Naslov
Applying Statistical Tool CLIPER in Forecasting Visibility at Airports

Autori
Jurković, Jadran ; Zoldoš, Marko ; Kos, Igor ; Juras, Josip ; Pasarić, Zoran

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, prošireni sažetak, znanstveni

Skup
WMO Aeronautical Meteorology Scientific Conference 2017

Mjesto i datum
Toulouse, Francuska, 06.11.2017. - 10.11.2017

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
visibility, airports, forecast, CLIPER

Sažetak
One of the biggest challenges in meteorology is predicting fog. Aviation meteorology has a special interest in visibility forecasts at airports since low visibility and ceiling conditions can cause many troubles to the traffic at airports and along routes. Theoretical background of processes in fog is well understood and availability of the measured data at airports is very good. But forecasting of fog at one site, such as airport, is difficult. Moreover, for example in Terminal Aerodrome Forecast (TAF), it is very hard to meet required visibility criteria stated in ICAO Annex 3. Usually, several approaches are used to forecast visibility operationally. Usage of models, post processing and statistical tools (or combination of them) are the most common ones. In Croatia Control we implemented CLIPER - simple statistical model for probabilistic short-range forecasting (nowcasting) proposed by Juras and Pasaric in 2006. Actual visibility is related to climatological visibility distribution for the present hour and month. The corresponding percentile of the visibility, in its equivalent normal distribution, is in the following hours slightly moved from starting percentile to the median, depending on strength of the correlation coefficient between hours. Forecast of visibility is given for the following 9 hours. This method should be helpful tool to the operational forecasters in forecasting especially low visibility. It is rather easy to apply it to all airports with METAR databases. In Croatia Control, operational forecasters use so called ‘fog panel’ that consists of several graphs which show present and recent measured visibility and climatological distribution. In addition to the median forecast of visibility (which is basically the 50th percentile), 50 % and 80 % confidence intervals are shown as well to provide a measure of the forecast uncertainty. Verification of results for last cold season at Zagreb Airport shows very good result in anticyclonic situation and especially for fog dissipation. Together with the observations and NWP models, this statistical method complements the visibility nowcasting methods.

Izvorni jezik
Engleski

Znanstvena područja
Geofizika

POVEZANOST RADA