RABENIAINA Anjara Davio Ulrick1, 2, a, RAHANTANOMENJANAHARY Telina Tsialonina Melinà1, 2, b, *, RAMAHEFARISON Heriniaina1, 2, c
RAHANTANOMENJANAHARY Telina Tsialonina Melinà
1 Biodiversity, AGROECOLOGY, EVOLUTION, and SYSTEMATICS OF EMBRYOPHYTES
2 Laboratory of Atmosphere, Climate and Ocean Dynamics, University of Antananarivo, Science and Technology, Antananarivo, Madagascar.
a[email protected], b[email protected], c[email protected]
*Corresponding Author
This study analyzes changes in temperature and vegetation cover in the Mahajanga II district (Madagascar) between 1994 and 2024. The spatial distribution of temperatures was determined using an autoencoder, while vegetation cover dynamics were assessed using NDVI indices obtained from Landsat images. The results show a significant increase in temperatures over the study period, more pronounced in inland areas than along the coast, with a cumulative increase of approximately 2°C for central areas and 1°C for coastal areas. At the same time, dense and very dense vegetation cover has declined sharply, replaced by very sparse vegetation and the expansion of bare soil, while water surfaces have also decreased. These changes indicate that vegetation loss contributes directly to local temperature increases by reducing the capacity of ecosystems to moderate the climate and maintain hydrological balance. The use of the autoencoder has made it possible to identify the areas most vulnerable to warming, providing a relevant tool for environmental planning and sustainable land management. The preservation and restoration of vegetation therefore appear to be essential measures for limiting local warming and maintaining the ecological functions of the district.
Vegetation cover, Temperature, NDVI, Autoencoder, Artificial Intelligence
RABENIAINA Anjara Davio Ulrick, RAHANTANOMENJANAHARY Telina Tsialonina Melinà, RAMAHEFARISON Heriniaina (2025). Impact of Changes in Vegetation Cover on Temperature in the Mahajanga II district, Madagascar. Journal of Artificial Intelligence and Systems, 7, 112–122. https://doi.org/10.33969/AIS.2025070107.
[1] Bonan, G. B. (2008). Forests and Climate Change: Forcings, Feedbacks, and the Climate Benefits of Forests. Science, 320(5882), 1444–1449.
[2] Courville, A., Goodfellow, I., & Bengio, Y. (2016). Deep Learning. MIT Press.
[3] Duveiller, G., Hooker, J., & Cescatti, A. (2018). The mark of vegetation change on Earth’s surface energy balance. Nature Communications, 9, 679. DOI: 10.1038/s41467-017-02810-8Killick, R., Fearnhead, P., & Eckley, I. A. (2012).
[4] Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. Science, 313(5786), 504–507.
[5] Jensen, J. R. (2015). Introductory Digital Image Processing: A Remote Sensing Perspective (4th Edition). Pearson.
[6] Peng, S. S., Piao, S., Zeng, Z., Ciais, P., Zhou, L., Li, L. Z. X., Myneni, R. B., Yin, Y., & Zeng, H. (2014). Afforestation in China cools local land surface temperature.
[7] Rousseeuw, P. J. (1987). Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20, 53–65.
[8] Rouse, J. W., Haas, R. H., Schell, J. A., & Deering, D. W. (1973). Monitoring vegetation systems in the Great Plains with ERTS. NASA SP-351, 309–317.
[9] Tucker, C. J. (1979). Red and photographic infrared linear combinations for monitoring vegetation. Remote Sensing of Environment, 8, 127–150.
[10] Xu, H. (2020). A review of vegetation indices based on Landsat imagery for assessing vegetation coverage. Remote Sensing Applications: Society and Environment, 17, 100291.