Skillful bias correction of offshore near-surface wind speed and wind direction forecasting based on a multi-task machine learning model

Dataset

1. observation data over 14 weather stations

Variables: hourly near-surface 2-min average wind speed, wind direction 

2. ECMWF-IFS forecast data over 14 weather stations

Variables: hourly predictors at surface level and upper level in next 48 hours (shown in Table 1. and Table 2.)

Table 1. ECMWF-IFS forecast data at surface level

Predictors

Abbreviation

Unit

Temperature at 2 m

2t

℃

Sea surface temperature

sst

℃

Dewpoint temperature at 2 m

2d

℃

Convective precipitation in the past hour

cp

mm

Mean sea level pressure

msl

hPa

Zonal component of wind speed at 10 m

10u

m s-1

Meridional component of wind speed at 10 m

10v

m s-1

Wind speed at 10 m

10ws

m s-1

Wind direction at 10 m

10wd

°

Zonal component of wind speed at 100 m

100u

m s-1

Meridional component of wind speed at 100 m

100v

m s-1

Wind speed at 100 m

100ws

m s-1

Wind direction at 100 m

100wd

°

Table 2. ECMWF-IFS forecast data at upper level

Predictors

Abbreviation

Unit

Relative humidity at xxx hPa

r_Lxxx

%

Temperature at xxx hPa

t_Lxxx

℃

Vertical velocity of wind at xxx hPa

w_Lxxx

Pa s-1

Zonal component of wind at xxx hPa

u_Lxxx

m s-1

Meridional component of wind at xxx hPa

v_Lxxx

m s-1

Wind speed at xxx hPa

ws_Lxxx

m s-1

Wind direction at xxx hPa

wd_Lxxx

°

3. key variables constructed by feature engineering

(1) sort-term statistics, including maximum, minimum, mean and variance of key variables (2t, 10u, 10v and 10ws) from ECMWF-IFS model during the next 48 hours,

 (2) long-term statistics, including mean and deviation of key variables (2t, 10u, 10v and 10ws) from ECMWF-IFS model during history 3-yr period (January 2020–December 2022),

 (3) thermodynamic factors,  including the low-level wind shear between 10ws and 100ws, vertical wind shear between 200 hPa and 850 hPa, the differences between sst and 2t.

Scripts

1. Random Forest model training code

2. LightGBM model training code

3. XGBoost model training code

4. TabNet-MTL model training code