Evaluation of Microphysical Schemes of High-Resolution WRF-ARW Model in Windspeed Forecasting over a Complex Terrain Wind-farm Cluster in India

doi:10.21203/rs.3.rs-3086895/v1

Download PDF

Research Article

Evaluation of Microphysical Schemes of High-Resolution WRF-ARW Model in Windspeed Forecasting over a Complex Terrain Wind-farm Cluster in India

https://doi.org/10.21203/rs.3.rs-3086895/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Microphysical sensitivity experiments were conducted to forecast complex terrain windspeed over a windfarm cluster in Maharashtra, India using high-resolution WRF-ARW model with 6 km outer and 2 km inner domain during the month of June 2022. The simulations were compared to hub-height wind measurements from wind turbines data. Results showed that the WSM5 microphysical scheme produced the minimum absolute error for complex mountainous terrain, while Ferrier performed the worst, and produced the largest error in wind speed at a height of 120 meters. Moreover, planetary boundary layer and topographic representation also play a vital role in modeling complex terrain wind forecasts. The performance of other physical variables for different microphysical schemes remained almost similar with minor fluctuations. Our experiments suggest that the adopting high-resolution WRF-ARW model with suitable combinations of physical parameterizations especially WSM5 microphysical schemes can significantly improve windspeed forecasting over complex terrain wind sites.

Complex terrain

Windspeed forecasting

high-resolution WRF-ARW

Microphysical schemes

Wind energy plays an increasingly important role in renewable energy production. The accurate prediction of wind power generation is significant to energy planning and power grid dispatch. In recent years due to advancements in the high-computing systems, numerical weather prediction (NWP) based meteorological models are being used in wind forecasting. A study by Mirjanovic and Chow (2014) showed that meteorological mesoscale models perform well on simple terrains; however, complex terrains benefit from high resolution modelling to reproduce local wind forces.

Although NWP models have been widely used to study the weather and climate, it inherently involves many sources of uncertainty as they cannot resolve all physical processes. Thus, parameterizations bring the effect of key physical processes, such as radiative heating, planetary boundary layers, cloud microphysics, etc. Different physical parameterization packages reproduce natural phenomena to varying degrees of accuracy, and choosing appropriate combinations is extremely important, as this decision strongly influences the model simulation results (Yu et al., 2011; Gao et al., 2016; Yang et al., 2017; Taraphdar et al., 2021; Gómez-Navarro et al., 2015; Stegehuis et al., 2015).

The complex terrain naturally causes variations in winds due to local effects, mainly because of region’s topography. In the complex terrain, slope and valley winds blow up and down during the day and night, respectively. Wind forecasting over a complex terrain has a variety of applications, especially in wind power sectors. Weather Research and Forecasting (WRF) is an NWP model which is extensively used in the forecasting of wind speed for a variety of purposes over a complex terrain (Carvalho et al., 2014; Fernandez et al., 2018; Jimenez and Dudhia 2013; Giannaros et al., 2017). The representation of the terrain in the model is also important for better simulation of wind flow in complex terrain (Carvalho et al., 2012b; Mughal et al., 2017). Their studies emphasized using the WRF model to reproduce the wind at lower heights near the surface in a complex terrain area.

Dynamical downscaling is crucial for increasing the model resolution to 1 km or even meters, especially for complex terrain cases (Ching et al., 2014; Seaman et al., 2012), and large-eddy simulations (LES) models (Chow and Street 2008). Previous studies have shown that higher resolution of the model gives better forecast of the wind pattern over the complex terrain (Carvalho et al., 2012b; Mughal et al., 2017; Fernandez-Gonzalez´ et al., 2018; Carvalho et al., 2012a; Samuelsen, 2007; Valkonen et al., 2020) but at the same time, it requires high computational power (Mass et al., 2002). However, after achieving a specific resolution in model, further increment in the resolution doesn’t reduce the errors in complex terrain (Solbakken et al., 2021). The WRF model can provide very high-resolution wind data, which is required by the wind power sector. A slight deviation from the actual wind speed induces a significant error in the wind energy calculations. Various experiments are conducted to set the model configuration for a specific location to forecast the wind, which is used in the power calculations. There is lack of evaluations in the complex terrain because of the lack of the data over there (Wagenbrenner et al., 2016). Accurate wind forecasting over complex terrain remains a challenge for the research communities, since more gusts present over a complex terrain results in frequent changes in the direction of the wind.

On the other hand, finding the suitable WRF physical parameterization schemes in different terrains is an active area of research. Yu et al. (2022) reported that the wind speed is the most sensitive to the planetary boundary layer (PBL) schemes, followed by radiation and microphysical schemes, while wind direction is less susceptible to variation of the physical parameterizations. The choice of cloud microphysics parameterization scheme also affects the performance of numerical models simulating wind patterns (Solbakken et al., 2021). Cloud microphysical processes such as moisture evaporation and condensation, etc. can affect atmospheric thermodynamic and dynamic interactions (Santos-Alamillos et al., 2013; Rajeevan et al., 2010; Li et al., 2020), and thereby affecting the vertical distribution of heat and the behavior of surface wind fields. Recently, Cheng et al. (2013) reported that predictions of summer wind speeds in northern Colorado were strongly affected by choice of cloud microphysical parameterization scheme and that the WRF double-moment 6-class (WDM6) scheme performed the best (Rajeevan et al., 2010).

Currently India has the fourth highest wind installed capacity in the world with a total installed capacity of 39.25 GW (as of 31st March 2021) and has generated around 60.149 billion Units during 2020-21. Maharashtra state has the third largest installed capacity of about 5 GW wind energy generation in India after Tamil Nadu, which has 9.6 GW, and Gujarat, which has 8.5 GW (https://mnre.gov.in/wind/current-status/). Wind sites located at the complex hill terrain largely fall under Maharashtra state in India. Wind power generators face significant difficulties in power estimation at these hilly sites, owing to the challenges these terrains pose to wind forecasting. So, we chose these sites which consists of a cluster of the wind farm in the western part of India at Lohara region in Maharashtra.

In this paper, we intend to find the best microphysical scheme for the WRF-ARW model, and also, we aim to provide best possible combination of physical schemes on complex terrain wind forecasting. Here, we evaluated the performance of eight microphysical schemes of the high-resolution WRF-ARW model. To address this issue, a total of 72 hours simulation was carried out with the initial condition at 12 UTC on June 5, 2022. Our results would provide a valuable evaluation of microphysical schemes of WRF-ARW model performance over complex-terrain wind forecasting. The rest of the paper is structured as follows: Section 2 describes the data and model configuration, and Section 3 discusses the results. Finally, Section 4 concludes the findings.

Hourly land surface temperature (LST), relative humidity (RH) at the surface, and height of PBL data from the European Centre for Medium Range Weather Forecasting (ECMWF) global reanalysis (ERA5, Hersbach et al., 2020) were used to compare with the model output. ERA5 horizontal resolution was 0.25^o x 0.25^o. ERA5 was taken due to the unavailability of supervisory control and data acquisition (SCADA) observation in RH, LST & PBL. WRF model (version 3.9.1.1) with an advanced research WRF (ARW) core was used for the simulation of wind over the complex hilly terrain wind farm location at Lohara (17.98^oN, 76.32^oE), which is a non-hydrostatic atmospheric model with terrain-following vertical coordinates (Skamarock et al., 2008). Results are compared with the SCADA observation from wind turbines of the specific wind sites in Maharashtra. SCADA data are regularly received by the company 50Hertz Ltd, located in New Delhi (https://re50hertz.in). They usually process observation and issue forecast for their clients.

Sensitivity experiments were conducted with eight different microphysical schemes to find the impact on wind forecasting by keeping the other combinations constant. Two nested domains with resolutions of 6 km for the outer and 2 km for the inner domain with 35 vertical levels were taken over the domain shown in Fig. 1.. The other physics schemes (cumulus, PBL, etc.) have been determined based on our previous sensitivity experiments, which are not discussed here. The Global Forecast System (GFS) dataset from National Canters for Environmental Prediction (NCEP) has a horizontal resolution of 0.25° x 0.25^o and 38 vertical levels and was used to define the initial and boundary conditions for all the simulations. The WRF model was initialized at 12 UTC on June 5th, 2022, and integrated until 12 UTC on June 8, 2022, with the first 6 hours treated as a spin-up period. June is the high-wind season for the Lohara site since the Indian summer monsoon commences at that time. Transition months from summer to monsoon ( May and June), when high-wind season starts, many ramp-up events due to sudden gusty wind occur during this time every year, which causes a lot of difficulties for wind-energy generators.

Mean absolute error (MAE) and mean absolute percentage error (MAPE) are used to present quantitative assessment in our study for model evaluation. MAPE is a measure of prediction accuracy of a forecasting method in statistics. It usually expresses the accuracy as a ratio defined by the formula:

$$MAPE=\frac{100\text{%}}{n}{\sum }_{t=1}^{n}\left|\frac{{A}_{t}-{F}_{t}}{{A}_{t}}\right|$$

where A_t is the actual value and F_t is the forecast value. Their difference is divided by the actual value A_t. The absolute value of this ratio is summed for every forecasted point in time and divided by the number of fitted points n.

Model physics combinations used in this study is shown in Table 1.

Table 1

Model setup and Experimental configuration
Exp Name	Resolution (km)	Cumulus	Boundary Layer	Microphysics
FERRIER	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	ETA Ferrier (Rogers et al., 2001)
GODDARD	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	Goddard (Tao et al., 1989)
THOMPSON	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	Thompson (Thompson et al. 2008)
LIN	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	Purdue Lin (Chen and Sun, 2002)
WSM3	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	WRF single-moment 3-class (WSM3, Hong et al., 2004)
WSM5	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	WRF single-moment 5-class (WMS5, Hong et al., 2004)
WSM6	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	WRF single-moment 6-class (WSM6, Hong and Lim, 2006)
KESSLER	6 and 2	Kain-Fritsch (Kain and Fritsch, 1990), 0	Yonsei University (YSU, Hong et al., 2006)	Kessler (Kessler, 1969)

In inner domain no cumulus scheme was used for its cloud-resolving scale resolution. Before explicitly looking into microphysical impact, we conducted several sensitivity experiments in model dynamics by considering other suggested physical schemes from various kinds of literatures. We found that gravity waves in complex terrain create sudden wind speed peaks. So, to contain these gravity waves, the following dynamical combinations at the namelist provided the best possible results in wind forecasting: hybrid_opt = 2, w_damping = 1, diff_opt = 2, and km_opt = 4. Also, in model domain settings, we find better results after applying topo_shading = 1, which takes care of topography shading. As spin-up is necessary to achieve the model’s stability, we use 6 hours as the model spin-up time here.

PBL schemes play a critical role in modulating mass, energy, and moisture fluxes between the land and atmosphere, which in turn influence the simulation of low-level temperatures, cloud formation, and wind fields (Falasca et al., 2021; Gholami et al., 2021; Gómez- Navarro et al., 2015; Jiménez and Dudhia, 2012; Gonçalves-Ageitos et al., 2015). Thus, we first conducted few PBL sensitivity experiments before starting our microphysical experiments with LES, Mellor-Yamada-Janjic (MYJ), and YSU schemes. We found that Yonsei University (YSU) does better than MYJ and LES (not shown here). LES sometimes performed better, sometimes not better than other existing PBL options. Gómez-Navarro et al. (2015) investigated the sensitivity of the WRF model to PBL schemes by simulating wind storms over complex terrain at a horizontal resolution of 2 km. In their study, the WRF model was combined with the MYJ scheme and overestimated wind speed by up to 100%; however, this bias was significantly reduced when the non-local scheme developed at YSU was used instead.

The experimental domain is shown in Fig. 1, where topographic height is shown in-set for the inner domain. Domain terrain height is more than 700 meters, a complex terrain site in Maharashtra state in India.

A total of eight sensitivity experiments have been conducted to find the impact of microphysical schemes on surface wind forecasting in the WRF-ARW model in complex terrain at a location of a wind turbine cluster. Here, we analyzed only the inner domain (2 km resolution) produced simulations. Figure 2 shows wind at 120-meter height time series of all experiments from 5th June 12 UTC to 8th June 12 UTC, 2022. SCADA data, which is our observation, is indicated in a black dotted line. The 3-day simulation performed reasonably well as compared to the observation. Although, two abnormal peaks in simulation at around the 26th and the 50th hour lead-time can be observed. These overestimations may be accompanied by high sensitivity to a slight fluctuation, while the WSM5 scheme simulated comparatively well with fewer fluctuations.

From Fig. 2, we see that WSM5 performed better than other microphysical schemes. Apart from windspeed, we also investigate other important physical variables which have direct/indirect impact on windspeed. To examine the performance of microphysical schemes on other physical variables, domain averaged LST, RH and PBL height have been investigated and compared against the ERA5 dataset, shown in Fig. 3. Windspeed is clearly linked with PBL height, RH, and LST. From Figs. 2 and 3, windspeed is almost proportional (inversely proportional) to RH (LST and PBL).

Interestingly, it can be seen that very high wind (an anomaly compared to SCADA data) is observed when PBL and LST are high, and RH is low. Generally, it is observed that lower LST, PBL height and high RH resulted in higher wind speed. As per the physical variables, high simulated wind speed at around 24th hours lead-time can be considered as the right response by the model. Where we can see SCADA observation showed a much lower windspeed, but the model showed a large overestimation. The model’s internal dynamics may play a significant role in this case. Overall, as per our model’s response to other physical variables like LST, RH, and PBL height, wind speed outcome can be regarded as a correct response. But a comparatively large underestimation can be observed in LST and PBL at around 30th and 53rd hours lead-time (Fig. 3), where we can also see a large overestimation in windspeed by the model.

Thus, model response was appropriate, but from 30th to 53rd lead-times LST and RH simulation show noticeable difference from the observation except PBL, where PBL height is correctly simulated in between this lead time. Therefore, it is seen that windspeed is highly related with the PBL height.

On the other hand, microphysical schemes represent the hydrometeor distribution of cloud, domain averaged reflectivity is also plotted to observe the hydrometeor distribution pattern (Fig. 4). Here, we see the THOMPSON, GODDARD, WSM5 and WSM6 schemes produced higher concentration of hydrometeor distribution, resulting in higher reflectivity. While FERRIER and KESSLER schemes did not produce much concentration of hydrometeors like others. This figure is the purpose for representation of hydrometeors by all the schemes. Here, we can see the response of WSM5 as compared to the THOMPSON in terms of hydrometeor distribution. Although, hydrometeor distribution directly impacts the cloud-forming process, but for windspeed it has an indirect effect, via cloud-PBL-temperature feedback. A study by D. Choudhury et al. (2017) also reported that for high-resolution WRF modeling, Thompson and Goddard schemes are suitable for tropical cyclone intensity forecasting; they can capture well the cloud-hydrometeor structure of a system.

Finally, to get a quantitative assessment, Table 2 documents the MAE and MAPE of all experiments. From Table 2, it can clearly see that WSM5 performs the best among all other microphysical schemes with 1.42 m/s MAE. Where 1.73, 1.5. 1.5. 1.67, 1.63, 1.72, and 1.65 m/s MAEs are observed for FERRIER, GODDARD, THOMPSON, KESSLER, LIN, WSM3, and WSM6, respectively. In terms of MAPE too, WSM5 produced the least error with 25.64%, while the maximum MAPE is found for Ferrier with 32.8%. Poor representation of hydrometeor in Ferrier and Kessler schemes, as seen in Fig. 4, can lead to higher MAE than other schemes.

Table 2

Model performance based on Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE)
EXPTS	FERRIER	GODDARD	THOMPSON	KESSLER	LIN	WSM3	WSM5	WSM6
MAE (m/s)	1.73	1.5	1.5	1.67	1.63	1.72	1.42	1.65
MAPE (%)	32.8	27.96	27.96	31.93	31.59	31.99	25.64	29.82

Summaries

Microphysical sensitivity experiments were conducted for complex terrain wind forecasting over a wind-turbine cluster site in Maharashtra in India. 6 km outer domain and 2 km inner domain resolution in two-way nesting mode have been used for WRF-ARW model simulation. After conducting some PBL sensitivity experiments, YSU is considered the PBL parameterization for both domains, KF is taken as the cumulus scheme for the outer domain, and none for the inner domain. It is well established that the PBL plays a very significant role in high-resolution wind forecasts by NWP models. After finding a suitable combination of dynamics and physics options, we design microphysical sensitivity experiments to choose the best microphysical scheme for such complex terrain wind forecasting. A total of eight sensitivity experiments have conducted to find the most suitable microphysical scheme for complex terrain wind modeling using WRF-ARW.

The experimental results showed that the WSM5 scheme produced the least MAE for such complex mountainous terrain, while FERRIER performed the worst, and produced the largest error in wind speed at 120-meter height. Moreover, for complex terrain wind forecast modeling, PBL and topography representation plays a vital role. The performance of other physical variables for different microphysical schemes remains almost similar with minor fluctuations. We know that large differences exist between the schemes for total water vapour, cloud water, and accumulated precipitation. The different distributions of atmospheric moisture strongly impact on both the shortwave and longwave downward fluxes at the ground.

From our experiments, we can say that the high-resolution WRF-ARW model with proper dynamical and physical scheme combinations significantly improves in wind forecasting over complex terrains wind turbine sites. However, this study has few potential limitations. Due to lack of availability of SCADA data in other weather variables (LST, RH, etc.), coarser-resolution ERA5 data have been compared to the high-resolution simulated parameters. And the sensitivity experiments conducted in the study focused on a narrow temporal and spatial scope, consisting of only a 3-day simulation and a small geographical region. Further studies will be required for longer-period simulation in multiple windfarm sites, especially over the intra-seasonal periods and with several other physical combinations of the WRF-ARW model to get more robust combinations.

Code and data availability

The Weather Research and Forecasting (WRF) model is freely available online and can be downloaded from the page: https://github.com/wrf-model/WRF. The ERA5 data is available at ECMWF (https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5).

Authors’ contributions

DC conceptualized the study and conducted the simulations. The analysis was carried out by both DC and BSK. The original draft of the paper was written by DC, all the authors took part in the edition and revision of it.

Competing interests

The authors declare that they have no conflict of interest.

Acknowledgements

We are thankful to the 50Hertz Ltd, New Delhi, India for providing the SCADA observation from wind turbines. We would like to express our sincere thanks to Mrs. Sanchita Saha for her English editing, Manish Paliwal at 50Hertz Ltd for his support during experiments. We also thank NCEP-NCAR for their WRF-ARW model and ECMWF for the ERA5 datasets. Finally, we are also thankful to the anonymous reviewer’s suggestions.

Carvalho D, Rocha A and Gomez-Gesteira M (2012a) Ocean surface wind simulation forced by different reanalyses: comparison with observed data along the Iberian Peninsula coast. Ocean Model 56, 31–42. https://doi.org/10.1016/j. ocemod.2012.08.002.
Carvalho D, Rocha A, Gomez-Gesteira M and Santos C (2012) A sensitivity study of the WRF model in wind simulation for an area of high wind energy. Environ.Model Softw. 33, 23–34. https://doi.org/10.1016/j.envsoft.2012.01.019.
Carvalho D, Rocha A, Gómez-Gesteira M and Silva Santos C (2014) WRF wind simulation and wind energy production estimates forced by different reanalyses: comparison with observed data for Portugal. App. Energy, 117, pp. 116-126, 10.1016/j.apenergy.2013.12.001.
Chen SH and Sun WY (2002) A one-dimensional time dependent cloud model.J. Meteorol. Soc. Japan Ser. II, 80, 99-118.
Cheng WYY, Liu Y, Zhang Y, Mahoney WP and Warner TT (2013) The impact of model physics on numerical 380 wind forecasts. Renew Energy, 55, 347-356, https://doi.org/10.1016/j.renene.2012.12.041.
Ching J, Rotunno R, LeMone M, Martilli A, Kosovic B, Jimenez PA and Dudhia J (2014) Convectivelyinduced secondary circulations in fine-grid mesoscale numerical weather prediction models. Mon Wea Rev, 142, 3284–3302.
Choudhury D and Das S (2017) The sensitivity to the microphysical schemes on the skill of forecasting the track and intensity of tropical cyclones using WRF-ARW model. J Earth Sys Sci, 126(4), DOI: 10.1007/s12040-017-0830-2.
Chow FK and Street RL (2008) Evaluation of turbulence closure models for large-eddy simulation over complex terrain: flow over Askervein Hill. . Appl Meteorol Climatol, 48, 1050–1065.
Falasca S, Gandolfi I, Argentini S, Barnaba F, Casasanta G, Di Liberto L, Petenko I and Curci G (2021) Sensitivity of near-surface meteorology to PBL schemes in WRF simulations in a port-industrial area with complex terrain. Atmos Res, 264, 105824, https://doi.org/10.1016/j.atmosres.2021.105824.
Fernandez-Gonzalez S, Martın ML, Garcıa-Ortega E, Merino A, Lorenzana J, Sanchez JL,Valero F and Rodrigo JS (2018) J Appl Meteoro. Climatol, 57 733–753 (Preprint https://doi.org/10.1175/JAMC-D-17-0121.1)
Gao X, Shi Y and Giorgi F (2016) Comparison of convective parameterizations in RegCM4 experiments over China with CLM as the land surface model. Atmos Ocean Sci Lett, 9, 246-254. 10.1080/16742834.2016.1172938.
Gholami S, Ghader S, Khaleghi-Zavareh H and Ghafarian P (2021) Sensitivity of WRF-simulated 10 m wind over the Persian Gulf to different boundary conditions and PBL parameterization schemes. Atmos Res, 247, 105147, https://doi.org/10.1016/j.atmosres.2020.105147.
Giannaros TM, Melas D and Ziomas I (2017) Performance evaluation of the Weather Reasearch and Forecasting (WRF) model for assessing wind resource in Greece. Renew Energy, 102 190–198. https://doi.org/10.1016/j.renene.2016.10.033
Gómez-Navarro JJ, Raible CC and Dierer S (2015) Sensitivity of the WRF model to PBL parametrisations and nesting techniques: evaluation of wind storms over complex terrain. Geosci Model Dev, 8, 3349-3363, 10.5194/gmd-8-3349-2015.
Gonçalves-Ageitos M, Barrera-Escoda A, Baldasano JM and Cunillera J (2015) Modelling wind resources in climate change scenarios in complex terrains. Renew Energy, 76, 670-678.
Hersbach H, Bell B, Berrisford P, Hirahara S, Horani A, Munoz-Sabater J, Nicolas J, Peubey C, Radu R, Schepers D et al. (2020) The era5 global reanalysis. Q. J. R. Meteorol Soc 146(730): 1999-2049
Hong SY and Lim J (2006) The WRF single-moment 6-class microphysics scheme (WSM6). Asia Pac J Atmos Sci, 42, 129-151.
Hong SY, Dudhia J and Chen SH (2004) A revised approach to ice microphysical processes for the bulk parameterization of clouds and precipitation. Mon Wea Rev, 132, 103-120.
Jiménez PA and Dudhia J (2012) Improving the representation of resolved and unresolved topographic effects on surface wind in the WRF model. J App Meteorol Climatol, 51, 300-316.
Jiménez PA and Dudhia J (2013) An evaluation of WRF’s ability to reproduce the surface wind over complex terrain based on typical circulation patterns. J App Meteorol Climatol, 52 1610 https://journals.ametsoc.org/doi/10.1175/JAMC-D-12-0266.1.
Kain JS and Fritsch JM (1990) A One-Dimensional Entraining/Detraining Plume Model and Its Application in Convective Parameterization. J Atmos Sci., 47, 2784–2802.
Kessler E (1969) On the distribution and continuity of water substance on atmospheric circulation. Meteorol Monogr, 10 (32), 84.
Li J, Ding C, Li F and Chen Y (2020) Effects of single- and double-moment microphysics schemes on the intensity of super typhoon Sarika 2016. Atmos Res, 238, 104894, https://doi.org/10.1016/j.atmosres.2020.104894.
Mass CF, Ovens D, Westrick K and Colle BA (2002) Does increasing horizontal resolution produce more skilful forecasts?: the results of two years of real-time numerical weather prediction over the pacific northwest. Bull Am Meteorol Soc, 83 (3), 407–430. https://doi.org/10.1175/1520-0477(2002)0832.3. CO2.
Mirjanovic SW N and Chow F (2014) Investigation of model parameters for high-resolution wind energy forecasting: A case study over simple and complex terrain. J Wind Eng Ind Aerodyn, 134, 10–24.
Mughal MO, Lynch M, Yu F, McGann B, Jeanneret F and Sutton J (2017) Wind modelling, validation and sensitivity study using Weather Research and Forecasting model in complex terrain. Environ Model Softw, 90, 107–125. https://doi.org/ 10.1016/j.envsoft.2017.01.009.
Rajeevan M, Kesarkar A, Thampi SB, Rao TN, Radhakrishna B and Rajasekhar M (2010) Sensitivity of WRF cloud microphysics to simulations of a severe thunderstorm event over Southeast India. Ann Geophys, 28, 603-619, 10.5194/angeo- 28-603-2010.
Rogers E, Black T, Ferrier B, Lin Y, Parrish D and DiMego G (2001) Changes to the NCEP Meso Eta Analysis and Forecast System: Increase in resolution, new cloud microphysics, modified precipitation assimilation, modified 3DVAR analysis. NWS Technical Procedures Bulletin, 488, 15.
Samuelsen EM (2007) EtdynamiskstudiumavstormenNarve - et kaldluftsutbruddiFinnmark - ved hjelpavobservasjonerognumeriskesimuleringer (In Norwegian).
Santos-Alamillos FJ, Pozo-Vázquez D, Ruiz-Arias JA, Lara-Fanego V and Tovar-Pescador J (2013) Analysis of WRF Model Wind Estimate Sensitivity to Physics Parameterization Choice and Terrain Representation in Andalusia (Southern Spain). J Appl Meteorol Climatol, 52, 1592-1609, 10.1175/jamc-d-12-0204.1.
Seaman NL. Gaudet BJ, Stauffer DR, Mahrt L, Richardson SJ, Zielonka JR and Wyngaard JC (2012) Numerical prediction of submesoscale flow in the nocturnal stable boundary layer over complex terrain. Mon Wea Rev, 140, 956–977.
Skamarock WC, Klemp JB, Dudhia J, Gill DO, Barker DM, Wang W and Powers JG (2008) Adescription of the Advanced Research WRF version 3. NCAR Technical note, 475+ STR.
Solbakken K, Birkelund Y and Samuelsen EM (2021) Evaluation of surface wind using WRF in complex terrain: Atmospheric input data and grid spacing. Environ Model Softw, 145, 105182, ISSN 1364-8152, https://doi.org/10.1016/j.envsoft.2021.105182.
Stegehuis AI,Vautard R, Ciais P, Teuling AJ, Miralles DG and Wild M (2015) An observation-constrained multi-physics WRF ensemble for simulating European mega heat waves. Geosci Model Dev, 8, 2285-2298, 10.5194/gmd-8-2285-2015.
Tao WK, Simpson J and McCumber M (1989) An ice-water saturation adjustment. Mon Wea Rev, 117, 231-235.
Taraphdar S, Pauluis OM, Xue L, Liu C, Rasmussen R, Ajayamohan RS, Tessendorf S, Jing X, Chen S and Grabowski WW (2021) WRF Gray-Zone Simulations of Precipitation Over the Middle-East and the UAE: Impacts of Physical Parameterizations and Resolution. J .Geophys Res Atmos, 126, e2021JD034648, https://doi.org/10.1029/2021JD034648.
Thompson G, Field PR, Rasmussen RM and Hall WD (2008) Explicit Forecasts of Winter Precipitation Using an Improved Microphysics Scheme. Part II: Implementation of a New Snow Parameterization. Mon Wea Rev, 136, 12, 5095-5115, https://doi.org/10.1175/2008MWR2387.1.
Valkonen T, Stoll P, Batrak Y, Køltzow M, Schneider TM, Stigter EE, Aashamar OB, Støylen E and Jonassen MO (2020) Evaluation of a sub-kilometre NWP system in an Arctic fjord-valley system in winter. Tellus Dynamic Meteorol Ocean., 72 (1), 1–21. https://doi.org/10.1080/16000870.2020.1838181.
Wagenbrenner S, Jason M, Forthofer1 K Brian Lamb S Kyle Shannon and Bret Butler W (2016) Downscaling surface wind predictions from numerical weather prediction models in complex terrain with WindNinja. Atmos Chem Phy., doi:10.5194/acp-16-5229-2016.
Yang B, Qian Y, Berg LK, Ma PL, Wharton S, Bulaevskaya V, Yan H, Hou Z and Shaw WJ (2017) Sensitivity of Turbine-Height Wind Speeds to Parameters in Planetary Boundary-Layer and Surface-Layer Schemes in the Weather Research and Forecasting Model. Bound.-Layer Meteorol, 162, 117-142, 10.1007/s10546-016-0185-2.
Yu E, Bai R, Chen X and Shao L (2022) Impact of physical parameterizations on wind simulation with WRF V3.9.1.1 over the coastal regions of North China at PBL gray-zone resolution. Geosci Model Dev, 15, 8111–8134, https://doi.org/10.5194/gmd-15-8111-2022.
Yu E, Wang H, Gao Y and Sun J (2011) Impacts of cumulus convective parameterization schemes on summer monsoon precipitation simulation over China. Acta Meteorol. Sin., 25, 581-592.

Download PDF

Version 1

posted

You are reading this latest preprint version

Evaluation of Microphysical Schemes of High-Resolution WRF-ARW Model in Windspeed Forecasting over a Complex Terrain Wind-farm Cluster in India

Status:

Version 1

Abstract

Figures

Introduction

Data And Model Configuration

Results

Summaries

Declarations

References

Status:

Version 1