The Zhegu area in southern Tibet is situated in the central and eastern part of the Tethys Himalayan tectonic belt, with the Kangbuzhenri area being abundant in gneissic granites. This study examines the petrology, chronology, and geochemistry of the Kangbuzhenri gneissic granite, providing insights into its Pan-African and Early Paleozoic geological evolution. The zircon U-Pb chronology indicates an upper intercept age of ~539 Ma, reflecting Pan-African orogenic events in the eastern part of the Tethys Himalayan tectonic belt, and a lower intercept age of ~144 Ma, representing a late tectonic–thermal event. Geochemically, the gneissic granites are calc-alkaline peraluminous rocks with high SiO₂ and Al₂O₃ contents and low TiO₂, P₂O₅, MgO, and FeO^T contents. The gneissic granites are enriched in LREE and LILEs (Rb, Pb, Th, U, etc.), but relatively depleted in HREE and HFSEs (Nb, Ti, P, etc.). Most of them show a weak negative δEu anomaly, except for two samples which show a significant negative δEu anomaly due to the crystallization of plagioclase. Based on the above study, most of the gneissic granites exhibited the characteristics of an I-type granite, while two of the samples were a highly differentiated I-type granite with S-type affinities. All the above characteristics indicate that the gneissic granite likely originated from the partial melting of crustal materials and sediments with a minor involvement of mantle-derived materials. Combined with the previous chronological studies, the Kangbuzhenri gneissic granites were formed in an extensional tectonic environment during post-collision orogeny and then they were influenced by the Kerguelen mantle plume tectonic–thermal event around ~144 Ma and the subsequent Southern Tibet Detachment System (STDS).

Cropland non-grain and non-agriculturalization monitoring (CNNM) is of great significance for ensuring national food security. Recently, semantic segmentation has been a promising approach that can fully exploit the advantages of high-resolution remote sensing imagery. It is of great significance to clarify the performance of semantic segmentation algorithms for CNNM and their influencing factors. This paper contributes to research on deep learning-based semantic segmentation models for CNNM by investigating the robustness and generalization ability of detection models in complex scenarios. To investigate these models, seven mainstream segmentation models were examined with respect to two self-constructed unmanned aerial vehicle (UAV)-based cropland monitoring datasets under various experimental settings. The experimental results reveal several key findings: For the datasets employed in the present study, remote sensing semantic segmentation models demonstrate superior performance relative to generic visual models. Increasing backbone depth does not guarantee significant performance improvements. The recognition challenges posed by task-specific categories such as agricultural facility land further underscore the need for tailored deep learning semantic segmentation algorithms for CNNM. These experimental results provide valuable insights into the practical performance of deep learning semantic segmentation models and offer useful guidance for future semantic segmentation research in the field of CNNM.