The photovoltaic (PV) industry boom has accelerated the need for accurately understanding the spatial distribution of PV energy systems. The synergy of remote sensing and artificial intelligence presents significant prospects for PV energy monitoring. Currently, numerous studies have focused on extracting rooftop PV systems from airborne or satellite imagery, but their small-scale and size-varying characteristics make the segmentation results suffer from PV internal incompleteness and small PV omission. To address these issues, this study proposed a size-aware deep learning network called Rooftop PV Segmenter (RPS) for segmenting small-scale rooftop PV systems from high-resolution imagery. In detail, the RPS network introduced a Semantic Refinement Module (SRM) to sense size variations of PV panels and reconstruct high-resolution deep semantic features. Moreover, a Feature Aggregation Module (FAM) enhanced the representation of robust features by continuously aggregating deeper features into shallower ones. In the output stage, a Deep Supervised Fusion Module (DSFM) was employed to constrain and fuse the outputs at different scales to achieve more refined segmentation. The proposed RPS network was tested and shown to outperform other models in producing segmentation results closer to the ground truth, with the F1 score and IoU reaching 0.9186 and 0.8495 on the publicly available California Distributed Solar PV Array Dataset (C-DSPV Dataset), and 0.9608 and 0.9246 on the self-annotated Heilbronn Rooftop PV System Dataset (H-RPVS Dataset). This study has provided an effective solution for obtaining a refined small-scale energy distribution database.

Soil erosion in granite-derived weathering mantles poses serious threats to slope stability and ecological sustainability in subtropical regions. While polyacrylamide (PAM) is widely used to improve soil structure, its concentration-dependent effects on multiple soil functions remain unclear. This study developed a multifunctional Soil Function Index (SFI) framework integrating erosion resistance (SFI₁), water regulation (SFI₂), and ecological function (SFI₃) to evaluate the effects of PAM application (0‰, 1‰, 3‰, 5‰, 7‰) on gully-prone sandy material. Herein, SFI₁ was quantified through shear strength (τ) and soil erodibility (K_r); SFI₂ was assessed using soil hydraulic parameters (saturated hydraulic conductivity and water retention curves) and SFI₃ was derived from the grass root system analysis. The results showed that SFI₁ and SFI₂ increased nonlinearly with PAM concentration, reaching maximum values of 0.983 and 0.980 at 7‰, with K_r reduced by 77.3% and non-capillary porosity (NAP) increased by 8.1%. In contrast, SFI₃ peaked at 0.858 under 3‰ and declined sharply to 0.000 at 7‰, due to micropore over-compaction, reduced aeration, and limited plant-available water. The total SFI exhibited a unimodal trend, with a maximum of 0.755 at 3‰, beyond which ecological suppression offset physical improvements. These findings demonstrate that PAM modifies soil multifunctionality through pore-scale restructuring, inducing function-specific thresholds and trade-offs. A PAM concentration of 3‰ is identified as optimal, achieving a balance between erosion control, hydrological performance, and ecological viability in the management of subtropical granite-derived sandy slopes.

Granites, widely distributed in the Earth’s crust, undergo pedogenic processes, shaping diverse soil-mantled landscapes influenced by climatic factors in different regions. Investigating the geochemical signatures in granite weathering profiles across varying climatic conditions provides valuable insights into the intricate interplay between weathering and landscape evolution. In this study, the geochemical features, particularly major and rare earth elements, and the weathering degree of granites across temperate to subtropical regions in China were examined. The results indicated significant variations in the geochemical characteristics of granite weathering profiles, both at a pedon and regional scale (p < 0.01). With increasing hydrothermal conditions from north to south, soil pH shifted from neutral to acidic, accompanied by the leaching of major elements (K₂O, Na₂O, CaO, and MgO) and the enrichment of Al and Fe. The total rare earth elements (∑REEs) ranged from 75 to 352 ppm, and light rare earth elements (LREEs) from 71 to 317 ppm, exhibiting less significant variations across the study area, while heavy rare earth elements (HREEs) showed higher concentrations in the subtropical region (3 to 35 ppm). Plagioclases dominated the weathering process in temperate regions, with K-feldspar progressively increasing and, eventually, dominating from temperate to subtropic regions, resulting in a shift in clay minerals from 2:1-type in the temperate to 1:1-type in the subtropic. The chemical index of alteration (CIA) and comprehensive weathering index (W) increased from fresh rock to residual soils along the weathering profiles and from north to south across the study area, ranging from 50.72 to 97.44 and 35.11 to 70.62, respectively. The intensified granite weathering degree was significantly influenced by climatic conditions (p < 0.05), especially the multi-year average precipitation (22.4%) and relative humidity (9.1%) (p < 0.01). Gully erosion on the granite weathering mantle was concentrated in granites with a comprehensive weathering index exceeding 52.51, and the spatial variation of the granite weathering degree aligned with the spatial distribution of gully density across the study area.