Deep learning models for enhanced forest-fire prediction at Mount Kilimanjaro, Tanzania: Integrating satellite images, weather data and human activities data
Bibliographic record
Abstract
Forest fires (FFs) are a growing threat to ecosystems and human settlements, particularly in vulnerable regions such as Mount Kilimanjaro, Tanzania. Accurate and timely fire prediction is essential to mitigate these risks and improve fire management strategies. This study develops and evaluates advanced Deep Learning (DL) models for FF prediction by integrating spatiotemporal vegetation indices, environmental data, and human activity indicators. Specifically, Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNNs), and Convolutional Long Short-Term Memory (ConvLSTM) models were employed to analyze Sentinel-2 satellite imagery and weather data, along with anthropogenic factors such as beekeeping, tourism, agriculture, and deforestation rates. ConvLSTM, a hybrid model designed to capture spatial and temporal data, achieved superior performance with an AUROC of 0.9785 and an F1 score of 0.9618, surpassing the LSTM and CNN models. By integrating human-induced activities with environmental data, these models provide accurate and actionable predictions for fire management in high-risk areas. This study demonstrates the potential of ConvLSTM in developing operational tools for early fire detection, improving resource allocation, and guiding preventive strategies in fire-prone regions such as Mount Kilimanjaro. • The study highlights the importance of incorporating local human-induced factors (such as beekeeping and tourism) into deep learning models for more reliable forest fire predictions. • Developed and evaluated ConvLSTM, LSTM, and CNN models for predicting forest fires using multi-source data (satellite imagery, weather data, and human activity data). • ConvLSTM outperformed other models with the highest accuracy (98.08%) and AUROC (97.78%), effectively capturing both spatial and temporal dynamics in fire prediction.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.004 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".