Model development in practice: a comprehensive update to the boundary layer schemes in HARMONIE-AROME cycle 40

Rooy, Wim C. de; Siebesma, Pier; Baas, Peter; Lenderink, Geert; Roode, Stephan R. de; Vries, Hylke de; Meijgaard, E. van; Meirink, Jan Fokke; Tijm, Sander; Veen, Bram van 't

doi:10.5194/gmd-15-1513-2022

Cited by 9 publications

(9 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model uses a semi-Lagrangian scheme on an Eulerian grid. The HARATU turbulence scheme (de Rooy et al, 2022;Lenderink & Holtslag, 2004) was used, which uses a prognostic equation for the turbulent kinetic energy (TKE). Shallow convection follows de Rooy et al (2022).…”

Section: Model Description and Set-upmentioning

confidence: 99%

“…The HARATU turbulence scheme (de Rooy et al, 2022;Lenderink & Holtslag, 2004) was used, which uses a prognostic equation for the turbulent kinetic energy (TKE). Shallow convection follows de Rooy et al (2022). Surface Externalisée (SURFEX) version 7.3 was used as a land surface model (Masson et al, 2013) with the land use classification from ECOCLIMAP II (Faroux et al, 2013).…”

Section: Model Description and Set-upmentioning

confidence: 99%

See 1 more Smart Citation

A One‐Year‐Long Evaluation of a Wind‐Farm Parameterization in HARMONIE‐AROME

Stratum

Theeuwes

Barkmeijer

et al. 2022

J Adv Model Earth Syst

View full text Add to dashboard Cite

The need to mitigate climate change will boost the demand for renewable energy and lead to more wind turbines both on‐ and offshore. In the near future, the effect these wind farms have on the atmosphere can no longer be neglected. In numerical weather prediction models wind‐farm parameterisations (WFP) can be used to model the effect of wind farms on the atmosphere. There are different modeling approaches, but the parameterization developed by Fitch et al. (2012) is mostly used in previous studies. It models the wind farm as a momentum sink and a source of power production and turbulent kinetic energy. In this paper, we have implemented the Fitch et al. (2012) WFP into HARMONIE‐AROME, the numerical weather prediction model that is currently used by at least 11 national weather services in Europe. We used HARMONIE‐AROME to make year‐long simulations for 2016 with and without the WFP. The results were extensively evaluated using lidar, tower and flight measurements at several locations near wind farms. Including the WFP reduces the model bias for wind speed near offshore wind farms. Wind farms not only affect wind, but also temperature and humidity, especially during stable atmospheric conditions: the enhanced mixing caused by the wind turbines reduces the stratification of temperature and humidity. Including the WFP in HARMONIE‐AROME results in a more realistic representation of the atmosphere near wind farms and makes it a more future‐proof model for weather forecasting.

show abstract

Section: Model Description and Set-upmentioning

confidence: 99%

Section: Model Description and Set-upmentioning

confidence: 99%

A One‐Year‐Long Evaluation of a Wind‐Farm Parameterization in HARMONIE‐AROME

Stratum

Theeuwes

Barkmeijer

et al. 2022

J Adv Model Earth Syst

View full text Add to dashboard Cite

show abstract

“…This concerns the scaling issue, mentioned in Section 1.2. The effects of the too low update frequency, which is generally in the order of 3-6 h (for instance, 6 h in the Netherlands and Belgium; Bubnová et al, 1995;Bengtsson et al, 2017;Termonia et al, 2018;de Rooy et al, 2022), are enhanced by the time it takes between the start of the model run and the arrival time of the results at the end users, which can add an additional 2-4 h. Consequently, within those 2-4 h (and during the 6-h validity of the forecast after that) the observation-based initial conditions of the NWP simulations have changed considerably, especially during convective rainfall events. This could cause forecast errors already at the start of the issue time of both the rainfall forecast and the subsequent hydrological forecast (Sun et al, 2014).…”

Section: Numerical Weather Predictionmentioning

confidence: 99%

Rainfall nowcasting for flood early warning

Imhoff¹

View full text Add to dashboard Cite

List of symbolsϵ k Perturbation field at spatial scale level k (-) λ ext k Ratio of the explained to the unexplained variance of the extrapolation component at scale level k (-) λ nwp k Ratio of the explained to the unexplained variance of the NWP component at scale level k (-) µ Mean of variable (L T −1 ) ρ Pearson's correlation (-) σ Standard deviation of variable (L T −1 ) a S Surface water area fraction in WALRUS (-) B Deterministic categorical bias (-) c D Channel depth parameter of WALRUS (L)

show abstract

“…Details on the initialization scheme are given by Bengtsson et al (2017). The HA model is well-tested, optimized for clouds and under continuous development (de Rooy, et al, 2021). The HA data set consists of atmospheric temperature, humidity, cloud cover at different heights and irradiance, as summarized in Table 5.1.…”

Section: Mesoscale Nwp Modelmentioning

confidence: 99%

“…The Harmonie-Arome (HA) mesoscale weather model (Bengtsson, et al, 2017) delivers operational forecasts at a 1h resolution, with horizontal resolution of 2.5 km, which will be used as a feature source in this work. The HA model is well-tested, optimized for clouds and under continuous development (de Rooy, et al, 2021). The HA data set consists of 12 parameters such as windspeed, at 100m, global horizontal irradiance and air temperature.…”

Section: Operational Weather Forecastsmentioning

confidence: 99%

Every ray counts! : Enabling large-scale solar PV integration by smart-grid applications combining fine-resolution meteorology and grid management

Kreuwel

View full text Add to dashboard Cite

Chapter 6 evaluates a year-long case study of operational congestion management on a city area scale, where congestion occurs when there is an excess of solar generation. An open-source machine learning framework is provided to Summary iv systematically analyse the predictability of load on a contingency. Grid operators have only recently started to apply day-ahead congestion management to locally balance the grid. While accurate forecasts of the net-load are required to stay costeffective and ensure grid-safety, the operational achievability of this accuracy must be proven. The net-load in this case study is a combination of energy consumption primarily households and small, medium businesses) and solar-and wind generation. This ML framework is applied to attribute of inaccuracies in the net-load forecast to errors in the forecasted weather parameters related to wind-and solar energy generation. Overall, the largest part of the errors in the net-load forecast can be attributed to errors in the wind forecast. However, we show that when the load approaches the grid safety limits, errors in the solar forecasts have the biggest impact on the accuracy of the net-load forecast.The results of this thesis, as summarized in Chapter 7, contribute to connecting the fields of meteorology and grid management. Specifically high-resolution irradiance forecasting and integration of solar PV by active grid balancing. These findings have led the grid operator involved in this work to adapt their worst-case scenarios of PV generation to include extreme values found under broken-cloud conditions. Moreover, a predictability analysis using the framework detailed in this work is performed for areas where congestion management is considered. Moving forward, we recognize the growing importance of weather forecasts in facilitating the energy transition. To bridge the temporal gap towards the next generation of highresolution, cloud-resolving weather models, we suggest the providers of weather forecasts to include post-processed results in a comparable way to the current model output, using proven models form the academic forecasting community. Most importantly, we recommend formulating joint projects between the fields of meteorology and grid management, to facilitate the collaboration between experts of these fields, which is desperately required for an efficient transition to a renewable-based society.v System operators recognize that in addition to reinforcement of the grid, they also need to make better use of the grids capacity by balancing local supply and demand (Schittekatte & Meeus, 2020). For the distribution grid, this balancing on a local level needs to occur at three distinct spatial scales (Figure 1.1, right). The smallest scale is the street scale (~20 households, ~0.1 x 0.1 km 2 ), which concerns a single low-voltage cable. The intermediate scale is the neighbourhood scale (~150 households, ~0.5 x 0.5 km 2 ), where the key challenges lie in serving several street scale areas using a single distribution transformer. The la...

show abstract

Model development in practice: a comprehensive update to the boundary layer schemes in HARMONIE-AROME cycle 40

Cited by 9 publications

References 61 publications

A One‐Year‐Long Evaluation of a Wind‐Farm Parameterization in HARMONIE‐AROME

A One‐Year‐Long Evaluation of a Wind‐Farm Parameterization in HARMONIE‐AROME

Rainfall nowcasting for flood early warning

Every ray counts! : Enabling large-scale solar PV integration by smart-grid applications combining fine-resolution meteorology and grid management

Contact Info

Product

Resources

About