Connection methods are essential for integrating environmental factors with machine learning models for landslide susceptibility assessments.However,current research does not consider the different characteristics of ...Connection methods are essential for integrating environmental factors with machine learning models for landslide susceptibility assessments.However,current research does not consider the different characteristics of continuity and discreteness within environmental factors and therefore does not analyze the suitability of various connection methods for different factor types.Moreover,the applicability of connection methods remains unclear when slope units are used as the basic assessment units.This study employed slope units as mapping units.The original data of 15 environmental factors,including 12 continuous and three discrete factors,and two connection methods,i.e.,frequency ratio(FR)and modified FR(MFR),were separately used to construct the input datasets for landslide susceptibility modeling.The performance of four widely used machine learning models,random forest(RF),support vector machine(SVM),logistic regression(LR),and multilayer perceptron(MLP),was analyzed to evaluate the suitability of the connection methods for landslide susceptibility mapping.The results show that,in contrast to the decision tree-based RF model,the use of different connection methods for different factor types significantly influences the results of nontree models,including SVM,MLP,and LR.SVM model is the most sensitive to factor types and connection methods.When the MFR is used as the connection method,it improves the mapping results,especially for the SVM model.This shows that it is essential to consider the different characteristics of the data and select an appropriate environmental factor connection strategy to increase the effectiveness of landslide susceptibility evaluation.Furthermore,this study explored the role of connective methods from a sample distribution perspective,providing a theoretical foundation for the more rational and effective integration of environmental factors.展开更多
Algal blooms in lakes have become a common global environmental problem. Nowadays, remote sensing is widely used to monitor algal blooms in lakes due to the macroscopic, fast, real-time characteristics. However, it is...Algal blooms in lakes have become a common global environmental problem. Nowadays, remote sensing is widely used to monitor algal blooms in lakes due to the macroscopic, fast, real-time characteristics. However, it is often difficult to distinguish between algal blooms and aquatic vegetation due to their similar spectral characteristics. In this paper, we used modified vegetation presence frequency index(VPF) based on Moderate-resolution Imaging Spectroradiometer(MODIS) imagery to distinguish algal blooms from aquatic vegetation, and analyzed the spatial and temporal variations of algal blooms and aquatic vegetation from a phenological perspective for five large natural lakes with frequent algal bloom outbreaks in China from 2019 to 2020. We simplified the VPF method to make it with a higher spatial transferability so that it could be applied to other lakes in different climatic zones. Through accuracy validation, we found that the modified VPF method can effectively distinguish between algal blooms and aquatic vegetation, and the results vary from lake to lake. The highest accuracy of 97% was achieved in Hulun Lake, where the frequency of algal outbreaks is low and the extent of aquatic vegetation is stable, while the lowest accuracy of 76% was achieved in Dianchi Lake, which is rainy in summer and the lake is small. Analyses suggests that the time period when algal blooms occur most frequently might not coincide with that when they have the largest area. However, in most cases these two are close in terms of time period. The modified VPF method has a broad scope of application, is easy to implement, and has a high practical value. Furthermore, the method could be established using only a small amount of measured data, which is useful for water quality monitoring on large spatial scales.展开更多
基金supported by the National Key Research and Development Program of China(No.2023YFC3007202)Joint Research Project on Meteorological Capacity Enhancement of the China Meteorological Administration(No.23NLTSZ009)Project of the Department of Science and Technology of Sichuan Province(No.2024YFHZ0098)。
文摘Connection methods are essential for integrating environmental factors with machine learning models for landslide susceptibility assessments.However,current research does not consider the different characteristics of continuity and discreteness within environmental factors and therefore does not analyze the suitability of various connection methods for different factor types.Moreover,the applicability of connection methods remains unclear when slope units are used as the basic assessment units.This study employed slope units as mapping units.The original data of 15 environmental factors,including 12 continuous and three discrete factors,and two connection methods,i.e.,frequency ratio(FR)and modified FR(MFR),were separately used to construct the input datasets for landslide susceptibility modeling.The performance of four widely used machine learning models,random forest(RF),support vector machine(SVM),logistic regression(LR),and multilayer perceptron(MLP),was analyzed to evaluate the suitability of the connection methods for landslide susceptibility mapping.The results show that,in contrast to the decision tree-based RF model,the use of different connection methods for different factor types significantly influences the results of nontree models,including SVM,MLP,and LR.SVM model is the most sensitive to factor types and connection methods.When the MFR is used as the connection method,it improves the mapping results,especially for the SVM model.This shows that it is essential to consider the different characteristics of the data and select an appropriate environmental factor connection strategy to increase the effectiveness of landslide susceptibility evaluation.Furthermore,this study explored the role of connective methods from a sample distribution perspective,providing a theoretical foundation for the more rational and effective integration of environmental factors.
基金Under the auspices of National Key Research and Development Project of China (No. 2021YFB3901101)National Natural Science Foundation of China (No. 41971322, 42071336, 42001311, 41730104)+2 种基金Jilin Provincial Science and Technology Development Project (No. 20180519021JH)Youth Innovation Promotion Association of Chinese Academy of Sciences (No. 2020234)China Postdoctoral Science Foundation (No. 2020M681057)。
文摘Algal blooms in lakes have become a common global environmental problem. Nowadays, remote sensing is widely used to monitor algal blooms in lakes due to the macroscopic, fast, real-time characteristics. However, it is often difficult to distinguish between algal blooms and aquatic vegetation due to their similar spectral characteristics. In this paper, we used modified vegetation presence frequency index(VPF) based on Moderate-resolution Imaging Spectroradiometer(MODIS) imagery to distinguish algal blooms from aquatic vegetation, and analyzed the spatial and temporal variations of algal blooms and aquatic vegetation from a phenological perspective for five large natural lakes with frequent algal bloom outbreaks in China from 2019 to 2020. We simplified the VPF method to make it with a higher spatial transferability so that it could be applied to other lakes in different climatic zones. Through accuracy validation, we found that the modified VPF method can effectively distinguish between algal blooms and aquatic vegetation, and the results vary from lake to lake. The highest accuracy of 97% was achieved in Hulun Lake, where the frequency of algal outbreaks is low and the extent of aquatic vegetation is stable, while the lowest accuracy of 76% was achieved in Dianchi Lake, which is rainy in summer and the lake is small. Analyses suggests that the time period when algal blooms occur most frequently might not coincide with that when they have the largest area. However, in most cases these two are close in terms of time period. The modified VPF method has a broad scope of application, is easy to implement, and has a high practical value. Furthermore, the method could be established using only a small amount of measured data, which is useful for water quality monitoring on large spatial scales.
