Centre de documentation
scientifique

Applying time-dependent variance-based global sensitivity analysis to represent the dynamics of an agent-based model of land use change / A. Ligmann-Zielinska in International journal of geographical information science IJGIS, vol 24 n°11-12 (december 2010)  

Public [article]  inInternational journal of geographical information science IJGIS > vol 24 n°11-12 (december 2010) . - pp 1829 - 1850 Titre : Applying time-dependent variance-based global sensitivity analysis to represent the dynamics of an agent-based model of land use change Type de document : Article/Communication Auteurs : A. Ligmann-Zielinska, Auteur ; L. Sun, Auteur Année de publication : 2010 Article en page(s) : pp 1829 - 1850 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] analyse de sensibilité [Termes IGN] changement d'occupation du sol [Termes IGN] complexité [Termes IGN] données spatiotemporelles [Termes IGN] incertitude des données [Termes IGN] modèle orienté agent [Termes IGN] variance Résumé : (Auteur) The growing body of knowledge on modeling land use systems points to epistemic uncertainty as one of the challenging obstacles in development and application of agent-based models (ABMs). To decrease outcome uncertainty, sensitivity analysis (SA) is performed as part of model verification and validation. Oftentimes, however, it is inadequately addressed, partly because of the lack of tools and techniques that focus on an explicit evaluation of ABM dynamics. The nonlinear processes, inherent in such models, necessitate longitudinal SA with time path investigation of input-output relationships of endogenous variables. In response to the outlined deficiencies, this study investigates the potential of time-dependent global sensitivity analysis (time-GSA) in examining the dynamics of outcome uncertainty of a simple ABM of land use change. Specifically, we apply first and total order sensitivity indices to decompose variance of output landscape fragmentation, apportioned to model inputs for multiple time steps and multiple realizations of the ABM. We focus the analysis on selected complex systems characteristics including preference uncertainty, path dependence, access to information, and magnitude of interactions and feedbacks. We conclude that the factor sensitivity measures vary significantly during model execution. Consequently, a static snapshot of ABM sensitivity, taken at the end of the simulation, is inadequate when deciding on factor prioritization and reduction. Assuming that ABM dynamics is a result of factor interaction, we observe a distinct time lag of nonlinearity, which unfolds after the formation of the seeds of development. Therefore, we argue for further application of time-GSA in ABM as one of the visual quantitative techniques contributing to evaluation of ABM nonlinearity. Numéro de notice : A2010-552 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : 10.1080/13658816.2010.490533 En ligne : https://doi.org/10.1080/13658816.2010.490533 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=30744 [article]

Exemplaires(2)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
079-2010071	RAB	Revue	Centre de documentation	En réserve L003	Disponible
079-2010072	RAB	Revue	Centre de documentation	En réserve L003	Disponible