Abstract
This report describes the development and evaluation of different methods for predicting the wind locally. The look-ahead time of the forecasts ranges from 3 to 36 hours. The main model developed here is based on forecasts from a numerical weather prediction model. In this study HIRLAM (HIgh Resolution Limited Area Model) has been used. These forecasts are transformed to the surface using the geostrophic drag law and the logarithmic wind profile in their neutral versions. To take local effects into account, corrections output form WA{sup s}P (Wind Atlas Analysis and Application Program) are used. The conclusion is that the model developed here performs significantly better than persistence. The model performs best when applied to sites in Northern Europe (having high wind speeds). When using MOS (Model Output Statistics) it is possible to improve the forecasts, mainly those that do not perform well. (au) (21 tabs., 67 ills., 81 refs.)
Citation Formats
Landberg, L.
Short-term prediction of local wind conditions.
Denmark: N. p.,
1994.
Web.
Landberg, L.
Short-term prediction of local wind conditions.
Denmark.
Landberg, L.
1994.
"Short-term prediction of local wind conditions."
Denmark.
@misc{etde_10149105,
title = {Short-term prediction of local wind conditions}
author = {Landberg, L}
abstractNote = {This report describes the development and evaluation of different methods for predicting the wind locally. The look-ahead time of the forecasts ranges from 3 to 36 hours. The main model developed here is based on forecasts from a numerical weather prediction model. In this study HIRLAM (HIgh Resolution Limited Area Model) has been used. These forecasts are transformed to the surface using the geostrophic drag law and the logarithmic wind profile in their neutral versions. To take local effects into account, corrections output form WA{sup s}P (Wind Atlas Analysis and Application Program) are used. The conclusion is that the model developed here performs significantly better than persistence. The model performs best when applied to sites in Northern Europe (having high wind speeds). When using MOS (Model Output Statistics) it is possible to improve the forecasts, mainly those that do not perform well. (au) (21 tabs., 67 ills., 81 refs.)}
place = {Denmark}
year = {1994}
month = {Mar}
}
title = {Short-term prediction of local wind conditions}
author = {Landberg, L}
abstractNote = {This report describes the development and evaluation of different methods for predicting the wind locally. The look-ahead time of the forecasts ranges from 3 to 36 hours. The main model developed here is based on forecasts from a numerical weather prediction model. In this study HIRLAM (HIgh Resolution Limited Area Model) has been used. These forecasts are transformed to the surface using the geostrophic drag law and the logarithmic wind profile in their neutral versions. To take local effects into account, corrections output form WA{sup s}P (Wind Atlas Analysis and Application Program) are used. The conclusion is that the model developed here performs significantly better than persistence. The model performs best when applied to sites in Northern Europe (having high wind speeds). When using MOS (Model Output Statistics) it is possible to improve the forecasts, mainly those that do not perform well. (au) (21 tabs., 67 ills., 81 refs.)}
place = {Denmark}
year = {1994}
month = {Mar}
}