Land use change is one of the main factors driving changes in terrestrial carbon storage, which comprises the storage of vegetation carbon and soil carbon. Selecting the Chengdu–Chongqing urban agglomeration (CCUA) as the study area, land use and carbon storage from 2010 to 2030 were analyzed by combining the Future Land Use Simulation (FLUS) model and the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model. The main types of land use in CCUA are farmland and forest. The conversion of farmland to built-up land was the most important form of land use transfer between 2010 and 2020. Each type of land use shows the smallest change under the ecological protection scenario, and the degree of the comprehensive land use dynamic is only 0.19%. Under the natural development scenario, the areas of built-up land, wetland, and forest land will increase in 2030. Under the urban development scenario, the built-up land area will increase by 751.24 km², an increase in more than 10.08%, but farmland, forest, and grassland will decrease. The spatial pattern of carbon storage is “high in the east and west, low in the middle”; farmland accounts for the largest proportion of carbon storage at over 60% of the total. Carbon storage decreased by 29.45 × 10⁶ Mg from 2010 to 2020. Grassland showed the most significant decrease in carbon storage, with the proportion decreasing from 7.49% in 2010 to 6.09% in 2020. In 2030, the total carbon storage will reach 1844.68 × 10⁶ Mg under the ecological protection scenario, slightly higher than that in 2020, while it will show a downward trend under the natural development and urban development scenarios.

Efficient and accurate vegetation type extraction from remote sensing images can provide decision makers with basic forest cover and land use information, and provides a reliable basis for long-term monitoring. With the development of deep learning, the convolutional neural network (CNN) has been used successfully to classify tree species in many studies, but CNN models have rarely been applied in the classification of vegetation types on larger scales. To evaluate the performance of CNN models in the classification of vegetation types, this paper compared the classification accuracy of nine dominant land cover types in Baishuijiang National Nature Reserve with four models: 3D-CNN, 2D-CNN, JSSAN (joint spatial–spectral attention network) and Resnet18, using sentinel-2A data. Comparing the difference in classification accuracy between the direct use of raw sentinel images and fused feature indices sentinel images, the results showed that adding feature indices can improve the overall accuracy of the model. After fusing the characteristic bands, the accuracy of the four models was improved significantly, by 5.46–19.33%. The best performing 3D-CNN model achieved the highest classification accuracy with an overall accuracy of 95.82% and a kappa coefficient of 95.07%. In comparison, 2D-CNN achieved an overall accuracy of 79.07% and a kappa coefficient of 75.44%, JSSAN achieved an overall accuracy of 81.67% and a kappa coefficient of 78.56%, and Resnet18 achieved an overall accuracy of 93.61% and a kappa coefficient of 92.45%. The results showed that the 3D-CNN model can effectively capture vegetation type cover changes from broad-leaved forests at lower elevation, to shrublands and grasslands at higher elevation, across a range spanning 542–4007 m. In experiments using a small amount of sample data, 3D-CNN can better incorporate spatial–spectral information and is more effective in distinguishing the performance of spectrally similar vegetation types, providing an efficient and novel approach to classifying vegetation types in nature reserves with complex conditions.

Given the constraints of limited cultivated land resources, ensuring and enhancing crop productivity are crucial for food security. This study takes Chengdu as a case study. Using the cultivated land productivity (CLP) evaluation model, we calculated the cultivated land productivity index (CLPI) and analyzed its spatial distribution characteristics. The Geographical Detector model was employed to identify the main factors influencing CLP, and corresponding countermeasures and measures were proposed based on the limiting degrees of these factors. The findings reveal that Chengdu’s CLP index ranges from 1231 to 3053. Global spatial autocorrelation analysis indicates a spatial agglomeration pattern in Chengdu’s overall crop productivity distribution. The local spatial autocorrelation analysis demonstrates that township (street)-level crop productivity in Chengdu is primarily characterized by “high–high”, “low–low”, and “low–high” clusters. Key factors influencing the spatial differentiation of CLP in Chengdu include the agronomic management level, soil bulk density, irrigation guarantee rate, soil body configuration, field slope, and farmland flood control standard. Interaction detection shows that there are both double-factor and nonlinear enhancements among the factors. Specifically, the interaction between soil bulk density and the agronomic management level among other factors have the most explanatory power for the spatial differentiation of CLP. The CLP in Chengdu is highly restricted by its technical level, with the agronomic management level severely limiting CLP by more than 50%. These research results provide a theoretical reference for regional high-standard farmland construction and the protection and utilization of cultivated land resources.

Exploring the spatiotemporal evolution pattern of agricultural functions, analyzing their tradeoff and synergies, carrying out multifunctional zoning, and determining the combination and expansion direction of agricultural functions in combination with regional development strategies are conducive to guiding the adjustment of agricultural structure and promoting the sustainable development of regional agriculture. In this context, based on the county scale and statistical data, this paper uses the agricultural function evaluation index system to measure and analyze the agricultural function index of Sichuan Province and its mutual relations. Spatial overlay analysis is used to analyze the agricultural function index for agricultural leading function zoning. Cluster analysis is used to evaluate the agricultural function results to explore the agricultural multifunctional zoning scheme of Sichuan Province. The results show that the spatial and temporal distribution of agricultural product supply, agricultural leisure, ecological services, employment and social security services are heterogeneous, and the agricultural multifunction index of Sichuan Province shows a spatial distribution pattern of high in the east and low in the west. The synergistic effect between the supply function of agricultural products and the function of employment and life security is the strongest. From 2010 to 2020, the relationships between the ecological service function and the supply function of agricultural products, the agricultural leisure function, employment, and the social security function change from irrelevant or there being a tradeoff effect to there being a significant synergistic effect. The leading areas of the ecological service function are mainly distributed in western and northern Sichuan. The leading areas of the agricultural product supply function are mainly distributed in eastern and southern Sichuan. Agricultural multifunctional zoning in Sichuan Province is divided into the agricultural leisure function, the agricultural product supply cooperative functional area, the weak cooperative functional area, the strong cooperative functional area, and the agricultural leisure priority functional area. The spatiotemporal heterogeneity of agricultural functions and the changes in tradeoffs and synergies in Sichuan Province have a significant impact on the development of agricultural functions. The research results can provide s theoretical reference for agricultural multifunctional zoning in the study area and provide guidance and suggestions for the sustainable development of agricultural economy and society in Sichuan Province.

Exploring the spatio-temporal evolution patterns of rapeseed production at the county level in Sichuan Province, China, and analyzing the influence of natural conditions and socioeconomic development based on regional spatial characteristics, can help guide the rational distribution of crop production and provide a reference for the high-quality and sustainable development of the local rapeseed industry. Based on panel data from 2001 to 2023, this study employs GIS spatial analysis to examine the spatio-temporal evolution of rapeseed production in Sichuan and applies a Geodetector model to identify factors influencing its spatial and temporal variations. The results reveal that rapeseed production in Sichuan is concentrated in three main production areas: the northeastern Sichuan region, the middle Sichuan hilly region, and the Chengdu Plain. The dynamic evolution exhibits a composite pattern characterized by the stability and expansion of core areas, alongside breakthroughs and growth in peripheral regions, with increased production observed across 134 counties. The spatial center of rapeseed production shows short-range fluctuations and distinct regional anchoring, oscillating among Santai County, Shehong City, and Daying County, tracing a “Z”-shaped trajectory. Over the 23-year period, the global Moran’s I index ranged from 0.464 to 0.558, indicating a significant spatial clustering trend in rapeseed output among adjacent counties. Local spatial autocorrelation patterns were predominantly H-H, L-L, and L-H clusters. Factor detection identifies labor force availability, fertilizer application intensity, and effective irrigated area as the most influential factors. Interaction detection results consistently exhibit a two-factor enhancement effect. To enhance the rapeseed industry’s performance and efficiency, it is recommended to stabilize production capacity in the three core production areas, leverage central regions to strengthen radiation to the surrounding counties, optimize resource allocation based on clustering patterns, and focus on improving key factors such as labor and irrigation, as well as their synergistic effects.