EGU24-8116, updated on 08 Mar 2024
https://doi.org/10.5194/egusphere-egu24-8116
EGU General Assembly 2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

IMERG-E and IMERG-L: A Comprehensive Evaluation of the Medicane Daniel in Thessaly, Greece

Evangelos Leivadiotis1, Silvia Kohnová1, and Aris Psilovikos2
Evangelos Leivadiotis et al.
  • 1Department of Land and Water Resources Management, Faculty of Civil Engineering, Slovak University of Technology in Bratislava, (Radlinskeho 11), Bratislava 81005, Bratislava
  • 2Department of Agriculture, Ichthyology and Aquatic Environment, School of Agricultural Sciences, University of Thessaly, N. Ionia Magnisias, 38446, Greece

On 4 September 2023, the area of Thessaly (Greece) experienced a catastrophic flood as a result of the Daniel hurricane sequence. This severe phenomenon is characterized by extreme rainfall records ranging from 305 mm to 1096 mm between 4 and 7 September, causing severe damage to infrastructure, agriculture and buildings. Seventeen casaulties were recorded. The aim of the study is to complete the integrated multi-satellite harvesting of the Global Precipitation Measurement Mission (IMERG) using 10 precipitation stations distributed in the Thessaly region. Specifically, two precipitation products (IMERG-E and IMERG-L) were used to evaluate the early and late extreme precipitation events of IMERG version 7. In order to obtain the rainfall data needed for the research, a time period of 4 September 2023 (0000UTC) to 7 September 2023 (2330UTC) was chosen. This window corresponds to the approximate time at which Daniel's storm convective zone was on the area of interest. The National Meteorological Agency collected six of the ten precipitation stations and four of the Public Electricity Agency. The evaluation process was divided into two parts: the first part aimed at estimating the total rainfall of IMERG-E and IMERG-L, and the second part aimed at estimating the total daily rainfall of both products. Two statistical assessment indicators were used: the Pearson correlation coefficient and the root mean square error (RMSE) to quantitatively assess the performance of satellite precipitation products using rain-gauge data. Firstly, the correlation coefficient between IMERG-E, IMERG-L and total precipitation at IMERG-E, IMERG-L and IMERG-L is -0.03 and 0.27, respectively. Early products did not correlate with ground data, but later versions showed weak positive linear relationships. The RMSE values are 0.8 and 0.52, respectively. The daily analyses of IMERG-E showed moderate negative correlations on September 4 (-0.29), September 5 (-0.15), and September 7 (-0.25), and moderate positive correlations on September 6 (0.37). In terms of daily performance, the correlation coefficients suggest weak positive correlations (0.22 in 4 September, 0.13 in 5 September, 0.23 in 7 September), with the exception of -0.3 in 6 September. RMSE values remain low (0.31 on 4 September, 0.34 on 5 September, 0.20 on 7 September), except for 6, September, where values (0.95) indicate high levels of error. Overall, the late version is more efficient than the early version, but there are rooms for improvements when the IMERG final version will be available.

How to cite: Leivadiotis, E., Kohnová, S., and Psilovikos, A.: IMERG-E and IMERG-L: A Comprehensive Evaluation of the Medicane Daniel in Thessaly, Greece, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8116, https://doi.org/10.5194/egusphere-egu24-8116, 2024.