<![CDATA[Background: Malaria remains a major health issue in Jayapura, where climatic and landscape variability creates uneven transmission risks. Early identification of vulnerable areas is essential for supporting targeted control strategies.Objective: This study aims to develop a Random Forest model integrating remote-sensing environmental variables to identify malaria-prone areas in Jayapura District and Jayapura City.Methods: Environmental predictors including rainfall, land surface temperature, slope, NDVI, humidity, land use, and population density were linked to confirmed malaria cases. Data were split into 70% training and 30% testing datasets. Model performance was evaluated using accuracy, sensitivity, macro-sensitivity, and macro-specificity, and the outputs were used to generate a spatial malaria risk map.Results: The Random Forest model achieved an overall accuracy of 0.667, sensitivity of 0.833, macro-sensitivity of 0.800, and macro-specificity of 0.867, indicating good capability in identifying areas with higher malaria vulnerability. Feature importance analysis showed that rainfall, land surface temperature, and slope were the most influential predictors of malaria risk. High-risk areas were concentrated in coastal and urban zones, while peri-urban agricultural areas showed moderate risk and high-elevation regions exhibited lower vulnerability.Conclusion: Integrating remote-sensing environmental data with epidemiological information allows the Random Forest model to capture key malaria-risk patterns in Jayapura. The resulting spatial risk map can support targeted vector control strategies, improved surveillance, and more efficient allocation of public health resources toward Indonesia’s Malaria Elimination 2030 goals. ]]>
Copyrights © 2026
