Assessment of Desertification Risk for Natural Pine Forest Lands and Prediction with Artificial Intelligence
JOURNAL OF SUSTAINABLE FORESTRY, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1080/10549811.2026.2699697
- Dergi Adı: JOURNAL OF SUSTAINABLE FORESTRY
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, BIOSIS, CAB Abstracts, Compendex, Environment Index, Geobase, Greenfile, Public Affairs Index, Academic Search Ultimate (EBSCO), Natural Science Collection (ProQuest), Engineering Source (EBSCO), Materials Science & Engineering Collection (ProQuest), Technology Collection (ProQuest)
- Ondokuz Mayıs Üniversitesi Adresli: Evet
Özet
The objective of the current study is to assess the desertification risk of natural pine forest areas by considering the indicators addressed in the DIS4ME (Desertification Indicator System for the Mediterranean Europe) approach, with a particular focus on the natural pine forests in the Engiz basin (1165.5 km2) of Samsun Province. In the assessment of desertification risk in the DIS4ME, eight indicators (precipitation, aridity, soil texture, stoniness, vegetation cover, density, slope, soil depth, and aspect) were taken into account. However, the indicator index values obtained from the DIS4ME approach were not directly taken from the model but were made more sensitive through the Fuzzy-AHP approach. To this end, 73 samples were taken from the field to determine the basic physico-chemical properties of soils in the basin. The desertification risk of the area within Engiz's natural pine forest was determined to range between 1.08 and 1.32, with the lowest and highest values respectively, and statistically, an R2 value of 0.99 was determined. Additionally, to apply the obtained values to similar ecological regions or large scale areas, the results were estimated using artificial neural networks (ANNs). Therefore, another distinctive approach of the current work is the use of ANN alone in conjunction with soil analysis to forecast desertification risk. In the present study, the obtained data were predicted using ANNs and the predicted results ranged between 1.09 and 1.32. It was observed that the region is generally at risk of desertification, and the ANNs result's showed a close parallel to the model results.