Comparison of precipitable water vapor data of ERA-Interim and NCEP-NCAR with radiosonde observations (Case study: Shiraz Station)

هواشناسی کشاورزی

Amin Shirvani¹  HassanNuroozi² 

[1] Associate professor, Water department and atmospheric ocean reserch center, College of agriculture, University of Shiraz, Shiraz, Iran ;M.Sc. Student in Agricultural Meteorology, Water department, College of agriculture, University of Shiraz, Shiraz, Iran ;

关键词: atmospheric moisture;    radiosonde;    reanalysis data;    trend;   

DOI  :  10.22125/agmj.2019.158865.1047

来源: DOAJ

【 摘 要 】

Water vapor plays an important role in the atmospheric water balance and precipitation formation process in the lower troposphere. In this study, trend of the total precipitable water vapor (TPWV) which is column-integrated water vapor, is analyzed using radiosonde data of synoptic Shiraz station during the period 1990-2017. Besides, the TPWV data from the ERA- Interim of the ECMWF and NCEP-NCAR reanalysis datasets were retrived and used for comparison. The Mann Kendall (M-K) test, showed no significant trend in TPW data for both Radiosonde and ECMWF data for all months, except for Radiosonde data of August. Generally a non-significant decreasing trend in radiosonde time series of TPWV was observed for most of months. However, both linear regression and M-K tests showed a significant increasing trend in annual TPW of NCEP-NCAR data. The rate of decrease and increase in PW for Radiosonde and NCEP-NCAR data are 0.29 and 0.52 mm per decade, respectively. For all months (except April), there is a positive significant Pearson correlation between Radiosonde TPW series and ERA- Interim at 5% significance level. However, the Pearson correlation coefficients were not significant for NCEP-NCAR. The root mean square error of PWV data and ERA- Interim is generally smaller than NCEP-NCAR. The latter, showed an overestimation comparing to radiosonde. In general, ERA- Interim data has higher accuracy and more agreement with observed Radiosonde data.

【 授权许可】

Unknown