Detekcija intenzivnih neviht na območju Slovenije

Osolnik, Matevž (author), Skok, Gregor (mentor), Šavli, Matic (comentor)

V okviru magistrske naloge sem razvil metodo za detekcijo intenzivnih neviht na območju Slovenije. Metoda detektira intenzivne konvektivne celice na podlagi meritev meteorološkega radarja in razelektritev. Uporabljena je bila rutina pysteps, ki temelji na tem, da v polju radarske odbojnosti najprej išče zaključena območja nad določenim pragom radarske odbojnosti, nato pa preko dodatnih kriterijev iz zaključenih območij določi konvektivne celice. Optimizacija in verifikacija rezultatov je bila izvedena na konvektivnih sezonah od maja do septembra v letih od 2020 do 2022. Rezultati metode so bili optimizirani in verificirani na meritvah toče na padavinskih in klimatoloških postajah Agencije Republike Slovenije za okolje (ARSO) in na podatkih o intervencijah Uprave Republike Slovenije za zaščito in reševanje (URSZR), ki vsebujejo intervencije, ki so se zgodile zaradi toče, vetroloma in poplav meteorne vode. V primeru podatkov o intervencijah sta bili optimizacija in verifikacija izvedeni le na izbranih 13 območjih, kjer je prišlo do zadostnega števila javljenih dogodkov. Iz rezultatov so bili izračunani verifikacijski indeksi POD (angl. Probability of detection), FAR (angl. False alarm rate), CSI (angl. Critical success index) in BIAS, ki bazirajo na kontingenčni tabeli. Preko teh indeksov je bila nato ovrednotena uspešnost metode in so bili določeni optimalni parametri. Pri verifikaciji algoritma na meritvah toče na meteoroloških postajah so pri optimalnih parametrih verifikacijski indeksi dosegli vrednost POD = 0,39, FAR = 0,82, CSI = 0,14 in BIAS = 2,20, pri verifikaciji na podatkih o intervencijah toče, vetroloma in poplav meteorne vode pa so verifikacijski indeksi znašali POD = 0,29, FAR = 0,61, CSI = 0,20 in BIAS = 0,73. Posebej je bilo analiziranih tudi nekaj nevihtnih dni, za katere je ARSO izdal poročilo o neurjih. Za te dni je verifikacijski indeks CSI znašal med 0,10 in 0,33 za verifikacijo toče na meteoroloških postajah, pri verifikaciji na podatkih o intervencijah toče, vetroloma in poplav meteorne vode pa je indeks dosegal vrednosti med 0,25 in 0,80. Razvita metoda se pri detekciji intenzivnih neviht v splošnem obnese dobro, rezultati verifikacije pa so primerljivi z rezultati podobnih študij, narejenih v Sloveniji in tujni.

• Language: Slovenian
• Keywords: nevihta, toča, vetrolom, poplava, radar, razelektritev, pysteps, verifikacija, optimizacija
• Work type: Master's thesis/paper
• Typology: 2.09 - Master's Thesis
• Organization: FMF - Faculty of Mathematics and Physics
• Year: 2023
• PID: 20.500.12556/RUL-146988
• COBISS.SI-ID: 156661763
• Publication date in RUL: 18.06.2023

Secondary language

• Language: English
• Title: Detection of intense thunderstorms over Slovenia

Abstract

In this master’s thesis, I developed a method for detecting intense thunderstorms in the area of Slovenia. The method detects intense convective cells based on meteorological radar measurements and lightning discharges. The pysteps routine was used, which searches for closed areas above a certain threshold of radar reflectivity in the radar reflectivity field, and then determines convective cells from the closed areas using additional criteria. The optimization and verification of the results were carried out for the convective seasons from May to September in the years from 2020 to 2022. The results of this method were optimized and verified on the hail measurements of the precipitation and climatological stations of the Slovenian Environment Agency (ARSO) and on the intervention data of the Administration for Civil Protection and Disaster Relief (URSZR), where the events of hail, severe winds and floods of stormwater were jointly analyzed. In the case of data on interventions, optimization and verification were performed only in selected 13 areas where a sufficient number of reported events occurred. From the results, the verification indices POD (Probability of detection), FAR (False alarm rate), CSI (Critical success index) and BIAS were calculated, which are based on a contingency table. Using these indices, the performance of the method was then evaluated and the optimal parameters were determined. When verifying the method on hail measurements at meteorological stations with optimal parameters, the verification indices reached the value POD = 0.39, FAR = 0.82, CSI = 0.14 and BIAS = 2.20, and when verifying data on interventions related to hail, severe winds and stormwater floods, the verification indices amounted to POD = 0.29, FAR = 0.61, CSI = 0.20 and BIAS = 0.73. A few stormy days for which ARSO issued a storm report were also analyzed separately. For these days, the CSI verification index was between 0.10 and 0.33 for the verification of hail at meteorological stations, while the index reached values between 0.25 and 0.80 when verifying data on hail, severe winds and stormwater floods interventions. The developed method generally performs well in the detection of intense thunderstorms, and the verification results are comparable to the results of other studies performed in Slovenia or abroad.

• Keywords: storm, hail, severe wind, flood, radar, lightning, pysteps, verification, optimization