|
|
Francone, C., Cassardo, C., Richiardone, R., & Confalonieri, R. (2012). Sensitivity Analysis and Investigation of the Behaviour of the UTOPIA Land-Surface Process Model: A Case Study for Vineyards in Northern Italy. Boundary-Layer Meteorology, 144(3), 419–430.
Abstract: We used sensitivity-analysis techniques to investigate the behaviour of the land-surface model UTOPIA while simulating the micrometeorology of a typical northern Italy vineyard (Vitis vinifera L.) under average climatic conditions. Sensitivity-analysis experiments were performed by sampling the vegetation parameter hyperspace using the Morris method and quantifying the parameter relevance across a wide range of soil conditions. This method was used since it proved its suitability for models with high computational time or with a large number of parameters, in a variety of studies performed on different types of biophysical models. The impact of input variability was estimated on reference model variables selected among energy (e.g. net radiation, sensible and latent heat fluxes) and hydrological (e.g. soilmoisture, surface runoff, drainage) budget components. Maximum vegetation cover and maximum leaf area index were ranked as the most relevant parameters, with sensitivity indices exceeding the remaining parameters by about one order of magnitude. Soil variability had a high impact on the relevance of most of the vegetation parameters: coefficients of variation calculated on the sensitivity indices estimated for the different soils often exceeded 100 %. The only exceptions were represented by maximum vegetation cover and maximum leaf area index, which showed a low variability in sensitivity indices while changing soil type, and confirmed their key role in affecting model results.
|
|
|
|
Hjelkrem, A. - G. R., Höglind, M., van Oijen, M., Schellberg, J., Gaiser, T., & Ewert, F. (2017). Sensitivity analysis and Bayesian calibration for testing robustness of the BASGRA model in different environments. Ecol. Model., 359, 80–91.
Abstract: Highlights • The parameters to be fixed were consistent across sites. • Model calibration must be performed separately for each specific case. • Possible to reduce model parameters from 66 to 45. • Strong model reductions must be avoided. • The error term for the training data were characterised by timing (phase shift). Abstract Proper parameterisation and quantification of model uncertainty are two essential tasks in improvement and assessment of model performance. Bayesian calibration is a method that combines both tasks by quantifying probability distributions for model parameters and outputs. However, the method is rarely applied to complex models because of its high computational demand when used with high-dimensional parameter spaces. We therefore combined Bayesian calibration with sensitivity analysis, using the screening method by Morris (1991), in order to reduce model complexity by fixing parameters to which model output was only weakly sensitive to a nominal value. Further, the robustness of the model with respect to reduction in the number of free parameters were examined according to model discrepancy and output uncertainty. The process-based grassland model BASGRA was examined in the present study on two sites in Norway and in Germany, for two grass species (Phleum pratense and Arrhenatherum elatius). According to this study, a reduction of free model parameters from 66 to 45 was possible. The sensitivity analysis showed that the parameters to be fixed were consistent across sites (which differed in climate and soil conditions), while model calibration had to be performed separately for each combination of site and species. The output uncertainty decreased slightly, but still covered the field observations of aboveground biomass. Considering the training data, the mean square error for both the 66 and the 45 parameter model was dominated by errors in timing (phase shift), whereas no general pattern was found in errors when using the validation data. Stronger model reduction should be avoided, as the error term increased and output uncertainty was underestimated.
|
|
