Riparian woodlands play a significant role in hosting plant diversity and maintaining soil and water resources. However, riparian woodlands are highly sensitive to fluvial and human disturbances, and most are now degraded as a result. In this study, we analyze variation in soil nutrients and plant diversity and their relationships in riparian woodlands along the middle and lower reaches of the Yellow River in China, based on field investigation. Our results indicate that soil nutrient content and plant diversity were lowest in plots located closest to the river. Specifically, we found a hump-shaped relationship with increasing distance from river. This result can be attributed to the intermediate disturbance hypothesis,which explains that species diversity increase at intermediate disturbance levels. However, total species richness within each subzone was greatest close to the river, due primarily to the high level of species turnover observed among plots located closest to the river. Soil nutrients in the riparian woodlands were positively correlated with plant diversity across all distances. Specifically, soil TC, NO3 --N, and A-P contents were significantly correlated with plant species richness and diversity. This relationship also conformed to a hump-shaped response curve between species richness and soil nutrients, though species richness increases with increasing nutrient levels. Results from this study can provide a basis for sustainable management of riparian ecosystems.

Wildfire risk assessment requires integrating heterogeneous geospatial data to capture complex environmental dynamics. This study develops a hierarchical multimodal fusion framework combining high-resolution aerial imagery, historical fire data, topography, meteorology, and vegetation indices within Google Earth Engine. We introduce three progressive fusion levels: a single-modality baseline (NAIP-WHP), fixed-weight fusion (FIXED), and a novel geographically adaptive dynamic-weight approach (FUSED) that adjusts feature contributions based on regional characteristics like human activity intensity or aridity. Machine learning benchmarking across 49 U.S. regions reveals that Support Vector Machines (SVM) applied to the FUSED dataset achieve optimal performance, with an AUC-ROC of 92.1%, accuracy of 83.3%, and inference speed of 1.238 milliseconds per sample. This significantly outperforms the fixed-weight fusion approach, which achieved an AUC-ROC of 78.2%, and the single-modality baseline, which achieved 73.8%, representing relative improvements of 17.8% and 24.8\%, respectively. The 10-meter-resolution risk heatmaps demonstrate operational viability, achieving an 86.27% hit rate in Carlsbad Caverns, NM. SHAP-based interpretability analysis reveals terrain dominance and context-dependent vegetation effects, aligning with wildfire ecology principles.