Analysis and Evaluation of Ecological Risk in Sulaymaniyah Province, Iraq Using Machine Learning Techniques and MODIS Sensor Images in 2023

Document Type : Original Article

Authors

1 University of Tabriz

2 university of tabriz_tabriz_ir

10.22034/rsgi.2026.65456.1118

Abstract

One of the fundamental tools for identifying environmental crises in an area is to prepare ecological risk maps for that geographical region. The main goal of this research is to evaluate and analyze ecological risk in the Sulaymaniyah Province of Iraq based on selected indicators. The data related to the research indicators were collected in the Google Earth Engine environment and extracted from the MODIS sensor through coding. The CRITIC method was used for weighting the research data, and the Random Forest and k-means methods in machine learning were used for reducing and clustering the research indicators. To ensure the existence of significant differences between the clusters obtained by the k-means method, Levin's, ANOVA, and Games-Howell post hoc tests were applied. The clusters in this research were categorized into 5 groups ranging from very high to very low risk according to the analyses. The results showed that 62% of the total area of Sulaymaniyah Province in Iraq is in a moderate to very low ecological risk status. Spatial analysis with hot spots and subsequent Moran's spatial autocorrelation showed that the spatial distribution of high ecological risk clusters is not normal and follows a clustered pattern. These findings emphasize the need for meticulous monitoring planning in the Sulaymaniyah Province of Iraq.

Keywords

Main Subjects

Volume 6, Issue 21
January 2026
  • Receive Date: 20 January 2025
  • Revise Date: 04 June 2025
  • Accept Date: 25 February 2026