000136340 001__ 136340
000136340 005__ 20240829112947.0
000136340 0247_ $$2doi$$a10.1007/s12518-024-00572-9
000136340 0248_ $$2sideral$$a139250
000136340 037__ $$aART-2024-139250
000136340 041__ $$aeng
000136340 100__ $$aKariminejad, Narges
000136340 245__ $$aDetection of land subsidence using hybrid and ensemble deep learning models
000136340 260__ $$c2024
000136340 5203_ $$aLand subsidence (LS) is among the most prominent forms of subsurface erosion and geomorphological hazards. This study used two deep learning (DL) models consisting of the hybrid CNN-RNN and ensemble DL (EDL) merged with two dense models. The main variables controlling LS (consisting of environmental, hydrological, hydrogeological, digital elevation model, and soil characteristics), were used as the input for the predictive DL models. Likewise, to establish the degree of performance of each parameter, different control points have been established. We then trained and tested our DL models using the receiver-operating characteristic-area under curve (ROC-AUC) and precision-recall plots. The measures based on the game theory consisting of permutation feature importance measure (PFIM) and SHapley Additive exPlanations (SHAP) were employed to assess the features relative importance and interpretability of the predictive model output. Our findings show that the ensemble CNN-RNN model performed well with the ROC-AUC curve (0.95) of class 1 (land subsidence) for training data for detecting and mapping land subsidence compared to EDL with the ROC curve (0.93) of class 1 (land subsidence) for training datasets. The CNN-RNN also performed well with the precision-recall curve (0.954) of class 1 for testing data for detecting and mapping land subsidence compared to the EDL model with the precision-recall curve (0.95) of class 1. The results of this research revealed that much of the study area is susceptible to land subsidence. The results of the model sensitivity analysis suggested that the groundwater drop rate is the most sensitive for the model. Based on the SHAP values, the groundwater drop rate was identified as the most contributed feature to the model output. The importance of this study is at a broader level, especially in arid and semiarid environments with similar geomorphological, and climatic conditions.
000136340 540__ $$9info:eu-repo/semantics/openAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/
000136340 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/publishedVersion
000136340 700__ $$aMohammadifar, Aliakbar
000136340 700__ $$aSepehr, Adel
000136340 700__ $$aGarajeh, Mohammad Kazemi
000136340 700__ $$aRezaei, Mahrooz
000136340 700__ $$0(orcid)0000-0001-8949-5676$$aDesir, Gloria$$uUniversidad de Zaragoza
000136340 700__ $$aQuesada-Román, Adolfo
000136340 700__ $$aGholami, Hamid
000136340 7102_ $$12000$$2427$$aUniversidad de Zaragoza$$bDpto. Ciencias de la Tierra$$cÁrea Geodinámica Externa
000136340 773__ $$g16 (2024), 593–610$$pAppl. geomat.$$tApplied geomatics$$x1866-9298
000136340 8564_ $$s5051321$$uhttps://zaguan.unizar.es/record/136340/files/texto_completo.pdf$$yVersión publicada
000136340 8564_ $$s2307385$$uhttps://zaguan.unizar.es/record/136340/files/texto_completo.jpg?subformat=icon$$xicon$$yVersión publicada
000136340 909CO $$ooai:zaguan.unizar.es:136340$$particulos$$pdriver
000136340 951__ $$a2024-08-29-11:26:06
000136340 980__ $$aARTICLE