Farmers’ Land Sustainability Improvement with Soil, Geology, and Water Retention Assessment in North Kazakhstan

Abstract

Land degradation issues are getting complicated worldwide. Kazakhstan’s land use has sharply deteriorated over several decades, necessitating comprehensive assessment and restoration. Farmlands in Kazakhstan are grappling with multiple challenges related to climate change, intense anthropogenic disturbances, and aggressive industrial agricultural practices involving monoculture crop production. Soil depletion is widespread in Kazakhstan due to flood erosion and drought expansion, causing desertification. The land sustainability of farmland improvement, including the soil, geology, and water retention assessment, is currently under investigation through our project activities in North Kazakhstan. Nature-based methods for forest plantation along contour strips and topography-based design landscapes are rarely applied or are absent in many rural areas these days. The land use issues have resulted in the loss of the soil moisture protective functions and a reduction in agricultural efficiency. Geodesy geomatics tools were applied for a topography investigation with digital elevation, digital terrain model preparation, and potential retention ponds’ location identification for managed aquifer recharge introduction. The combination of effective water accumulation methods, considering topography, with the development of protective forest shelterbelts should enhance the land use strategies for sustainable development. This strategy is expected to reduce soil erosion, promote moisture accumulation, by improving the soil’s quality as a sponge in water collection, and increase crop yields. Alongside this, a system for developing the retention ponds with managed aquifer recharge locations for proper water collection to improve the agrolandscapes was presented.

1. Introduction

Urbanization, population growth, and intensive agriculture with monocrops have increased demand for sustainable rational land use and ecosystem restoration programs. The disruption of natural connections within the terrestrial water cycle, including changes in evaporation, transpiration, and atmospheric moisture convergence, has a substantial impact on the water balance, biodiversity, and stability of soil systems. A range of studies demonstrates that local changes in vegetative cover can improve restoration of small streams and rivers. A decrease in transpiration due to deforestation leads to reduced rainfall in the Amazon. At the same time, forest restoration on China’s Loess Plateau first decreased and then intensified atmospheric moisture convergence [1,2,3]. These challenges are particularly acute in arid and semi-arid regions such as Kazakhstan, Central Asia, where increasing land degradation, soil salinization, and water scarcity are already threatening agricultural productivity and rural livelihoods.

Afforestation restores the natural water cycle: moisture evaporating from leaf surfaces increases atmospheric humidity, which initiates cloud formation and increases precipitation. This process, in turn, enhances moisture convergence and creates a positive feedback loop where forest masses stabilize and even amplify the local and regional water cycle [4]. A crucial aspect lies in the fact that atmospheric dynamics vary depending on the initial state of the atmosphere: in a dry atmosphere, an increase in transpiration merely redistributes moisture, whereas in an atmosphere close to saturation, evapotranspiration is capable of simultaneously increasing both precipitation and moisture influx [5]. Thus, tree plantations not only prevent soil degradation and landscape destruction but also form a self-sustaining mechanism of moisture circulation, ensuring the long-term sustainability of ecosystems.

In recent decades in many countries, special attention has been devoted to adaptation of nature-based solutions to improve ecosystems, to improve soil retention capacity so it can be used as a sponge, and to increase the water resources’ sustainability. Among these solutions is the traditional practice of contour-strip land organization, which involves the placement of tree and shrub plantings along the contours of the relief. This practice reduces soil erosion by forming contour-aligned ridges that act as barriers to surface runoff, decreasing the flow velocity and increasing infiltration into the small depressions created by crop rows. Contour farming has become widespread globally as an effective method for controlling soil erosion and preventing the siltation of water bodies [6,7,8]. The primary goal of filter strips is to reduce the number of suspended solids and sediment in the runoff [9]. These measures not only stabilize the hydrological regime but also contribute to the restoration of soil fertility and local bio-productivity. Recent studies demonstrate that local changes in vegetation can restore degraded hydrological systems and improve microclimatic conditions, thereby increasing land resilience and stabilizing water balances. For example, deforestation in the Amazon leads to reduced transpiration and, consequently, decreased precipitation, while large-scale forest restoration on the Loess Plateau in China has altered patterns of atmospheric moisture convergence. These examples highlight the global significance of vegetation–climate interactions and demonstrate the potential of nature-based solutions to mitigate the adverse impacts of climate change on land and water systems [10].

Kazakhstan, Central Asia, represents one of the world’s most vulnerable regions to climate change and unsustainable land use practices [9,10,11,12]. Over the past few decades, this region has experienced growing water shortages, degradation of irrigated and rainfed lands, and loss of ecosystem services due to excessive groundwater extraction, reduced snow accumulation, and soil salinization [13,14,15]. Sustainable management of land and water resources is therefore becoming a key strategic priority for the region’s agricultural development and climate adaptation policies.

One promising approach involves the integration of managed aquifer recharge (MAR) and contour-strip land organization as part of nature-based land management systems. These methods enhance infiltration, control erosion, and improve groundwater replenishment through the interaction of vegetation, soil, and hydrological processes [16,17,18,19]. The use of Geographic Information Systems (GISs) and remote sensing (RS) technologies enables multi-criteria assessments to identify suitable areas for MAR implementation, considering topographic, soil, geological, and climatic parameters [20,21,22,23,24,25]. Despite growing global experience, there remains a significant knowledge gap regarding how these combined landscape–hydrological approaches can be adapted to the semi-arid agricultural systems of northern and central Kazakhstan, where precipitation is limited, snowmelt is the main water source, and soils are prone to salinization. Addressing this gap is essential for designing land use strategies that ensure both agricultural productivity and long-term ecosystem sustainability.

The objectives of this study are therefore (1) to assess the potential of integrating forest contour-strip land organization with MAR methods in the Akmola Region of central–northern Kazakhstan; (2) to identify the most suitable zones for groundwater replenishment using GIS–RS–AHP analysis; (3) to evaluate the influence of shelterbelts on snow retention, soil moisture accumulation, and crop yield stability; and (4) to develop recommendations for sustainable land and water management strategies under increasing climate variability. By linking ecosystem restoration, hydrological modeling, and geospatial analysis, this study contributes to the global discourse on sustainable agriculture and climate-resilient land management in dryland regions. The results will provide valuable insights into how landscape-based MAR and contour-strip practices can enhance land–water interactions and support agricultural sustainability in Central Asia [26,27,28,29,30,31,32,33].

2. Materials and Methods

2.1. Study Area

The Akmola Region is in the north–central part of Kazakhstan, covering an area of approximately 146.2 thousand km² (Figure 1). The region’s climate is sharply continental, characterized by cold, prolonged winters and hot, relatively short summers. The average temperature in January ranges from −16 to −18 °C, and in July, it ranges from +19 to +21 °C. Annual precipitation fluctuates between 250 and 350 mm. Evaporation significantly exceeds precipitation, which creates a moisture deficit and poses challenges for agriculture [34]. The soil cover is diverse and generally favorable for farming. Ordinary and southern chernozems constitute the majority, possessing high natural fertility. Solonetzes and salinized soils, which limit the productivity of agricultural lands, are found in depressed relief elements. Long-term land exploitation and the manifestation of water and wind erosion have led to the deterioration of some soil conditions [35,36,37]. Agricultural lands comprise about 80% of the Akmola Region’s territory. The main cultivated crops are spring wheat, barley, sunflower, flax, lentil, and alfalfa. Crop yields are highly dependent on weather conditions, primarily the moisture supply during the growing season. Along with farming, livestock production is developed, particularly cattle and sheep breeding [38]. Protective forest plantations, shelterbelts, play a significant role in stabilizing agrolandscapes. They were established during the Soviet period as an element of contour-strip land organization for snow-water retention. The forest shelterbelts helped reduce soil erosion, improve moisture accumulation capacity, and increase crop yields. However, the forest shelterbelts’ condition has deteriorated during the last four decades due to the lack of maintenance, disturbances by fires, and the aggressive industrial agricultural monocrop expansion [39].

2.2. Overall Methodology

For the identification of potential MAR sites, GIS RS data tools were employed. The related processing steps of the methodology are presented in Figure 2.

This methodology (Figure 2) consists of the following major steps:

(1)

Data Acquisition: Collection of RS and hydrometeorological data.

(2)

Thematic Layer Generation: Creation of thematic layers (slope, land use, soil salinity, soil type, geology, and precipitation).

(3)

GIS Processing and Preparation: Processing and preparation of layers within the GIS environment.

(4)

Indicator Standardization: Standardization of indicators into a 5-class scale.

(5)

Criteria Weight Determination: Determination of criteria weights using the Innovative Groundwater Solutions (INOWAS) platform.

(6)

Multi-Criteria Analysis: Application of the AHP method.

(7)

Potential Mapping: Contour-strip land organization with MAR potentiality.

The main project activities are related to Earth remote sensing (ERS), hydrometeorological data collection, and GIS application for data processing combined with field geodesy work to determine the potential areas suitable for MAR and contour-strip land organization. The initial stage involved the collection of baseline data. RS datasets were used as the main input to generate layers for terrain slope and land use. Soil salinity, soil cover characteristics, geology, and hydrometeorology were incorporated. Precipitation distribution and the groundwater replenishment process were analyzed. GIS tools were applied for the preparation and unification of all thematic layers, bringing them to a common coordinate system, spatial resolution, and format. All input datasets were checked for compatibility in the subsequent analysis. In the next steps, indicators were provided. The values for each factor, including slope, land use type, and soil salinity level, were transformed into a five-point scale, reflecting the degree of favorable conditions from very low to very high. After standardization, the determination of criteria weights was performed using the INOWAS platform. This stage established the relative significance of each factor participating in the analysis. For instance, terrain slope or geological conditions might exert a greater influence on the aquifer recharge than land use type. The AHP was applied to integrate all indicators into a unified system. This method enabled the simultaneous consideration of multiple criteria to form the final MAR suitability map, based on a multi-criteria evaluation. The concluding stage was the identification of territories with high and low MAR potential. The results of the analysis were presented in the contour-strip topographic layers, with the most promising and least favorable zones for MAR locations.

The following are the main groups of datasets that were collected and prepared for the assessment using the AHP-MCDA methods: geology, slope, precipitation, soils, salinity, and land use. A similar approach was adapted from previous studies on the application of AHP–MCDA [40,41,42,43,44]. Within this project, in comparison with the previous studies, we are more focused on the farmlands and land use strategies for sustainable development for Kazakhstani farmers.

Land use maps were constructed using ESA WorldCover data for the Akmola Region, North Kazakhstan. The land use map was processed and used as one of the input layers in the subsequent AHP-MCDA modeling analysis (Figure 3). The primary source used was the global product ESA WorldCover v200 (2024), a land cover map with a 10 m spatial resolution, developed by the European Space Agency (ESA) based on Sentinel-1 and Sentinel-2 satellite data. This product provides 11 land use categories and ensures high detail for regional analysis. The ESA WorldCover scene was loaded and processed within the Google Earth Engine (GEE) environment. The administrative mask of the Akmola Region, based on the Food and Agriculture Organization (FAO) of the United Nations Global Administrative Areas (GADM), namely, the FAO GADM dataset (Level 1, 2015), was used to define the study boundaries. The WorldCover scene was clipped, after which a mask for agricultural lands (Class 40—Cropland) was constructed, and all land use classes were visualized using the official color palette.

The map (Figure 3) displays the main types of land use, including forests, vegetation, built-up areas, water bodies, wetlands, and agricultural lands. The processing resulted in a land use map for the entire Akmola Region. The map reflects a high proportion of agricultural land use in the central and southern parts of the region, as well as the presence of areas with natural vegetation in the north and near water bodies. The raster layer was exported in GeoTIFF format with a 10 m spatial resolution and EPSG: 4326 coordinate system, allowing it to be used in any desktop or cloud GIS for further analysis. Additionally, a script was prepared on the Google Earth Engine platform, which can be used for reanalysis, data updating in subsequent years, and publication as part of web maps and geoportal services. The resulting map is used in the next phase of the project. The land use type serves as one of the criteria in the AHP-MCDA methodology, alongside slope, soil types, geology, and salinity level. Areas classified as agricultural lands with low slopes and favorable soil characteristics are prioritized for MAR implementation, which should enhance the water-holding capacity of agrolandscapes and the sustainability of agricultural production in arid climates. The map illustrates the spatial distribution of land use types across the Akmola Region based on the global ESA WorldCover product for 2024. District boundaries are indicated by red lines, allowing the land use structure to be correlated with the administrative division of the region. Forest cover, represented in dark green, is primarily concentrated in the northern and northeastern parts of the region. Meadows and herbaceous vegetation are highlighted in light green, distributed mainly in the east and northeast. A significant portion of the territory is occupied by agricultural lands, shown in brown, mostly confined to the central and southern districts. Built-up areas are marked in red and localized as isolated patches corresponding to urban and rural settlements. Water bodies are indicated in blue and are predominantly concentrated in the south of the region, where large lakes and reservoirs are located. Herbaceous wetlands (swamps), shown in light blue, occur fragmentarily, mainly in the northeastern part. Overall, the map demonstrates the dominance of agricultural land use in the central part of the region while maintaining forest and wetland ecosystems in the north and east, and the significant role of water resources in the southern part of the region.

Soil salinity maps were generated using Sentinel satellite data for the Akmola Region. Sentinel-2 satellite data from the summer of 2024 was used as the source material for creating the maps. The images have a 10 m spatial resolution, which allows for a detailed reflection of the spatial heterogeneity of soil salinity. All materials were standardized by the European Petroleum Survey Group (EPSG) database of coordinate reference systems and coordinate transformations, namely, the Spatial Reference List EPSG: 4326 (WGS84) coordinate system. Data was loaded and processed using the GEE platform, which provides direct access to the satellite observation archive and cloud processing capabilities for large datasets. The atmospherically corrected Sentinel-2 imagery (COPERNICUS/S2_SR_HARMONIZED) was used in this work for soil salinity monitoring. Bands selected from Sentinel-2 include Blue (B2), Green (B3), Red (B4), Near-Infrared (B8), and Short-Wave Infrared (B12). The salinity index was calculated as follows [45]:

$$SSI8={\displaystyle \frac{B3\times B2}{B4}}$$

This index aims to highlight soil salinity indicators, as the spectral responses in these channels reflect elevated salt content. The images were processed on the GEE platform. The Support Vector Machine (SVM) classification method was applied. This is a powerful machine learning tool for multiclass classification based on spectral features, trained on prepared ground truth points labeled with salinity levels (None, Slight, Moderate, Strong, Very Strong). This method for soil salinity classification showed the best accuracy result among three classification types tested: Artificial Neural Network (ANN), Random Forest (RF), and Support Vector Machines (SVMs). As a result of the processing, classification maps with soil salinity levels were obtained for the studied territories: the Akmola Region, the KazGer farm, and the N.P.C. named after A.I. Barayev. All maps are suitable for spatial analysis in the context of MAR (Figure 4).

In Akmola region, the climate is extremely continental, arid, with hot summers and cold winters. The climate is extremely continental, belongs to the West Siberian climatic region of the temperate zone. The daily and annual temperature amplitudes are very large through the region (Figure 5).

The average monthly and annual maximum air temperatures (°C) according to data from the Kokshetau meteorological station for the period 1960–2022 were as follows: annual −8.1 °C; January: −10.7 °C; February: −9.1 °C; March: −1.9 °C; April: 10.4 °C; May: 19.4 °C; June: 24.7 °C; July: 25.8 °C; August: 23.4 °C; September: 17.4 °C; October: 8.1 °C; November: −2.2 °C; and December: −8.0 °C.

The average monthly and annual maximum air temperatures (°C) according to data from the Borovoe meteorological station (SCFM) for the period 1960–2022 were as follows: annual −8.2 °C; January: −9.5 °C; February: −8.3 °C; March: −1.5 °C; April: 10.7 °C; May: 19.1 °C; June: 24.0 °C; July: 25.1 °C; August: 23.3 °C; September: 16.9 °C; October: 8.5 °C; November: −2.5 °C; and December: −7.6 °C (Appendix A,

Table A1. Average monthly climate parameters and calculated values of radiation and evapotranspiration for Burabay farm in 2024.

Month	Minimum Temperature, °C	Maximum Temperature, °C	Relative Humidity, %	Wind Speed, km/day	Sundial, Watch	Radiation, MJ/m2/day	Evapotranspiration–Radiation, mm/day
January	−17.3	−8.2	81	345	3.1	3.3	0.31
February	−17.2	−8.6	75	409	5.6	6.8	0.43
March	−8.9	0	78	329	6.5	11.1	0.77
April	2.8	12.8	72	306	8.5	17.0	2.27
May	4.1	16.1	68	305	9.4	20.9	3.25
June	15.0	25.6	74	264	13.2	26.8	4.91
July	15.4	24.9	81	239	11.2	23.6	4.32
August	13.4	21.4	81	250	9.5	19.1	3.31
September	6.3	16.4	71	259	9.1	14.8	2.51
October	−1.1	9.3	73	239	6.5	8.4	1.25
November	−7.0	0.1	83	345	3.8	4.0	0.49
December	−11.3	−5.0	83	373	1.1	1.9	0.36
Average	−0.5	8.7	77	305	7.3	13.1	2.02

,

Table A2. Average monthly and annual maximum air temperature (°C) for 2004–2024.

Month	I	II	III	IV	V	VI	VII	VIII	IX	X	XI	XII	Year
MC Kokshetau	−10.7	−9.1	−1.9	10.4	19.4	24.7	25.8	23.4	17.4	8.1	−2.2	−8.0	8.1
MS SKFM Borovoe	−9.5	−8.3	−1.5	10.7	19.1	24.0	25.1	23.3	16.9	8.5	−2.5	−7.6	8.2

and

Table A3. Average monthly and annual minimum air temperature (°C) for 2004–2024.

Month	I	II	III	IV	V	VI	VII	VIII	IX	X	XI	XII	Year
MC Kokshetau	−19.2	−18.5	−11.4	−0.5	6.4	11.9	14.0	11.7	6.3	−0.3	−9.4	−15.9	−2.1
MS SKFM Borovoe	−18.9	−18.1	−11.4	−1.3	4.9	10.1	12.4	10.3	4.6	−1.1	−10.4	−16.3	−2.9

).

The basis of agroclimatic zoning was the thermal and moisture availability of the territory, namely, the moisture coefficient for the vegetatively active period (May–August) and the sum of active air temperatures above 10 °C, averaged over a long-term period.

The “normal” value refers to the long-term average for the period 1961–1990. The mean annual amount of atmospheric precipitation for this period was 319 mm.

According to data from the Kokshetau and Borovoe (SCFM) meteorological stations for 1960–2022, the average monthly precipitation (mm) was as follows: January—13 and 12; February—12 and 11; March—12 and 14; April—18 and 21; May—29 and 35; June—41 and 37; July—72 and 72; August—44 and 39; September—24 and 26; October—23 and 27; November—17 and 22; and December—13 and 14 mm, respectively (Figure 6).

The annual amount of precipitation according to the Kokshetau and Borovoe (SCFM) meteorological stations for this period was 318 mm and 330 mm, respectively; from November to March, the values were 67 mm and 73 mm; and from April to October, the values were 251–257 mm (Appendix A,

Table A4. Average monthly precipitation (mm) for 2004–2024.

Station	Months													Season
	I	II	III	IV	V	VI	VII	VIII	IX	X	XI	XII	Year	XI-III	IV-X
MC Kokshetau	13	12	12	18	29	41	72	44	24	23	17	13	318	67	251
MS SKFM Borovoe	12	11	14	21	35	37	72	39	26	27	22	14	330	73	257

).

There are 22 meteorological stations (MSs) and 10 agrometeorological posts (AMPs) of the RSE “Kazhydromet” of the Republic of Kazakhstan operating in the territory of Akmola Region. To characterize the climatic conditions of the studied areas, data from three meteorological stations with continuous long-term observation series were used. It should be noted that, according to the requirements of the World Meteorological Organization (WMO), a long-term observation period of at least 30 years is necessary for reliable climate characterization. Accordingly, to determine the current climatic conditions, we used meteorological data from the RSE “Kazhydromet” of the Republic of Kazakhstan spanning more than 30 years, covering the periods 1960–2022, 1960–2024, and 2008–2024.

For the automatic meteorological station (AMS) Stepnyak (Akmola Region, Birzhan sal District) and AMS Shortandy (Akmola Region, Shortandy District), the available data were insufficient for calculating climatic parameters.

The maps of thermal and moisture availability for the Akmola Region were developed based on data from meteorological stations across the territory and were further refined using the WorldClim Version 2.1 high-resolution global climate data for the period 2000–2024. Based on this data, an automation process was implemented due to the large volume of information (Figure 7).

The presented scheme illustrates the process of automated processing of WorldClim v2.1 climate data for calculating the thermal resources of the Akmola Region during the vegetation period (April–September). It visually represents the sequence of operations, including raster combination (Raster Calculator), clipping by regional boundaries (Clip), data reclassification (Reclassify), and filtering (Majority Filter). It provides convenience and mobility:

–

Convenience: The scheme clearly demonstrates the logic and sequence of automated operations, making it easier for other users or researchers to reproduce the procedure.

–

Mobility: The model can be easily adapted to other regions and periods—it is sufficient to replace the input rasters and boundary layers. This ensures that the calculation process is fast, reproducible, and scalable.

–

Practical advantage: Automation of large climate datasets significantly reduces analysis time and minimizes the risk of human error during manual processing.

2.3. Methods

Despite the growing implementation of MAR technologies, their current recharge volumes still replenish only a small fraction of the increasing global groundwater demand [46,47,48,49]. This limitation is largely due to the restricted availability of surplus surface water suitable for recharge, the scarcity of appropriate sites for infiltration, and insufficient infrastructure for water diversion and distribution [50]. Additional challenges include legal constraints on water use [51], economic feasibility, and institutional barriers [52]. Consequently, many researchers and practitioners continue to view MAR as an expensive technology with a relatively high level of potential risk [53].

One of the most promising directions is agricultural managed aquifer recharge (Ag-MAR)—an innovative water-spreading approach that utilizes surface (aboveground) water to replenish groundwater reserves [54,55,56,57,58]. Ag-MAR, also referred to as agricultural groundwater banking, on-farm recharge, or flood-flow capture, is designed to use excess surface water during times of abundance (e.g., rainy seasons, snowmelt, or reservoir releases) for infiltration through the soil profile across agricultural lands, thereby enhancing groundwater recharge (

Table 1. Normalization of criteria for multivariate analysis of the suitability of a territory for potential water accumulation zones (MAR).

Criterion	Class Criteria	Standards (Class)
Geology	Wend (V)	1
	Ordovician Lena-Dolina stage (O2ld)	2
	Ordovician Lena-Dolinian–Kugartian (O2ld-k); Ordovician granitoids and dacites (γδO3); Lower Visean Carboniferous 1-2 (C1v1-2); Lower Bashkir Carboniferous (C1I); Middle–Upper Carboniferous (C2-3)	3
	Upper Ordovician (O3); Middle–Upper Ordovician (O2-3); Ordovician Cougartian (O2k); Devonian Frasnian–Famennian (D3fm); Carboniferous Lower Visean–Serpukhovian (C1v2-s); Middle Moscow Carboniferous (C2bm)	4
	Devonian, middle–upper (D2-3)	5
Slope	>5°	1
	2–5°	2
	2–0.5°	3
	0.2–0.5°	4
	0–0.2°	5
Precipitation	<379 mm	1
	379–392 mm	2
	392–405 mm	3
	405–418 mm	4
	>418 mm	5
Soil	Salt marshes	1
	Salt licks, solonetz	2
	Dark-gray forest, meadow	3
	Leached chernozems	4
	Chernozems	5
Soil salinization	Very high (EC 8–15 dS/m)	1
	High (EC 4–8 dS/m)	2
	Moderate (EC 2–4 dS/m)	3
	Weak (EC 0.75–2 dS/m)	4
	None (EC < 0.75 dS/m)	5
Land use	Permanent reservoirs	1
	Forest	2
	Sandy and rocky areas, sparse vegetation	3
	Arable land	4
	Natural grass pastures	5

) [59].

Unlike traditional MAR systems that rely on specifically constructed infiltration basins, the Ag-MAR approach integrates recharge processes within existing agrolandscapes. The use of surface water as a primary resource makes the method more environmentally sustainable and economically efficient, as it allows simultaneous management of both land and water resources without requiring separate infrastructure. Thus, Ag-MAR contributes to the harmonization of surface water and groundwater interactions and enhances the overall resilience of agricultural landscapes to climatic variability [8,60,61].

The method was implemented on the INOWAS platform. During the analysis, the slider was moved in the direction of a more important criterion, and the distance from the center determined its priority level. The final weights were calculated automatically and showed that the highest value is for geology (40.63%), and the lowest is for soil salinity (5.97%) (Figure 8).

The primary goal of this review is to synthesize current and past research related to MAR to present the modern state of knowledge on this topic, identify research gaps, and define potential synergies, tradeoffs, and the future direction for the development of MAR. In addition, this review offers a conceptual framework for analyzing the key elements and mechanisms that influence MAR implementation (

Table 2. Weights of criteria for assessing MAR potential.

Criteria	Weight, %
Slope	22.90
Land use	10.03
Soil	17.44
Soil salinization	5.97
Geology	40.63
Precipitation	3.02

).

The Kazger farm is situated in the southwestern part of the Birzhan-Sal District. The geology is characterized by loess-like loams and loamy deposits, underlain by carbonate marbles (Figure 7). A thematic geological map covering the boundaries of the KazGer company territory was prepared based on digitized data and vector layers (Figure 9). The primary goal of this stage was to present the spatial distribution of geological formations within the site and lay the foundation for further structural and resource analysis. Structure and Content of the Map: The map is based on digitized stratigraphic polygons, each corresponding to a specific geological age and lithological composition.

A thematic geological map covering the boundaries of the KazGer farm territory deposits ranges from the Late Precambrian to the Late Paleozoic. The geological cross-section is characterized by an alternation of terrigenous, carbonate, and, to a lesser extent, magmatic rocks, reflecting a change in tectonic regimes and marine facies conditions (

Table 3. Lithostratigraphic complexes by geochronological systems in Figure 8, the geological map of the KazGer farm territory.

Designation	Type	%	Lithology
	Ordovician, Middle, Lensko–Dolinsky Stage (OldLd)	27.4	deep-water clay shales, siltstones, siliceous deposits.
	Ordovician, Lensko–Dolinsky-Kugartsky (OldLd-k)	21.4	alternation of shales, tuffites, sandstones, possible basalt interlayers. interpretation: transition zone between tiers, possible signs of volcanic activity.
	Middle–Upper Devon (d2-3)	11.6	limestones, dolomites, more rarely sandstones and tuffaceous deposits. features: there are reef structures (bioherms, biostromes).
	Ordovician, Middle-Upper (O2-3)	11.1	limestones, dolomites, sometimes sandstones and siliceous shales. setting: fault-lined near-fault marine basins (possibly rift or transtension basins).
	Ordovician, Lower, Formation 3 (O13)	7.5	mainly siltstones, mudstones, and limestones; sandstones and siliceous shales occur in some places. shelf and coastal–marine basins, sedimentation in relatively shallow-water conditions.
	Ordovician, Kugart Stage (OkK)	4.4	limestones, carbonate–silicate rocks. environment: platform marine sediments (stable shelf).
	Carboniferous, Lower Visean 1-2 (CvV1-2)	3.9	carbonate–clay deposits (limestones, dolomites, clay shales). setting: marine, possibly with signs of lagoons.
	Carboniferous, Lower, Bashkir (Inzersky) Stage (CII)	3.8	coal-bearing strata, sandstones, siltstones. facies: lagoon–marine (alternating coastal and shallow-water deposits).
	Lower Visean 2–Serpukhov Carboniferous (CvV2-s)	3.4	fluvial–marine deposits (limestones, clay shales, carbonaceous interlayers). potential: promising for hydrocarbons.
	Devon, French–Famennian (D3fm)	3.0	limestones, dolomites, marls. conditions: carbonate platforms (shallow-sea basins).
	Vendian (V)	2.5	metamorphosed schists, quartzites, phyllites, and subordinate basalts. geological role: forms the foundation of the Paleozoic sedimentary cover.

).

The geological structure of the area studied is represented by sedimentary rocks of Paleozoic age, mainly Ordovician, Devonian, and carbonate–terrigenous formations. The section is based on limestones, dolomites, clay shales, siltstones, and sandstones, with siliceous and tuffaceous interlayers in some places. The most significant formations in terms of area include the following:

• Deep-water clay shales and siltstones of the OldLd formation O₂ld—27.4%. These deposits are represented by siliceous interlayers and were formed in a deep-sea basin.
• Siltstone and clay rocks of the OldLd-k formation—21.4%. There is an alternation of shales, tuffites, sandstones, and possible basalt interlayers, which indicates a transition zone between tiers and possible volcanic activity.
• Carbonate rocks of the D_2-3 formation—11.6%. This section is represented by limestones, dolomites, and more rarely sandstones and tuffaceous deposits. A special feature is the presence of reef structures (biotherm, biostrom).
• Limestones, dolomites, and siliceous shales of the O_2-3 formation—11.1%. These were formed in near-fault marine basins, under rift or transgressive conditions.
• Siltstones and mudstones of the O1₃ formation—7.5%. These sediments are typical of shelves and coastal–marine basins, which indicates shallow-water sedimentation conditions.
• Limestones of the O_k formation—4.4%. These are represented by carbonate–silicate rocks typical of platform deposits of a stable shelf.
• Carbonate–clay formation CvV1-2—3.9%. This section includes limestones, dolomites, and clay shales, probably of marine origin, with signs of lagoon conditions.
• Coal-bearing strata and sandstones of the C₁B formation C₁b(Lower Bashkir stage)—3.8%. These are formed in a lagoon–marine environment with alternating coastal and shallow-water sediments.
• Fluvial–marine deposits of the CvV2-s formation—3.4%. These include limestones, clay shales, and carbonaceous interlayers that are promising for hydrocarbons.
• Carbonate rocks of the D₁Fm formation—3.0%. The section is composed of limestones, dolomites, and marls that reflect the conditions of carbonate platforms in shallow marine basins.
• Conglomerates and metamorphosed shales of the V Formation—2.5%. These represent the basement of the Paleozoic sedimentary cover, composed of quartzites, phyllites, and subordinate shales.

The geological structure of farmland shows the presence of rocks ranging from carbonate (limestones, dolomites, marls) to terrigenous (clays, siltstones, sandstones) and coal-bearing strata. From a hydrogeological point of view,

• Sandstones and siltstones (O1₃, C₁b, C₁v2-s₂, D_2-3 ơ) have high permeability and can be considered as a promising reservoir for the accumulation and filtering of moisture;
• Limestone and dolomite (O_2-3 ơ, O_k, D₁fm, D_2-3 ơ) in the fractured zones can form a natural aquifer;
• Clay and mudstone strata (OldLd, OldLd-k, CvV1-2, V) are characterized by low filtration and can serve as a screen or water barrier, contributing to the preservation of moisture in the underlying horizons.

Thus, the combination of permeable sand–siltstone and carbonate rocks with clay screening layers creates favorable conditions for the accumulation of moisture and the formation of local aquifers. This allows us to consider this site as a promising location for MAR, being used for the underground storage facilities, keeping the soil moisture in favorable conditions. The soil cover is formed mainly by ordinary chernozems, characterized by a high thickness of the humus horizon (50–60 cm) and a humus content of up to 5–5.5% (Figure 10).

These soils (Figure 10) are characterized by high productivity, providing wheat yields above the regional average. However, a significant portion of the land is in drainage depressions where solonetzes and solonetzic chernozems have developed. In spring, these areas are prone to stagnant waterlogging and periodic flooding. The occurrence of shallow groundwater (2–4 m) with mineralization up to 2–3 g/L creates conditions for secondary salinization (Figure 11 and Figure 12).

For a more comprehensive and accurate analysis of soil salinity, seasonal variations characteristic of different times of the year were considered. In particular, the integration of RS data for spring, summer, and autumn makes it possible to identify the seasonal dynamics of salinity, reflecting the processes of salt accumulation and leaching during different vegetation periods. This approach contributes to the development of a more detailed map of salinity dynamics that considers key natural phenomena, including salinization after spring snowmelt and changes caused by autumn precipitation.

Using the Google Earth Engine platform, classification maps of soil salinity levels were generated for the study area for each season—spring, summer, and autumn—of 2024. The maps classify soils into five salinity categories: no salinity (blue), low (light blue), moderate (green), high (yellow), and very high (red) (Figure 11 and Figure 12).

The maps for the autumn and spring periods show that a high proportion of soils with elevated salinity levels are concentrated in areas with intensive agricultural activity, which may indicate improper and inefficient use of agricultural lands (Figure 13).

When comparing the maps, significant seasonal variations are observed in the distribution and intensity of soil salinity across the study area.

Spring and Autumn Periods:

• During the spring season (Figure 14), areas with high and very high salinity levels (yellow and red) dominate, particularly in the western part of the region.
• A considerable proportion of highly saline soils are found on agricultural fields, which may be associated with snowmelt and the concentration of salts on the soil surface because of evaporation.

Summer Period:

In the summer season (Figure 14), the salinity level significantly decreases across most areas, with an expansion of zones characterized by no or weak salinity (blue and light blue).

Moderate salinity (green) occupies the largest part of the territory, while the extent and concentration of highly and very highly saline zones are reduced.

This trend may be explained by the natural leaching of salts through precipitation and the increased activity of vegetation, which helps lower the soil salt content and can also affect the salinity index derived from remote sensing data.

Based on this data, processing was carried out to create an integrated map of the seasonal dynamics of soil salinity levels within KazGer farm using QGIS 3.40 software. This approach enabled the integration of multi-seasonal information and the identification of spatial temporal variations in soil salinity, significantly improving the quality of land resource monitoring and analysis.

The results obtained from the combined seasons demonstrate considerable spatial and temporal variability in soil salinity—patterns that cannot be captured when using single-season data alone (Figure 14).

The color scale of the map reflects the degree of seasonal dynamics of soil salinity:

• Blue corresponds to areas with no salinity changes throughout the three seasons (stable soil conditions).
• Light blue indicates minor dynamics, where salinity fluctuations are small and may be associated with localized seasonal processes.
• Green represents moderate seasonal dynamics, typical for zones moderately influenced by salinization factors.
• Yellow denotes strong changes, indicating pronounced seasonal variations in salinity likely related to the combined effects of natural and anthropogenic factors (e.g., spring snowmelt and summer precipitation).
• Red highlights areas with very strong seasonal dynamics, where the amplitude of salinity changes reaches maximum values, suggesting high soil sensitivity to seasonal hydrogeochemical processes and potential degradation risks.

The region exhibits significant spatial heterogeneity in the seasonal dynamics of soil salinity.

Most of the territory is characterized by moderate and low dynamics, indicating relatively stable soil conditions. In the farm, a sharp contrast in productivity is evident: on chernozems, cereal yields reach 15–18 dt/ha (decitons per hectare), while on salinized lands, they are no more than 8–10 dt/ha (Figure 15 and Figure 16). Forest shelterbelts are used to stabilize the agrolandscapes, but their condition has deteriorated in recent years.

The land structure of the KazGer farm in 2024 is characterized by a significant diversity of agricultural crops and the presence of fallow lands. The main cultivated areas are occupied by cereal crops, predominantly wheat and barley. These crops take up the largest contours by area, reflecting the farm’s traditional specialization in grain production. In addition, the land use structure includes industrial and leguminous crops—flax, lentil, and sunflower—distributed relatively uniformly within the land use boundary. A significant portion of the territory is represented by fallow lands, which play an important role in maintaining soil fertility, reducing degradation processes, and restoring agroecosystems. Their presence indicates the use of crop rotation schemes and agrotechnical practices aimed at sustainable land use.

In 2025, the land use structure of the KazGer farm underwent certain changes compared to 2024. Alfalfa was added to the previously cultivated crops, reflecting the expansion of the spectrum of perennial forage crops. The inclusion of alfalfa in the cropping structure indicates an increased role for feed production and an orientation toward the development of the livestock sector. The main crops remain wheat and barley, which occupy significant areas, although their spatial distribution varies. Flax, lentil, and sunflower are also retained in the structure, confirming the stable practice of combining cereal and industrial crops. The presence of fallow lands remains an important element of the agrolandscapes, ensuring the preservation of fertility and sustainable land use. Overall, the land use structure of the KazGer farm is distinguished by a balance between the production of cereal crops, industrial plantings, and the maintenance of a portion of the fallow land. The combined farm strategy is oriented simultaneously toward the production of commodity grain and the implementation of elements of sustainable agriculture.

3. Results

Assessment of the potential MAR replenishment on the KazGer farm with AHP-MCDA was carried out. The input data for the six main criteria were applied: slope, land use, soil type, soil salinity, geology, and precipitation.

• Slope. The slope of the terrain has a significant impact on the processes of infiltration and surface runoff. More gently sloping areas contribute to the accumulation of moisture and recharge of groundwater, while steep slopes allow rapid runoff of precipitation, reducing the potential for erosion damage. The global model FABDEM (Forest and Buildings removed Digital Elevation Model (DEM)) was used to analyze the terrain, which is an improved version of the Copernicus DEM data with the effects of vegetation and buildings removed, in preparing the digital terrain model (DTM).
• Land Use. Land use is an important factor in determining potential groundwater recharge zones. Agricultural land and pastures contribute to better infiltration compared to urban areas, where the surface is covered with impermeable materials. Land use data was taken from the ESA WorldCover 2024 global dataset, which provides classification with a resolution of 10 m.
• Soil. The type of soil determines its water permeability and ability to retain moisture. Sandy and sandy loam soils have high infiltration potential, while clay soils limit water penetration.
• Soil Salinity. Soil salinity has a direct impact on water permeability and on the possibility of using these areas for groundwater replenishment. High salt concentrations can limit filtration and degrade the quality of water entering the aquifer.
• Geology. The geological factor is key in determining the ability of rocks to accumulate and transfer groundwater. For the purposes of MAR modeling, all stratigraphic units were classified according to their water permeability and filtration capacity. Class 5—rocks with developed karst and fracturing, which have very high permeability. Class 4—carbonate and sandy rocks with good primary and secondary porosity. Class 3—sedimentary and magmatic complexes with mixed filtration properties. Class 2—clay and siliceous rocks with low water permeability. Class 1—dense igneous and metamorphic rocks with minimal filtration capacity. The largest area of the pilot zone belongs to the high (37.06%) and moderate (27.59%) classes, which indicates favorable conditions for the formation of underground runoff.
• Precipitation. Data from the ERA5-Land Monthly Aggregated-ECMWF Climate Reanalysis service was used to analyze precipitation. Within the pilot zone, 10 control points were identified, for which the values of total annual precipitation over the past 10 years were uploaded. Based on this data, the long-term average value for each point is calculated. The data source is the Historical Weather API, which is based on modern reanalysis models that combine information from weather stations; satellite observations; data from buoys and aircraft; and radar measurements. The applied model has a spatial resolution of 9 km, which makes it possible to consider local features of precipitation distribution in areas with complex terrain. As a result of the analysis, the average annual precipitation ranged from 366.63 mm to 445.07 mm.

The methodology was supplemented with a detailed description of approaches to data correction. To improve accuracy and reliability, the ERA5 data was corrected by Empirical Quantile Matching (EQM) based on local observations from the Borovoe SPM station. The validation results presented below (after correction, NBIAS = 0.00, NSE = 0.69) confirm the elimination of systematic bias and the suitability of the corrected data for further analysis.

Precipitation

To analyze the spatial and temporal distribution of precipitation in the pilot zone of KazGer farm, we used monthly aggregated ERA5 data from the ECMWF Copernicus Climate Reanalysis service. The original data was uploaded for 10 control points over a 25-year period (2014–2023).

Validation and Bias Correction

To ensure the accuracy of the results and to consider local features of precipitation distribution, the systematic bias correction of ERA5 data was performed using a reference series of observations from the Borovoe Stationary Meteorological Station (SPM).

• Primary Bias Analysis: Comparison of the initial ERA5 series and the observation series showed a significant systematic overestimation of precipitation by an average of USD 22 (NBIAS = 0.22). Figure 1, Figure 2 and Figure 3 clearly show that the average monthly precipitation profile of ERA5 systematically exceeds the observation profile. At the same time, the efficiency of the model was at a satisfactory level (NSE = 0.57), and the coefficient of determination was 0.704, which justifies the applicability of rank-based correction methods (Figure 17).
• Correction Method: Empirical Quantile Matching (EQM) was applied to eliminate systematic error and match the distributions. The correction function was calibrated based on data from the Borovoe SFM station and then applied to all 10 control points within the pilot zone (local offset correction method).
• Validation Results (After Correction): The correction (EQM) resulted in a significant improvement in data quality. The systematic bias was eliminated (NBIAS = 0.00), and the model efficiency increased to NSE = 0.69, which corresponds to the classification of “Very good” data quality.

Visualization of totals. After correction, the average monthly precipitation profiles are almost identical (Figure 18). The coefficient correlation between the series remains high throughout the year, and the scatter plot (Figure 19) confirms a high linear relationship

The Consistency Index (CI) for the analysis was 0.073, which is less than 0.1 and indicates the consistency of the decisions made. After standardization of all layers and determination of weights, they were combined into a single model reflecting the integral potential of MAR. The final map was calculated using the following equation:

«GWRSM = Slope × 0.2290 + LandUse × 0.1003 + Soil × 0.1744 + SoilSalinity × 0.0597 + Geology × 0.4063 + Precipitation × 0.0302GWRSM = Slope\times 0.2290 + LandUse\times 0.1003 + Soil\times 0.1744 + SoilSalinity\times 0.0597 + Geology\times 0.4063 + Precipitation\times 0.0302GWRSM = Slope × 0.2290 + LandUse × 0.1003 + Soil × 0.1744 + SoilSalinity × 0.0597 + Geology × 0.4063 + Precipitation × 0.0302»

The result is shown as a map with an index rating from 1 to 5, where

1—very low MAR potential;

5—very high MAR potential (Figure 20).

The results of the analysis showed that the zones with the lowest potential (Class 1) occupy only 0.03% of the total area. These sites are characterized by unfavorable conditions—local elevations and geology—that are unsuitable for realization of land use.

At the same time, the areas with the highest potential (Class 5) make up 12% of the territory. They are mainly concentrated in the central and southern parts of the study area, where favorable geological and topographical conditions prevail, such as flat terrain and permeable soils that promote infiltration.

In general, the map shows that a significant part of the territory has a medium-to-high internal MAR potential, which indicates that this area is promising for activities to replenish groundwater reserves using MAR. At the same time, it is important to consider the soil properties and heterogeneity of the geological structure when allocating priority zones for the implementation of MAR projects.

The DTM shows the terrain using a color scale (from green to red, indicating low and high areas, respectively) and horizontal lines. We also noted that this DTM was created based on horizontal lines, which we vectorized at intervals of 2.5 m to increase capacity for farmers to improve the landscape and land use strategies (Figure 21).

Moreover, the aerial survey was carried out on the territory of the KazGer farmland by using drones, unmanned aerial vehicles. Based on the materials obtained, in-house data processing was performed, because of which a digital terrain model (DTM) with a spatial resolution of 0.5 m/pixel was built, and orthomosaics were created for a part of the farmland (Figure 22) for comparative analysis.

The obtained DTM is considered one of the key criteria in the MAR for strengthening the topographic study of farmland. The high spatial resolution (0.5 m in plane and height) provides analysis of the topography and local changes in the relief landscape in more detail, to prepare recommendations for improvement, land use strategies for sustainable development, and potential MAR storage areas in more detail (Figure 23).

To justify the placement of the forest shelterbelts on the territory of the KazGer farm, a digital terrain model (DTM) was used. This DTM made it possible to identify the main features of the microrelief, including absolute elevation marks, and areas with depressions and elevated elements. The areas of low relief that are most susceptible to water erosion and stagnation of meltwater or rainwater, as well as elevated areas with increased vulnerability to wind soil erosion, were identified. In these zones, the feasibility of placing forest shelterbelts for various purposes was calculated.

–

In elevated areas, forest belts were oriented across the prevailing winds, which reduces wind speed and protects crops from deflation.

–

In lowlands, forest belts were formed to stabilize slopes and regulate surface runoff, which helps prevent soil erosion and improves the water balance of the agricultural landscape.

Thus, the use of a DTM made it possible to develop a higher-resolution scenario to design forest shelterbelts that consider the morphometry of the terrain. This approach should provide more sustainable protection of agricultural land from degradation processes and contribute to the formation of a sustainable land use structure (Figure 24).

The wind rose of the designed protective forest strips within the KazGer farm area, Akmola Region, is characterized by a predominance of westerly and northwesterly winds. The average wind speed is 1.09 m/s, the maximum is 4.3 m/s, and the climate is moderately windy, with no frequent storm events (Figure 25).

The wind rose chart shows that the most frequent winds blow from the west and northwest. Southern and eastern directions are rarely observed. This wind pattern indicates the need for protective measures against westerly winds, especially to protect farmland and human settlements.

The prevailing wind direction is westerly (about 25–30% of observations), followed by northwesterly (15%). Northerly winds account for about 10%, while eastern and southern winds account for less than 5%. Therefore, the main wind protection strips should be oriented perpendicularly to the direction of westerly winds, from north to south (

Table 4. Design parameters of protective forest strips.

Parameter	Recommendation
Lane type	Combined (tree and shrub)
Width	20–30 m
Number of rows	5–7 rows, alternating between high and low rocks
Planting density	70–80% (moderate to avoid swirl zones)
Distance between lanes	250–400 m for farmland, 100–150 m around settlements
Rock height	10–15 m

).

Frost- and drought-resistant types of tree and shrub species are suitable for the conditions of North Kazakhstan: trees—hanging birch (Betula pendula), balsam poplar (Populus balsamifera), ash-leaved maple (Acer negundo), and Scots pine (Pinus sylvestris); and shrubs—Tatar honeysuckle (Lonicera tatarica), Caraganaarborescens, arborescens May rose (Rosa majalis), and blood-red hawthorn (Crataegus sanguinea). Protective forest belts reduce wind speed by 40–60%, increase soil moisture, prevent wind erosion, create a favorable microclimate, and reduce dust in the air. Their use is especially important in areas with open terrain and farmland. The use of combined tree and shrub stands will provide effective soil protection, improve the microclimate, and increase agricultural productivity (Figure 26).

The current stage of the project report assessment is outlined as follows:

1.

Condition of Forest Belts and Yield (Appendix A,

Table A5. Crop yields of LLP A.I. Baraev Scientific and Production Center for Agricultural Development for 2023.

№	Culture	Variety	Reproduction	Area, ha	Yield, c/ha	Gross Harvest, c
1	2	3	4	5	6	7
8	wheat	Shortandinskaya 95 st	GR 2	9.0	6.4	58.0
35	wheat	Shortandinskaya 95 st	GR 3	57.0	5.7	326.0
35	wheat	Shortandinskaya 95 st	s/elite	209.0	9.4	1968.0
40	wheat	Shortandinskaya 95 st	elite	114.0	9.5	1088.0
34	wheat	Shortandinskaya 95 st	elite	200.0	6.9	1376.5
11	wheat	Shortandinskaya 95 st	elite	208.0	8.6	1778.5
3	wheat	Shortandinskaya 95 st	green manure	2.0	0	0
	Total			799.0	8.3	6595.0
8	wheat	Shortandinskaya 2014	GR 2	16.0	10.6	169.0
7	wheat	Shortandinskaya 2014	GR 3	20.0	10.1	201.5
34	wheat	Shortandinskaya 2014	s/elite	174.0	9.8	1707.0
9	wheat	Shortandinskaya 2014	s/elite	26.0	10.1	262.0
5	wheat	Shortandinskaya 2014	s/elite	20.0	7.7	153.5
10	wheat	Shortandinskaya 2014	elite	113.0	11.2	1268.5
6	wheat	Shortandinskaya 2014	elite	31.0	12.3	380.5
	Total			400.0	10.4	4142.0
8	wheat	Shortandinskaya 2012	GR 2	8.0	15.4	123.0
10	wheat	Shortandinskaya 2012	GR 3	23.0	7.7	176.5
10	wheat	Shortandinskaya 2012	s/elite	50.0	9.9	492.5
8	wheat	Shortandinskaya 2012	elite	44.0	6.6	289.0
7	wheat	Shortandinskaya 2012	elite	203.0	10.2	2063.0
5	wheat	Shortandinskaya 2012	elite	35.0	9.3	324.5
9	wheat	Shortandinskaya 2012	elite	191.0	7.8	1495.5
40	wheat	Shortandinskaya 2012	commodity	8.0	8.7	69.5
8	wheat	Shortandinskaya 2012	commodity	20.0	5.0	100.5
3	wheat	Shortandinskaya 2012	commodity	2.0	4.0	8.0
	Total			584.0	8.8	5142.0
11	wheat	Astana 2	elite	100.0	10.5	1047.0
9	wheat	Astana 2	s/elite	4.0	10.9	43.5
9	wheat	Astana 2	commodity	7.0	0.9	6.3
3	wheat	Astana 2	commodity	2.0	4.0	8.0
	Total			113.0	9.8	1104.8
11	wheat	Astana	s/elite	75.0	11.1	835.5
11	wheat	Astana	elite	56.0	10.7	596.5
4	wheat	Astana	commodity	14.0	5.6	79.0
	Total			145.0	10.4	1511.0
10	wheat	Taimas	GR 2	5.0	11.2	56.0
10	wheat	Taimas	GR 3	24.0	9.6	230.5
10	wheat	Taimas	s/elite	38.0	7.0	267.0
	Total			67.0	8.3	553.5
2	wheat	Independence 20	GR 3	5.0	13.1	65.5
2	wheat	Independence 20	s/elite	20.0	8.3	165.5
2	wheat	Independence 20	elite	25.0	9.5	238.0
	Total			50.0	9.4	469.0
1	2	3	4	5	6	7
8	wheat	Akmola 2	GR 3	11.0	11.1	122.5
5	wheat	Akmola 2	s/elite	20.0	7.7	153.5
5	wheat	Akmola 2	elite	24.0	7.7	184.0
8	wheat	Akmola 2	elite	12.0	9.2	110.5
40	wheat	Akmola 2	commodity	4.0	4.5	18.0
	Total			71.0	8.3	588.5
5	wheat	Celina 50	s/elite	15.0	5.4	81.0
5	wheat	Celina 50	elite	36.0	7.3	261.0
	Total			51.0	6.7	342
8	wheat	Asyl Sapa	GR 2	17.0	8.0	136.5
7	wheat	Asyl Sapa	GR 3	20.0	5.8	115.0
7	wheat	Asyl Sapa	s/elite	15.0	7.6	114.5
5	wheat	Asyl Sapa	elite	28.0	9.3	261.0
	Total			80.0	7.8	627.0
8	durum wheat	Korona	GR 2	6.0	10.4	62.5
10	durum wheat	KOrona	GR 3	18.0	8.3	149.0
9	durum wheat	Damsinskaya amber	s/elite	0.5	7.6	3.8
2	durum wheat	Damsinskaya 2017	GR 2	6.5	3.6	23.5
2	durum wheat	Lavina	GR 2	2.0	5.1	10.2
2	durum wheat	Lavina	GR 3	10.0	3.5	35.3
	Total			43.0	6.6	284.3
				2403.0	8.9	21,359.0
5	barley	Virgin Lands 2005	s/elite	28.0	8.3	233.5
8	barley	Astana 2000	elite	20.0	8.3	166.5
40	barley	Astana 2000	1 rep	175.0	11.7	2043.0
40	barley	Astana 2000	commodity	12.0	9.7	116.5
4	barley	Astana 2000	commodity	15.0	8.2	123.5
	Total			250.0	10.7	2683.0
7	oats	Bitik	s/elite	24.0	4.8	114.5
9	oats	Bayzat	elite	2.0	13.8	27.5
1	oats	Duman	commodity	1.0	2.5	2.5
	Total			27.0	5.4	144.5
3	flax	Kustanai amber	commodity	2.0	2.5	5.0
8	flax	Kustanai amber	commodity	42.0	2.7	111.5
3	flax	Kustanai amber	commodity	42.0	1.9	80.0
11	flax	Kustanai amber	commodity	33.0	2.1	70.0
6	flax	Kustanai amber	commodity	25.0	2.0	50.0
2	flax	Kustanai amber	commodity	25.0	2.7	67.0
1	flax	Kustanai amber	commodity	55.0	1.9	105.5
	Total			224.0	2.2	489.0
3	buckwheat	Shortandinskaya 4	GR 3	5.0	2.0	10.0
3	buckwheat	Shortandinskaya 4	GR 3	19.0	2.1	40.0
7	buckwheat	Shortandinskaya 4	s/elite	44.0	0.7	30.3
6	buckwheat	Shortandinskaya 4	s/elite	16.0	0.7	11.2
2	buckwheat	Shortandinskaya 4	s/elite	50.0	1.6	81.5
	Total			134.0	1.3	173.0
40	rape	Maikudyk	GR 3	12.0	0.7	8.0
10	rape	Osiris	GR 3	14.0	5.7	79.5
	Total			26.0	3.4	87.5
11	mustard	Commodity	commodity	90.0	8.0	723.5
	Total			90.0	8.0	723.5
2	Sudanese grass	Nika	GR 2	14.0	3.7	51.5
2	Sudanese grass	Nika	GR 3	11.0	5.7	62.5
	Total			25.0	4.6	114.0
34	sunflower	Zhaidarman	GR 3	13.0	2.4	31.0
	Total			13.0	2.4	31.0
4	peas	Kasib	commodity	5.0	6.2	31.0
3	peas	Aksai mustachioed	commodity	2.0	5.0	10.0
	Total			7.0	5.9	41.0
	Sum			3199.0	8.1	25,845.6

and

Table A6. Crop yields of A.I. Baraev Scientific and Production Center for Agricultural Production for 2024.

№	Culture	Variety	Reproduction	Area, ha	Yield, c/ha	Gross Collection, t
1	2	3	4	5	6	7
40	wheat	Shortandinskaya 95 st	GR 2	12.7	17.3	22.0
40	wheat	Shortandinskaya 95 st	GR 3	19.9	16.1	31.9
39	wheat	Shortandinskaya 95 st	elite	4.5	15.3	6.9
35	wheat	Shortandinskaya 95 st	elite	267.0	13.1	351.0
34	wheat	Shortandinskaya 95 st	elite	382.0	13.2	502.8
40	wheat	Shortandinskaya 95 st	elite	177.4	15.7	277.7
40	wheat	Shortandinskaya 95 st	s/elite	28.0	21.5	60.2
	Total			891.5	16.0	1252.6
5	wheat	Shortandinskaya 2014	GR 3	15.0	25.5	38.3
5	wheat	Shortandinskaya 2014	GR 2	10.0	27.6	27.6
5	wheat	Shortandinskaya 2014	elite	15.0	27.3	41.0
8	wheat	Shortandinskaya 2014	elite	70.0	13.8	96.5
39	wheat	Shortandinskaya 2014	elite	4.5	16.3	7.3
6	wheat	Shortandinskaya 2014	s/elite	4.0	11.6	4.6
10	wheat	Shortandinskaya 2014	s/elite	113.0	23.0	268.8
	Total			231.5	20.7	484.3
4	wheat	Shortandinskaya 2012	GR 2	8.0	32.8	26.3
10	wheat	Shortandinskaya 2012	GR 3	20.0	15.6	31.3
10	wheat	Shortandinskaya 2012	elite	62.0	15.2	94.2
7	wheat	Shortandinskaya 2012	elite	200.0	19.7	394.0
10	wheat	Shortandinskaya 2012	s/elite	66.0	22.4	147.7
39	wheat	Shortandinskaya 2012	commod	4.5	9.6	4.3
	Total			360.5	19.2	697.9
11	wheat	Astana 2	elite	26.0	11.3	29.5
11	wheat	Astana	s/elite	240.0	17.3	415.6
3	wheat	Astana	s/elite	6.0	22.0	13.2
6	wheat	Astana	s/elite	25.0	21.9	54.7
7	wheat	Astana	elite	44.0	12.6	55.5
9	wheat	Astana	elite	157.0	16.8	263.7
4	wheat	Astana	commod	12.0	2.6	3.2
	Total			510.0	14.9	835.4
39	wheat	Taimas	GR 2	10.0	32.5	32.5
39	wheat	Taimas	GR 3	20.0	33.1	66.1
39	wheat	Taimas	s/elite	50.0	38.3	191.6
4	wheat	Taimas	s/elite	24.0	14.3	34.4
39	wheat	Taimas	s/elite	54.5	22.5	10.1
	Total			158.5	28.1	334.9
5	wheat	Akmola 2	GR 3	26.0	21.6	56.3
5	wheat	Akmola 2	s/elite	18.0	31.1	55.95
	Total			44.0	26.3	112.2
5	wheat	Virgin Land 50	s/elite	36.0	18.6	66.85
4	wheat	Asyl Sapa	GR 2	7.0	30.9	21.65
5	wheat	Asyl Sapa	GR 3	28.0	24.3	68.15
3	wheat	Asyl Sapa	s/elite	17.0	14.5	24.6
	Total			88.0	22.0	181.2
2	durum wheat	Crown	GR 3	36.0	29.7	106.9
8	durum wheat	Crown	GR 3	16.0	22.8	36.5
39	durum wheat	Crown	s/elite	4.5	14.4	6.5
8	durum wheat	Damsinskaya 2017	GR 2	11.0	32.6	35.8
8	durum wheat	Damsinskaya 2017	GR 3	8.0	25.6	20:4
39	durum wheat	Damsinskaya 90	GR 3	4.5	14.2	6.4
	Total			80.0	23.3	212.6
				2364.0	29.4	4111.0
2	barley	Tselinny 60	elite	186.0	26.1	486.45
8	barley	Tselinny 60	elite	20.0	23.0	46.05
39	barley	Tselinny 60	commodity	4.5	38.0	17.1
4	barley	Virgin Lands 2005	s/elite	23.5	27.6	64.85
8	barley	Astana 2000	elite	42.0	38.0	159.7
	Total			276.0	30.5	774.1
5	oats	Bayzat	GR 2	5.0	27.2	13.6
5	oats	Duman	GR 3	10.0	34.3	34.3
39	oats	Duman	commodity	4.5	22.0	9.8
7	oats	Bitik	GR 3	20.0	33.5	67.0
5	oats	Bitik	s/elite	52.5	34.0	178.4
	Total			92.0	30.2	303.2
5	buckwheat	Shortandinskaya 4	s/elite	22.0	12.60	27.8
9	buckwheat	Shortandinskaya 4	s/elite	25.0	14.60	36.5
3	buckwheat	Shortandinskaya 4	commodity	16.0	16.50	26.4
	Total			98.0	14.0	133.1
7	sunflower	Kun-Nur	GR 3	16.0	7.1	11.3
6	sunflower	Zhaidarman	GR 3	31.0	5.5	17.1
				47.0	6.3	28.5
4	peas	Status	commodity	2.0	4.5	0.9
4	peas	Orys	commodity	26.0	6.2	16.2
				28.0	5.3	17.1
10	rape	Maykudyk	s/elite	24.0	11.2	26.8
5	flax	Kustanai amber	commodity	35.0	12.1	42.4
5	safflower	commodity	commodity	13.0	3.6	4.7
9	lentils	commodity	commodity	40.0	14.8	59.4
	Total			112.0	10.4	133.3
	Sum			3017.0	18.0	5500.3

, Figure A1, Figure A2, Figure A3, Figure A4, Figure A5, Figure A6, Figure A7, Figure A8 and Figure A9).

–

Loss of Forest Belt Function: Field-protective forest belts in the investigated farms of the Akmola Region, including A. Barayev, Yesil-Agro, Rodina, KazGer, established in the 1970s–1980s, have lost their original purpose due to a lack of maintenance and reconstruction.

–

Need for Reconstruction: A forestry and ecological assessment was conducted, which showed that the forest belts require fundamental reconstruction (cutting down up to 80% of dry wood, thinning, planting new seedlings).

–

Yield Growth: A significant increase in the average yield of agricultural crops was noted in 2024 (up to 18 c/ha) compared to 2023 (8.1 c/ha), which may be partially attributed to improved moisture availability, although the forest belts currently do not have a significant impact in their present state.

2.

Application of Geospatial Technologies (GIS and RS) (Appendix A,

Table A7. Classification of land use categories by infiltration conditions for assessing MAR potential.

Class	Land Use Categories	Characteristic
5	meadow lands, shrubs	the most favorable areas with high permeability and minimal restrictions for infiltration
4	agricultural land	agricultural land with relatively high infiltration capacity
3	bare/sparse vegetation	moderate conditions: limited permeability of soils and vegetation
2	grassy wetlands, forest/tree cover	low permeability due to swampiness or dense forest vegetation
1	permanent water bodies, built-up areas	the least favorable areas where infiltration is practically impossible

,

Table A8. Requirements for planting material (saplings) used for forest restoration and afforestation.

№	Name of Breeds	Forest Vegetation Zones	Age, Years	Variety	Thickness of the Trunk at the Root Collar, Not Less Than, mm	Height of Aboveground Part, Not Less Than, cm
1	2	3	4	5	6	7
1	Silver birch Betula pendula Roth	all zones	3–4	1	8	50
				2	5	35
2	Small-leaved elm (elm) Ulmus pinnatoramosa	forest-steppe, steppe	2–3	1	8	55
				2	6	40
3	Common pear Pirus communis L.	all zones	2–3	1	7	45
				2	5	30
4	Caragana arborescens Caragana Arborescens.	all zones	3–4	1	6	35
				2	4	25
5	Norway maple Acer platanoides L.	all zones	3–4	1	8	35
				2	6	25
6	Small-leaved linden Tilia cordata Mill.	all zones	3–4	1	9	50
				2	5	30
7	Sea buckthorn Hippophae ramnoides L.	all zones	3–4	1	9	35
				2	7	25
8	Rowan tree Sorbus aucuparica L.	all zones	3–4	1	9	35
				2	7	25
9	Scots pine Pinus silvestris L.	forest-steppe	3–4	1	8	25
				2	5	20
10	White poplar Populus alba L.	forest-steppe, steppe	2–3	1	10	100
				2	7	70
11	Black poplar Populus nigra L.		2	1	7	80
				2	6	60
12	Wild apple tree Malus silvestris (L.)	all zones	2–3	1	8	45
				2	6	30
13	Common ash Flaxinus excelsior L.	all zones	3–4	1	9	35
				2	7	25

,

Table A9. Trees and shrubs recommended for the creation of protective forest belts and biogroups in the Akmola region.

№	Names of Trees and Shrubs	Indexes
1	Beresa pendula (Betula pendula)	BB
2	Scots pine (Pinus sylvestris)	So
3	Elaeagnus angustifolia	Lh
4	Siberian elm (Ulmus pumilia)	Vz
5	White poplar (Populus alba)	Tb
6	Siberian rowan (Sorbus sibirika)	PC
7	Currants (Ribes)	Cm

,

Table A10. Principal Component Classification for LVI–IPCC calculation.

Factors and Main Components of IPCC
Contact	Human–wildlife conflict Natural disasters and climate change
Adaptive capacity	Life support strategies Natural resources Social media Infrastructure Socio-demographic profile Earth Finance and income
Sensitivity	Agriculture and food security Healthcare Type of housing Water resources and sanitation

,

Table A11. Structure of the Livelihood Vulnerability Index (LVI).

IPCC Factors	Main Components—14	Examples of Subcomponents—56
Exposure	natural disasters and climate change	frequency of droughts, hail, livestock losses from predators
Sensitivity	health; agriculture and food security; water resources and sanitation	time to market; access to clean water; availability of permanent housing
Adaptive Capacity	socio-demographic profile; livelihood strategies; social networks; natural resources; infrastructure; finance and income	level of education; access to credit; income diversification

,

Table A12. LVI by components and subcomponents for districts of Akmola Region.

Component/Subcomponent	Data Source	Birzhan Sal	Burabaysky	Tselinogradsky	Shortandinsky	General (cf.)
1	2	3	4	5	6	7
SDP (socio-demographic profile)	statistics + survey	0.389	0.702	0.705	0.669	0.616
- % of female heads of households	statistics	0.767 (76.7%)	0.935 (93.5%)	0.904 (90.4%)	0.885 (88.5%)	0.873
- Population density (people/km2)	statistics	0.182 (1.1)	0.909 (11.9)	0.864 (11.0)	0.818 (5.7)	0.693
- Size DH (persons)	survey	0.217 (1.3)	0.261 (1.6)	0.348 (2.1)	0.304 (1.8)	0.283
LS (livelihood strategies)	statistics	0.500	0.429	0.626	0.526	0.520
- Cultivated area/person (ha/person)	statistics	0.654 (0.48)	0.571 (0.47)	0.812 (0.51)	0.948 (0.53)	0.746
- Cost of labor (tenge/worker)	statistics	0.612 (2803)	0.613 (2643)	0.641 (2762)	0.531 (2027)	0.599
- Number of livestock (heads)	statistics	0.235 (98,351)	0.103 (80,740)	0.426 (419,387)	0.100 (62,162)	0.216
Health	statistics for the region	0.258	0.258	0.258	0.258	0.258
- % stunting <5 years	S (UNICEF MICS 2024)	0.162 (16.2%)	0.162 (16.2%)	0.162 (16.2%)	0.162 (16.2%)	0.162
- Full vaccination 15–26 months, %	S (UNICEF MICS 2024)	0.380 (62%)	0.380 (62%)	0.380 (62%)	0.380 (62%)	0.380
- ECD index (%)	S (UNICEF MICS 2024)	0.139 (86.1%)	0.139 (86.1%)	0.139 (86.1%)	0.139 (86.1%)	0.139
- Low birth weight, %	S (UNICEF MICS 2024, North Kaz)	0.090 (9%)	0.090 (9%)	0.090 (9%)	0.090 (9%)	0.090
- Improved water sources, %	S (UNICEF MICS 2024)	0.060 (94%)	0.060 (94%)	0.060 (94%)	0.060 (94%)	0.060
Food	statistics	0.404	0.430	0.463	0.388	0.421
- Grain yield, c/ha	statistics	0.421 (8.9)	0.421 (7.5)	0.447 (9.1)	0.368 (5.2)	0.414
- Vegetable yield, c/ha	statistics	0.387 (171.7)	0.439 (175.4)	0.479 (95.8)	0.409 (163.6)	0.429
1	2	3	4	5	6	7
Water	data from the Ministry of Emergency Situations	0.220	0.340	0.170	0.280	0.252
- % with water deficit	survey	0.200 (20%)	0.300 (30%)	0.150 (15%)	0.250 (25%)	0.225
- Flood adjustment	data from the Ministry of Emergency Situations	+0.020 (1)	+0.040 (2)	+0.020 (1)	+0.030 (1.5)	+0.028
Social networks	A (survey, 1.9/4.6/4.8/4.9)	0.525	0.525	0.550	0.550	0.538
- % with technologies	A (poll, 4.6)	0.450 (0.55)	0.500 (0.50)	0.450 (0.55)	0.400 (0.60)	0.450
- Financial literacy, %	A (poll, 1.9/4.8/4.9)	0.600 (0.40)	0.550 (0.45)	0.650 (0.35)	0.700 (0.30)	0.625
Exposure (climate risks)	data from the Ministry of Emergency Situations	0.305	0.405	0.255	0.355	0.330
- Frequency of floods (cases)	data from the Ministry of Emergency Situations	0.300 (1.5)	0.400 (2)	0.250 (1.25)	0.350 (1.75)	0.325
- Climate risks (low precipitation index)	Kazhydromet	0.310 (0.155)	0.410 (0.205)	0.260 (0.13)	0.360 (0.18)	0.335
Final LVI	aggregation	0.380	0.447	0.433	0.415	0.419
LVI-IPCC	aggregation	−0.045	0.040	−0.025	0.035	0.001

and

Table A13. Design data for organizing work on contour-strip organization of the territory using the Lukin–Potapenko–Zverev method.

Minimum plot	100 hectares	Explanation. The method requires digging windrow ditches per hectare: 50 m wide, 80 cm wide, and 1.5 m deep. In general, windrow ditches must be dug at intervals of 100 to 500 m, depending on the surface slope. The windrow ditches are planted with trees and shrubs at the top and bottom, and the bottom is lined with biomass. A topographic survey is required for planning, based on which the territory is planned. Soil monitoring is required at the entrance and exit for assessment. Monitoring visits are carried out throughout the entire cycle.
Cycle	3 years
Territory	Pastures/arable lands
Distance between stripes, meters	300

).

–

Comprehensive Survey: A reconnaissance survey of forest belts and terrain was conducted by using GIS and RS technologies.

–

MAR Assessment: The field-protective potential of the forest belts was determined using the MAR technology and the Lukin–Potapenko–Zverev model.

3.

Water Resources and Economic Effect (Appendix A,

Table A14. Questionnaire for households.

A. General information about the farm

1. Please indicate your age: years

2. Your gender: (a) Male (b) Female

3. Your level of education: (a) General average (b) Secondary specialized (c) Higher agricultural (d) Higher other

4. Number of family members, including you: people.

5. Number of workers in your household (except family members): _ people.

6. Total area of your agricultural land: ha.

7. Main type of agricultural activity: (a) Plant growing (b) Animal husbandry (c) Mixed type of farming

B. Vulnerability to climate risks

8. What climate events have most often negatively impacted your farm over the past 5 years? (You can choose several options). (a) Drought (b) Excessive precipitation/flooding (c) Strong winds/dust storms (d) Sudden temperature changes/freezing (e) Other (specify): _____________________

9. How often have droughts caused significant crop losses over the past 5 years? (a) Every year (b) once every 2–3 years (c) Rarely or almost never happens

10. Average level of crop losses on the farm due to weather conditions over the past 5 years: (a) Less than 10% (b) 10–30% (c) 30–50% (d) More than 50%

C. State of the economy’s natural resources (sensitivity)

11. How would you rate the general condition of the soils on your farm? (a) Good (b) Satisfactory (c) Unsatisfactory (requires improvement)

12. What are the main soil problems you notice on your plots? (You can choose several options). (a) Wind erosion (b) Water erosion (c) Formation of ravines (d) Soil salinization

(e) Lack of moisture (f) Loss of fertility (decrease in humus) (g) Others (specify):

13. Do you have enough water for farming? (a) There is enough water (b) There is a periodic deficit (c) The water shortage is constant and significant

14. Do you have forest shelterbelts in your fields? (a) Yes (b) No (if no, go to the next block)

15. How do you assess the condition of these forest belts today? (a) Good (b) Satisfactory (c) Poor (significant degradation or drying out)

16. What are the main causes of forest belt degradation on your farm? (You can choose several options.) (a) High costs for their maintenance and care (b) Age of trees (c) Drought and lack of moisture (d) Deforestation due to expansion of cultivated areas (e) Felling due to negative impact on the start of sowing work (f) Damage by insect pests (g) Other (specify): ____________________

17. What moisture-saving technologies are used in the fields? (Several options are possible) (a) Drip irrigation (b) Mulching (c) Soil fissure formation (d) Planting forest belts (e) Snow retention (f) Other (specify): ____________________ (g) No, it is not used.

D. Readiness to adapt (adaptive capacity)

18. Are you familiar with the method of restoring the hydrological surface (ditches, tree planting, contour-strip organization of fields)? (a) Yes, I know him well. (b) I heard, but I don’t know the details. (c) No, I’m not familiar.

19. Are you interested in implementing this method on your farm? (a) Yes, definitely (b) Possibly, with support (c) No, I’m not interested.

20. What support do you need to participate in such projects? (a) financial support (b) Training and consultations (c) Machinery and equipment (d) Seedlings and planting material (e) Other (specify):

,

Table A15. Questionnaire for experts.

Information about the expert:

Full name:

Profession/specialization:

Work experience in the specialty: years

1. Shortandy District - What are the key features of the relief and soil-hydrological conditions in the Shortandy district that should be considered when planning windrow ditches and shelterbelts? - Which areas in the vicinity of the KazGer farm are the most promising for the implementation of the method of restoring the hydrological surface (considering the contour-strip organization)?

2. Birzhan salt district - What natural and landscape factors (e.g., topography, soil, water regime) typical for the Birzhan Sal region should be considered when organizing and planting forest belts? - In what specific places in the specified rural districts would you recommend creating new or restoring existing forest belts, considering climatic and soil characteristics?

3. Tselinogradsky district - What are the specific natural conditions of the Tselinograd region (relief, soil type, water availability, etc.) that are important for the successful application of the method of restoring the hydrological surface and creating shelter belts? - Can you indicate specific areas of the Tselinograd district that are most suitable for the implementation of adaptation measures (windrow ditches, tree planting)?

4. Burabay district - What soil and hydrological characteristics (e.g., proximity to groundwater, presence of seasonal reservoirs) should be considered in the Burabay district when selecting sites for creating windrow ditches and shelterbelts? - Which area in the vicinity of the Burabay district is most promising for the restoration of the hydrological regime and shelterbelt forest plantations?

5. General recommendations (for all specified areas) - Which tree and shrub species do you consider most suitable for planting shelterbelts in the specified areas of the Akmola region and why? - What technical or natural risks should be considered when implementing the method of restoring the hydrological surface and creating forest belts in the specified areas?

6. What measures do you recommend to include in the project to ensure the long-term sustainability and effectiveness of the created shelter belts (monitoring, maintenance, regular care)?

7. Additional comments and suggestions Please provide any additional arguments and comments that you believe are important to consider when implementing the project in the specified districts and settlements of the Akmola region.

and

Table A16. Principal Component Classification for LVI–IPCC calculation.

Factors and Main Components of IPCC
Contact	Human–wildlife conflict Natural disasters and climate change
Adaptive capacity	Life support strategies Natural resources Social media Infrastructure Socio-demographic profile Earth Finance and income
Sensitivity	Agriculture and food security Healthcare Type of housing Water resources and sanitation

, Figure A10, Figure A11, Figure A12, Figure A13, Figure A14, Figure A15 and Figure A16).

–

Identification of Water Accumulation Sites: Geodetic and topographic surveys were carried out, which made it possible to determine potential locations for the accumulation of drainage meltwater on agricultural lands.

–

Agroclimatic Improvement: Forest shelterbelts in combination with MAR and considering the terrain are expected to lead to improved climatic conditions in the region and increased crop yields in the future.

4.

Additional Research (Appendix A,

Table A17. Organoleptic indicators of surface waters.

Research Objects	Smell, Points	Transparency, cm	Color, Degree	Suspended Solids, mg/dm3
Reservoir №1	0	>20	<20	<0.25
Reservoir №2	0	>20	<20	<0.25
Reservoir №3	0	>20	<20	<0.25
Reservoir №4	0	>20	<20	<0.25
Reservoir №5	0	>20	<20	<0.25
Reservoir №6	0	>20	<20	<0.25
Reservoir №7	0	>20	<20	<0.25
Reservoir №8	0	>20	<20	<0.25
Reservoir №9	0	>20	<20	<0.25
Reservoir №10	0	>20	<20	<0.25
Reservoir №11	0	>20	<20	<0.25
Reservoir №12	0	>20	<20	<0.25

,

Table A18. Water mineralization indicators of water bodies.

Object of Study	Standardized Indicators, mg/dm3								GH, mg-eq/L
	SO42−	Cl−	Ca2+	Mg2+	∑Na + K	HCO3−	Dry Remainder	Totall Mineralization
Reservoir №1	38.0	14.0	52.0	6.0	43.0	232.0	270.0	386.0	3.1
Reservoir №2	19.0	35.0	66.0	27.0	53.0	391.0	396.0	591.0	5.5
Reservoir №3	19.0	14.0	36.0	6.0	34.0	183.0	202.0	293.0	2.3
Reservoir №4	9.0	14.0	44.0	11.0	39.0	256.0	246.0	374.0	3.1
Reservoir №5	19.0	7.0	52.0	6.0	25.0	220.0	220.0	329.0	3.1
Reservoir №6	14.0	18.0	38.0	17.0	34.0	244.0	244.0	366.0	3.3
Reservoir №7	215.0	585.0	160.0	85.0	285.0	391.0	1527.0	1723.0	15.0
Reservoir №8	7.0	16.0	34.0	6.0	18.0	146.0	155.0	228.0	2.2
Reservoir №9	14.0	11.0	36.0	5.0	18.0	146.0	158.0	231.0	2.2
Reservoir №10	1681.0	3368.0	441.0	425.0	1786.0	287.0	7845.0	7989.0	57.0
Reservoir №11	1680.0	3368.0	441.0	437.0	1766.0	293.0	7840.0	7986.0	58.0
Reservoir №12	96.0	195.0	52.0	33.0	152.0	268.0	662.0	796.0	5.3

,

Table A19. Content of biogenic substances in water bodies.

Object of Study	Biogenic Substances, mg/dm3
	NO3−	NH4+	Ptotal
Reservoir №1	0.6	1.33	1.431
Reservoir №2	0.4	0.23	0.330
Reservoir №3	<0.3	0.21	0.111
Reservoir №4	0.5	0.40	0.469
Reservoir №5	<0.3	0.20	0.152
Reservoir №6	0.7	0.54	13.026
Reservoir №7	1.9	-	-
Reservoir №8	<0.3	-	-
Reservoir №9	<0.3	-	-
Reservoir №10	<0.3	-	-
Reservoir №11	0.9	-	-
Reservoir №12	0.4	-	-

,

Table A20. Content of heavy metals and arsenic in water bodies.

Object of Study	Standardized Indicators, mg/dm3
	Al3+	Be2+	∑Fe2, Fe3+	Pb2+	Hgtotal	Zn2+	Sr2+	Astotal
Reservoir №1	10.400	0.0006	0.16	0.006	0.00117	0.0430	0.2674	<0.005
Reservoir №2	1.276	<0.0001	-	<0.001	0.00069	0.0103	0.6398	<0.005
Reservoir №3	0.708	<0.0001	-	<0.001	0.00044	0.0110	0.2553	<0.005
Reservoir №4	0.652	<0.0001	0.05	<0.001	0.00020	0.0101	0.2962	<0.005
Reservoir №5	0.457	<0.0001	0.16	<0.001	0.00018	0.0124	0.2545	<0.005
Reservoir №6	19.675	<0.0001	0.70	0.013	0.00032	0.1175	0.7207	<0.005
Reservoir №7	0.085	<0.0001	<0.05	<0.001	-	<0.005	1.9410	<0.005
Reservoir №8	0.123	<0.0001	<0.05	<0.001	-	<0.005	0.2125	<0.005
Reservoir №9	0.131	<0.0001	-	0.0012	-	<0.005	0.4729	<0.005
Reservoir №10	0.092	<0.0001	-	<0.001	-	<0.005	6.0340	<0.005
Reservoir №11	0.349	<0.0001	-	<0.001	-	0.0051	6.0380	<0.005
Reservoir №12	0.080	<0.0001	-	<0.001	-	<0.005	0.8249	<0.005

,

Table A21. Pesticide content in water bodies, mg/dm³.

Object of Study	g-HCH (Lindane)	DDT (Sum of Isomers)	2, 4D
Reservoir №7	<0.00001	<0.00001	<0.0003
Reservoir №8	<0.00001	<0.00001	<0.0003
Reservoir №9	<0.00001	<0.00001	<0.0003
Reservoir №10	<0.00001	<0.00001	<0.0003
Reservoir №11	<0.00001	<0.00001	<0.0003
Reservoir №12	<0.00001	<0.00001	<0.0003

,

Table A22. Oxygen conditions of water bodies.

Object of Study	CODbihr, mgO2/dm3	BOD5, mgO2/dm3	BOD20, mgO2/dm3	pH Environment
Reservoir №1	67.8	50.4	-	6.9
Reservoir №2	75.2	44.8	-	7.5
Reservoir №3	76.2	33.4	-	7.3
Reservoir №4	75.7	30.4	-	7.1
Reservoir №5	54.4	23.6	-	7.0
Reservoir №6	52.7	35.2	-	6.7
Reservoir №7	47.7	6.7	34.6	6.4
Reservoir №8	12.6	1.3	5.9	7.3
Reservoir №9	15.7	2.7	4.5	7.5
Reservoir №10	45.6	2.9	5.1	7.7
Reservoir №11	27.7	3.9	8.9	7.8
Reservoir №12	33.3	9.3	12.9	7.9

,

Table A23. Gross content of chemical elements in soil samples (mg/kg) taken from dried-up meltwater ponds on the agricultural lands of KazGer farm.

№	Element	Symbol	MAC mg/kg	Clarke According to Wedepohl, mg/kg	Total Content of Elements in Samples, mg/kg
					№1	№2	№3	№4
1	Copper	Cu	33.0	30.000	32.26	37.18	21.99	33.48
2	Cadmium	CD	0.5	-	<3.00	<3.00	<3.00	<3.00
3	Lead	Pb	32.0	15.000	22.02	25.85	22.04	21.12
4	Manganese	Mn	1500.0	690.000	647.71	920.21	775.59	617.88
5	Arsenic	As	2.0	1.700	14.27	15.15	13.35	20.33
6	Aluminum	Al	-	78,300	80,693.59	72,287.15	59,135.18	69,340.57
7	Bismuth	Bi	-	0.200	<1.00	<1.00	<1.00	<1.00
8	Iron	Fe	-	35,400	48,164.24	40,175.00	36,744.65	38,934.95
9	Cobalt	Co	-	12.000	18.66	16.72	15.23	15.51
10	Hafnium	Hf	-	3.000	<5.00	<5.00	<5.00	<5.00
11	Nickel	Ni	-	44.000	43.41	32.83	30.67	33.39
12	Tin	Sc	-	14.000	<3.00	<3.00	<3.00	<3.00
13	Silver	Ag	-	0.060	<1.00	<1.00	<1.00	<1.00
14	Thallium	Tl	-	1.300	<10.00	<10.00	<10.00	<10.00
15	Chromium	Cr	-	70.000	112.62	83.65	77.33	99.20
16	Zinc	Zn	-	60.000	106.18	96.44	73.58	90.69
17	Vanadium	V	-	95.000	139.20	109.53	99.17	115.87
18	Antimony	Sb	-	-	<5.00	<5.00	<5.00	<5.00
19	Tungsten	W	-	1.300	2.23	2.01	2.03	2.13
20	Tantalum	Ta	-	3.400	<10.00	<10.00	<10.00	<10.00
21	Thorium	Th	-	11.000	<5.00	<5.00	<5.00	<5.00
22	Beryllium	Be	-	2.000	2.00	2.00	2.00	2.00
23	Ytterbium	Yb	-	3.400	3.00	<3.00	<3.00	<3.00
24	Yttrium	Y	-	34.000	27.31	29.07	20.85	25.45
25	Lanthanum	La	-	44.000	29.13	33.24	19.79	26.00
26	Scandium	Sc	-	14.000	17.03	13.34	11.14	13.74
27	Cerium	Ce	-	75.000	66.92	69.51	50.30	59.21
28	Lithium	Li	-	30.000	42.52	27.89	28.17	28.07
29	Barium	Ba	-	590.000	534.60	657.10	495.10	489.00
30	Strontium	Sr	-	290.000	92.09	111.10	92.46	92.53
31	Gallium	Ga	-	17.000	15.14	12.82	12.52	12.85
32	Germanium	Ge	-	1.300	<5.00	<5.00	<5.00	<5.00
33	Indium	In	-	0.070	<5.00	<5.00	<5.00	<5.00
34	Zirconium	Zr	-	160.000	83.41	72.78	69.24	78.10
35	Molybdenum	Mo	-	1.000	<3.00	<3	5.45	3.89
36	Niobium	Nb	-	20.000	9.94	8.38	8.88	9.36
37	Tellurium	Te	-	0.002	<20.00	<20.00	<20.00	<20.00
38	Titanium	Ti	-	4700.000	5765.40	4952.60	4645.40	5494.00
39	Phosphorus	P	-	810.000	1119.84	1698.10	896.83	1048.64
40	Bor	B	-	9.000	56.04	75.01	40.85	105.40

and

Table A24. Exceedance factor of total concentrations of elements in samples taken from dried meltwater ponds on the agricultural lands of KazGer farm.

Object of Study	CMPCI/MPC (CCi/Cph)	Object of Study	CMPCI/MPC (CCi/Cph)
Dry Reservoir №1	CMPC (As) = 7.0 CCi (Fe) = 1.3 CCi (Co) = 1.3 CCi (Cr) = 1.6 CCi (Zn) = 1.7 CCi (V) = 1.4 CCi (W) = 1.7 CCi (Sc) = 1.2 CCi (Li) = 1.4 CCi (Ti) = 1.2 CCi (P) = 1.4 CCi (B) = 6.2	Dried Reservoir №3	CMPC (As) = 6.7 CCi (Co) = 1.3 CCi (Cr) = 1.1 CCi (Zn) = 1.2 CCi (V) = 1.0 CCi (W) = 1.5 CCi (Mo) = 5.4 CCi (P) = 1.1 CCi (B) = 4.5
Dried Reservoir №2	CMPC (As) = 7.5 CMPC (Cu) = 1.2 CCi (Co) = 1.4 CCi (Cr) = 1.1 CCi (Zn) = 1.6 CCi (V) = 1.1 CCi (W) = 1.5 CCi (Ba) = 1.1 CCi (Ti) = 1.0 CCi (P) = 2.0 CCi (B) = 8.3	Dried Reservoir №4	CMPC (As) = 10.0 CMPC (Cu) = 1.0 CCi (Co) = 1.3 CCi (Cr) = 1.4 CCi (Zn) = 1.5 CCi (V) = 1.2 CCi (W) = 1.6 CCi (Mo) = 3.9 CCi (Ti) = 1.1 CCi (P) = 1.3 CCi (B) = 11.7

, Figure A17 and Figure A18).

–

Monitoring of agrochemical soil indicators of agricultural lands and chemical analysis of the quality of natural surface waters were conducted.

–

A sociological survey of the rural population was carried out, and the vulnerability index of households to climate change was calculated.

–

A comparative analysis of the agroclimatic conditions of the region for the period from 1960 to 2024 was performed.

4. Discussion

The study of land use integrated with the analysis of topography, soil, and hydrogeology for the localized MAR locations and the integration of forest shelterbelts contributes to the refinement of land use strategies for sustainable development. This strategy has been successfully employed globally, but requires promotion in Kazakhstan, Central Asia. In Russia’s chernozem zone, the restoration of shelterbelts combined with contour-strip land organization has significantly reduced wind and water erosion, improved the water balance, and increased cereal crop yields by 15–25%. A comprehensive approach, including soil recultivation and slope terracing, showed particularly high effectiveness [8,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60]. In the Canadian Prairies, MAR technologies are actively used in conjunction with shelterbelts and pasture systems. Such solutions have improved the soil’s water-holding capacity, reduced seasonal moisture deficits, and enhanced the productivity of agricultural lands, especially under unstable precipitation regimes. In the northwestern regions of China, the integration of shelterbelts, terraced slopes, and MAR systems has reduced soil salinization and restored agricultural land productivity. The combination of agrolandscape measures and MAR technologies has proven highly effective in water resource management and adapting agriculture to frequent droughts [8,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60]. Comparative analysis indicates that the approaches being implemented in the Akmola Region align with global best practices and possess the potential for being scaled to other agricultural regions of Kazakhstan, offering significant potential for sustainable agriculture and climate change adaptation. The ongoing project work on the contour-strip organization of farmlands with the introduction of MAR and shelterbelts contributes to the following:

✔

Improved soil water regime: Local topographic depressions and areas with permeable soils ensure the accumulation and retention of meltwater.

✔

Reduced erosion: The contour-strip organization of lands and the shelterbelt system protect soils from deflation (wind erosion) and water erosion, reducing the loss of the fertile layer.

✔

Increased crop yields: Zones with high MAR potential demonstrate better productivity indicators due to improved moisture availability and microclimate regulation.

✔

Slowed soil degradation: Analysis of soil salinity maps with the implementation of shelterbelts and rational land use contributes to a substantial reduction in secondary salinization, especially in areas of intensive industrial farming.

The complex application of contour-strip land organization with MAR technologies and the introduction of forest shelterbelts could contribute to the refinement of land use strategies for sustainable development with effective adaptation to climate change in the northern regions of Kazakhstan, where spring floods alternate with summer drought periods.

Kazakhstan, a country with a vast territory and diverse, often inefficient, water use methods and technologies, experiences significant seasonal variations in precipitation, creating challenges in water resource management. Situated at the end of several transboundary river basins, Kazakhstan has problems with surface water, connected to its dependence on neighboring countries. High water losses and pollution create challenges for Kazakhstan’s sustainable development. Climate fluctuations, characterized by increased flood peaks in early spring and severe droughts in summer, further exacerbate these problems. During spring floods, residents of Kazakhstan often struggle to cope with the excess flood water, and in attempting to dispose of it, often try to pass the flood water on somehow, transferring the flood problems to their neighbors at lower elevations. Without well-established, rational programs for conserving flood water near their homes and communities for later use in the summer, these flood water passing actions are irrational. During droughts, the same local people may have critical water shortage issues, which impacts the local sustainability in rural farms. MAR application expansion could be a more efficient and sustainable approach for many of Kazakhstan’s regions to improve water security. MAR allows replenishment of groundwater resources and can support the expansion of sustainable water reuse practices throughout Kazakhstan. By strategically capturing and filtering excess water during periods of flood-water abundance, MAR can both reduce flood risks and conserve water for summer droughts. The localized MAR technologies that combine snow- and flood-water harvesting with runoff management can be widely implemented throughout Kazakhstan. This can be achieved through the introduction of forest shelterbelts and contour-strip land organization, sequentially connected in the final section to the retention ponds, effectively creating a distributed network of water storage ponds with reasonable flow rates for storing flood water and recharging groundwater aquifers.

Melting-snow floods are one of the main emergency disaster event issues in Kazakhstan. The implementation of efficient water collection systems for snowmelt and flood relocation could be one of the key strategies for improving water sustainability in most of Kazakhstan’s regions. This strategy should be rationally implemented by following nature, respecting the water’s natural flow movement. Minimal adjustments would be reasonable to adapt to the natural terrain, by following the local topography, while avoiding direct confrontation with nature by building big dams against direct movement of a large amount of flood water. The natural water movements of streams would be rational to keep, with the basic adaptation activities, allowing the melted snow to move and be stored gradually, before potential widespread big flooding may happen. This strategy would allow control of the potential movement of a large amount of water, to help snow melting activities, by directing and accumulating water for further seepage into underground aquifers with connected MAR technologies. The snow-flood collection system could be designed based on the terrain and the local topography by using various types of retention ponds. Below is a basic scenario for snow-flood collection in a retention pond during the early-spring period.

Trees are rain makers through evapotranspiration activities. Trees produce hydroscopic microorganisms within their leaves in volatile organic matter. These microorganisms and particles drift up into the atmosphere, forming a microenvironment like cation–anion microparticles, providing precipitation nuclei needed to condense water vapor into droplets forming clouds and then rain. Micro energy, like positive “+” energy, attracts micro minus “−” energy, like a magnet, creating a microparticle of precipitation. Trees release chemical compounds, called volatile organic compounds (VOCs), which can form aerosol particles that act as cloud condensation nuclei (CCNs). These hygroscopic particles attract water vapor, allowing it to condense and form clouds, which eventually lead to precipitation. The microorganisms that live on or in leaves also contribute to this process. Trees emit VOCs: Trees release a large variety of VOCs into the atmosphere. The type and number of VOCs emitted depend on factors like the tree species, season, and environmental conditions. VOCs form particles: These VOCs can react with other atmospheric chemicals like ozone layers and form new aerosol particles. Microorganisms on leaves: Microbes on the surfaces of leaves are also a source of aerosols and contribute to the atmospheric particle load. Particles become CCNs: The resulting particles are tiny, often solid or liquid specks that are highly effective at attracting water molecules, making them hygroscopic. Cloud formation: These particles serve as the “seeds” for clouds by providing a surface on which water vapor can condense to form tiny droplets. Rainfall: As these droplets grow and become saturated, they fall as rain [62,63,64,65].

5. Conclusions

This study provides a comprehensive framework for the sustainable management of dryland agricultural systems in the Akmola Region of Kazakhstan through the integration of managed aquifer recharge (MAR) principles, topographic and hydrogeological assessment, and landscape-based land use planning. By combining field observations, GIS-based analysis, and multi-criteria decision-making (AHP–MCDA), the research identified areas most suitable for implementing localized MAR systems integrated with forest shelterbelts and contour-strip land organization.

The results demonstrate that the coordinated use of MAR with soil and water conservation measures could improve the agroecological balance of farmlands by enhancing groundwater replenishment, optimizing soil moisture retention, and reducing erosion and salinization. In particular, the restoration and strategic placement of forest shelterbelts were shown to stabilize the microclimate, mitigate wind and water erosion, and contribute to long-term productivity growth. Salinity and land use maps developed in this study serve as vital decision-support tools for spatial planning and sustainable agricultural management.

Practically, the approach contributes to the refinement of land use strategies and supports the transition toward climate-resilient and resource-efficient agriculture in Kazakhstan’s northern drylands. The methodology can guide policymakers, land managers, and local communities in planning MAR-oriented interventions that combine hydrological, geomorphological, and ecological criteria for maximum efficiency.

Future research should focus on expanding the application of this framework to different agroecological zones of Kazakhstan to evaluate its transferability and scalability. It is also recommended to integrate this approach into national land restoration and climate adaptation programs, such as the development of regional strategies for drought resilience and sustainable groundwater management. Long-term monitoring and modeling will be essential to quantify the cumulative impacts of MAR–forest shelterbelt systems on soil fertility, water availability, and carbon sequestration potential.

In summary, the proposed integrated land management approach—combining MAR technologies, agroforestry elements, and geospatial analysis—provides a scientifically grounded and adaptable solution for achieving sustainable land and water resource use in semi-arid regions of Kazakhstan and beyond.