As climate change accelerates, wildfires are increasing in frequency, scale, and intensity, posing severe threats to ecosystems, biodiversity, and human societies. Rising global temperatures, shifting precipitation patterns, and prolonged droughts have made many regions increasingly fire-prone. At the same time, wildfire-induced habitat loss has become a major driver of species endangerment and extinction. In this context, there is an urgent need for accurate and spatially explicit wildfire prediction models to inform ecological conservation strategies under future climate scenarios. This project proposes the development of a deep learning framework based on Convolutional Neural Networks (CNNs) to predict wildfire burned areas using multi-source spatiotemporal data. Our approach integrates historical burned area records (e.g., GFED4), high-resolution climate reanalysis datasets (e.g., ERA5), vegetation indices (such as NDVI and LAI), and land cover classifications. The CNN model will be trained to learn complex spatiotemporal patterns from past fire events and generalize them to predict fire-prone areas under future climate projections, including SSP2-4.5 and SSP5-8.5 scenarios. In addition to fire prediction, this study places particular emphasis on assessing the ecological consequences of increased wildfire activity. By overlaying the predicted burned areas with species habitat maps from databases such as the IUCN Red List, we aim to identify biodiversity hotspots at elevated future fire risk. This will allow us to quantify potential species loss and prioritize areas for conservation interventions. The integration of AI-based prediction with biodiversity impact analysis represents an innovative step toward more comprehensive, data-driven environmental planning. The expected outcomes of the project include: (1) a robust CNN-based wildfire prediction model trained on diverse environmental datasets; and (2) future burned area datasets under different climate scenarios. These results will support policy-making in climate adaptation, wildfire management, and biodiversity conservation. To accomplish this work, we request access to storage resources to support the project. Training deep neural networks on large-scale spatiotemporal datasets requires substantial data storage capacity. With adequate storage to complement GPU computing, this research will contribute to a deeper understanding of climate-driven wildfire dynamics and their ecological consequences, ultimately informing long-term strategies to mitigate biodiversity loss.