Hybrid Computational Intelligence Models for Improvement Gully Erosion Assessment

被引:37
作者
Arabameri, Alireza [1 ]
Chen, Wei [2 ,3 ,4 ]
Lombardo, Luigi [5 ]
Blaschke, Thomas [6 ]
Dieu Tien Bui [7 ]
机构
[1] Tarbiat Modares Univ, Dept Geomorphol, Jalal Ale Ahmad Highway, Tehran 9821, Iran
[2] Minist Nat Resources, Key Lab Coal Resources Explorat & Comprehens Util, Xian 710021, Peoples R China
[3] Xian Univ Sci & Technol, Coll Geol & Environm, Xian 710054, Peoples R China
[4] Shaanxi Prov Key Lab Geol Support Coal Green Expl, Xian 710054, Peoples R China
[5] Univ Twente, Fac Geoinformat Sci & Earth Observat ITC, Drienerlolaan 5, NL-7522 NB Enschede, Netherlands
[6] Univ Salzburg, Dept Geoinformat Z GIS, A-5020 Salzburg, Austria
[7] Duy Tan Univ, Inst Res & Dev, Da Nang 550000, Vietnam
基金
奥地利科学基金会;
关键词
land degradation; soil conservation; ensemble; machine learning; environmental management; LANDSLIDE SUSCEPTIBILITY ASSESSMENT; EVIDENTIAL BELIEF FUNCTION; LOGISTIC-REGRESSION; SPATIAL PREDICTION; CONDITIONING FACTORS; STATISTICAL-MODELS; EVENTS APPLICATION; HAZARD ASSESSMENT; NEURAL-NETWORKS; RANDOM FOREST;
D O I
10.3390/rs12010140
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Gullying is a type of soil erosion that currently represents a major threat at the societal scale and will likely increase in the future. In Iran, soil erosion, and specifically gullying, is already causing significant distress to local economies by affecting agricultural productivity and infrastructure. Recognizing this threat has recently led the Iranian geomorphology community to focus on the problem across the whole country. This study is in line with other efforts where the optimal method to map gully-prone areas is sought by testing state-of-the-art machine learning tools. In this study, we compare the performance of three machine learning algorithms, namely Fisher's linear discriminant analysis (FLDA), logistic model tree (LMT) and naive Bayes tree (NBTree). We also introduce three novel ensemble models by combining the aforementioned base classifiers to the Random SubSpace (RS) meta-classifier namely RS-FLDA, RS-LMT and RS-NBTree. The area under the receiver operating characteristic (AUROC), true skill statistics (TSS) and kappa criteria are used for calibration (goodness-of-fit) and validation (prediction accuracy) datasets to compare the performance of the different algorithms. In addition to susceptibility mapping, we also study the association between gully erosion and a set of morphometric, hydrologic and thematic properties by adopting the evidential belief function (EBF). The results indicate that hydrology-related factors contribute the most to gully formation, which is also confirmed by the susceptibility patterns displayed by the RS-NBTree ensemble. The RS-NBTree is the model that outperforms the other five models, as indicated by the prediction accuracy (area under curve (AUC) = 0.898, Kappa = 0.748 and TSS = 0.697), and goodness-of-fit (AUC = 0.780, Kappa = 0.682 and TSS = 0.618). The analyses are performed with the same gully presence/absence balanced modeling design. Therefore, the differences in performance are dependent on the algorithm architecture. Overall, the EBF model can detect strong and reasonable dependencies towards gully-prone conditions. The RS-NBTree ensemble model performed significantly better than the others, suggesting greater flexibility towards unknown data, which may support the applications of these methods in transferable susceptibility models in areas that are potentially erodible but currently lack gully data.
引用
收藏
页数:25
相关论文
共 82 条
[1]   Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS) [J].
Allouche, Omri ;
Tsoar, Asaf ;
Kadmon, Ronen .
JOURNAL OF APPLIED ECOLOGY, 2006, 43 (06) :1223-1232
[2]   A novel ensemble bivariate statistical evidential belief function with knowledge-based analytical hierarchy process and multivariate statistical logistic regression for landslide susceptibility mapping [J].
Althuwaynee, Omar F. ;
Pradhan, Biswajeet ;
Park, Hyuck-Jin ;
Lee, Jung Hyun .
CATENA, 2014, 114 :21-36
[3]  
[Anonymous], 2013, WORLD SCI
[4]   Gully Head-Cut Distribution Modeling Using Machine Learning Methods-A Case Study of NW Iran [J].
Arabameri, Alireza ;
Chen, Wei ;
Blaschke, Thomas ;
Tiefenbacher, John P. ;
Pradhan, Biswajeet ;
Dieu Tien Bui .
WATER, 2020, 12 (01)
[5]   Comparison of machine learning models for gully erosion susceptibility mapping [J].
Arabameri, Alireza ;
Chen, Wei ;
Loche, Marco ;
Zhao, Xia ;
Li, Yang ;
Lombardo, Luigi ;
Cerda, Artemi ;
Pradhan, Biswajeet ;
Dieu Tien Bui .
GEOSCIENCE FRONTIERS, 2020, 11 (05) :1609-1620
[6]   Application of Probabilistic and Machine Learning Models for Groundwater Potentiality Mapping in Damghan Sedimentary Plain, Iran [J].
Arabameri, Alireza ;
Roy, Jagabandhu ;
Saha, Sunil ;
Blaschke, Thomas ;
Ghorbanzadeh, Omid ;
Dieu Tien Bui .
REMOTE SENSING, 2019, 11 (24)
[7]   Spatial Pattern Analysis and Prediction of Gully Erosion Using Novel Hybrid Model of Entropy-Weight of Evidence [J].
Arabameri, Alireza ;
Cerda, Artemi ;
Tiefenbacher, John P. .
WATER, 2019, 11 (06)
[8]   Gully erosion susceptibility mapping using GIS-based multi-criteria decision analysis techniques [J].
Arabameri, Alireza ;
Pradhan, Biswajeet ;
Rezaei, Khalil ;
Conoscenti, Christian .
CATENA, 2019, 180 :282-297
[9]   An ensemble model for landslide susceptibility mapping in a forested area [J].
Arabameri, Alireza ;
Pradhan, Biswajeet ;
Rezaei, Khalil ;
Lee, Saro ;
Sohrabi, Masoud .
GEOCARTO INTERNATIONAL, 2020, 35 (15) :1680-1705
[10]   Spatial prediction of gully erosion using ALOS PALSAR data and ensemble bivariate and data mining models [J].
Arabameri, Alireza ;
Pradhan, Biswajeet ;
Rezaei, Khalil .
GEOSCIENCES JOURNAL, 2019, 23 (04) :669-686