%0 Journal Article
%T An¨¢lise de diferentes m¨¦todos de interpola  o para a precipita  o pluvial no Estado de Goi¨¢s = . Analysis of different spatial interpolation methods for rainfall in the state of Goi¨¢s
%A Elis Dener Lima Alves
%A Francisco Arthur Silva Vecchia
%J Acta Scientiarum : Human and Social Sciences
%D 2011
%I 
%X A espacializa  o de vari¨¢veis meteorol¨®gicas, especialmente a precipita  o pluvial, necessita de estudos constantes, visando ao aperfei oamento dos m¨¦todos de interpola  o e desenvolvimento de mapas. Neste contexto, o presente trabalho objetivou avaliar a acur¨¢cia de diferentes interpoladores (inverso quadr¨¢tico da distancia, krigagem, m¨ªnima curvatura e triangula  o) sobre a precipita  o anual do Estado de Goi¨¢s. Para isto foram utilizadas informa  es de 70 postos pluviom¨¦tricos. Os interpoladores foram avaliados por meio do erro m¨¦dio quadr¨¢tico ¨C RMSE, do desvio das m¨¦dias ¨C MBE, do coeficiente de determina  o (r2) e do coeficiente de correla  o de Pearson (r2). Os resultados mostraram que, entre os m¨¦todos analisados, o mapa resultante da triangula  o n o permitiu a extrapola  o para fora do dom¨ªnio dos postos pluviom¨¦tricos amostrais, obtida a melhor acur¨¢cia pelo m¨¦todo da m¨ªnima curvatura, seguido pelo m¨¦todo do inverso quadr¨¢tico da distancia. The spatial distribution of meteorological variables, especially rainfall, requires constant study, aimed at improving methods for interpolation and development of maps. In this context, this study aimed to evaluate the accuracy of different interpolation methods (inverse square distance, kriging, minimum curvature and triangulation) on the annual rainfall in the state of Goi¨¢s. To that end, data from 70 rainfall stations were used. The interpolators were assessed using the root mean square error (RMSE), mean bias error (MBE), the coefficient of determination (r2) and the Pearson correlation coefficient (r). The results showed that among the methods analyzed, the map resulting from triangulation did not allow extrapolation outside the domain of rainfall sampling stations. The best accuracy was obtained by the minimum curvature method, followed by the inverse square distance method.
%K interpoladores
%K precipita  o
%K mapas
%K interpolators
%K rainfall
%K maps
%U http://www.periodicos.uem.br/ojs/index.php/ActaSciHumanSocSci/article/view/13815/pdf