Mapping potential desertification-prone areas in North-Eastern Algeria using logistic regression model, GIS, and...

Area of Science:

• Environmental Science
• Geospatial Analysis
• Ecology

Background:

• Desertification is a critical environmental challenge, particularly impacting Algeria.
• Assessing desertification sensitivity is crucial for effective land management and conservation efforts.

Purpose of the Study:

• To evaluate desertification-prone areas in North-Eastern Algeria (Tebessa province).
• To apply logistic regression modeling and geomatics for desertification risk assessment.

Main Methods:

• Utilized geomatics-based approaches and logistic regression modeling (LRM).
• Selected key indicators: Topsoil Grain Size Index (TGSI), Normalized Difference Vegetation Index (NDVI), Aridity Index (AI), and Anthropic pressure on the steppe environment (APSE).
• Analyzed land surface conditions related to soil, vegetation, climate, and human impact.

Main Results:

• Aridity Index (AI) and TGSI were identified as the primary drivers of desertification risk.
• NDVI and APSE were found to be secondary contributing factors.
• The logistic regression model achieved a satisfactory accuracy with an Area Under Curve (AUC) of 0.94.
• Desertification risk increased along a North-to-South gradient, with 58.72% of the area classified as very high risk.

Conclusions:

• Logistic Regression Model (LRM) is an effective tool for monitoring desertification, especially in data-scarce regions.
• The study provides a scientific basis for developing and implementing policies to combat desertification in high-risk areas.
• Understanding the spatial distribution of desertification risk is essential for targeted interventions.