A quasi-linear relationship between GPS-derived precipitable water vapor and surface vapor pressure observed on dry days in spring and autumn: a case study at Saga Plain in northern Kyushu Island, Japan

Abstract

Surface vapor pressure (SVP) is a significant meteorological variable that expresses the status of atmospheric moisture near the ground surface. Its lower values characterize dry atmospheric conditions, and are strongly correlated with the occurrence of days with lower daily minimum temperatures from autumn to the following spring, which frequently induces frost damage in field crops. It is therefore important to obtain information regarding SVP to assess such frosty events. Recent advances in GPS technology have enabled us to infer the atmospheric column-integrated water vapor or precipitable water vapor (PWV) as GPS-derived PWV (GPS-PWV). However, few studies have applied GPS-PWV to such low-temperature-induced events to estimate SVP. In this study, we analyzed GPS and meteorological data that were routinely collected for about 7 years at a plain in the northern part of Kyushu Island, Japan. A quasi-linear relationship between daily-averaged values of GPS-PWV and SVP was obtained specifically under dry conditions in spring and autumn. We related GPS-PWV to SVP by linear regression and theoretically examined their relationship. A simple formula was developed to estimate SVP from GPS-PWV on a daily basis under such conditions. The estimated vapor pressures agreed acceptably with observed values.