Statistical atmospheric downscaling for rainfall estimation in Kyushu Island, Japan

摘要

The present paperdevelops linear regression models based on singular value decomposition (SVD)with the aim of downscaling atmospheric variablesstatistically to estimate average rainfall in the Chikugo River Basin, KyushuIsland, southern Japan, on a 12-hour basis. Models weredesigned to take only significantly correlated areas into account in thedownscaling procedure. By using particularly precipitable water incombination with wind speeds at 850 hPa, correlation coefficients betweenobserved and estimated precipitation exceeding 0.8 were reached.The correlations exhibited a seasonal variation with higher values during autumnand winter than during spring and summer. The SVD analysispreceding the model development highlighted three important features of therainfall regime in southern Japan: (1) the so-called Bai-ufront which is responsible for the majority of summer rainfall, (2) the strongcirculation pattern associated with autumn rainfall, and(3) the strong influence of orographic lifting creating a pronounced east-westgradient across Kyushu Island. Results confirm the feasibility ofestablishing meaningful statistical relationships between atmospheric state andbasin rainfall even at time scales of less than one day. style="line-height: 20px;">Keywords: atmospheric downscaling, precipitation, rainfall, singular value decomposition, southern Japan