African Journal of
Agricultural Research

  • Abbreviation: Afr. J. Agric. Res.
  • Language: English
  • ISSN: 1991-637X
  • DOI: 10.5897/AJAR
  • Start Year: 2006
  • Published Articles: 6862

Full Length Research Paper

Deforestation modeling using logistic regression and GIS (Case study: Northern Ilam forests, Ilam Province, Iran)

Saleh Arekhi1* and Ali Akbar Jafarzadeh2
1Department of Forest and Rangeland, University of Ilam, Ilam, Iran. 2Deparment of Forestry, University of Sari, Sari, Iran.
Email: [email protected]

  •  Accepted: 17 January 2012
  •  Published: 19 March 2012


This study aims to predict spatial distribution of deforestation and detects factors influencing forest 
degradation of Northern forests of Ilam province. For this purpose, effects of six factors including 
distance from road and settlement areas, forest fragmentation index, elevation, slope and distance from 
the forest edge on the forest deforestation are studied. In order to evaluate the changes in forest, 
images related to TM1988, ETM+
2001 and ETM+
2007 are processed and classified. There are two classes 
as, forest and non-forest in order to assess deforestation factors. The logistic regression method is 
used for modeling and estimating the spatial distribution of deforestation. The results show that about 
19,294 ha from forest areas are deforested in the 19 years. Modeling results also indicate that more 
deforestation occurred in the fragmented forest cover and in the areas of proximity to forest/non forest 
edge. Furthermore, slope and distance from road and settlement areas had negative relationships with 
deforestation rates. Meanwhile, deforestation rate is decreased with increase in elevation. Finally, a 
simple spatial model is presented that is able to predict the location of deforestation by using logistic 
regression. The validation was also tested using ROC approach which was found to be 0.96. 
Key words: Deforestation modeling, remote sensing, logistic regression, Zagros forests, Ilam province, Ilam.