Master Thesis: Emilie Hermansson – Niels Bohr Institute - University of Copenhagen

Niels Bohr Institute > Calendar > 2010 > Master Thesis: Emilie ...

Master Thesis: Emilie Hermansson

A numerical study of a Piteraq-event in Greenland

the talk will be in Swediskh

The ice sheet in Greenland and its topography largely influence the wind system and weather situation over Greenland. When air is cooled on the ice sheet it will descent down the slope towards the coastline due to the gravitational force and thus form a katabatic wind field. A Piteraq is a phenomenon which occurs when a low pressure system enhances the katabatic flow in Greenland. The most extreme Piteraq ever to be documented occurred at the east coast in Tasiilaq the 6th of February 1970, when wind gusts up to 72 m/s were measured.

These winds nearly destroyed the entire town and it is crucial to be able to correctly forecast events like this. The aim of this master thesis is thus to compare the operational DMI-HIRLAM K05 hydrostatic weather prediction model with two non-hydrostatic HARMONIE models - one with the same horizontal resolution as the K05 model (5 km) and the other with a horizontal resolution of 2.5 km - in order to see which model best predicts this topographical event. The three models captured the location of the Piteraq well, but they all underestimated the wind speeds during the time when the storm had reached its most intense stage. The non-hydrostatic models proved to forecast the wind speeds better than the hydrostatic model. The wind speed predictions produced by the HARMONIE model with the coarsest resolution deviated 11.3 m/s from the maximum observed wind speed of 54 m/s at 18 UTC while the other HARMONIE model deviated 24.2 m/s and the K05 model 30.3 m/s. The hydrostatic approximation was probably not valid during the evening when the extreme wind speeds were observed which might have resulted in the inability of the K05 model to correctly forecast these wind speeds. Furthermore; the K05 model did not parametrise the gravity wave drag which might have been another reason for the underestimation of wind speed, neither did the HARMONIE model with horizontal resolution of 2.5 km which could have been the reason why it, despite its higher resolution, predicted a larger deviation from the observations than the other non-hydrostatic model did.

Aksel Walløe Hansen NBI and Niels Woetmaann, DMI