Atgriešanās līmeņa aprēķināšana ekstremāliem nokrišņiem Latvijā
Loading...
Date
Authors
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
Latvijas Universitāte
Language
N/A
Abstract
Darbā tika aplūkota ekstremālo nokrišņu atgriešanās līmeņa aprēķināšana ar statistiskās modelēšanas
palīdzību. Tas tika veikts ar divām dažādam metodēm, kas ietver sadalījuma piemeklēšanu
datiem. Pirmā metode piemeklē sadalījumu gada maksimālām novērotām vērtībām (bloku
maksimumu metode), otrā — datiem, kas pārsniedz noteikto slieksni (maksimumu virs sliekšņa
metode). Lai pārbaudītu vai atrastais sadalījums atbilst datiem, tika veikti atbilstošie statistiskie
testi. Tika salīdzināti abu metožu rezultāti novērotiem nokrišņu datiem Latvijā 1961.-2010.
periodā. Iegūtie rezultāti ir svarīgi, jo tiem ir liela nozīme lēmumu pieņemšanas procesos, kas
ir saistīti ar vadības, plānošanas, projektēšanas un būvniecības jomām, un palīdz novērtēt būvkonstrukciju
izturību ekstremālos apstākļos.
Statistical modeling was used to estimate return level of extreme rainfall. It was done by two different probabilistic approaches involving fitting of probability distributions. One method fits distribution to the recorded annual maximum rainfall data (block maxima method), other - data that exceeds certain threshold (peak over threshold method). Corresponding statistical tests were used for checking the adequacy of fitting of probability distributions to the recorded data. Results were compared for both methods for precipitation series recorded in Latvia in 1961- 2010. Obtained results are essential since play a key role in decision making processes in the wide fields of management, planning, design and construction in order to estimate the ability of building structures to survive under extreme conditions.
Statistical modeling was used to estimate return level of extreme rainfall. It was done by two different probabilistic approaches involving fitting of probability distributions. One method fits distribution to the recorded annual maximum rainfall data (block maxima method), other - data that exceeds certain threshold (peak over threshold method). Corresponding statistical tests were used for checking the adequacy of fitting of probability distributions to the recorded data. Results were compared for both methods for precipitation series recorded in Latvia in 1961- 2010. Obtained results are essential since play a key role in decision making processes in the wide fields of management, planning, design and construction in order to estimate the ability of building structures to survive under extreme conditions.