Spatial and Temporal Evolution Analysis Model of Biodiversity with Multi-source Heterogeneous Data Fusion

Biodiversity conservation faces the challenges of data acquisition and analysis, which restricts the accurate implementation of conservation decision-making. In this paper, a spatial-temporal evolution analysis model of biodiversity is constructed by fusing heterogeneous data from multiple sources, which innovatively combines deep learning and knowledge graph technology to achieve the deep fusion of data from multiple sources, such as satellite remote sensing, infrared camera, and ground survey. The model performs well in animal target detection, biodiversity index prediction and trend analysis, and successfully identifies multiple biodiversity hotspots and provides early warnings. The study provides a novel technical path for biodiversity monitoring and conservation, which is of great practical significance for promoting ecological environment management.