Variogram as a tool for assessing the quality of climate models

• Autor(a) principal: Zanetti, Vitor Baccarin
• Data de Publicação: 2017
• Outros Autores: Chou, Sin Chan, Gandini, Maria Luiza Teófilo, Lyra, André
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Revista Interdisciplinar de Pesquisa em Engenharia
• Texto Completo: https://periodicos.unb.br/index.php/ripe/article/view/21612
• topic: Variogram. Geostatistics. Climate model. Model quality assessment.

Climate models are very sensitive to spatial resolution. Their skill must always be verified, as they involve several phenomena which take place in different scales. For that reason, some of those phenomena must be adequately parameterized, with appropriate techniques of upscaling. The proposal of this work is to present the variogram as a tool for assessing the quality of climate models, based on comparison of model results with different spatial discretization. Results of the ETA Model from INPE are presented in two different levels of discretisation: for resolutions higher than 5 km, to which non-hydrostatic models must be taken into account, and for resolution lower than 8 km, to which hydrostatic models are suited. Variograms for 36 km, 18 km, 4 km, 2 km and 1 km are calculated and their results are discussed, together with other metrics for quality assessment of forecast models. Variograms showed that there is an impact of grid coarseness over these numerical models, which was less noticeable in plots of precipitations for coarser grids.