%0 Journal Article %T Evaluating the Performance of Era-5 Re-Analysis Data in Estimating Daily and Monthly Precipitation, Case Study; Ardabil Province %J Iranian Journal of Soil and Water Research %I University of Tehran %Z 2008-479X %A azizi mobaser, javanshir %A Rasoulzadeh, Ali %A rahmati, akbar %A shayeghi, afshin %A Bakhtar, Aydin %D 2021 %\ 01/20/2021 %V 51 %N 11 %P 2937-2951 %! Evaluating the Performance of Era-5 Re-Analysis Data in Estimating Daily and Monthly Precipitation, Case Study; Ardabil Province %K Bias correction %K Satellite %K Precipitation %K contingency table %R 10.22059/ijswr.2020.302176.668600 %X Inappropriate distribution of precipitation measurement stations has led to the use of gridded precipitation datasets, consisting of satellites, reanalysis and ground-based precipitation datasets in recent years. In this study, one of the important precipitation products named Era5 has been evaluated in Ardabil province. The observation data were first interpolated during the 2004–2014 statistical period and compared with Era5 data based on daily, monthly and annual time scales. Evaluations were performed using RMSE, correlation coefficient and contingency table indices consisting of POD, FAR, CSI and POFD. The results showed that the correlation coefficient for Era5 at the daily time scale was above 0.75 for most of the cells and RMSE was below 3 mm. Also, the correlation coefficient for monthly time scale was above 0.8 and the RMSE was below 20 mm in most of the cells. Evaluation using contingency table indices showed that POD index in the studied cells ranged from 0.7 to 0.85, FAR ranged from 0 to 0.25, POFD ranged from 0.1 to 0.2 and the CSI was in the range of 0.4 to 0.5. Precipitation values of both precipitation sources were classified into 6 classes using Ward cluster analysis method. The results of k-means method and Wilkes-Lambda model confirmed the classification accuracy and the difference between the means of the clusters. In general, it can be concluded that the Era5 precipitation product at both daily and monthly time scales can be used as an appropriate alternative to data scarce regions after bias correction. %U https://ijswr.ut.ac.ir/article_77716_8f7af8edfb094df09a3c6e22535b5f06.pdf