Wind atlas of the Northern European Seas based on Envisat ASAR, QuikSCAT and ASCAT
Hasager, Charlotte1; Mouche, Alexis2; Badger, Merete1; Karagali, Ioanna1; Driesenaar, Tilly3; Stoffelen, Ad3; Bingöl, Ferhat1; Peña, Alfredo1

In the EU project NORSEWiND (Northern Seas Wind Index database, which lasted from 2008 to 2012 there was a goal of contributing a satellite-based wind atlas for the Northern European Seas. The effort included collection of more than 9000 Envisat ASAR WSM wide swath mode scenes covering the Baltic Sea, Irish Sea and North Sea during the years 2002 to 2012. The near-real-time processing to wind maps was done at CLS and DTU Wind Energy using various CMOD’s and with a priori input of wind directions from the ECMWF and NOGAPS models. At CLS the in-house processing was used. At DTU Wind Energy the Johns Hopkins University Applied Physics Laboratory processing was used. Also QuikSCAT ocean wind vector maps from Remote Sensing Systems and ASCAT from KNMI OSI-SAF were used to assess the wind climate. In the NORSEWiND project an intense measurement campaign with ten ground-based wind profiling lidars on offshore platforms was conducted. Also collection and analysis of meteorological data from around 15 offshore meteorological masts was performed. The wind lidars observed wind vectors at various heights above sea level from around 80 m to more than 100 m, i.e. around the hub-height of modern wind turbines. Data were stored each 10 minutes but temperature profile observations were not available at these sites. At the meteorological masts wind speed and wind direction were observed at various heights and at some masts also temperatures. At none of the meteorological masts winds were observed at 10 m but only at much higher levels (around 30 m and up). Therefore vertical extrapolation of winds was necessary in order to compare the mast observations with the 10-m satellite-based winds. Due to variations in atmospheric stability the vertical extrapolation of winds includes a correction (when possible). Analysis of the wind data are presented in Peña et al. 2012. The meteorological data from 10 masts in the Baltic Sea (Hasager et al. 2011) and 5 masts in the North Sea were used for comparison analysis with Envisat ASAR. The study also include inter-comparison of results between various CMOD’s. The variations between CMOD5 and CMOD-IFR were only minor. It was decided to reprocess all maps with CMOD-IFR and use ECMWF wind directions as input to obtain a homogenous data set. This massive processing was done by CLS. Finally, the resulting wind fields were used in the S-WAsP software at DTU Wind Energy to calculate maps of the wind energy statistical parameters: mean wind speed, Weibull scale and shape parameters, energy density and uncertainty estimates, please refer to Hasager et al. 2012 for details. The number of overlapping scenes ranges from a few hundred in the Irish Sea to more than 1400 in parts of the North Sea and Baltic Sea. The more overlapping scenes, the less uncertainty is there in the resulting maps. The spatial resolution of the final maps is 2 km by 2 km. The maps are freely available through the web-sites, and The analysis of QuikSCAT ocean winds included wind speed and direction comparisons against observations from selected offshore meteorological masts. The resulting wind maps of variations in winds across the Northern European seas in time and space are presented (Karagali et al. 2012 and Karagali et al. 2013). The analysis of ASCAT include comparison the mesoscale model WRF results (see Hahmann et al. 2012) on the modeling, and the comparison showed small variations (Hasager et al. 2012). The satellite-based wind atlas is the first version for an extended offshore area. However, issues remain with the appropriate sampling of climatology, SAR wind direction error projection onto wind speed and contamination by rain and structures such as ships and offshore wind turbines at sea, all leading to variable local biases in the diverse satellite systems. The presentation will focus on the Envisat ASAR wind processing methodology, the in-situ comparison results and the final wind resource maps from Envisat ASAR, ASCAT and QuikSCAT. References: Hahmann AN, Lange J, Pena Diaz A, Hasager CB 2012. The NORSEWInD numerical wind atlas for the South Baltic. DTU Wind Energy. 53 p. DTU Wind Energy E; No. 0011(EN). Hasager et al. 2012 Norsewind satellite wind climatology, DTU Wind Energy-E-0007(EN), Roskilde, Denmark Hasager, C.B., Badger, M., Peña, A, Larsén, X.G. 2010 SAR-based wind resource statistics in the Baltic Sea, Remote Sens. 2011, 3(1), 117-144 ; doi:10.3390/rs3010117 Karagali I., Badger, M., Hahmann, A., Peña, A., Hasager. C., Sempreviva, A.M. 2013 Spatial and temporal variability in winds in the Northern European Seas, Renewable Energy, in press Karagali I., Peña, A., Badger, M., Hasager. C. 2012 Wind characteristics in the North and Baltic Seas from the QuikSCAT satellite, Wind Energy, DOI: 10.1002/we.1565 Peña Diaz, A., Mikkelsen, T., Gryning, S-E., Hasager, C. B., Hahmann, A. N., Badger, M., Karagali, I., & Courtney, M. (2012). Offshore vertical wind shear: Final report on NORSEWInD’s work task 3.1. DTU Wind Energy. DTU Wind Energy E; No. 0005.