Uncertainty propagation in models driven by remotely sensed data

Often, little importance is given to the problem of how uncertainty propaga tes in models driven by remotely sensed data, and what the effects of uncer tainty might be on the output of these models. In this paper, a general pro cedure to support a characterisation of uncertainty in the generation of re mote sensing (RS) products is proposed. The procedure can be used with mode ls characterised by any degree of complexity and driven by any kind of data . It provides two useful tools to analyse models: uncertainty analysis (UA) , which allows the assessment of the uncertainty associated with model outp ut, and sensitivity analysis (SA), which is useful to determine how much ea ch source of uncertainty contributes to model output uncertainty. Uncertain ty modelling, i.e. finding suitable tools to represent uncertainty, is a ke y step in performing UA and SA. A general error model for quantitative rast er data is described. Different applications of UA and SA are proposed, and , in the last part of the paper, an example of UA and SA on a model for bur ned area detection is discussed. (C) 2001 Elsevier Science Inc. All rights reserved.