Evaluation of quantitative precipitation forecasts over southern South America
This paper examines current status of Quantitative Precipitation Forecasts (QPFs) over southern South America and particularly over Argentina, of 36 and 60-hour forecasts corresponding to the LAHM/CIMA regional model and the NCEP global medium-range forecast model (MRF) during a three-month period....
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | JOUR |
| Materias: | |
| Acceso en línea: | http://hdl.handle.net/20.500.12110/paper_00049743_v52_n2_p81_Saulo |
| Aporte de: |
| id |
todo:paper_00049743_v52_n2_p81_Saulo |
|---|---|
| record_format |
dspace |
| spelling |
todo:paper_00049743_v52_n2_p81_Saulo2023-10-03T14:03:12Z Evaluation of quantitative precipitation forecasts over southern South America Saulo, A.C. Ferreira, L. error analysis numerical model performance assessment precipitation (climatology) weather forecasting Argentina South America This paper examines current status of Quantitative Precipitation Forecasts (QPFs) over southern South America and particularly over Argentina, of 36 and 60-hour forecasts corresponding to the LAHM/CIMA regional model and the NCEP global medium-range forecast model (MRF) during a three-month period. Two indexes of skill for QPFs, the Equitable Threat Score (ETS) and the bias score, have been calculated over two subregions. Better ETS were obtained with MRF for predicting areas with precipitation intensities ranging from weak to moderate, while LAHM/CIMA performs better for higher amounts of precipitation. Both models show a tendency to a loss of accuracy as precipitation increases. Bias scores are relatively small for both models, except at larger rain limits, where LAHM/CIMA tends to overestimate the areas and MRF to underestimate. Limitations include diversity of precipitation regimes over the larger domain, and the scarcity of observations. While the former is common to similar studies, the latter becomes critical over southern South America and could affect the representativeness of interpolated fields. Generating a gridded dataset from an uneven (and coarse) observational network tends to spread precipitation leading to spurious rain in weak precipitation areas, and unreliable scores. The scores varied significantly when they were calculated against individual station observations, giving a more realistic depiction of model performance. These findings suggest that measures of skill should be applied and interpreted with caution, since different forecast verification strategies may lead to contradictory results. JOUR info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by/2.5/ar http://hdl.handle.net/20.500.12110/paper_00049743_v52_n2_p81_Saulo |
| institution |
Universidad de Buenos Aires |
| institution_str |
I-28 |
| repository_str |
R-134 |
| collection |
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) |
| topic |
error analysis numerical model performance assessment precipitation (climatology) weather forecasting Argentina South America |
| spellingShingle |
error analysis numerical model performance assessment precipitation (climatology) weather forecasting Argentina South America Saulo, A.C. Ferreira, L. Evaluation of quantitative precipitation forecasts over southern South America |
| topic_facet |
error analysis numerical model performance assessment precipitation (climatology) weather forecasting Argentina South America |
| description |
This paper examines current status of Quantitative Precipitation Forecasts (QPFs) over southern South America and particularly over Argentina, of 36 and 60-hour forecasts corresponding to the LAHM/CIMA regional model and the NCEP global medium-range forecast model (MRF) during a three-month period. Two indexes of skill for QPFs, the Equitable Threat Score (ETS) and the bias score, have been calculated over two subregions. Better ETS were obtained with MRF for predicting areas with precipitation intensities ranging from weak to moderate, while LAHM/CIMA performs better for higher amounts of precipitation. Both models show a tendency to a loss of accuracy as precipitation increases. Bias scores are relatively small for both models, except at larger rain limits, where LAHM/CIMA tends to overestimate the areas and MRF to underestimate. Limitations include diversity of precipitation regimes over the larger domain, and the scarcity of observations. While the former is common to similar studies, the latter becomes critical over southern South America and could affect the representativeness of interpolated fields. Generating a gridded dataset from an uneven (and coarse) observational network tends to spread precipitation leading to spurious rain in weak precipitation areas, and unreliable scores. The scores varied significantly when they were calculated against individual station observations, giving a more realistic depiction of model performance. These findings suggest that measures of skill should be applied and interpreted with caution, since different forecast verification strategies may lead to contradictory results. |
| format |
JOUR |
| author |
Saulo, A.C. Ferreira, L. |
| author_facet |
Saulo, A.C. Ferreira, L. |
| author_sort |
Saulo, A.C. |
| title |
Evaluation of quantitative precipitation forecasts over southern South America |
| title_short |
Evaluation of quantitative precipitation forecasts over southern South America |
| title_full |
Evaluation of quantitative precipitation forecasts over southern South America |
| title_fullStr |
Evaluation of quantitative precipitation forecasts over southern South America |
| title_full_unstemmed |
Evaluation of quantitative precipitation forecasts over southern South America |
| title_sort |
evaluation of quantitative precipitation forecasts over southern south america |
| url |
http://hdl.handle.net/20.500.12110/paper_00049743_v52_n2_p81_Saulo |
| work_keys_str_mv |
AT sauloac evaluationofquantitativeprecipitationforecastsoversouthernsouthamerica AT ferreiral evaluationofquantitativeprecipitationforecastsoversouthernsouthamerica |
| _version_ |
1807320957228941312 |