Publicação
A multimethod interpolation approach for mapping the spatial distribution of rainfall in southwest Iberian Peninsula
| Resumo: | Eight spatial interpolation models were used to map the spatial distribution of precipitation in the southwestern sector of the Iberian Peninsula (22330 km2) over 40 years (1980/1981-2019/2020). Rainfall data from 103 meteorological stations were used to generate the interpolation models, namely inverse distance weight (IDW) with 6, 12 and 24 points, regression spline (RS), thin spline (TS), universal kriging with spherical and Gaussian variogram (UK_Sphe and UK_gauss, respectively) and multilinear regression (MR), based on physiographic and geographic variables. Furthermore, 32 rainfall stations were used to assess the performance of the previous methods through 7 statistical metrics Ordinary Least Squares (OLS), Root Mean Square Error (RMSE), Normalized root mean square error (NRMSE), Coefficient of determination (R2), Nash-Sutcliffe efficiency coefficient (NSE), Mean Absolute Error (Bias and MAE). Based on these metrics, UK_gauss and IDW_6 provided the best adjustments, whereas MR presented the highest errors. All methods were suitable to predict the spatial distribution of rainfall, but adjustments are conditioned by the features of the study area, gauge density and gauge spatial distribution. |
|---|---|
| Autores principais: | Ruiz-Ortiz, Verónica |
| Outros Autores: | Maria Neto Paixão Vazquez Fernandez Martins, Helena; Granja Martins, Fernando Miguel; Vélez-Nicolás, Mercedes; Isidoro, Jorge; García-López, Santiago |
| Assunto: | Rainfall Inverse distance weight Spline Universal kriging Multilinear regression Southwestern Iberian Peninsula |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | artigo |
| Tipo de acesso: | acesso restrito |
| Instituição associada: | Universidade do Algarve |
| Idioma: | inglês |
| Origem: | Sapientia - Universidade do Algarve |
| Resumo: | Eight spatial interpolation models were used to map the spatial distribution of precipitation in the southwestern sector of the Iberian Peninsula (22330 km2) over 40 years (1980/1981-2019/2020). Rainfall data from 103 meteorological stations were used to generate the interpolation models, namely inverse distance weight (IDW) with 6, 12 and 24 points, regression spline (RS), thin spline (TS), universal kriging with spherical and Gaussian variogram (UK_Sphe and UK_gauss, respectively) and multilinear regression (MR), based on physiographic and geographic variables. Furthermore, 32 rainfall stations were used to assess the performance of the previous methods through 7 statistical metrics Ordinary Least Squares (OLS), Root Mean Square Error (RMSE), Normalized root mean square error (NRMSE), Coefficient of determination (R2), Nash-Sutcliffe efficiency coefficient (NSE), Mean Absolute Error (Bias and MAE). Based on these metrics, UK_gauss and IDW_6 provided the best adjustments, whereas MR presented the highest errors. All methods were suitable to predict the spatial distribution of rainfall, but adjustments are conditioned by the features of the study area, gauge density and gauge spatial distribution. |
|---|
Atividades financiadas
Não foram encontrados projetos financiados.