Fig. 1
Various parameterised and measured drag coefficients (data were reproduced except Östergarnsholm data and Southern Ocean data). Red+signs are data measured from the ocean side of the air–sea interface (Jarosz et al., 2007); black squares are GPS dropsonde data (Powell et al., 2003); blue×signs are eddy-correlation data from Östergarnsholm for wind speeds over 15ms−1 (Högström et al., 2008); black+signs are data from the Southern Ocean (Sahlée et al., 2012); cyan×signs are high wind speed CBLAST data (Bell et al., 2012); black×signs are typhoon data from the Pacific Ocean (Potter et al., 2014).
Fig. 2
Comparison of wave state impact on the drag coefficient in the newly proposed parameterisation: (a) parameterisation with eqs. (5) and (9); (b) parameterisation with eqs. (5), (9) and (10).
Fig. 3
The domain of the RCA model used in this study (red box area); the red×is the FINO1 site; the blue box area is the area shown in Figs. 17 and 18; the blue×indicates the centre of storm Uill at time 2012-01-03:12 (discussed in Section 6.3).
Table 1. Wind stress and heat fluxes parameterisations for the various simulations
Exp-1Eq. (19)RCABasic experimentsExp-2Eqs. (5) and (6)RCABasic sea spray parameterisationExp-3Eqs. (5) and (9)RCAWave-state-dependent SSGF impact on wind stressExp-4Eqs. (5), (9) and (10)RCAWave-state-dependent SSGF and impact on wind stressExp-5Eq. (19)Andreas et al., 2014Sea spray impact on heat fluxesExp-6Eqs. (5), (9) and (10)Andreas et al., 2014Full coupled case (Exp-4+Exp-5)
Fig. 4
Modelled wind speed results compared with FINO1 measurements at a height of 33 m: (a) January 2005, (b) January 2007 and (c) January 2012.
Fig. 5
Statistical results for wind speed measured at a height of 33m (a) mean error, (b) mean absolute error and (c) root mean square difference.
Table 2. Comparative statistics for the wind speed results at a height of 33 m; bias or mean error (ME), mean absolute error (MAE), root mean square error (RMSE) and Pearson correlation coefficient (R)
ME0.130.090.470.560.200.62MAE1.831.881.861.951.852.03RMSE2.472.552.592.742.532.79R0.880.870.870.860.870.86
Table 3. Comparative statistics for the temperature results at a height of 100m; bias or mean error (ME), mean absolute error (MAE), root mean square error (RMSE) and Pearson correlation coefficient (R)
ME−0.59−0.58−0.65−0.68−0.44−0.41MAE0.990.991.041.060.870.87RMSE1.401.401.441.451.311.31R0.870.870.860.870.870.87
Table 4. Comparative statistics for the different wind speed range results at a height of 33 m; bias or mean error (ME), mean absolute error (MAE) and root mean square error (RMSE)
ME
Exp-11.460.790.52−0.08−0.65−1.67 Exp-41.821.160.780.310.04−0.75 Exp-51.721.020.55−0.06−0.68−1.57 Exp-61.871.140.790.420.16−0.55MAE
Exp-11.732.221.941.661.272.05 Exp-42.012.412.181.821.171.76 Exp-51.882.271.831.741.352.06 Exp-62.032.612.091.871.501.75RMSE
Exp-12.152.822.672.351.612.52 Exp-42.493.233.042.681.562.27 Exp-52.452.942.532.511.732.58 Exp-62.613.322.982.651.992.31
Fig. 6
Same as in Fig. 4, but for temperature measured at a height of 100m compared with the second model layer (about 100 m).
Fig. 7
Statistical results for temperature measured at a height of 100 m: (a) mean error, (b) mean absolute error and (c) root mean square difference.
Fig. 8
The storm tracks represented by the minimum sea level pressure every three hours: (a) Gero, (b) Erwin/Gudrun, (c) Kyrill, (d) Ulli, (e) Patrick and (f) Klaus.
Table 5. The mean absolute error (MAE) of the minimum sea level pressure centre (km)
Gero187.92189.86182.32150.99170.48184.31Erwin172.49174.53174.34172.85174.91180.54Kyrill106.56110.58119.84131.80101.88132.21Ulli250.64265.42254.23259.64230.51199.48Patrick193.08245.76197.37197.88194.17186.84Klaus266.72270.49251.58250.69268.47269.11Average196.23209.44196.61193.98190.07192.08
Fig. 9
The minimum sea level pressure of different storms over time: (a) Gero, (b) Erwin/Gudrun, (c) Kyrill, (d) Ulli, (e) Patrick and (f) Klaus.
Table 6. The mean absolute error (MAE) of the minimum sea level pressure (hPa)
Gero6.626.415.785.696.105.06Erwin2.452.432.352.362.352.16Kyrill3.673.653.333.283.723.21Ulli3.783.652.992.683.512.28Patrick7.617.706.837.276.545.33Klaus2.582.602.582.532.532.41Average4.454.413.983.974.123.41
Fig. 10
The maximum wind speed at 925 hPa of different storms over time: (a) Gero, (b) Erwin/Gudrun, (c) Kyrill, (d) Ulli, (e) Patrick and (f) Klaus.
Table 7. The mean absolute error (MAE) of the maximum wind speed at 925 hPa (ms−1)
Gero4.584.123.413.344.383.34Erwin2.962.932.542.523.122.53Kyrill2.162.141.791.822.111.66Ulli1.711.691.811.971.531.98Patrick3.863.712.812.933.432.40Klaus1.761.701.591.571.541.24Average2.842.722.322.362.682.19
Fig. 11
The simulation results from the control experiment (Exp-1) at time 2012-01-03:12: (a) wind speed (m s−1) at 10m and sea level pressure (h Pa), (b) significant wave height (m), (c) air temperature at 2m (°C), (d) humidity at 2m (g kg−1), (e) latent heat flux (W m−2) and (f) sensible heat flux (W m−2). The black×is the centre of the storm in this time step.
Fig. 12
Difference in sea level pressure (hPa) from that of Exp-1 at time 2012-01-03:12: (a) Exp-3 – Exp-1, (b) Exp-4 – Exp-1, (c) Exp-5 – Exp-1 and (d) Exp-6 – Exp-1. The red×is the centre of the storm in this time step.
Fig. 13
Difference in wind speed at 10m (m s−1) from that of Exp-1 at time 2012-01-03:12: (a) Exp-3 – Exp-1, (b) Exp-4 – Exp-1, (c) Exp-5 – Exp-1 and (d) Exp-6 – Exp-1. The red×is the centre of the storm in this time step.
Fig. 14
Difference in heat flux (i.e. sensible heat flux and latent heat flux) (W m−2) from that of Exp-1 at time 2012-01-03:12: (a) Exp-3 – Exp-1, (b) Exp-4 – Exp-1, (c) Exp-5 – Exp-1 and (d) Exp-6 – Exp-1. The red×is the centre of the storm in this time step.
Fig. 15
Difference in wave age from that of Exp-1 at time 2012-01-03:12: (a) Exp-3 – Exp-1, (b) Exp-4 – Exp-1, (c) Exp-5 – Exp-1 and (d) Exp-6 – Exp-1. The red×is the centre of the storm in this time step.
Fig. 16
Difference in wind wave age from that of Exp-1 at time 2012-01-03:12: (a) Exp-3 – Exp-1, (b) Exp-4 – Exp-1, (c) Exp-5 – Exp-1 and (d) Exp-6 – Exp-1. The red×is the centre of the storm in this time step.
Fig. 17
Difference in humidity from that of Exp-1 at different heights at time 2012-01-03:12: (a) Exp-1, (b) Exp-4 – Exp-1, (c) Exp-5 – Exp-1 and (d) Exp-6 – Exp-1.
Fig. 18
Difference in temperature from that of Exp-1 at different heights at time 2012-01-03:12: (a) Exp-1, (b) Exp-4 – Exp-1, (c) Exp-5 – Exp-1 and (d) Exp-6 – Exp-1.