A Combined Physical–Statistical Approach for the Downscaling of Model Wind Speed

Rooy, Wim C. de; Kok, Kees

doi:10.1175/1520-0434(2004)019<0485:acpaft>2.0.co;2

Cited by 46 publications

(25 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are systematic errors between the model forecasts and the observed wind at an individual site. A solution to correct these errors is to downscale coarse resolution NWP outputs statistically to generate finer scale forecasts of local variables (Wilby and Wigley, 1997;Wilby et al, 2004;de Rooy and Kok, 2004;Gutiérrez et al, 2004;Pryor et al, 2005;Salameh et al, 2008). One commonly used statistical downscaling approach is the regression model, which directly quantifies the relationship between the NWP model outputs (predictors) and locally observed meteorological variables (predictands) using linear or non-linear methods, such as multiple linear regression (Beyer et al, 1999), artificial neural networks (Barbounis and Theocharis, 2006), or support vector machines (Mohandes et al, 2004).…”

Section: Introductionmentioning

confidence: 99%

Surface Wind Speed Prediction in the Canadian Arctic using Non-Linear Machine Learning Methods

et al. 2011

View full text Add to dashboard Cite

Two non-linear, machine-learning/statistical methods, i.e., Bayesian neural network (BNN) and support vector regression (SVR), plus multiple linear regression (MLR), were used to forecast surface wind speeds at lead times of 12, 24, 48 and 72 h. Three different schemes, a statistical downscaling model (Scheme 1) using daily reforecast data from the National Centers for Environmental Prediction (NCEP) Global Forecasting System (GFS), an autoregressive model (Scheme 2) based on past wind observations, and a full model (Scheme 3) combining the two, were investigated in this study for the October-March winds from two meteorological stations in the Canadian Arctic (Clyde River and Paulatuk). At very short lead times, Scheme 2 provides better wind speed prediction than Scheme 1, but its forecast scores decrease rapidly with lead time. Scheme 3 generally performs best, especially at shorter lead times. All the linear and non-linear downscaling methods have significantly higher forecast scores at the two stations than the GFS reforecast. The non-linear methods tended to have slightly better forecast scores than linear methods (MLR and the linear version of SVR).There is particular interest in high-wind events, defined as having wind speeds over 22 knots (11.3 m s −1). After rescaling, the continuous wind predictions from Scheme 3 were classified into two types -high-wind event or nonevent. For high-wind event forecasting, the non-linear methods have marginally better binary forecast scores than the linear methods for Clyde River but not for Paulatuk. The alternative approach of using support vector classification (SVC) did not perform better, but weighting the high-wind events more heavily than the non-events during model training improved the binary forecast scores.RÉSUMÉ [Traduit par la rédaction] Nous avons utilisé deux méthodes d'apprentissage automatique/statistiques non linéaires, c'est-à-dire un réseau neuronal bayesien (RNB) et la régression des vecteurs de support (RVS), en plus de la régression linéaire multiple (RLM) pour prévoir les vitesses du vent de surface à 12, 24, 48 et 72 heures dans le futur. Nous avons examiné, dans cette étude, trois schémas différents, soit un modèle de réduction statistique (schéma 1) utilisant les données quotidiennes reprévues du Global Forecasting System (GFS) des NCEP (National Centers for Environmental Prediction), un modèle autorégressif (schéma 2) basé sur les observations passées du vent et un modèle complet (schéma 3) combinant les deux, pour les vents d'octobre à mars à deux stations météorologiques dans l'Arctique canadien (Clyde River et Paulatuk). Pour des échéances très courtes, le schéma 2 produit de meilleures prévisions de vitesse du vent que le schéma 1 mais ses scores de prévision diminuent rapidement à mesure que l'échéance recule. Le schéma 3 donne habituellement les meilleurs résultats, surtout pour les échéances plus courtes. Toutes les méthodes de réduction linéaires et non linéaires ont des scores de prévision nettement plus élevés aux deux stati...

show abstract

Section: Introductionmentioning

confidence: 99%

Surface Wind Speed Prediction in the Canadian Arctic using Non-Linear Machine Learning Methods

et al. 2011

View full text Add to dashboard Cite

show abstract

“…The wind fields (speed and direction) were produced applying a multi-step procedure proposed by De Rooy and Kok (2004). The regional climate model COSMO-CLM provided a first guess of the field of hourly wind speed and wind direction of 2.8-km resolution.…”

Section: Wind Componentsmentioning

confidence: 99%

“…At a certain height (e.g. the blending height), the average wind profiles above a certain grid node and a station located within that grid node converge to a common wind speed (De Rooy and Kok 2004).…”

Section: Wind Speedmentioning

confidence: 99%

“…(3.13) and (3.14), the local wind speed can be calculated: Estimator A does not depend on the stability function Ψ M and produces overestimated/underestimated u 10,loc for stable/ unstable conditions (De Rooy and Kok 2004). To account for atmospheric stability, the following assumption was made: Transforming model wind speed at 140 m, considering the model roughness length but not the stability function, the estimated u 10 will deviate from the modelled wind speed at 10 m. This deviation corresponds to the stability correction to be applied when transforming wind speed from 140 to 10 m and is simply added to estimator A: Estimation of the roughness length Before the wind transformation could be applied, roughness lengths valid for the model grid, the 1-km 2 grid and the station locations had to be derived.…”

Section: Wind Speedmentioning

confidence: 99%

“…Statistical downscaling of model wind, a relatively recent development, allows wind to be derived at the local scale. De Rooy and Kok (2004), for example, introduced a combined physical/statistical downscaling approach, where the total error between model and observations is decomposed into a small-scale representation mismatch and a large-scale model error.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

High-resolution grids of hourly meteorological variables for Germany

Krähenmann

Brienen

Imbery

et al. 2016

Theor Appl Climatol

View full text Add to dashboard Cite

We present a 1-km 2 gridded German dataset of hourly surface climate variables covering the period 1995 to 2012. The dataset comprises 12 variables including temperature, dew point, cloud cover, wind speed and direction, global and direct shortwave radiation, down-and up-welling longwave radiation, sea level pressure, relative humidity and vapour pressure. This dataset was constructed statistically from station data, satellite observations and model data. It is outstanding in terms of spatial and temporal resolution and in the number of climate variables. For each variable, we employed the most suitable gridding method and combined the best of several information sources, including station records, satellite-derived data and data from a regional climate model. A module to estimate urban heat island intensity was integrated for air and dew point temperature. Owing to the low density of available synop stations, the gridded dataset does not capture all variations that may occur at a resolution of 1 km 2 . This applies to areas of complex terrain (all the variables), and in particular to wind speed and the radiation parameters. To achieve maximum precision, we used all observational information when it was available. This, however, leads to inhomogeneities in station network density and affects the long-term consistency of the dataset. A first climate analysis for Germany was conducted.

show abstract

Can dynamically downscaled windstorm footprints be improved by observations through a probabilistic approach?

2014

View full text Add to dashboard Cite

Windstorms are a main feature of the European climate and exert strong socioeconomic impacts. Large effort has been made in developing and enhancing models to simulate the intensification of windstorms, resulting footprints, and associated impacts. Simulated wind or gust speeds usually differ from observations, as regional climate models have biases and cannot capture all local effects. An approach to adjust regional climate model (RCM) simulations of wind and wind gust toward observations is introduced. For this purpose, 100 windstorms are selected and observations of 173 (111) test sites of the German Weather Service are considered for wind (gust) speed. Theoretical Weibull distributions are fitted to observed and simulated wind and gust speeds, and the distribution parameters of the observations are interpolated onto the RCM computational grid. A probability mapping approach is applied to relate the distributions and to correct the modeled footprints. The results are not only achieved for single test sites but for an area-wide regular grid. The approach is validated using root-mean-square errors on event and site basis, documenting that the method is generally able to adjust the RCM output toward observations. For gust speeds, an improvement on 88 of 100 events and at about 64% of the test sites is reached. For wind, 99 of 100 improved events and~84% improved sites can be obtained. This gives confidence on the potential of the introduced approach for many applications, in particular those considering wind data.

show abstract

A Combined Physical–Statistical Approach for the Downscaling of Model Wind Speed

Cited by 46 publications

References 13 publications

Surface Wind Speed Prediction in the Canadian Arctic using Non-Linear Machine Learning Methods

Surface Wind Speed Prediction in the Canadian Arctic using Non-Linear Machine Learning Methods

High-resolution grids of hourly meteorological variables for Germany

Can dynamically downscaled windstorm footprints be improved by observations through a probabilistic approach?

Contact Info

Product

Resources

About