Extend abstract

Background and Objective: Land use change is one of the most critical environmental challenges on a global scale and is considered a key factor in global ecosystem changes. The destructive effects and excessive use of the natural environment by humans are one of the main factors in environmental degradation and land use change. In many countries, due to a lack of awareness of its destructive consequences and inadequate planning, this phenomenon has become one of the most critical risks. In Iran, the inappropriate distribution of population, the urban physical development, and the undesirable use of the natural environment by humans are among the main reasons for land use changes. Northern Iran is one of the central regions for land use changes due to the high concentration of population and agricultural activities, which requires the identification and evaluation of the current patterns of changes in land use categories. Furthermore, in order to create a future perspective trend of LULC changes, it is necessary to model the trend of these changes. Today, preparing LULC maps using satellite images is one of the most effective solutions for understanding the environmental change conditions and an efficient tool for ecological planning. Landsat images are considered one of the valuable tools in this field due to their continuous monitoring of land surface changes on a long-time scale (since 1972) and suitable spatiotemporal resolution.

Materials and Methods: In the present research, the trend of LULC changes in Nowshahr county in northern Iran was evaluated using Landsat multispectral images over a 30-year time scale (1994, 2009, and 2024) with a time interval of 15 years. To monitor changes, first, pre-processing of images, including geometric, radiometric, and atmospheric corrections, was performed, and then land use maps of the study area were prepared by applying the Maximum Likelihood algorithm in five categories: water bodies, agriculture, vegetation, bare lands, and built-up. Subsequently, by combining the effective criteria in land use changes and LULC maps in 1994, 2009, and 2024 in the Cellular Automata-Logistic Regression (CA-LR) model, the modeling process was carried out, and the LULC changes of Nowshahr County were modeled until 2039.

Results and Discussion: Our findings, with an overall accuracy of more than 89% and a Kappa coefficient of more than 86%, indicated that during the study period, vegetation and bare lands in the region decreased by 124.61 km2 and 148.24 km2, respectively, and agriculture and built-up areas increased by 224.15 km2 and 28.83 km2, respectively. The most significant decrease in vegetation occurred in the central to northern parts of Nowshahr County between 1994 and 2009, and the most significant increase in agriculture occurred in the central and southern parts between 2009 and 2024. Also, based on the output of the CA-LR model, it is predicted that the trend of changes until 2039 will be in line with previous periods, with less intensity. In this period (2024 to 2039), agriculture and built-up areas will grow by 59.91 km2 and 13.73 km2, respectively. On the other hand, vegetation and bare lands will decrease by 26.57 km2 and 47.05 km2, respectively.

Conclusion: The trend of past, present, and future LULC changes in Nowshahr County demonstrates that this region is one of the most prone areas to environmental crises caused by land use changes, which requires the adoption of appropriate environmental planning to reduce and control LULC changes, and this planning should prioritize vegetation and agriculture. Also, the evaluation of the outputs of the CA-LR model of the trend of LULC changes, and its compliance with validated land use maps, indicates that this model, with a kappa coefficient of more than 93%, is considered one of the most efficient models for modeling land use changes alongside traditional models such as CA-Markov. The use of this model in a similar study is recommended to develop the methodology and research literature of LULC studies.