Comparison of Local Mean Temperature Equations for GPS-based Precipitable Water Vapor Determination

Abstract

The mean temperature equation is a key factor in calculating GPS meteorological information. A local mean temperature equation should be used to improve accuracy of GPS PWV (Precipitable Water Vapor). In this paper, four local mean temperature equations, HP, HP_M, HPt_Y, and HPt_M from Ha & Park (2008) were used to analyze the effects of local models in determining GPS PWV. Four different sets of GPS PWVs were compared with radiosonde PWV to validate the accuracies of local models. GPS PWVs of four local models have similar trends compared against radiosonde PWV. The bias and RMS error were the same level: the bias is ˜ 3 mm and the RMS is ˜ 3.6 mm after the bias was removed. Especially, with HPt_Y and HPt_M models one can obtain accurate PWVs even without surface temperature measurements. And we investigated dry bias of radiosonde measurements depending on sensor types and observation time at Sokcho weather station. After the radiosonde sensor equipment was changed from RS80-15L to GRS DFM-06, dry bias of radiosonde PWV decreased about 18.2 % during daytime (KST 09:00), and 16.1 during nighttime (KST 21:00).

Keywords

• GPS
• local mean temperature equation
• precipitable water vapor
• radiosonde
• dry bias