An evaluative technique for drought impact on variation in agricultural LULC using remote sensing and machine learning

Mustapha, Musa; Zineddine, Mhamed

doi:10.1007/s10661-024-12677-0

Cited by 5 publications

References 99 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Assessing drought stress impact on olive trees using AI and Sentinel‐2 satellite imagery

Majikumna,

Zineddine,

Alaoui

2024

World Water Policy

View full text Add to dashboard Cite

Olive tree cultivation in the Mediterranean region faces significant challenges due to drought stress, highlighting the critical role of water resources in sustaining agricultural productivity. This study proposes a novel approach to evaluate the impacts of drought on olive trees by analyzing LULC variations from 2018 to 2023. We integrated Sentinel‐2 data for LULC classification with the Vegetation Health Index (VHI) for drought assessment. Three machine learning algorithms, namely Support Vector Machine (SVM), K‐Nearest Neighbor (KNN), and Random Forest (RF), were employed to assess LULC classification, with RF demonstrating the best overall performance. Our results show that both RF and KNN achieved significantly better performance when combining spectral bands with spectral indices, compared to using spectral bands alone, and both outperformed SVM. The LULC analysis reveals an initial growth in the olive cultivation area, peaking at 148.573 km2 in 2020, followed by a decline to 124.276 km2 in 2022, with a slight recovery to 129.485 km2 in 2023. This trend suggests that while the initial expansion was driven by favorable conditions, subsequent reductions may be linked to climate variability, resource limitations, or economic factors. This observation is supported by a Pearson correlation coefficient of 0.74 between VHI and rainfall. To promote sustainable olive cultivation, the study recommends optimizing irrigation practices, enhancing climate and vegetation monitoring, investing in drought‐resistant olive varieties, and adopting precision agriculture technologies.

show abstract