The purpose of this review is to systematize data on the level of soil pollution and the rate of erosion processes in the Republic of Kazakhstan (RK) and to identify river basins with the high probability of secondary soil pollution due to the lateral migration of pollutants transported together with sediments. Using the Google Scholar, Springer Link, Science Direct, Scopus, and RSCI search engines, publications on soil pollution with heavy metals (HMs), radionuclides of technogenic and natural origins, oil and petroleum products, and rocket fuel components, as well as characterizing the rates of water and wind erosion and the degree of soil erosion in the RK were collected. On the basis of these data, thematic maps of the RK showing river basins with increased concentrations of pollutants in soils were compiled. Using the maps of soil pollution and water and wind erosion, river basins with the high probability of secondary soil pollution due to the transport of contaminated soil particles by water-erosion processes and local transport by wind were identified. It was found that the most intense secondary pollution, primarily of alluvial soils, can occur in the basins of the Irtysh and Shu rivers. It should be taken into account that existing estimates of water erosion rates are based on calculations using erosion models, while estimates of wind erosion rates are based on semiquantitative data. In this regard, research is needed to verify calculations using erosion models based on field data.

For the first time, an electronic vector map of small river basins and their interfluves with a regional level of spatial detail (1 : 1 000 000) has been created using GIS technology for the territory of mainland Russia with a total area of almost 17 million km². The GMTED2010 global model is used as a digital elevation model. The total number of basin geosystems is 388 627 and their area averages 47.8 km². The selected basin geosystems are used as operational–territorial units, in relation to which the geodatabase was created, characterizing the natural resource potential and geoecology of Russia. The open access River Basins of European Russia geoportal was created for a large part of the country, where all information thus obtained is posted. The article provides examples of solutions based on the previously formed GIS and the basin approach. A digital mosaic of small river basins makes it possible to “collect” territories of different scales (from local to transregional) and different taxonomies (from catchments of large rivers to federal districts of Russia) for geographical analysis. Such examples are given with the creation of specialized GIS for the great rivers of Siberia: the Ob and Lena, and a number of federal districts: the Volga and Siberian districts. Based on the map thus created of basins and GIS, a number of major geographical and geoecological problems are solved: the evaluation of current intensity of soil erosion, density of gully erosion, modeling of river runoff and anthropogenic impact on basin geosystems.

The paper presents a hybrid machine learning model for the spatial segmentation of soils by salinity using multispectral satellite data from Sentinel-2 and climate parameters of the ERA5-Land model. The proposed method aims to solve the problem of accurate soil cover segmentation under climate change and high spatial heterogeneity of data. The approach includes the sequential application of unsupervised learning algorithms (K-Means, hierarchical clustering, DBSCAN), the XGBoost model, and a multitasking neural network that performs simultaneous classification and regression. At the first stage, pseudo-labels are formed using K-Means, then a probabilistic assessment of object membership in classes and ensemble voting of clustering algorithms are carried out. The final model is trained on an extended feature space and demonstrates improved results compared to traditional approaches. Experiments on a sample of 33,624 observations (23,536—training sample, 10,088—test sample) showed an increase in the Silhouette Score value from 0.7840 to 0.8156 and a decrease in the Davies–Bouldin Score from 0.3567 to 0.3022. The classification accuracy was 99.99%, with only one error in more than 10,000 test objects. The results confirmed the proposed method’s high efficiency and applicability for remote monitoring, environmental analysis, and sustainable land management.

In the article the presentation of estimation methods of potential soil loss in the conditions of Siberia with application of geographical information systems is resulted. For the reference area of the Marha river basin, which is a part of the Lena river catchment, there was created a specialized geographic information database of potential soil erosion, with scale of 1: 1,000,000. Digital elevation model "GMTED2010" and the hydroset layer corresponding to the scale of 1: 1,000,000 are taken to calculate the soil loss values. The formation of the geobase data is considered in detail being constructed on the basis of the multiplicative structure which reflects the main parameters of the relief (slope steepness, exposition, slope length, erosion potential of the relief), soil, climatic characteristics and modern types of land cover. At the quantitative level with sufficiently high degree of spatial detail results were obtained for calculating the potential erosion of soils. The average value of potential soil loss in the basin without taking into account the factor of land cover types, was 12.6 t/ha/yr. The calculations carried out, taking into account the types of land cover obtained from remote sensing data from outer space resulted in an appreciable reduction of the soil loss values (0.04 t/ha/yr.).

In order to determine the impact of urbanization on raptors in the semi-desert conditions of southwestern Kazakhstan, an analysis of the spatio-temporal distribution of raptors is presented for the first time based on the results of surveys of the avifauna of Kyzylorda. Eight species of raptors were recorded: field Hen Harrier (Circus cyaneus), Marsh Harrier (C. aeroginosus), Eurasian Sparrowhawk (Accipiter nisus), Long-Legged Buzzard (Buteo rufinus), Eurasian Buzzard (B. buteo), Steppe Eagle (Aquila nipalensis), Eurasian Hobby (Falco subbuteo), and Common Kestrel (F. tinnunculus). The probability of raptors being present was negatively associated with dense urban low-rise buildings with limited greenery in the bay and the new part of the city. At the same time, the dense urban development with little or no greenery in the old central part of the city provides adequate habitat (including a foraging base and nesting sites) for the Common Kestrel. Raptor presence was positively associated with the Syrdarya River floodplain and wasteland with small groups of trees and/or shrubs. The landfill site located on the north-eastern edge of the city serves as a feeding ground for the Long-Legged and Eurasian Buzzards, while the airport area is inhabited by the Eurasian Buzzard, Steppe Eagle, and Common Kestrel. Based on this study, we would recommend that enterprises (e.g., grain storage facilities, airports) and local executive bodies who are interested in the conservation of raptors and regulating the population of the pigeons around their territories should retain or plant more native vegetation and shrubs and preserve areas with green spaces.

The environmental component of Sustainable Development for large regions of the Earth can be assessed through the evaluation of the natural resource potential of the territory. The methodological challenge of such assessments is always determined by the type of operational-territorial unit within which the geodatabase is formed. This article details the possibilities of using the basin approach as such units. This approach is one of the most important in humid climate regions where a river network has been developed. Using the example of the Ob’ river basin in Northern Eurasia, the article illustrates the application of the basin approach to assess the environmental determinants of Sustainable Development. The studies were conducted in three stages: formation of an GIS database of basin geosystems of the Ob’ river basin; creation of a geospatial database on the natural resource potential in the small river basins; selection of criteria and assessment of anthropogenic load on the basin geosystems of the Ob’. A total of 30,738 small river basins were delineated automatically based on GMTED DEM, with a mean area of 66 km². GIS integrated geoinformation represents the natural and anthropogenic characteristics of river basins. The assessment of the environmental state of the territory should considerthe types and strengths of anthropogenic loads. For this purpose, the integral indicators used, which directly or indirectly reflect anthropogenic impact: population density, road network density, and the percentage of arable land in the total area. The final indicator of anthropogenic load was calculated as a linear combination of specific variables and ranked into six categories. Thematic and complex maps were created, allowing us to identify the natural background in which the geosystems of small river basins are formed and function, as well as the types and strength of anthropogenic loads on the territory.