Estudio comparativo de tres modelos climáticos en Argentina
; Ferrelli, Federico
; Piccolo, Maria Cintia 
Otros responsables
Fecha
04/2017
Editorial
Universidade Federal do Rio de Janeiro. Instituto de Geociências
Revista
Anuário do Instituto de Geociências
Título del libro
Nombre del evento
Fecha del evento
Institución organizadora
ISSN
0101-9759
eISSN
1982-3908
ISBN
Idioma
Español
Tipo del evento
Clasificación temática
Palabras clave
Resumen
En algunas regiones del planeta, la escasez de estaciones meteorológicas con largas series de datos, ocasiona problemas para la caracterización del clima. En este contexto, adquieren importancia los modelos numéricos de obtención de datos climáticos. Uno de los mayores problemas que se presentan en Argentina es la desigual distribución espacial de las estaciones meteorológicas con registros de 50 años, lo que dificulta el estudio de su clima. El objetivo de esta investigación fue establecer para Argentina el modelo numérico que mejor represente las variables de: precipitación, humedad relativa, temperatura media, máxima y mínima del aire. Para ello, se analizaron datos históricos in situ pertenecientes al Servicio Meteorológico Nacional para el período 1950-2010 y se consideraron los resultados obtenidos de los modelos Reanálisis (NCEP/NCAR), WorldClim y el NEW01. Para su comparación se aplicaron diferentes técnicas estadísticas (análisis de varianza, test de diferencia mínima significativa, análisis de regresión, coeficientes de correlación, determinación y concordancia). Además, se utilizaron métodos de interpolación para la representación gráfica de los resultados mediante un SIG. Los modelos resultaron útiles para caracterizar variables climáticas en zonas carentes de registros. El WorldClim tuvo un mejor ajuste para los datos de temperatura media del aire, mientras que el NEW01 fue mejor para la caracterización de las precipitaciones.
The lack of meteorological data, with a long time series in some areas all over the world, generate difficulties when we have to characterize the climate. In this context, climate models are relevant to obtain these type of data. Particularly, in Argentina there are no meteorological stations that are distributed equidistantly with records over fifty years, therefore it climatic characterization is complicated. In this context, the aim of this study was to set for Argentina the climatic model that best represent the following variables: precipitation, relative humidity, mean, minimum and maximum temperature. To carry it out, historical meteorological data (from Servicio Meteorológico Nacional) registered In Situ and those who belong to three climatic models (Reanalysis NCEP/NCAR, WorldClima and NEW01) were analyzed for the period 1950-2010. For comparison, different statistical techniques were applied (Analysis of the variance, Fisher tests, regression analysis, correlation coefficients, determination and consistency). In addition, interpolation methods were applied in order to analyze the spatial distribution of each meteorological variable. As a result, climatic models are useful to characterize the climate in those areas with lack of information. Wordlclim has the best fit for mean temperature, while NEW01 for precipitation.
The lack of meteorological data, with a long time series in some areas all over the world, generate difficulties when we have to characterize the climate. In this context, climate models are relevant to obtain these type of data. Particularly, in Argentina there are no meteorological stations that are distributed equidistantly with records over fifty years, therefore it climatic characterization is complicated. In this context, the aim of this study was to set for Argentina the climatic model that best represent the following variables: precipitation, relative humidity, mean, minimum and maximum temperature. To carry it out, historical meteorological data (from Servicio Meteorológico Nacional) registered In Situ and those who belong to three climatic models (Reanalysis NCEP/NCAR, WorldClima and NEW01) were analyzed for the period 1950-2010. For comparison, different statistical techniques were applied (Analysis of the variance, Fisher tests, regression analysis, correlation coefficients, determination and consistency). In addition, interpolation methods were applied in order to analyze the spatial distribution of each meteorological variable. As a result, climatic models are useful to characterize the climate in those areas with lack of information. Wordlclim has the best fit for mean temperature, while NEW01 for precipitation.
Colecciones
Aprobación
Revisión
Complementado por
Referenciado por
Licencia Creative Commons
Excepto donde se indique lo contrario, la licencia de este ítem se describe como by-nc-sa
