The objective of this study is to quantify the magnitude of the spatial variations of the monthly precipitation and relate them to the regional temporal variations of the monthly precipitation at the spatial scale of the climate division of Oklahoma. The precipitation variations within a climate division are assumed to consist of three components: the systematic spatial variations, the mean temporal variations of the climate division, and random variations. The systematic spatial variations are defined with the long-term precipitation gradient. The mean temporal variations of the climate division are represented by the temporal variations of the spatial average of the precipitation observed at the stations included in the climate division; this average is called divisional precipitation. The random variations are estimated with the differences between the standardized values of station and divisional precipitation. This study shows that the magnitude of the random variations of the monthly precipitation of the nine climate divisions of Oklahoma is significant compared to the regional temporal variations. The quantification of the magnitude of the random variations is critical for the use of regional precipitation forecasts, because it allows one to define ranges of local precipitation around the divisional precipitation, and then to quantify the increase of the risk taken by local users of the regional precipitation forecasts.
Monthly precipitation, spatial variation, temporal variation, climate division, divisional value, Oklahoma.
F. Rossel, 8 rue des Lézards, 37600 Loches, FRANCE