Accurate identification of tea plantation distribution is critical for optimizing agricultural practices, informing land-use policies, and preserving ecological balance. However, challenges persist in mountainous regions with persistent cloud cover and heterogeneous vegetation, where conventional methods relying on single-source remote sensing features face limitations due to spectral confusion and information redundancy. This study proposes a novel framework integrating multisource remote sensing data and feature optimization to address these challenges. Leveraging the Google Earth Engine (GEE) cloud platform, this study synthesized 108 spectral, textural, phenological, and topographic features from Sentinel-1 SAR and Sentinel-2 optical data. SVM-RFE (support vector machine recursive feature elimination) was employed to identify the optimal feature subset, prioritizing spectral indices, radar texture metrics, and terrain parameters. Comparative analysis of three classifiers, namely random forest (RF), support vector machine (SVM), and decision tree (DT), revealed that RF achieved the highest accuracy, with an overall accuracy (OA) of 95.03%, a kappa coefficient of 0.95. The resultant 10 m resolution spatial distribution map of tea plantations in Shiyan City (2023) demonstrates robust performance in distinguishing plantations from forests and farmlands, particularly in cloud-prone mountainous terrain. This methodology not only mitigates dimensionality challenges through feature optimization but also provides a scalable solution for large-scale agricultural monitoring, offering critical insights for sustainable land management and policy formulation in subtropical mountainous regions.

The accurate mapping of crop planting patterns is vital for sustainable agriculture and food security, particularly in regions with complex cropping systems and limited cloud-free observations. This research focuses on the Jianghan Plain in southern China, where diverse planting structures and persistent cloud cover make consistent monitoring challenging. We integrated multi-temporal Sentinel-2 and Landsat-8 imagery from 2017 to 2021 on the Google Earth Engine platform and applied a sample migration strategy to construct multi-year training data. A random forest classifier was used to identify nine major planting patterns at a 10 m resolution. The classification achieved an average overall accuracy of 88.3%, with annual Kappa coefficients ranging from 0.81 to 0.88. A spatial analysis revealed that single rice was the dominant pattern, covering more than 60% of the area. Temporal variations in cropping patterns were categorized into four frequency levels (0, 1, 2, and 3 changes), with more dynamic transitions concentrated in the central-western and northern subregions. A multiscale geographically weighted regression (MGWR) model revealed that economic and production-related factors had strong positive associations with crop planting patterns, while natural factors showed relatively weaker explanatory power. This research presents a scalable method for mapping fine-resolution crop patterns in complex agroecosystems, providing quantitative support for regional land-use optimization and the development of agricultural policies.

The nonagricultural conversion of cultivated land (NACCL) poses a major challenge to cultivated land protection and food security in intensive agricultural regions. However, the reliable identification of NACCL in complex agricultural plains remains difficult because cultivated land, aquaculture ponds, shrimp–rice fields, greenhouses, water bodies, and rural settlements are often spatially interlaced and spectrally confused. Taking the Jianghan Plain, a major grain-producing region in central China, as the study area, this study used Sentinel-2 time-series imagery, percentile-based temporal metrics, spectral indices, and random forest classification on the Google Earth Engine platform to generate annual landuse maps from 2017 to 2021. The classification results achieved overall accuracies of 86%–92%, with Kappa coefficients of 0.77–0.87, and the producer’s accuracy of cultivated land reached 96.23%–100%. Compared with the ESRI 10 m land-use product, the classification results provided finer spatial details for cultivated land boundaries, indicating the reliability and applicability of the remote sensing interpretation method in complex agricultural plains. Based on the classification results, this study further diagnosed NACCL by combining landuse transition analysis, grid-based nonagricultural conversion rate measurement, and patch-scale cultivated-land overlay. It’s shown that patch-scale diagnosis could distinguish unchanged, changed, and suspected changed patches, thereby revealing both confirmed conversion and uncertainty caused by mixed land-use signals. Finally, multiscale geographically weighted regression (MGWR) was used to explain the spatial differentiation of NACCL. The results showed that NACCL was spatially heterogeneous and scale-dependent: natural factors exerted relatively broad and stable constraints, accessibility factors influenced conversion around transport corridors and urban fringes, and socioeconomic factors showed stronger local variation. This study provides a reliable remote sensing interpretation method, a patch-scale diagnostic perspective, and multiscale driving evidence for refined cultivated-land protection and high-risk area monitoring.

Accurate mapping of rubber plantations is essential for sustainable land management and forest conservation in tropical regions. However, existing methods face two major challenges: persistent cloud cover limits the effectiveness of optical remote sensing in tropical areas, and regional phenological heterogeneity hinders the transferability of fixed-parameter approaches. This study proposes a Phenology-Adaptive Rubber Plantation Mapping (PARM) framework that integrates Sentinel-1 SAR time-series data with optimally selected spectral indices through a cascading constraint architecture. The framework operates as a structurally coherent system wherein SAR-derived phenological anchors explicitly govern downstream optical analysis across three internally dependent modules. First, three key phenological nodes—leaf-off start (LOS), fastest greening point (FGP), and full canopy point (FCP)—are extracted directly from SAR VH-polarization backscatter time series, enabling cloud-independent extraction of phenological temporal anchors. Second, the Jeffries–Matusita (JM) distance, evaluated within SAR-constrained phenological windows, is employed to identify the optimal vegetation and water indices for each region from six candidate spectral indices. Third, a time-weighted Rubber Plantation Discrimination Index (RPDI) is constructed using the selected indices and locally extracted phenological nodes, thereby amplifying the coupled signals of canopy greenness and moisture dynamics during critical phenological transitions. The framework was validated in Hainan Island and Vietnam, two regions with contrasting phenological regimes, using a spatial-block partitioning protocol (leave-one-subregion-out combined with DBSCAN-based clustering) designed to prevent samples from the same plantation from occurring in both training and test subsets. Within the Dynamic World forest mask, PARM achieved overall accuracies of 92.04% and 91.24%, respectively (93.57% and 91.00% on the fully held-out Qionghai City and Gia Lai province subregions), with Kappa coefficients exceeding 0.81 in both regions, consistently outperforming schemes based on raw spectral bands, individual spectral indices or direct multi-source time-series stacking. Error structure analysis revealed that residual classification failures are primarily associated with landscape fragmentation, stand immaturity, and residual cloud contamination, delineating the generalizability boundaries of the framework. These results demonstrate that tightly coupling SAR-based phenological characterization with adaptive optical index selection through a cascading constraint architecture provides a reliable foundation for rubber plantation mapping in cloud-prone tropical regions.