Monitoring the Health Status of Ghouts in El Oued Region (Southeast Algeria) Using Support Vector Machine Models

Absztrakt

El Oued, located in the Souf region (southeastern Algeria), is characterized by its unique architecture and palm groves. These are cultivated in depressions excavated into the sandy soil to access the water table, thereby enabling date palm cultivation. These depressions, known as ‘Ghouts’, constitute an exceptional agricultural innovation. In 2011, the Food and Agriculture Organization of the United Nations (FAO) recognized the Ghout system as a Globally Important Agricultural Heritage System (GIAHS) due to its profound historical, socio-economic, and cultural significance. However, this valuable agricultural heritage, essential for the livelihoods of the inhabitants, is currently threatened by various factors of degradation, including the rising or lowering of the water table, extreme climatic conditions, rural exodus, groundwater pollution, urban expansion, and more. This article examines the contribution of artificial intelligence to the monitoring of Ghout health status in the municipality of El Oued, employing the Support Vector Machine (SVM) model. The research focuses particularly on these agricultural systems that are currently threatened by severe degradation. The results reveal a steady decline in the Ghouts’ area, shrinking from 61.79 hectares in 2015 to 23.90 hectares by 2023. A significant degradation was observed between 2018 and 2019, with a loss of 17.65 hectares documented within a single year. The factors contributing to this degradation include climate change, increasing urbanization, and the evolution of agricultural practices that have altered the structure and function of these essential agricultural systems. Furthermore, this comprehensive analysis expands upon the initial findings by providing detailed SVM methodology documentation, enhancing uncertainty quantification, presenting a comparative analysis with alternative machine learning methods, and integrating a socio-economic analysis with policy implications and long-term monitoring recommendations for the GIAHS conservation.