Contribution of consensus methods to resolve sources of uncertainty in suitability maps modeling: application in the Zahrez El Gharbi,Steppe of Algeria |
| |
| Authors: | Boudjema Sehl Mohamed Said Guettouche Hocine Ait Mouheb María Teresa Camacho Olmedo |
| |
| Institution: | 1.Laboratoire de Géomorphologie et Géorisques, Faculté des Sciences de la Terre, de Géographie et Aménagement du Territoire,Université des Sciences et de la Technologie Houari Boumediene (USTHB),Algiers,Algeria;2.Laboratoire Ecologie Végétale et Environnement, Faculté des Sciences Biologiques,Université des Sciences et de la Technologie Houari Boumediene (USTHB),Algiers,Algeria;3.Departamento de Análisis Geográfico Regional y Geografía Física,University of Granada,Granada,Spain |
| |
| Abstract: | Modeling techniques have the power to predict land use change. Our research had three main aims: firstly, to compare and contrast the predictive performance of individual models versus consensus methods; secondly, to compare the different consensus methods; and finally, to compare their predictive performance for generalized and marginal land uses in the environment in the study area. Seven individual models were used to generate the suitability maps for each type of land use in Zahrez El Gharbi (Algeria), characterized by the suitability values for their occurrence. Seven consensus methods were created by combining the individual models. The predictive performance of the individual models and of the consensus methods was assessed by calculating the area under the curve (AUC). For each type of land use, the predictive quality of the consensus methods was considerably better than that of the individual models. The weighted averages method (WA) proved the most efficient of the seven consensus methods. The types of land use have a significant influence on the predictive quality, in that the AUC values increase in line with increases in the marginality of the particular type of land use. The improvements in predictive quality achieved by consensus methods in general and the weighted averages method (WA) in particular would substantially resolve the different sources of uncertainties resulting from the use of individual models and from the environmental characteristics of the different types of land use. |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
