Kazakhstan, the largest landlocked country in the world, positioned in the center of Eurasia, is characterized by a cold and arid climate. The availability and quality of its water resources play a crucial role in its socioeconomic development, agricultural productivity, and ecosystem sustainability for the entire region (Zhupankhan et al., 2018). Despite its importance, comprehensive baseline data and detailed basin-wide information of Kazakhstan’s river basins remain incomplete or fragmented (Yapiyev et al., 2017a). Understanding the hydrological dynamics, spatial variability, and regional characteristics of Kazakhstan’s river systems is essential for effective water management, policy-making, and climate change adaptation strategies (Saidaliyeva et al., 2024). These river basins span different climatic regions, from arid and semi-arid zones in the South to more temperate areas in the North, leading to highly variable hydrological regimes (Yapiyev et al., 2021). The previous works on water resources of Kazakhstan relied on gray literature (Zhupankhan et al., 2018), without detailed appraisal of water balance and its components (Chen et al., 2018) or a country-based regional perspective on water storage changes (Deng and Chen, 2017). Effective water resources management requires a better understanding of the water balance and its components (precipitation, evaporation, runoff and storage), which is still lacking. The growing impacts of climate change, coupled with increasing water demands, are further straining the country’s water resources, necessitating a detailed understanding of basin-level characteristics to inform sustainable management practices. To address this knowledge gap, we provide a data-based hydro-basin perspective on water resources and hydrology of large river basins of Kazakhstan.

This paper is structured as follows. Section 2. First, we provide a brief historical background, overview of water management basins of Kazakhstan and regional hydrological basins, providing their main geographic characteristics (subsection 2.1. and 2.2). Next, we provide basin-wise information on hydroclimate, bioclimate zones, consider seasonal cycles of temperature, precipitation and streamflow, as well as surface runoff, actual (ET) and potential evaporation (PET), landcover (subsections 2.3 and 2.4). In Section 3, we discuss the implications of our classification (regionalization) for water research and management in Kazakhstan and wider Central Asia (CA), based on the information provided in the Section 2. Finally, we outline the knowledge gaps and suggest future directions for water research in the region.

In the seminal work on regional climate modeling Mannig et al. (2013) divided Central Asia into eight subregions. The main clusters covering selected region are Northern CA (NCA, mean Ta and total P of 2.6°C and 346 mm/year, with P peaking in July), miscellaneous transition zones (mean Ta and total P of 8.4°C and 304 mm/year, with P peaking in March), arid basins of CA (mean Ta and total P of 9.8°C and 119 mm/year, with uniform P distribution). Jiang et al. (2020) separated CA into northern and southern areas based on the same seasonality principles (Figures 4A,B). In the latest IPCC report (IPCC, 2021) the CA region is presented in four subregions: Eastern Europe (EEU), Western Siberia (WSB), Western Central Asia (WCA), and Eastern Central Asia (ECA). These differences between different approaches to regionalization also shows that Kazakhstan and Central Asia are not an independent uniform region but is embedded in other regions (Yapiyev et al., 2017a). For example, geographically, the Ural (Zhaiyk) river is considered to be a European river (Yarushina et al., 2009; Keune and Miralles, 2019) and Ertis is the largest tributary of the Ob River in Siberia (Yapiyev et al., 2017a; Huang et al., 2021). The eastern part of the Caspian basin (Figure 1B) is an inland sink of the Murghab-Harirud river system that territorially can be considered to be in the Middle East (Marti et al., 2023; Fallah et al., 2024).

The arid zone of CA and Kazakhstan is located in the cold and dry desert zone (BWk zone in Koppen-Geiger, Figure 3) with seasonally uniform precipitation. However, in the Northen – Southern classification, the Northern Kazakhstan rivers situated in this zone, such as Zhem (Emba) in Zhaiyk-Caspian (Laiskhanov et al., 2025), Torgai and Sarysu have the nival (snow-dominated) flow regime (Figure 4C). In contrast, the arid lowlands of the south region, including the Aral Sea basin, Shu-Talas and Ily-Balkhash (Figure 1B), are dominated by discharges in mountain headwaters with nival-glacial river regimes (Figure 4D). According to Beckinsale (2021) river regime classification, the study region (Figure 1B) belongs to microthermal (one or more months with mean temperatures below −3 C⁰, and snow cover duration for a month or more) and mountain regimes (microthermal regimes on high mountains, above 1,500 m a.s.l,). In this classification, the North region has microthermal regime Dfc (violent nival spring maximum, strong winter minimum) and the South region belongs to mountains regimes HN (nivale or highland snow) and HG (gletscher or highland ice). In Soviet classification Kazakhstan was predominantly classified to have spring flood type and warm season high-flow in the mountain regions (Zaikov, 1946).

This river basin classification has significant implications for water research and management in Kazakhstan and Central Asia. Similar to Yao et al.’s (2021) approach, Central Asia was classified into Northern (NCA), Southwestern (SWCA), and Southeastern (SECA) subregions based on Köppen-Geiger zones, aridity index, precipitation distribution, geomorphological landscapes, and land cover conditions. This classification helped assess extreme precipitation, forecasted to increase in the region until the end of the century, with the largest changes in NCA and SECA. Studies on climatology and moisture sources in precipitation in Central Asia were aided by similar regionalization (Jiang et al., 2020; Peng et al., 2020, 2024).

In terms of agricultural water management, the North Region is dominated by rain-fed agriculture (Figure 5), mostly grains (Schierhorn et al., 2020), relying on summer rain and spring snowmelt for soil moisture replenishment (Li et al., 2020). Crops in the South Region (Figure 5) fully rely on irrigation during the vegetation period in summer; the water is provided from river with glaciated headwaters (Li et al., 2020). Both regions rely on the dominance of grasslands for livestock farming (Chen T. et al., 2019) (Figure 5). The crucial difference in agriculture implies different approaches to water management and climate change adaptation (Li et al., 2020). It was estimated that in the rain-fed North under 2.0°C warming scenario increase in precipitation will not be sufficient to meet crop water requirements in the period by end of the century (Li et al., 2020). This will require development of drought resistant crops and adoption of wider use of climate-resilient irrigation. Decreased runoff from glacierized headwaters in the South, will require incremental (such as improvement in existing irrigation systems) and transformative (drip irrigation technologies, self-organized governance system) adaptation actions (Saidaliyeva et al., 2024).

Flood risk management and mitigation differ significantly between the North and South regions. In the Northern part, the spring snowmelt flooding is a natural hazard resulting from the natural river regime (Figure 4C) and is exacerbated by climate change impacts, such as earlier and more pronounced spring flood peaks (Didovets et al., 2024). The catastrophic snowmelt flood in spring 2024 that affected the Northern region, including southwestern Russia, and north and northwest Kazakhstan. The event was compounded by record high snowmelt, antecedent soil moisture anomalies, extreme rainfall, and high air temperatures in March 2024 (Laiskhanov et al., 2025; Zhang et al., 2025). Evidence suggests that settlement expansion in flood-prone zones played an important role in the 2024 flood disaster (Laiskhanov et al., 2025), aligning with global trends (Rentschler et al., 2023). In the South, common flood hazards include glacial lake outburst floods, pluvial floods from extreme precipitation, intense glacial-snowmelt events and related debris flows and landslides (Kapitsa et al., 2017; Saidaliyeva et al., 2024; Shahgedanova et al., 2024). Thus, flood risk assessments, adaptation, and mitigation measures must account for such regional differences (Ceresa et al., 2025). For example, the AI flood forecasting tool recently developed by Google (Nearing et al., 2024) must be adapted differently for the North and the South of Kazakhstan. Addressing the geohazards and climate risks in both regions will require multi-sectoral approaches including land and water management, early-warning systems and capacity building, and improved monitoring (Yapiyev et al., 2017a; Saidaliyeva et al., 2024; Shahgedanova et al., 2024), taking into account regional differences.

The proposed water resources framework for Kazakhstan directly supports efforts toward achieving the Sustainable Development Goals (SDGs), particularly: SDG 6 (Clean Water and Sanitation) SDG 13 (Climate Action), SDG 2 (Zero Hunger), and SDG 15 (Life on Land). Regionalization reveals crucial differences in climate change adaptation needs across Kazakhstan’s Northern and Southern hydrological regions (SDG 13). Understanding regional hydrological characteristics is essential for developing effective strategies to combat climate change and its impacts on water resources (SDG 6). The identified differences in agricultural water management between the rain-fed North and irrigation-dependent South regions stress the need for tailored approaches to ensure food security and sustainable agriculture (SDG 2). While achieving human-centered SDGs is important, the growing body of research on the region (Saidaliyeva et al., 2024; Song et al., 2025) emphasizes the importance of water resources for ecosystem sustainability and ecological conservation (SDG 15). The detailed basin-level transboundary approach is better suited for the protection, restoration, and promotion of sustainable use of terrestrial ecosystems. Baseline basin information and characteristics are essential for setting the hydrologically relevant boundary conditions (Tarasova et al., 2024). Catchment regionalization has already proved to be essential to conduct global, regional and local water research (Meybeck et al., 2013; Pool et al., 2021; Zhao et al., 2024; Kiraz-Safari et al., 2025). Such background information can serve as a starting point for setting and exploring one’s research hypotheses and questions (Wade et al., 2024; Zhao et al., 2025) and for countries to engage in climate-proofed water agreements.

Overall, current and future water availability and quality are the most important knowledge gaps in Kazakhstan and wider Central Asia. Our data and the wider literature show that the region is historically climatologically water-limited and climate change makes accurate future predictions even more difficult. Moreover, it is not addressed in detail in terms of impacts and adaptation (Vakulchuk et al., 2022; Saidaliyeva et al., 2024). While transboundary water sharing and management has been a hot point for discussion in the South region, the North basins shared with the Russian Federation are still largely ignored. Apart from water availability, water quality is also a pressing issue. There is growing interest in the South (e.g., Yapiyev et al., 2021; Wade et al., 2024), but larger areas in the North have less data and research projects.

Groundwater presents a potentially untapped reserve to compensate water deficits during droughts but has the largest knowledge gap (Gafurov et al., 2019; Yapiyev et al., 2021). Groundwater sustains river flows, especially during dry periods (Xie et al., 2024) but very little is known about the links between surface and groundwater flows (Yapiyev et al., 2020). Hydrogeological regionalization for Kazakhstan and Central Asia was developed in the Soviet period (e.g., Lagutin, 2020), yet, little progress in groundwater research was achieved thereafter. A recent study (Sallwey, 2024) has shown that Kazakhstan has high potential for implementing managed aquifer recharge (MAR) systems but more in-depth research is needed. Kazakhstan and Central Asia is the region with the most expansive groundwater-dependent ecosystems (Rohde et al., 2024) but yet little is known about water movement in the subsurface and ecohydrology (plant water use).

Recent literature reviews (Yapiyev et al., 2021; Saidaliyeva et al., 2024) show that water security is a dominant topic, while freshwater ecosystems and biodiversity (Song et al., 2025) are underrepresented in the scientific literature for the region. Kazakhstan and Central Asia are dominated by non-perennial streams (Messager et al., 2021; Krabbenhoft et al., 2022). This results in a high degree of endorheism (Yapiyev et al., 2017a) even in the river basins connected to ocean outlets are embedded regions with inland sinks (Figure 1B). Studying regional hydrology and water resource management must take into account the presence of such closed drainage areas (Prusevich et al., 2024). Endorheic areas are water-scarce, thus, the possibility to do economic activities is very limited; furthermore, these regions are also very sensitive to climatic changes (Yapiyev et al., 2024). Nevertheless, the available topography data such as Digital Elevation Models have at best 25 m spatial resolution (Hawker et al., 2022), which often is not sufficient for hydrological modeling tasks such as flood forecasting and risk assessment, especially for relatively flat regions in the north (Table 1).

The region has experienced an increase in both population and urbanization, but urban water infrastructure and management practices are still inadequate to deal with residential floods and other natural hazards, as well as to propose climate change adaptation in the cities (Laiskhanov et al., 2025). Soil hydrology has been receiving increased attention in recent years (Yapiyev et al., 2018; Siegfried et al., 2024) but together with groundwater is also mostly unexplored topic.

Snow (and snowmelt) is the most critical resource for water supply and quality of both surface water, soil moisture and groundwater (Mashtayeva et al., 2016; Fugazza et al., 2020; Khanal et al., 2021). Nevertheless, snow related hydrological processes and their representation in hydrological models are yet poorly captured and researched.

Another knowledge gap relevant to Kazakh and Central Asian hydrological studies is the lack of knowledge on runoff formation processes and runoff components. Assessment of seasonal contribution of runoff components is necessary to understand runoff formation mechanisms and the dynamics of water resources, to estimate the influence of climate change on water regime and to develop adaptation strategies in addition to management plans. There is a limited number of studies which directly address hydrograph separation of South Central Asian rivers considered in this work, such as Ala-Archa river (He et al., 2018, 2020; Tokarev et al., 2024), Naryn River (Hill et al., 2017), Syrdarya and Amudarya (Khanal et al., 2021). While the majority of studies on hydrograph separation for Central Asian region focus on the rivers in the Chinese part of Central Asia, such as Urumqi river (Sun et al., 2016), Tarim River Basin (Fan et al., 2016; Sun et al., 2018), river runoff in Tianshan Mountains (Chen H. et al., 2019). Overall, results indicate that meltwater (from snow and glacier) and groundwater comprise a significant portion of annual runoff. While the glacier (melt) contribution is often emphasized as important for surface water resources in glaciarized catchments of South region (Pritchard, 2019; Tillakarim et al., 2024), snowmelt (Kraaijenbrink et al., 2021) and groundwater (Khanal et al., 2021) are the main contributors to the total runoff. Literature review highlights the absence of such studies for the Northern Kazakh river basins, which are especially reliant on winter snowpack for runoff generation.

Water research studies and development of underlying theories, concepts, process formalization are based on results from data-rich Europe, North America, China and Australia (Kratzert et al., 2023; Tarasova et al., 2024). Accordingly, the applicability of modelling tools for Kazakhstan is often quite limited. For example, relatively recently published exercise of application of the global hydrological model World-Wide HYPE, has explicitly stated that the model failed to represent hydrographic situation in CA basins (Arheimer et al., 2020). Emerging AI modeling approaches can also be leveraged for hydrological modeling (e.g., Kratzert et al., 2018), drought and flood forecasting, and other related environmental approaches in Kazakhstan, but this will require better access to ground observations, and proper set-up and model training taking into account regional features.

Water and related environmental studies for Central Asia and Kazakhstan often rely on national country borders as boundary conditions (e.g., Hu et al., 2021). This limits the validity and practical applicability of such works. While considering Central Asia, the research domains are either limited to national boundaries in Kazakhstan, Uzbekistan, Turkmenistan, Kyrgyzstan, sometimes together with Northwest China (or the Tarim basin). This division constitutes only the South Region or South Central Asia (SCA), while ignoring transboundary rivers shared by Kazakhstan and Russia in the North which can be considered Northen Central Asia (NCA) (Mannig et al., 2013; Yapiyev et al., 2017b). The focus of international studies and projects has a strong SCA focus such as mountains (e.g., Saidaliyeva et al., 2024), glaciers (Kapitsa et al., 2020), water policy (Menga, 2017) and Aral Sea basin (Xenarios et al., 2019). On the other hand, the projects funded in Kazakhstan are usually have strong national focus. We tried to show that for Kazakhstan and Central Asia water, climate and environmental research and management activities require on one hand more nuanced and on the other broader approaches, taking into account the hydrological, climatological and biophysical boundaries, instead of narrower ‘national borders’. Such approaches must consider specific regional traits such as seasonality of water and climate cycle, land cover and biome zonation. This will allow us to address quickly changing change and related challenges of the 21st century.