پایش و شبیه سازی تغییرات کاربری و پوشش اراضی در حوضه آبریز کارون بزرگ (مقاله علمی وزارت علوم)
درجه علمی: نشریه علمی (وزارت علوم)
آرشیو
چکیده
در سال های اخیر افزایش جمعیت و گسترش شهرنشینی، تغییرات گسترده ای در کاربری و پوشش اراضی ایجاد کرده است. این تغییرات، پیامدهای زیانباری همچون کاهش تنوع زیستی، بیابان زایی و جنگل زدایی را به دنبال داشته و اکوسیستم حوضه های آبریز را با تهدیدات جدی زیست محیطی مواجه کرده است. ازاین رو، پایش تغییرات کاربری و پوشش اراضی در این مناطق اهمیت ویژه ای دارد. پژوهش حاضر به مدل سازی تغییرات کاربری و پوشش اراضی در حوضه آبریز کارون بزرگ می پردازد تا اثرات زیست محیطی این تغییرات تحلیل و پایش شود. در ابتدا، نقشه های کاربری و پوشش اراضی منطقه از داده های Globeland30 تهیه شد. سپس، عوامل مؤثر بر این تغییرات تحت سناریوی رشد اراضی انسان ساخت استخراج شد. در ادامه ضمن تهیه لایه های موردنیاز این عوامل در نرم افزار GIS، با استفاده از دو روش PSI و MEREC، دو نقشه مبتنی بر سناریوی رشد اراضی انسان ساخت تولید شد. برای ارزیابی و بهبود مدل سازی، دو رویکرد مورد مقایسه قرار گرفتند. در رویکرد اول، خروجی روش PSI با مدل ترکیبیSVM-CA-Markov ادغام شد و در رویکرد دوم، خروجی روش MEREC با همان مدل ترکیب شد. مدل سازی تغییرات کاربری و پوشش اراضی این حوضه برای سال 2020 انجام و نتایج حاصل با استفاده از منحنی ROC صحت سنجی شد. نتایج نشان داد که الگوریتم مبتنی بر خروجی روش MEREC و مدل ترکیبیSVM-CA-Markov، با مقدار AUC برابر با 0.89، از دقت بالاتری برخوردار است. با توجه به نتایج صحت سنجی، برای پیش بینی تغییرات کاربری و پوشش اراضی در افق 2040 از الگوریتم MEREC+SVM-CA-Markov استفاده شد. نتایج نهایی نشان داد که وسعت اراضی انسان ساخت تا سال 2040 به بیش از 1862 کیلومترمربع خواهد رسید. اگرچه این گسترش می تواند به رونق اقتصادی منطقه کمک کند، اما پیامدهای زیست محیطی متعددی، ازجمله تخریب منابع طبیعی و افزایش فشار بر اکوسیستم منطقه را به دنبال خواهد داشت. ازاین رو، تدوین برنامه های مدیریتی بهینه برای کاهش اثرات زیان بار این تغییرات ضروری است.Monitoring and simulation of land use and land cover changes in the Great Karun Basin
Introduction In recent decades, population growth and urbanization have led to extensive changes in land use and land cover (LULC), resulting in consequences such as deforestation, desertification, greenhouse gas emissions, increased surface temperature, reduced biodiversity, and decreased ecosystem service quality. The driving forces behind LULC changes are not uniform across different parts of the world and are heavily dependent on human factors. While human factors play a primary role in these changes, other factors such as climate change and biophysical characteristics can also influence the intensity and trend of these changes. Spatial-temporal analysis of LULC changes using remote sensing (RS) and geographic information systems (GIS), provides crucial information for natural resource management and sustainable development. The use of satellite data obtained from remote sensing and multi-criteria decision-making (MCDM) analyses based on GIS can significantly improve the decision-making process in land use and land cover management. These technologies enable the identification and monitoring of both minor and major changes on the earth's surface with high precision. Various models, such as CA-Markov and machine learning algorithms, are employed to predict LULC changes and make effective land management decisions. These studies can offer important policy recommendations for ecosystem management and natural disaster monitoring. Previous studies have predominantly focused on smaller scales, such as sections of a basin, cities, or sub-basins. However, the current research is conducted on a larger scale, specifically the Great Karun Basin. In this study, the modeling scenario combines PSI and MEREC models, which are efficient multi-criteria decision-making (MCDM) methods based on GIS. Materials & Methods In this research, to model land use/land cover (LULC) changes in the Great Karun Basin, the driving factors of built-up areas growth were first identified, and digital maps of these criteria were prepared in a GIS environment. Then, using the PSI and MEREC methods, each criterion was weighted, and two maps were produced for the potential growth of built-up areas for the year 2010. Multi-temporal LULC maps from the Globeland30 data were used as input for the model. The LULC changes in the region for the year 2020 were simulated using two algorithmic combinations: PSI+SVM-CA-Markov and MEREC+SVM-CA-Markov in the TerrSet software. Finally, the simulation's accuracy was validated using the ROC curve, and the optimal algorithm for simulating LULC in 2040 was selected. Results & Discussion In this study, the driving factors for the growth of built-up areas in the Great Karun Basin were weighted using the PSI and MEREC methods. The results indicated that the criterion of distance from built-up areas had the highest importance, while the distance from power transmission lines had the least importance. Potential growth maps for the year 2010 were prepared, showing that both models indicate the northeast-southwest axis and the eastern areas of the region have suitable potential for the growth and development of built-up areas. Subsequently, potential land use conversion maps for the year 2010 were created using the PSI+SVM-CA-Markov and MEREC+SVM-CA-Markov algorithms. The LULC changes for the year 2020 were simulated, and the results were validated against the 2020 ground truth map. Validation using the ROC curve showed that the MEREC+SVM-CA-Markov algorithm, with an AUC of 0.89, was more accurate than the PSI+SVM-CA-Markov algorithm, with an AUC of 0.86. Finally, using the MEREC+SVM-CA-Markov algorithm, the LULC changes for the year 2040 were simulated. The results showed that built-up areas would increase from 843.9075 square kilometers in 2020 to 1862.487 square kilometers in 2040, with the most significant growth occurring towards the southeast, southwest, and northeast of the region. Additionally, a significant decrease in forest areas was observed in this basin, highlighting the need for sustainable environmental management strategies. Conclusion Land use and land cover changes are serious environmental concerns today, directly caused by human activities. These changes cause ecological disturbances, climate changes, and destructive effects on the environment and ecosystems. LULC modeling is a fundamental tool for natural resource management that helps organizations adopt optimal strategies for basin land management. The catchment area of the Great Karun River is one of the strategic areas of Iran, where the importance of modeling land use changes is very high. The results of this research can help reduce the ecological, economic, and social consequences caused by land use changes in the Great Karun Basin. Finally, it is recommended that future research incorporates more comprehensive decision-making criteria and compares the results of the algorithm used in this study with other land use and land cover change modeling algorithms to enhance the accuracy of the outcomes. Additionally, incorporating various climate change scenarios into the modeling process could facilitate a more precise assessment of the impacts of such changes on ecosystems and land use/land cover. This, in turn, would enable the development of more effective management strategies for the conservation of natural resources and sustainable development.
تبلیغات
