Environmental numerical models are deterministic tools widely used to simulate and predict complex systems. However, they are unsatisfying since they do not provide information about the uncertainty associated with their predictions. Conversely, uncertainty assessment of model outputs can be useful to guide environmental agencies in improving computer models. We propose a Bayesian hierarchical model to obtain spatially varying uncertainty associated with a numerical model output. We show how we can learn about such uncertainty through suitable stochastic data fusion modeling using some external validation data. The model is illustrated by providing the uncertainty map associated with a temperature output over the northeastern United States.
Paci, L., Cocchi, D., Gelfand, A., Quantifying uncertainty associated with a numerical model output, in Cabras, C. S., Di Battista, D. B. T., Racugno, R. W. (ed.), Proceedings of SIS 2014, CUEC, Cagliari 2014: 1- 6 [http://hdl.handle.net/10807/98624]
Quantifying uncertainty associated with a numerical model output
Paci, LuciaPrimo
;
2014
Abstract
Environmental numerical models are deterministic tools widely used to simulate and predict complex systems. However, they are unsatisfying since they do not provide information about the uncertainty associated with their predictions. Conversely, uncertainty assessment of model outputs can be useful to guide environmental agencies in improving computer models. We propose a Bayesian hierarchical model to obtain spatially varying uncertainty associated with a numerical model output. We show how we can learn about such uncertainty through suitable stochastic data fusion modeling using some external validation data. The model is illustrated by providing the uncertainty map associated with a temperature output over the northeastern United States.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.