Zoning and Identification of Factors Affecting Illegal Livestock Grazing in Golestan National Park Using Logistic Regression

Introduction: A comprehensive understanding of human behavioral drivers is pivotal for the effective conservation of habitats and protected areas. Environmental crimes encompass a wide spectrum of activities, including habitat destruction, vegetation clearance, deforestation, livestock grazing, hunting and wildlife trafficking, fishing, indiscriminate waste disposal, mining, and more. By discerning patterns in these violations, managers can allocate resources strategically to those groups exhibiting the least compliance with conservation regulations and implement more effective protection strategies. Golestan National Park, given its significant conservation value at both national and regional levels, is subject to a myriad of human-induced violations that inflict substantial physical, ecological, and economic damages. Consequently, precisely identifying areas susceptible to such violations can be a pivotal step in elucidating and enhancing the public's comprehension of the imperative to protect this region. Moreover, it can inform environmental policymakers and political and judicial officials. Owing to the complexity inherent in identifying and locating instances of illegal livestock grazing and the influence of diverse environmental and anthropogenic factors, a logistic regression model was employed to delineate crime zones within the study area. This quantitative approach provides a means to measure the prevalence of offenses and to identify practical solutions for mitigating crime within the designated region.Materials and Methods: Golestan National Park, Iran's first national park, has been under protection since 1957. To zone and identify the factors influencing illegal livestock grazing in Golestan National Park, a logistic regression model was applied, using the TerrSet IDRISI software. Logistic regression is a suitable model for zoning, providing an equation to predict and explain the changes in a dependent variable based on independent variables using existing data. This method uses maximum likelihood estimation to find the best factors that fit the model. The regression equation predicting the probability of illegal livestock grazing on the border of Golestan National Park was obtained from logistic regression modeling using independent physiographic, vegetation, and human variables.Results: Results from the model and the regression equation predicting the probability of illegal livestock grazing within Golestan National Park showed that the most influential factor was the distance from the road, with a coefficient of -5.679. Altitude with a coefficient of 4.192, ruggedness with a coefficient of 4.051, and distance from springs with a coefficient of -2.34 were the next most important factors. The Aspect factor with a coefficient of 0.0015 was determined to be the least influential independent variable. The coefficients of the slope, distance from the road, distance from Ranger’s stations, and distance from water sources (rivers and springs) were negative, indicating a decrease in the probability of violations with increasing distance from these variables. The Pseudo R2 index was 0.2402, the ROC index was 0.8916, and the Chi-Square index was 8493.8447, confirming the acceptable fit of the model and indicating a very good and high accuracy of the model execution and a strong relationship with the probability values obtained from the logistic regression model. The removal of each independent variable individually and the comparison of sensitivity analysis using statistical indices confirmed the independent variable of distance from the road as the most important factor affecting the occurrence of illegal grazing.Discussion: The results of the logistic regression model provided the best fitting function to describe the relationship between the factors influencing illegal livestock grazing and to predict its occurrence. Accordingly, almost all areas around Golestan National Park with access roads are susceptible to the presence of illegal herders. Important factors such as a lack of enough rangers, and a lack of ranger stations, have also been very influential in exacerbating the occurrence of illegal grazing. Moreover, the lack of a defined boundary for the protected areas has caused illegal herders to identify suitable habitats on the margins of the roads around Golestan National Park and enter the park for illegal grazing. Therefore, by identifying the centers of violations and increasing the efficiency of patrolling in these areas, it is possible to significantly prevent the occurrence of illegal grazing and help improve the quality of the ecosystem by restoring damaged areas.