نوع مقاله : پژوهشی
نویسندگان
1 دانش آموختة دکتری علوم و مهندسی آبخیز (حفاظت آب و خاک) گروه مهندسی منابع‌طبیعی، دانشکدة کشاورزی و منابع‌طبیعی، دانشگاه هرمزگان، بندرعباس، ایران
2 استاد گروه مهندسی منابع‌طبیعی، دانشکده کشاورزی و منابع‌طبیعی، دانشگاه هرمزگان، بندرعباس، ایران
3 دانشیار گروه مهندسی منابع‌طبیعی، دانشکده کشاورزی و منابع‌طبیعی، دانشگاه هرمزگان، بندرعباس، ایران
4 کارشناس اداره کل منابع‌طبیعی و آبخیزداری استان هرمزگان، بندرعباس، ایران
5 کارشناس شرکت سهامی آب منطقه ای استان هرمزگان، بندرعباس، ایران
چکیده
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Introduction and Goal
Over the past few decades, subsidence has become a major problem on a global scale. Given the increase in this phenomenon in the country, predicting and spatial modeling of land subsidence and identifying areas prone to subsidence are essential to reduce the negative effects of this environmental impacts. Given the threats and destructive effects of land subsidence on water and soil resources, managing this phenomenon and prevent its spread is a key issue in the sustainable development of the country. Subsidence studies is essential to gain insight, identify research gaps, improve methodology, and ensure that new research contributes to the existing knowledge base. In this regard, the Kerdi Shirazi Plain is of great importance due to the presence of the Mourkerdi Forest Reserve, its biodiversity, and its agricultural status, and given that subsidence is expanding in this region, identifying areas prone to subsidence risk is essential to combat this phenomenon and reduce the damages caused by it. The main objective of this study was to develop a spatial model for subsidence risk using GLM machine learning model in the studied area. Therefore, in this study, for the first time, a machine learning model is used to identify areas prone to land subsidence risk in the Kerdi Shirazi Plain. Also, the contribution and relative importance of various factors controlling subsidence were quantitatively determined using the Cforest machine model.
Materials and Methods
In this study, to prepare a land subsidence risk map in the study area, a database related to the factors controlling this phenomenon was first prepared. In this regard, the existing map of subsidence in the area were prepared by conducting field visits and collection data related to the presence or absence of subsidence in the ArcGIS software. After identifying the most important factors controlling subsidence, the relationship between the effective variables and subsidence points with and without it was examined using the GLM machine learning model. The output of the prediction model (values 0 to 1) was classified into five subsidence risk classes including very low risk (0 - 0.2), low (0.2 - 0.4), moderate (0.4 - 0.6), high (0.6 - 0.8) and very high (0.8 - 1) and presented as a subsidence risk map. The Cforest machine models is the best model for determining the importance of variables controlling various hazards, especially subsidence. The efficiency of this model is higher and its less error is lower compared to other models. Therefore, the Cforest model was used to determine the relative importance of each of the effective and restraining factors of this phenomenon.
Results and Discussion
The performance of the GLM model in predicting subsidence risk was evaluated using the area under the AUC curve. The area under the ROC curve, was found to be 0.99. This data indicates the excellent performance of the GLM model in identifying subsidence points. Based on the results of this model, 2180 and 441 hectares of the total area were in the very low and low subsidence sensitivity classes. On the other hand, 402, 447 and 1113 hectares of the total area were in the moderate, high and very high subsidence sensitivity classes, respectively. Also, 24.3% of the total study area has a very high susceptibility to subsidence risk. The central parts of the region with agricultural, horticultural and forest uses adjacent to agricultural and horticultural lands, share a common aquifer, and the risk of land subsidence was very high. Also, according to the results of the relative importance of variables, three main variables, including land use, groundwater level, and groundwater drawdown were among the most important variables controlling subsidence risk in the study area. The results of this study of variables controlling the risk of land subsidence showed for the first time that this phenomenon can be a serious threat to forest lands, especially forests in arid and semi-arid regions.
Conclusion and Suggestions
Based on the results obtained, the sensitivity to land subsidence risk in the central parts of the region (agricultural and horticultural uses and forest use in the vicinity of agricultural and horticultural lands) was very high and high. The results of the study of variables controlling the risk of subsidence in the studied area showed that the reason for the increase in groundwater exploitation was the development of agricultural and horticultural activities in the Kerdi Shirazi plain. Therefore, in order to reduce the negative effects of land subsidence, it is recommended to prevent activities that increase the exploitation of groundwater resources. It is also suggested that watershed management activities (flood spreading) be carried out upstream of the studied area in order to recharge the regional aquifer.
کلیدواژه‌ها [English]
)