A Systematic Review of Digital Twin for Forest Protection

The global forest landscape is currently navigating an unprecedented crisis of ecological stability, with data from Global Forest Watch indicating that forest loss in 2024 surged by 80% compared to 2023. This acceleration, driven by anthropogenic climate change and intensifying wildfire regimes, has exposed the fundamental limitations of traditional monitoring paradigms. Conventional approaches relying on static Geographic Information Systems (GIS) suffer from high latency and an inability to provide the bidirectional feedback loops necessary for real-time intervention. In response, this study proposes a comprehensive five-layer cyber-physical architecture for Forest Digital Twins (FDTs). Following a PRISMA-guided methodology, the research evaluates 54 high-impact studies to address the "Reality Gap" between virtual simulations and physical ecosystems. Findings indicate that Level 3 bidirectional DTs—exemplified by initiatives such as Destination Earth (DestinE) and the TEMA Project—significantly improve risk assessment. Quantitative analysis demonstrates that Attention-based Long Short-Term Memory (ALSTM) networks achieve 96.72% accuracy in wildfire prediction, while DT-guided autonomous reforestation reduces per-tree costs from $3.75 to $0.11. The review concludes that integrating "space–air–ground" data technology chains with secure blockchain ledgers provides a resilient pathway for next-generation forest stewardship, moving the field from reactive observation to proactive, high-precision guardianship.