A Gridded Weather Generator SPAGETTA: Towards the finer resolution

0479085 - ÚFA 2018 DE eng A - Abstrakt
Dubrovský, Martin - Dabhi, H. - Huth, Radan - Rotach, M. W.
A Gridded Weather Generator SPAGETTA: Towards the finer resolution.
EMS Annual Meeting Abstracts, Vol. 14. Berlín: European Meteorological Society, 2017. EMS2017-760-3.
[EMS Annual Meeting and European Conference for Applied Meteorology and Climatology. 03.09.2017-07.09.2017, Dublin]
Institucionální podpora: RVO:68378289
Klíčová slova: stochastic weather generator * SPatial GEneraTor for Trend Analysis ( SPAGETTA) * gridded weather generator * climate change scenarios
Kód oboru RIV: DG - Vědy o atmosféře, meteorologie
http://meetingorganizer.copernicus.org/EMS2017/EMS2017-760-3.pdf

SPAGETTA is a gridded (multisite) multivariate parametric stochastic weather generator: precipitation occurrence
and amount are modelled by Markov chain and Gamma distribution, and the non-precipitation variables are modelled
by a first-order autoregressive model conditioned on precipitation occurrence. The spatial coherence of all
variables is modelled following Wilks’ (2009) approach. Development of SPAGETTA started in 2016 and was motivated
by the need to have a generator that will be able to produce realistic high-resolution gridded weather data
(representing both present and future climates) for use in hydrological modelling in complex Alpine terrain (Ötztal
Valley area, Austria). In the first stage of developing the generator we used gridded E-OBS daily data (Haylock
et al, 2008) to calibrate the generator, and single RCM simulation (taken from the CORDEX database, EUR44
domain, RCP8.5 emissions) to develop the climate change scenarios for perturbing the WG parameters. The generator
was validated in terms of selected validation characteristics (focusing on the generator’s ability to reproduce
spatial temperature patterns; large-area hot days and hot spells were included) and the effect of the climate change
was assessed (an emphasis was put on the effect of changes in the spatial and temporal structure in weather series).
Now the experiment is
extended by including more validation characteristics (focus on the spatial indices related to extreme temperature,
precipitation and drought events) and more RCMs for developing the climate change scenarios (to account for the
modelling uncertainty in the future climate projection).
Trvalý link: http://hdl.handle.net/11104/0275103