Wetland ecosystem service function provides and maintains the Earth’s life system, which supports human and social development. However, in recent years, with the intensification of human social activities, the wetland area in northeast China has been reduced, and wetland ecosystem service function has been damaged. This paper evaluates the ecosystem service function of wetlands in northeast China based on the InVEST model, taking 40 prefecture-level cities as the evaluation unit, calculating the carbon stock, soil retention, and habitat quality of the wetlands in the study area and analyzing the drivers of changes in ecosystem service function using the PLUS model. The following results were obtained: temporally, the wetland carbon stock decreased from 754 Tg in 2000 to 688 Tg in 2020; the wetland soil retention increased from 24,424 Tg in 2000 to 33,160 Tg in 2010, and then decreased to 28,765 Tg in 2020; and the quality of wetland habitats was roughly unchanged. The wetland habitats in the study area were categorized into 5 types, classified as I, II, III, IV, or V, and the spatial changes in the 40 prefecture-level cities in northeast China were analyzed. The driving factors affecting the change in the wetland ecosystem service function were further analyzed, mainly focusing on changes in the wetland area itself. The influence of other land-use types and the influence of related policies were analyzed in three aspects, among which the GDP and spatial density of the population are social factors, and the elevation and slope are natural factors that provide larger contributions to the change in wetland area. The reduction in forest and grassland areas and the increase in cultivated land and construction land areas have a negative effect on the ecosystem service function of wetlands, and the implementation of relevant wetland protection policies promotes the ecosystem service function of wetlands. According to the problems faced by wetlands in different regions, the government formulates strategies that are in line with local development, with a view to implementing wetland ecological development in the northeast region in the new context, which will help to realize intensive land use and stimulate the vitality of the region.

The accurate prediction of soil organic matter (SOM) content is important for sustainable agriculture and effective soil management. This task is particularly challenging due to the variability in factors influencing SOM distribution across different cultivated land types, as well as the site-specific responses of SOM to remote sensing data and environmental covariates, especially in the black soil region of northeastern China, where SOM exhibits significant spatial variability. This study evaluated the variations on the importance of different remote sensing imagery and environmental covariates in different cultivated land zones. A total of 180 soil samples (0–20 cm) were collected from Youyi County, Heilongjiang Province, China, and multi-year synthetic bare soil images from 2014 to 2022 (focusing on April and May) were acquired using Google Earth Engine. Combining three types of environmental covariates such as drainage, climate and topography, the study area was categorized into dry field and paddy field. Then, the SOM prediction model was constructed using random forest regression method and the accuracy of different strategies was evaluated by 10-fold cross-validation. The findings indicated that, (1) in the overall regression analysis, combining drainage and climate variables and multi-year synthetic remote sensing images of May could attain the highest prediction accuracy, and the importance of environmental covariates was ranked as follows: remote sensing (RS) > climate (CLI) > drainage (DN) > Topography (TP). (2) Zonal regression analysis was conducted with a high degree of precision, as evidenced by an R² of 0.72 and an impressively low RMSE of 0.73%. The time window for remote monitoring of SOM was different for dry field and paddy field. More specifically, the optimal time frames for SOM prediction in dryland were identified as April and May, while those for paddy fields were concentrated in May. (3) In addition, the importance of diverse environmental covariates was observed to vary with the cultivated land types. In regions characterized by intricate topography, such as dry fields, the contributions of remote sensing images and climate variables assumed a heightened importance. Conversely, in paddy fields featuring flat terrain, the roles of climate and drainage variables played a more substantial role in influencing the outcomes. These findings underscore the importance of selecting appropriate environmental inputs for improving SOM prediction accuracy.

Soil organic matter (SOM) is very important to the quality evaluation of cultivated land, especially in fertile black soil areas. Many studies use remote sensing images combined with different machine learning algorithms to predict the regional SOM content. However, the information provided by remote sensing images in different time windows is very different. Taking Youyi Farm, a typical black soil area in Northeast China, as the research area, this study obtains all available Sentinel-2 images covering the research area from 2019 to 2021, calculates the spectral index of single-phase and multi-temporal synthesis images, takes the spectral index and band of each image as the input, and employs the random forest regression algorithm to evaluate the performance of SOM prediction using remote sensing images with different time windows. The results show that: (1) the accuracy of SOM prediction using image band and spectral index is generally improved compared to using only the band; (2) when using single-phase images, the R² range of SOM prediction using image band and spectral index is from 0.16 to 0.59 and the RMSE ranges from 0.82% to 1.23%; When using multi-temporal synthesis images, the R² range of SOM prediction using image band and spectral index is from 0.18 to 0.56 and the RMSE ranges from 0.85% to 1.19%; (3) the highest accuracy of SOM prediction using synthetic images is lower than that of single-phase images; (4) the best time window of the bare soil period in the study area is May. This study emphasizes the importance of the time window to SOM prediction. In subsequent SOM prediction research, remote sensing images with appropriate time windows should be selected first, and then the model should be optimized.

Soil organic matter (SOM) is a key indicator of cropland quality and carbon cycling. Accurate SOM mapping is essential for sustainable soil management and carbon sink assessment. This study investigated the effects of interannual climatic variability on SOM prediction using remote sensing and machine learning. Youyi Farm in the Sanjiang Plain, Heilongjiang Province, was selected as the study area, covering three representative years: 2019 (flood), 2020 (normal), and 2021 (drought). Based on multi-temporal Sentinel-2 imagery and environmental covariates, Random Forest models were used to evaluate single- and dual-period combinations. Results showed that combining bare-soil and crop-season images consistently improved accuracy, with optimal combinations varying by year (R² = 0.544–0.609). Incorporating temperature, precipitation, and elevation enhanced model performance, particularly temperature, which contributed most to prediction accuracy. Feature selection further improved model stability and generalization. Spatially, SOM showed a pattern of higher values in the northeast and lower in the central region, shaped by topography and cultivation. This study innovatively integrates interannual climatic variability with remote sensing temporal combination and feature selection, constructing a climate-adaptive SOM mapping framework and providing new insights for accurate inversion of cropland SOM under extreme climates, highlights the importance of multi-temporal imagery, environmental factors, and feature selection for robust SOM mapping under different climatic conditions, providing technical support for long-term cropland quality monitoring.

The accurate mapping of soil texture, a key determinant of soil’s hydrological and nutritional behavior, is essential for agricultural drought assessment, yet the application of multi-temporal satellite data for this purpose remains largely unexplored. In this study, we first identified the optimal prediction period by evaluating the performance of single-date imagery (satellite images captured on individual observation dates). Subsequently, dual-phase imagery (DPI) was developed to increase mapping accuracy. Finally, these refined predictions quantified soil texture’s response to drought and its corresponding thresholds. Results demonstrated that: (1) the bare soil period in April provided peak prediction accuracy for all texture fractions (Sand: R² = 0.617, RMSE = 10.21%; Silt: R² = 0.606, RMSE = 8.648%; Clay: R² = 0.604, RMSE = 1.945%); (2) Significant accuracy gain from DPI using April-August imagery fusion (Sand: R² = 0.677, RMSE = 9.386%; Silt: R² = 0.660, RMSE = 8.034%; Clay: R² = 0.658, RMSE = 1.807%); (3) sand content was the most critical factor influencing crop drought stress, with a threshold of 31%. By integrating multi-temporal satellite observations with quantitative drought evaluation for high-resolution soil texture mapping and precision agricultural management in Northeast China’s black soil region.

Ensuring stable soybean supply requires understanding how production patterns evolve and whether policy consistently dominates production dynamics. This study examines 36 prefecture-level units in Northeast China from 2000 to 2024 by integrating production growth decomposition, the production concentration index, center-of-gravity and standard deviational ellipse models, spatial autocorrelation (Moran’s I), and the geographical detector. The results indicate the following: (1) Production fluctuated upward, with its spatial core consolidating in the northern high-latitude zone and the regional centroid moving 123.3 km northeastward overall. Concentration increased more than sown-area concentration, indicating a gradual, spatially uneven recovery in yield contributions rather than a completed shift to yield-led growth. (2) Spatial differentiation was fundamentally governed by hydrothermal endowments rather than monotonic policy dominance. Annual mean temperature consistently exhibited the strongest independent explanatory power (reaching q = 0.39 in 2024), followed by agricultural machinery power (q = 0.22) and fertilizer input (q = 0.24). (3) The influence of agricultural fiscal support intensity was stage-specific and conditional (significant in 2010 and 2020, but waning to q = 0.14 in 2024), functioning primarily through synergistic interactions with natural endowments and input levels. Consequently, regional soybean layout reflects a multifactor interplay where climatic suitability sets the production ceiling, while factor configuration and targeted fiscal incentives regulate localized capacity. Policy interventions should shift from uniform area expansion subsidies toward differentiated, climate-resilient input optimization in core and marginal regions.