The Environmental Justice Atlas lists a total of 60 environmental conflicts in Chile, of which ~73% are related to mineral extraction and water issues (EJAtlas, 2022). Problems of water scarcity, water insecurity and deficient water quality in the drylands of north-central Chile are generally affected by significant climatic variability, including a prolonged drought period since 2010 (Muñoz et al., 2020). Moreover, various impacts of land use systems in different parts of the country, especially agriculture and mining together with specific aspects of water legislation critically affect precarious human-water relations (Larraín and Schaeffer, 2015). Increasing pressure on scarce water resources leads to conflicts between different water user groups, such as agricultural producers, mining companies, and local communities across the region.

The legal base of water use in Chile is regulated in the Water Code, which was passed in 1981 under the dictatorship of Augusto Pinochet (Bauer, 1998, 2015; Budds, 2013, 2020). The aims of the Water Code have been the promotion of export-oriented agriculture, mining, and energy production through hydropower. Following a neoliberal logic, its guiding principle is the complete privatization of water rights and its concurrent decoupling from land ownership. Provided there are no restrictions for a catchment area and no objections from third parties, water rights are granted free of charge by the responsible National Water Directorate (Dirección General de Aguas; DGA) and these titles are subsequently tradable and inheritable. The rights to use surface water and groundwater are separate despite the connectivity of these hydrological systems. Several studies on water conflicts and governance in Chile have proven severe inequalities with accumulation of water rights by powerful actors and corresponding lack of water access by less influential groups and communities (Budds, 2009, 2020; Prieto, 2015, 2016; Usón et al., 2017). These socio-political interventions have been accompanied by adverse environmental impacts. Critiques have led to minor modifications of the Water Code in 2005, which included imposing fees for non-use of water rights to reduce speculation (Budds, 2013).

These characteristic regional development trajectories raise a number of questions, which require integrated conceptual approaches (Wesselink et al., 2017). Since 2012, the research field of socio-hydrology includes socioeconomic dynamics in the analysis of human-water interactions (Nüsser et al., 2012; Sivapalan et al., 2012). Mainly advanced by authors with a background in the natural or engineering sciences, socio-hydrology understands water and society as coupled systems characterized by specific interactions and feedback loops (Sivapalan et al., 2012; Sivapalan, 2015; Troy et al., 2015; Blair and Buytaert, 2016; Ross and Chang, 2020). Causal relationships are mostly based on pathways, dependencies and watershed modeling to assess water-related processes, associated risks (Viglione et al., 2014; Di Baldassarre, 2017; Schmidt et al., 2020) and future scenarios (Pande and Savenije, 2016; Pande and Sivapalan, 2017; Roobavannan et al., 2017). An important aspect of socio-hydrological studies is the consideration of processes at different temporal and spatial scales (Sivapalan et al., 2014), in some cases with a focus on the particularities of mountain environments (Nüsser, 2017).

Assessments often rely on quantitative data, which are increasingly complemented by qualitative methods. Socio-hydrology has frequently been criticized for its strong reliance on numerical models and its neglect of the plurality of human agency, perceptions and values (Wesselink et al., 2017). In a similar vein, critical comments refer to a disregard of local or indigenous knowledge on water-related topics (Troy et al., 2015; Nüsser and Baghel, 2016) and “non-scientific” meanings of water (Krueger et al., 2016). Despite ongoing conceptual debates and advances, the integration of social science approaches and methods into socio-hydrology remains a major research challenge. Authors argue for more balanced perspectives in interdisciplinary human-water studies and combined methodological approaches to adequately address human agency and governance (Di Baldassarre, 2017; Haeffner et al., 2021; Yu et al., 2022).

In the social sciences, the term socio-hydrology is less frequently used. Instead, alternative framings of integrated concepts have been proposed as hydrosocial approaches, which are mainly rooted in critical geography, political ecology and science and technology studies (Swyngedouw, 1997; Bakker, 2009). Their focus is the politicized character of water governance related to power asymmetries and inequalities in resource access. Conceptually, the hydrosocial cycle (Linton and Budds, 2014) departs from a relational perspective to overcome the dualism between nature and culture. Water is understood as a hybrid where nature and society constantly co-produce each other. Typical fields of research are irrigation and supply infrastructures, and the uneven access to water, with several regional examples from Chile (Boelens, 2014; Linton and Budds, 2014; Prieto, 2015; Usón et al., 2017). The hybrid nature of human-water relations has also been conceptualized as hydrosocial territories (Boelens et al., 2016) or explored under the umbrella of waterscapes to investigate the interrelated nature of water, power and capital which produce uneven socio-ecological conditions (Budds and Hinojosa, 2012; Karpouzoglou and Vij, 2017; Flaminio et al., 2022). However, several hydrosocial studies have not amply addressed the hydrological dimensions at the human-water interface.

Under the umbrella term of socio-hydrology, water management has been investigated in specific historical, cultural, economic and environmental settings. Especially mountain regions are characterized by complex water-related processes and social constellations, leading to site-specific particularities (Carey et al., 2017; Nüsser and Schmidt, 2017; Nüsser et al., 2019b). Therefore, case study approaches are indispensable to get deeper insights into entangled socio-hydrological interactions. As in previous studies (Nüsser et al., 2012, 2019a; Parveen et al., 2015), the socio-hydrological approach is used as a framework to address these complex relations across different spatial, temporal and social scales. An improved understanding of mountain waterscapes, that encompasses spatial, material, and discursive facets of human-water dynamics, requires identification of the main hydrological components and actor constellations. In the case of the upper Huasco valley, socio-hydrological interactions are characterized by water abstraction from cryosphere runoff for mining and agriculture in the context of neo-liberal water governance (Figure 1).

Departing from an integrated concept of mountain waterscapes, this article investigates socio-hydrological transformations in the upper Huasco valley against the background of mining impacts and agricultural changes, which increase pressure on scarce water resources and are a potential threat to water quality. The case study aptly exemplifies the impact of the expansion of extractivism and agriculture on local water resources. Since the early 2000s, the upper Huasco valley has received international attention due to the conflict over the construction and operation of the Pascua Lama open-pit mine, for the extraction of gold, silver and copper. In Chile, the economic relevance of the mining sector is large, with a share of 14.6% of the GDP in 2021. The country is a world leader in copper production with a volume of over 5580 tons in 2021 and belongs to the top 5 gold-producing countries in the world (SERNAGEOMIN, 2022, p. 9). Metal extraction is always associated with a high demand for water and energy as well as a risk of water contamination by toxic waste products from mining operations (Budds and Hinojosa, 2012; Valdés-Pineda et al., 2014; Zanetta-Colombo et al., 2022). At the same time, the Huasco valley is characterized by a mix of small-scale and export-oriented agricultural land use. In Chile, the area under fruit crop cultivation has been enlarged significantly and fruit exports have increased by 18% between 2008 and 2018. Chile is the global leader in table grape exports and ranks second in blueberry and cherry exports (ODEPA, 2019). Agriculture is the most important water consuming economic sector in the country, accounting for about 75% of the total water demand (Oyarzún et al., 2008). While export-oriented fruit cultivation amounts to 60% of the total agricultural production, agriculture only contributes to < 3% of the Chilean GDP, but it retains primary importance for rural livelihoods (ODEPA, 2019).

Previous studies on the Huasco valley have mostly focused on distinct aspects, such as the prominent conflict over the Pascua Lama mining project (Urkidi, 2010; Urkidi and Walter, 2011; Bottaro et al., 2014; Li, 2018), water quality (Strauch et al., 2009; Zang et al., 2018) and glacier changes in the upper catchment (Nicholson et al., 2009; Rabatel et al., 2011; Staub and Munos, 2016; Hess et al., 2020). In the present study, we analyse and explore the multi-dimensional land use changes from a socio-hydrological perspective using quantitative and qualitative methods. After an introduction to the study area, a description of the empirical social science and remote sensing methods is given. The precarious drinking water situation is taken as a starting point for an assessment of recent land use changes in the context of mining and agricultural expansion and related discourses over water scarcity and quality. Own results are contextualized in a broader literature review.

The Huasco valley stretches from the main Andean Cordillera to the Pacific Ocean. The main range of the Andes marks the border to Argentina and exceeds heights of 6,000 m a.s.l. Located about 200 km to the south of Ojos del Salado (6,893 m a.s.l.), Chile's highest peak, the Huasco valley is part of the transition zone between the arid Norte Chico and the hyperarid Norte Grande with the Atacama Desert. The average annual precipitation amounts to 49 mm and the mean annual temperature reaches 17.8 °C at Santa Juana station (560 m a.s.l.) downstream of the town Alto del Carmen (1965–2015; DGA, 2016). Predominantly influenced by westerlies, about 81–85% of the annual precipitation occurs during the austral winter between May and August (Salas et al., 2016). During the dry summer months and in times of drought, the Rìo Huasco is almost exclusively fed by meltwater from the cryosphere (Falvey and Garreaud, 2007; Favier et al., 2009; Vicuña et al., 2012; Azócar et al., 2017). The glacierized area in the upper catchment amounts to more than 24 km² in the year 2016, while 80% of the ice-bodies are smaller than 0.1 km² (Hess et al., 2020), often classified as glacierets or névés (Nicholson et al., 2009).

Annual precipitation exhibits pronounced interannual variability influenced by El Niño Southern Oscillations (ENSO) (Fiebig-Wittmaack et al., 2012; Masiokas et al., 2020). Rare extreme summer precipitation events regularly cause floods and landslides, such as in 2015 with heavy precipitation and flooding in the Huasco watershed (Salas et al., 2016; Meltzer et al., 2021). Recurring and often extensive drought periods adversely affect water resources. Since 2010, a so-called mega drought has affected Chile, which is the longest period of continuous rainfall deficits ever recorded (Garreaud et al., 2020). This unprecedented event has aggravated water scarcity in different parts of the country (Aldunce et al., 2017; Muñoz et al., 2020).

The case study examines the upper catchment of Río Huasco covering an area of about 7,000 km², upstream of the town Alto del Carmen at the confluence of the two main tributaries Río del Carmen from the southeast and Río del Tránsito from the northeast at an elevation of 815 m a.s.l. (Figure 2). Characterized by steep slopes, the topography of the upper Huasco valley (Huasco Alto) limits the suitability for settlements and agricultural areas (Figure 3). The population of the municipality of Alto del Carmen, which comprises the upper Huasco valley, amounts to 5,754 inhabitants in 2021 living in dispersed rural settlements (BCN, 2021). Approximately 20% are members of the Diaguita community, officially recognized as an indigenous community since 2006 (Molina Otárola and Campos Muñoz, 2017). Irrigated agriculture along the riverbed, where table grapes, avocados, and grapes for viniculture are cultivated as main crops, is the most important income source. Between 2009 and 2013, employment in the agricultural sector increased from 52% to 68% (BCN, 2021). Population growth and steadily expanding irrigation by export-oriented agricultural companies contribute to rising water demand (Wagnitz et al., 2014).

To assess the complex dynamics and processes in the upper Huasco waterscape, this study used a set of complementary methods. Interview and survey data from different subprojects within a wider research project on environment and health in arid regions have been reanalysed and complemented by an analysis of social media data and a remote sensing analysis. Qualitative interviews with stakeholders were conducted between 2014 and 2016 to gain information on water use, perception of water quality, water governance, water supply, and agricultural practices. Relevant place-based and non-place-based stakeholders included farmers, policy makers, researchers, NGO members, employees of mining or agribusiness companies (with more than 20 year-round employees), independent consultants and members of the Rural Drinking Water Committees (Comité de Agua Potable Rural; CAPR). Interviews were conducted in Spanish by a German PhD student who was in several occasions accompanied by a Chilean research assistant from Santiago. Interviews have not been recorded digitally. We assessed the data based on Kuckartz (2014) qualitative content analysis using thematic analysis as the basic method. The analysis of qualitative interview data was performed software-supported using the program MaxQDA, complemented by information from official statistical data, (unpublished) reports, and media reports. For detailed information on water demand, water quality, (public) health issues and drinking water systems, a standardized household survey was conducted in 25 settlements of the upper Huasco valley. The survey was realized within the wider project context by a German medical student and a Chilean research assistant from Santiago and also included sections on a flood event in 2015 and risk adaptation strategies (see Annex 3 in Meltzer et al., 2021). The questionnaire covered data from 262 households. Random sampling was implemented in settlements with more than 20 households, while solitary houses and households in smaller hamlets were all approached. We aimed for an even distribution over the two main tributary valleys to ensure representative results. In total, 342 households were approached and 262 interviews were realized. A pre-test was performed a few weeks prior to the data collection in the same region, but outside the actual study area. All interviews were carried out in Spanish. Written informed consent to participate in the questionnaire survey was provided by the participants.

Both researchers who conducted the interviews were female, European, young academics. They spent several months in the study area and were not actively involved in the conflictive situation. The study design was discussed with researchers at the Heidelberg Center for Latin America in Santiago de Chile. Both research assistants were female, Chilean, young graduates from a well-known university in Santiago and paid by the project. They also helped to cross-check cultural appropriateness and wording of the survey. This can have implications on all steps in the research process, including the participation of households and interview partners as well as the production of knowledge.

The collected interview and survey data were combined with an analysis of Twitter tweets for a broader assessment of views from the communities and local as well as non-local stakeholders. Moreover, it allowed to cover a longer time-span even during the Covid pandemic. In social media networks, people express their views, including political ones, openly. Although social media analyses are less inclusive than other social research approaches, given the limitations in access for marginalized people (Ash et al., 2018; Fearnley and Fyfe, 2018; Buckingham et al., 2020), the method has untapped potential in combination with other social science methods. Twitter has been selected as one of the most popular social media sites providing a potentially large set of data (Fearnley and Fyfe, 2018). It provides an interesting additional perspective by detecting main arguments in the discourses over water by local and non-local Twitter users. For this analysis, all tweets with the hashtags agua/water, glaciares/glaciers, sequía/drought (to capture tweets in Spanish and English) combined with #pascualama and #huasco (to restrict tweets to the study area) have been extracted from the database Twitter between 2013 (the halt of the Pascua Lama mining project) and 2022. Results were subsequently analyzed using the software MaxQDA, while individual users were anonymised.

The results from social research methods were combined with remote sensing approaches for rigorous triangulation. Multi-temporal satellite imagery (Corona, Landsat and Sentinel-2) were used to map agricultural areas (Table 1). Using a standardized semi-automatic approach based on a Normalized Difference Vegetation Index (NDVI) NDVI = (NIR – R) / (NIR + R), dense irrigated vegetation cover was mapped on Landsat images (path: 233, row: 80) taken in December by applying a threshold of NDVI > 0.4. In order to investigate the expansion of agricultural area, the classified Landsat images were combined on a decadal basis (1984–1989, 1990–1999, 2000–2009, 2010–2021). To reduce the effect of misclassified pixels caused by image errors, pixels which were classified as vegetation covered on only one single dataset were discarded. In order to extend the observation period, a panchromatic Corona image from 1973 was co-registered to the Landsat images. In a second step, Sentinel 2 data were used to investigate the impact of the current mega drought (Garreaud et al., 2020) on the agricultural production. The classified images (NDVI > 0.4) were summarized into three classes (cultivated in ≥ 6 years; cultivated in 3–5 years; cultivated ≤ 2 years between 2015 and 2021).

To identify glacier changes in the vicinity of the Pascua Lama mine project, cloud-free satellite imagery from the end of the ablation period were used to map the ice-covered area in 1986, 2000, and 2022. A standardized semi-automatic threshold approach based on the red/shortwave infrared band ratio was applied to delineate glacier boundaries, as ice and snow are characterized by a high reflection in visible light and high absorption in the shortwave infrared range (Paul et al., 2009; Hess et al., 2020). The classified ice-covered areas were transformed into vector data and objects smaller than 0.01 km² were deleted. The infrastructure of the Pascua Lama mine (roads, compounds, tailings) were digitized manually for different observation periods from satellite data.

In the context of the recent mega drought, the Chilean Water Directorate (DGA) declared the upper Huasco catchment as a water scarcity zone in the summer 2021/2022 (MOP, 2022). The area is facing problems of water provision for human consumption and irrigation purposes. Drinking water supply in the upper Huasco valley depends on groundwater. As in other rural areas of Chile, drinking water supply is mainly organized by local Agua Potable Rural committees (CAPR), collective institutions initiated in 1964. An CAPR is composed of elected community members, who are assisted by government organizations (Donoso and Vicuña, 2016). Each of the 20 CAPRs in the upper Huasco valley is responsible for infrastructure maintenance and water quality monitoring. Their size varies between 45 and 329 households covering about 95% of the total population (MOP, 2016). These rural institutions are independent in terms of administration, finances, and technical aspects. As some local CAPRs have reported in interviews, financial limitations to maintain and extend infrastructure for water provision, about 120 households (higher numbers after floods or during droughts) depend on drinking water supply by trucks, which are organized by the municipality Alto del Carmen. These trucks are filled from different wells in the upper Huasco valley. The declared drinking water provided by CAPRs and trucks is not only used for household purposes, but also for irrigation (42% of the surveyed households, n = 262) and livestock (32% of the surveyed households). Only few households either depend on own bore wells, springs, or additional water from neighbors.

Despite the crucial role of groundwater for drinking water supply and the fact that several studies have examined aquifers, their extensions and interconnections (e.g., DGA, 2004; CNR, 2006; CAZALAC, 2012), members of local CAPRs stated that they regularly have no or only little information on groundwater systems and recharge. In consequence and in addition to the lack of a public register of bore wells, the CAPRs do not know how many farmers access the same aquifer and how much water will be available. In addition, major concerns by the local population brought forward in interviews are negative impacts of mining activities on water quality and diverse adaptation needs in agricultural land use. The following sections contextualize the prominent mining conflict with regional water governance and agricultural land use change to shed light on the entangled socio-hydrological dynamics in the waterscape.

The study shows how water insecurities in the rural area of the upper Huasco valley have mostly been associated with large-scale landscape interventions in the context of the Pascua Lama mine and agricultural land use changes leading to an increase in water demand. Agricultural land use has expanded significantly until 2015, while the mega drought has strongly affected the region over the past years. Water availability largely depends on the cryosphere, which has faced significant changes attributable to climate change and the Pascua Lama mining project. The project has experienced massive resistance resulting from severe violations of environmental regulations and is emblematic for problems in water governance. It remains to be a prominent example of socio-hydrological conflicts in Chile. While assessing the nexus of land use change, cryosphere dynamics and diverse actors, this study highlights the need to consider integrative approaches in water governance reforms. A cross-sectional involvement of actors, participation of local population and civil society actors, close cooperation between different government actors and NGOs are important steps toward environmental justice.

As access to water is generally influenced by the political framing of water privatization, conflicts throughout Chile are often linked to the unequal distribution of water rights and an overuse of the resource (Larraín and Schaeffer, 2015). Recent attempts to reform environmental governance in the context of a new constitution argued for a more just and equitable model of water governance. Based on understanding water as a public good and access to water as a human right, the prioritization of resource use for human consumption was proposed. Focusing on organizational reforms in rural water governance, suggested steps included the installation of watershed committees at the local scale and the establishment of a National Water Agency (Agencia Nacional de Aguas, ANA). In October 2022, the electorates voted for a “rejection” of the proposal. Yet, Chile's water crisis remains a major concern. The impacts of the recent halt of the Pascua Lama project on the development of socio-hydrological interactions of the Huasco valley has yet to be seen.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. The patients/participants provided their written informed consent to participate in this study.

JD, MN, and SS: conceptualization, design, writing—review and editing, and contributing to manuscript revision. JD, SS, and CZ: methodology. CZ: field surveys. JD and SS: writing—original draft preparation. JD and MN: supervision of the study. All authors contributed equally to the article and approved the submitted version.