing the values of the two parameters using a double-loop algorithm.

The simulation for future runoff projections is carried out for the period of 2080–2099 with MIROC5, provided based on ISI-MIP, in accordance with the emissions scenario RCP8.5. The adjustment of this data to the basin scale is performed for 0.5-degree resolution surrounding the study basin using the observed current data.

The results show that, a) the two parameters for low flow have inverse relationship, b) the two parameters for floods have the relationship which is represented by exponential function. c) The snowfall is conspicuously decreased and it will be affected to the water resources of the snowy region in Japan.

The essential “water tower” role played by mountains is compromised by climate change and human development. Misalignments in various socio-ecological dimensions threaten adaptive capacity and resilience in mountain water-dependent regions. Interdisciplinary research in Utah’s Wasatch Mountains reveals a complex set of mid-elevation dynamics and stakeholder perspectives complicating water resource planning at local and state levels. Rapid urban development and population growth in the region point to water demand exceeding supply in the near future. Climate change is already influencing snowpack levels, snow water equivalent, and phase changes in mountain precipitation. Winter forest evapotranspiration rates present unexpected water loss with warming air temperatures. Mountain water quality is deteriorating due to up-slope nitrogen deposition as well as mid-elevation grazing, fire, and residential development. Multiple data sources point to diverse and conflicting stakeholder perspectives throughout the region suggesting considerable work to be done to find common ground for water management and planning in this dynamic water tower system. We explore the opportunities and constraints related to a range of adaptation pathways being considered and attempted at local, regional and state government scales, including water reuse, water transfers and pipelines, new reservoirs, water banking, and water conservation promotion.

Peak water describes the hydrological response of glacier fed rivers to climate change, indicating that warming first drives increasing discharge until a glacier melt threshold is surpassed and discharge falls below pre-climate forcing values. Although the physical principles of peak water are well understood and accepted, there remains little empirical work evaluating what peak water means for residents of high mountain communities at the frontlines of glacial change. In response, this study­­––drawing on 160 household interviews, 30 key informant interviews, and 4 focus groups––characterizes and compares lived experiences of peak water in remote communities in the upper Manaslu region of the Nepal Himalaya and the Cordillera Huayhuash region of the Peruvian Andes. It problematizes the peak water model by revealing socially-relevant hydrological changes that are obscured when peak water per se is used to delineate potential vulnerabilities to glacio-hydrological change. To this end, the study highlights unanticipated vulnerabilities and adaptations among communities situated on both the rising and falling limb of the peak water profile, using these findings to argue for a broadened and more critical approach to the study of life in rapidly changing high mountain watersheds. These insights emerge at the intersection of complex social-cultural and glacio-hydrological realities and indicate the importance of situating peak water dynamics in specific socio-ecological contexts. The presentation concludes by discussing locally-identified adaptation options and needs as well as lessons learned for future research.

The new paradigm of ‘socio-hydrology’ calls for the consideration of the entire social-ecological system and its dynamics, thereby also requesting for the use of new modelling approaches. In a joint cooperation of hydroclimatologists and ecologists, we developed the agent-based model (ABM) Aqua.MORE. It is designed for simulation of water fluxes in a coupled water supply/demand system at the catchment scale. The ABM approach represents the relevant real-world entities as individual agents: These are the natural resource flow (‘water agents’) on the one hand, and various socio-economic actors (‘manager agents’ and ‘user agents’) with specific behaviours on the other. Aqua.MORE includes variabilities in water availability as well as socio-economic effects, and their mutual interactions and feedback relations. It therefore provides a framework for comprehensive scenario testing, including the effects of climate driven trends and shifts in natural water supply on the complex water demand and supply dynamics. Aqua.MORE is suitable for interdisciplinary resource studies, addressing a wide range of researchers of different fields (e.g. hydrology, sociology, ecology).

With Aqua.MORE (Agent based Modelling of Resources in Environmental Systems), we present a novel approach to simulate water fluxes in a coupled supply/demand system, including variabilities in water availability as well as socio-economic effects, and their mutual interactions and feedback relations. In the Aqua.MORE agent-based modelling (ABM) approach, water resources and socio-economic actors are represented by individual interacting agents. Aqua.MORE provides a modelling framework, where the main principles of water supply and water abstraction as well as a monitoring system for scarcity situations are formulated. The adaption for varying case studies, however, can be performed by every model user. As Aqua.MORE relies on functionalities embedded in the open source programming platform NetLogo, the creation and characterization of site-specific agents and their interaction patterns is very simple.

To demonstrate the model strengths and application possibilities, we realized two different use cases in the Upper Ötz Valley (Austria) and the Matsch Valley (Italy). These two model regions have been selected as they mostly differ in both natural water availability and in their main water demand sectors.

Recent climatic changes alter the hydrological cycle in many ways. Particularly regions dominated by snow and ice undergo substantial changes reshaping the character of their entire hydrological regime. This study aims at a better understanding of 1) the long-term impact of changes in snow cover on river runoff with implications for water availability and seasonality of floods in nival and pluvio-nival flow regimes and 2) interactions between changes induced by modified snow covers with changes caused by anthropogenic river regulations and changes in precipitation. We analyze hydro-climatological observational data in daily resolution recorded in Central Europe and the Alpine region dating back into the 19th century. Focus is on the alpine part of the Rhine river basin. Highly resolved non-linear trends are determined combining quantile sampling with moving average trend statistics and empirical mode decomposition. Preliminary results hint at a strong decrease in seasonality of river runoff in snow-dominated river basins. We attribute this redistribution of annual flow from summer to winter mainly to changes in snow cover and the constructions of dams and reservoirs for high-head storage hydropower. We hypothesize that this alpine signal propagates downstream influencing complex pluvio-nival runoff regimes and overlaps with changes in evaporation and precipitation. Investigations of changes in precipitation and the frequency and characteristics of weather types hint at a recent intensification of the hydrological cycle.

In the Andes of Peru, runoff from glaciers plays an important role in the water cycle, accounting for up to 50% of the water supply during the dry season. However, climate change and the associated strong glacier shrinkage are expected to reduce the water supply, in particular on the long term. Additionally, increasing demand from the agriculture, domestic and hydropower sectors are exacerbating the water supply-demand problem. Usually, modelling the future of water supply-demand system is based on simple rules-of-thumb on how changes in population size affect water demand. While the water availability depends on hydrological processes, water usage depends more on social, economic, political and cultural processes with complex interactions behind decision-making. Some studies even propose that socioeconomic drivers could be more important than changes in the physical processes.

In this context, we build upon socio-hydrology concepts and combine hydrological and social science modelling in our interdisciplinary research project. On the one side, we are assessing water availability and demand through different hydrological models with different levels of complexity to identify the uncertainties arising in mountain regions due to scarce data. On the other side, we use the sustainable livelihood approach to identify how the water users configure their capitals (human, physical, economic, natural and social) for decision-making on water resources management. Together, we can then identify water supply-demand scenarios that account for both hydrological and socioeconomic factors. We are focusing also on the integration of multipurpose projects (one project for different water usages) which requires full understanding of water supply, demand and decision-making.

In the Santa River basin (SRB, Peru), glacial ice and lakes represent a crucial source for Andean livelihoods and cultural values. However, shrinking glaciers and growing lakes as well as increasing water demand and changes in socio-ecological systems are altering spatiotemporal water availability and exacerbating human vulnerability with pronounced hydrological risks particularly in the dry season. In this context, we analyze hydrological and socioeconomic changes of the small-scale Quillcay catchment (QC) in the headwaters of the SRB compared to the entire basin. While the QC is characterized by traditional small-scale agriculture and particularly important for domestic water supply, the SRB includes important hydropower schemes in the middle basin and huge agro-export industries situated at the basin outflow in the dry coastal areas. Both require a constant water amount from rivers and reservoirs. Hence, water-related activities in the lower basin considerably exert pressure on water availability and allocation further upstream. Furthermore, we found that glacier melt contributes up to 70% (60%) to upstream SRB (QC) river runoff, and up to 50% in the downstream coastal areas during the dry season. Projections of future water availability suggest a strong reduction in dry season runoff. Our integrated research identifies major management challenges due to seasonally declining water supply and rising water demand, potentially exacerbating existing competition over water and, hence, enhancing conflicts among upstream and downstream water users. In this context, assessments for risk management under high uncertainty increasingly need to integrate low and high flow risks into one assessment. Investments in water governance and more inclusive decision-making mechanisms as well as equal water access will be key to prevent social conflicts. More sociohydrologic research is needed, including improved hydroclimatic, water demand and socioeconomic monitoring and more in-depth understanding of the complexities of water supply-demand systems and multiple users’ perspectives.

Lake Poopó lies in the Bolivian Altiplano at an altitude of 3,686 meters high. It used to be the second largest lake in the central Andes, covering an area of around 3,000 km², with a maximum depth of less than 4 m. In December 2015, Lake Poopó had completely dried up, only partially recovering in the following years.

When the lake dried up, explanations in the public debate involved five factors, ranging from water withdrawal for mining and the expansion of irrigation agriculture (mainly for quinoa) to natural climate variability, the occurrence of an ENSO event and climate change. However, very little research has been done on the roles played by each of those factors. Thus, the aim of our study is to analyze how a vulnerable socio-ecological system is impacted by various stressors and their multiple interactions.

On the basis of an extensive literature review, including technical documents and other grey literature, a series of expert interviews, as well as field visits to the lake regions, we conclude that even though the complete drying of Lake Poopó at the end of 2015 was triggered by a very strong ENSO event during the second half of the year in combination with an unusually late start of the rainy season, only the interactions of these climatic stressors with the other socio-economic factors can fully explain the drying event.

As part of the results of our study we have produced a schematic overview showing the main feedbacks and interrelations existing between the factors behind the drying of Lake Poopó. Furthermore, our results indicate that an integrated perspective of the Lake Poopó socio-hydrological system, can best serve design and implementation of sustainable pathways that would allow the recovery of Lake Poopó as Bolivia´s second largest body of water.

In Agriculture, sustainable use of soil and water resources are a present day challenge and require detailed knowledge of soil biophysical properties and water needs. Irrigation practices and soil managment are more and more critically discussed due to ongoing climatic changes and rising public sensitivity toward sustainable usage of soil and water resources.. Permanent crops commonly require irrigation and this highly depends on soil and topographical location. Thus the knowledge of spatial variability of soil properties and the temporal variability of water needs is a key element in soil and water management. This study covered the Europe’s largest apple-growing area: the Venosta/Vinschgau and Adige/Etsch Valleys, in the Province of Bolzano/Bozen, South Tyrol, Italy. South Tyrol has a typical continental alpine precipitation regime, with low total annual precipitation (450–850 mm) unsufficient for orchards cultivaiton which gets compensated by a large scale irrigation management.

To optimise irrigation practice an objective and quantitative base of information was established . Soil moisture and soil water potential were continuosly monitored in 17 different locations and soil properties were geo-referenced and spatialized using a modern
geostatistical method. To assess whether the zones with shallow aquifer influence soil water availability, areas of potenzial groundwater influence were retrieved. Finally apple orchards water needs were modelled with a water balance hydrological model.
The preliminary results highlight the complex interplay of metereological varibility, soil properties, ground water influence, and farmer irrigation practices. In many locations, irrigation largely exceeds plant water needs while in other locations measurements show potential water stresses. This multidisciplinar approach proves to be a successul framework, highlighting the need to further investigate, monitor and model the complex interplay of the spatial and temporal ecohydrologcial processes in apple orchards fields.

Mountains act as water towers for downstream populations because the buffering capacity of snow and ice helps to sustain water resources during dry periods. Though often considered in isolation, it is the interactions between the climate, glacier, surface water and groundwater systems that shape the impact of climate change on mountain water resources.

My research focuses on mountain hydrological systems in the Peruvian Andes, where climate warming and retreating glaciers are raising concerns over future water resources in the mountains and on the arid coast. Specifically, I have investigated the important role of groundwater in buffering dry season streamflow in the Andes, and how this buffering capacity is likely to change over time. Furthermore, I have studied groundwater based climate change adaptation strategies, human interventions that seek to increase groundwater recharge to buffer dry season flows.

In the Shullcas Watershed in central Peru, 44% of glacier area disappeared between 1984 and 2011 and the nearby city of Huancayo suffers dry season water shortages with less than 50% households surveyed having 24 hour access to water. However, available discharge records do not show significant decreasing trends in average dry season discharge meaning that population growth is likely a larger contributor to water stress in this area. On the other hand, our hydrological modelling results project that decreasing glacier melt and increasing evapotranspiration will have a large impact (~-25%) on water availability by the end of 21^st century.

Evaluating climate impacts on Andean hydrology using a mountain systems approach

Lauren Somers^1*, Jeffrey McKenzie¹, Bryan Mark², Pablo Lagos³, Gene-Hua Crystal Ng⁴, Andrew Wickert⁴, Christian Yarleque⁵, Yamina Silva⁵, Michel Baraer⁶

¹McGill University

²The Ohio State University

³Universidad Nacional de San Marcos

⁴University of Minnesota

⁵Instituto Geofísico del Perú

⁶École de Technologie Supérieure

Mountains act as water towers for downstream populations because the buffering capacity of snow and ice helps to sustain water resources during dry periods. Though often considered in isolation, it is the interactions between the climate, glacier, surface water and groundwater systems that shape the impact of climate change on mountain water resources.

We use GSFLOW (http://water.usgs.gov/ogw/gsflow/), a combined surface water and groundwater flow numerical model, which we couple to a glacier melt module, to simulate the complete hydrological system of the Shullcas Watershed in the Peruvian Andes. We calibrate the model using six stream gauges and two clusters of groundwater table wells. We then apply statistically downscaled climate projections to simulate the impacts of climate change on the glacier area, stream discharge and groundwater flow from 1960 to 2099.

The results demonstrate that the hydrologic system is very integrated between surface water, groundwater, glacial melt, seasonality and climate change. Results indicate that groundwater discharge is the dominant source of streamflow during the dry season and that only approximately three percent of groundwater discharge to the stream is glacially sourced. For most of our modelled carbon emission scenarios, the Huaytapallana glaciers are projected to disappear completely by the end of the 21^st century. As a result of decreasing meltwater and increasing evapotranspiration, groundwater discharge to the stream and streamflow are projected to decrease by 23 and 36 percent respectively. While groundwater provides and initial buffer to decreasing meltwater production, this buffering effect diminishes as increasing ET reduces groundwater recharge. These results provide important information for water managers and demonstrate the increasing relative importance of groundwater in the Andes.

Climate change has impacts on the alpine cryosphere, hydrosphere, and biosphere. This calls for mitigation actions through renewable energy production such as hydropower and for adaptation to the occurrence of regional water scarcity. One potential strategy of jointly addressing mitigation and adaptation challenges is the use of multi-purpose water reservoirs, which should be operated for multiple services. However, water shortage does spatially often not co-occur with available, stored water. It is mostly expected in lowland catchments, where water demand is highest, and less in headwater catchments where the large hydropower reservoirs are located. Water transfers from highland to lowland could potentially alleviate water shortage situations. However, institutional barriers often hinder the implementation of such transfers.

This paper examines the challenge of highland-lowland water transfers using a case study in the canton of Bern, Switzerland. The lowlands of this region are characterized by a water-intense agriculture while a new hydropower-reservoir is planned in the headwater part of the region. We assess how a multi-purpose reservoir-lake cascade could be implemented in order to cover downstream water demand by upstream water supply. Therefore, we first identify regions prone to water shortage by quantifying water supply and demand. Second, we collect and evaluate data about the institutions and governance of this region through interviews and document analysis.

Our results show that a downstream water deficit could be covered by upstream water supply by the new reservoir. However, the actual regulations and governance processes do not allow for such an approach. This means that lowland water shortage is not necessarily a hydrological issue, but a management problem which needs to integrate various state and non-state actors, as well as water users from different sectors. A well-managed multi-purpose reservoir-lake cascade could be a promising solution for addressing both mitigation and adaptation challenges arising from climate change.

Mountains are particularly affected by the effects of climate change. However, impacts on mountain streams remain poorly understood due to: i) a lack of long-term records, ii) difficulties in identifying significant changes in flood activity in existing series and firmly attributing these to climate change because of the heterogeneity of the sources available and to a lack of adapted statistical methodologies, and iii) changing behavior from one catchment to another as a result of differences in size, elevation and presence/absence of significant glacier coverage. For instance, in the Alps, Bard et al. (2012) did not detect any clear evolution in flood intensity at the scale of large catchments, but significant changes in runoff regime may exist in small, glaciered watersheds (Bavay et al., 2013). As an attempt to fill these gaps, this study documents historical floods in two small, non-contiguous watersheds, namely the Vièze and Saltina (Valais, Switzerland). We combined different types of sources: historical archives, regional newspapers and tree-ring records. This multiproxy approach enabled the reconstruction of flood events dating back to the 14^th century. Descriptive statistical analyses of the chronologies of recorded floods and their links to past climatic conditions and statistical modeling of chronology-climate linkages demonstrate non-stationary with enhanced flood activity during wet periods and a linkage with temperature variations. However, interestingly, the response of the two catchments to warming is not unequivocal: whereas since the end of the LIA, Saltina flood activity has notably decreased, a significant increase occurred for the Vieze. Even if incompleteness of sources on the past may partially explain these results, they also reflect differentiate hydrological responses to ongoing changes that should be acknowledged for making realistic future prognoses.