Back to Search Start Over

How inaccurate offshore wind observations impact the quality of operational numerical weather prediction.

Authors :
Køltzow, Morten
Øiestad, Merete
Azad, Roohollah
Rugaard Furevik, Birgitte
Olsen, Anne‐Mette
Remes, Teresa
Source :
Quarterly Journal of the Royal Meteorological Society. Oct2024, Vol. 150 Issue 765, p5337-5355. 19p.
Publication Year :
2024

Abstract

At Norwegian offshore platforms, wind speed is observed at heights between approx. 35 and approx. 150 m a.s.l. These observations are used to estimate 10‐m wind speed, by a power law with a constant wind shear coefficient. The resulting 10‐m wind speed is then distributed to the Global Telecommunication System (GTS) and used for assimilation and verification purposes by the meteorological community. However, multiple methods with varying results exist for the reduction of wind speed from the original measurement height to 10 m. The resulting estimate of 10‐m wind speed is therefore sensitive to the choice of method. The use of such observations may therefore be suboptimal in forecast analysis and make the forecast verification difficult to interpret. This study investigates how the use of these observations impacts forecasts from the operational regional forecast system used at MET Norway and the Global Integrated Forecast System operated by ECMWF for the period 2017–2023. Both forecast systems assimilated the observations over major parts of the period and the presented results indicate that the use of these observations increased the initial errors in the forecasts. This is confirmed in a data denial experiment over a shorter period of time. The verification results show that both forecast systems underestimate the wind speed at the original measurement heights, while the results for 10‐m verification vary with the interpolation method between original sensor height and 10 m. The forecasts show in general more wind than the GTS‐distributed estimates and are in better agreement with more advanced interpolation methods. The verification results show a strong dependency on the vertical stability of the atmosphere. The major differences between the forecast systems are that the regional system shows less bias, while the standard deviations of the errors are slightly lower in the global system. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
00359009
Volume :
150
Issue :
765
Database :
Academic Search Index
Journal :
Quarterly Journal of the Royal Meteorological Society
Publication Type :
Academic Journal
Accession number :
181275495
Full Text :
https://doi.org/10.1002/qj.4872