Rogier

WAsP 12.8 has been released. We have done extensive validation of the new model and have gathered 109 tall masts from 4 different continents to test the new model. For each of these sites the ERA5 stability data is extracted an used as input for WAsP12.8. The previous version 12.7 with default heat fluxes all over the world is also tested. At each masts we do 3 cross-predictions, i.e. using the input at one height and predicting at another height, comparing with observations at that height. Some sites also include horizontal cross predictions from one mast to another one. You can find the full report with all the error metrics here: https://www.wetterzentrale.de/validation_report_WASP12_8.html

Hi Keegan, For license issues it is probably better to use the form on the website: https://www.wasp.dk/support because we can check your license status there. Regards

Hi Alfonso, Yes for some reason the permissions had changed. I have updated the link above now. There is also some new instruction videos added. Regards

Hi, Yes that is a rather tricky situation. Regardless what you do you will probably have high flow-modelling uncertainty with a forest on a hill. Using the CORINE data you will have to add a displacement height, that is not done automatically because it depends on the height of the forest. You can try to use those global forest data to estimate a forest height, but I don't have experience with that myself. If you really want an accurate forest height you can in some countries download data from lidar scans (e.g sweden, denmark, finland). If you have lidar data you can also separate the orography and roughness better, because SRTM data include to some extent the tree heights within the elevation data because they are measured by satellite (i.e. it is a digital surface model and not a digital terrain model). If your wind farm is directly in the forest the displacements will for sure have big impact. If they are further away it is mostly the roughness that will have a impact. Some more links regarding forest modelling and WAsP: https://wes.copernicus.org/articles/6/1379/2021/ https://wes.copernicus.org/articles/3/353/2018/ Regards Rogier

Hi, Yes coincidentally I just gave a course with that. You will need the same scripts as I included above. I copy the instructions below: https://data.dtu.dk/articles/software/Using_QGIS_to_create_WAsP_maps/20495178 There is a video how-to here: https://panopto.dtu.dk/Panopto/Pages/Viewer.aspx?id=f855a532-b99c-44af-b105-b08300954b2d You may also need the introduction to QGIS to give some background: https://panopto.dtu.dk/Panopto/Pages/Viewer.aspx?id=870b0413-4956-4b3f-a9b2-b0830094e790

Yes these new GML files can contain displacements heights, which is a feature which is only available from WAsP 12.7 onwards. But I am happy you got it to work.

Hi Gururaj, My best guess is that you didn't unpack the file CORINE.gpkg from the ZIP file. That also has to be in the script directory as explained on page 2. Regards Rogier

I posted some instructions in the other thread about this topic.

Hi Antonio, There is some experimental helper scripts in the free software QGIS attached. You can install these files per the instructions in the attached document and then look for a script called 'Polygons to roughness lines'. The instructions attached are for a new release of WAsP that can deal with displacement height, but if you are dealing with roughness length you only need the 'z0' field. So to save a WAsP .map file you will need to use the script 'Save WAsP roughness layer' instead of the 'Save WAsP landcover layer' that is described in the document. Regards All files are in this folder: https://data.dtu.dk/articles/software/Using_QGIS_to_create_WAsP_maps/20495178 The experimential helper scripts are called "wasp_scripts.zip" (installation instructions in the documents below) A full workflow to make both elevation and roughness maps in QGIS is included in "Creating a WAsP elevation and roughness map using QGIS"

Yes sorry, I meant interannual variation, because I though you wanted to calculate the AEP (annual energy production) for the whole year based on two months, but after reading your question again it is only for two months. Then your approach is fine.

If there is no annual variation in your wind resource then it is as simple as you describe: you just use the tabfile to calculate the AEP and and divide by 6. There is also technical and electrical losses in the wind farm that are unaccounted for in WAsP, so you will have to estimate those as well.

Hi Filiberto, You are right that you shouldn't apply the high resolution maps from WAsP here. What you actually want is the roughness and elevation map that were used in the numerical model that is used by Vortex (WRF). At DTU we call this a 'generalization' (see here: https://orbit.dtu.dk/files/244458690/DTU_Wind_Energy_E_0218.pdf), i.e. cleaning the numerical model output so that you get something similar as you get when you clean an observed wind climate of local effects. This process is in fact what is done in the global wind atlas (https://globalwindatlas.info/), i.e. we run WRF for a long period (10 yrs) and generalize using the modelled topography. There is a new feature in WAsP 12 where you can import this generalized wind climate directly from the global wind atlas. This is the way that I would use numerical model data for preliminary wind resource mapping. If you really want to use the data you have you would need to get hold of the roughness and elevation map of the WRF setup that vortex used to do it 'correctly'..

Yes, I would do both, adding elevations to the existing map and extracting a height from your input wind climate. You then have to make sure to have the correct displacement height for all your turbines as well. Adding/subtracting the height from the turbines/climate is likely to have the biggest impact. And make sure to double check because I know from experience it is easy to make a mistake in this manual process.

Hi, We are currently working on a forest implementation where it will be possible to easily ingest displacement heights into WAsP. We have recently published a paper about this that might be useful: https://wes.copernicus.org/preprints/wes-2021-28/ All these routines are unfortunately not yet available in a user-interface, but planned probably somewhere this year. As a rule of thumb you can use what is written there: displacement = (2/3)*tree height and roughness = 0.1*tree height. You could use these rules to create a roughness and displacement map from the layer you found. The new satellite based data layers that are presented in that paper are available here: https://windsight.dhigroup.com/ In the paper we used a displacement from a forest that is at least 2.5 m high so in your case of 7 m trees displacement height could already become important (although still quite small).

Ok, so in that case the reason why you won't get the same values is indeed due the the modelling. The most common causes are the roughness length of the terrain (if it is too high you will get a too high wind resource). Also the default stability used in WAsP could be off (mostly the offset heat flux in profile model). If these things are wrong you will not be able to reproduce the wind climate from one height to another, so it is worth spending some time investigating what could changed.

Hi Lucas, I am not sure if I understand your question correctly, but it seems like you are applying the same time series twice to get a GWC? You will need to use the 'true' height where your time series was measured and insert this in the WAsP program. I.e. if you have a measured time series from a height of 60 m, you will get different results when you use this time series with an input height of 100 m. You will need the GWC that was generated using the real height of 60 m, otherwise WAsP will think the wind speed of 60 m was valid at 100, giving you much lower values in the GWC.

The main reference is the European wind atlas: https://orbit.dtu.dk/en/publications/european-wind-atlas(335e86f2-6d21-4191-8304-0b0a105089be).html More detailed publications on specific topics can be found here: https://www.wasp.dk/publications Hope this helps

Hi, I am not sure what you mean: when I try to assign more than 5 roughness value in map editor 12 it works just fine. See attached screenshot for assigning new roughness to lines.

Dear Lino, I am not fully sure if I understand your question, but WAsP can give you an AEP for a given input wind speed distribution. It cannot give you a 10 min timeseries of power production of each turbine. Regards Rogier

Hi, You are right that you can have similar mean shear by using different combinations of a heat flux offset/rms. Usually we recommend to just change the offset heat flux and make sure that your wind measurements profile fits that value. You can see some examples of how this looks like for example Fig. 6 in this report: http://orbit.dtu.dk/files/110948908/DTU_Wind_Energy_E_0072.pdf

Dear Dhanraj, If there is errors in the 10 m master plus nearby obstacles I would use the 40 m height, because it will be much less sensitive to the local roughness. I don't think you can do a model validation (I assume you mean fitting the profile to the measurements) with such a data set. Regards Rogier

Hi Dhanraj, No the distance up to which a observed wind climate is valid depends mostly on the complexity and similarity of the terrain and on mesoscale wind effects. Dependent on those two factors can vary indeed roughly between 0 an 50 km. As far as I know there is no reference for this, it is just based on the physical assumptions used in the model. You can check mesoscale gradients in wind speed on the global wind atlas (https://globalwindatlas.info), which is based on simulations with a resolution of 9 km. So even with that resolution it is still possible that some mesoscale effects are not fully resolved. Regarding microscale effects we usually use the 'similarity principle': if sites are very similar in type of terrain you can extrapolate over larger distances, if not, the distance becomes smaller.

Dear Tomas, I cannot reproduce your error: for me the calibration of image in the map editor seems to work fine. I think it is better if you send your map, image and coordinates to the WAsP support (waspsupport@dtu.dk) so we can take a look and try to reproduce the error. Best Regards Rogier

Dear Manasi, I am not sure if I get your question correct, but if you are trying to digitize a roughness map, it is not supposed to depend on height. If you see a forest in whatever you are using to digitize you would put a certain roughness area there using the map editor. If you are trying the digitize an elevation map you just have to assign the height above sea level to the line. Best regards, Rogier

Dear Ata, We usually recommend to take 12 months of data, even if you have more available. This is to avoid any seasonal dependence in the sampled wind data: for example in Denmark the wind speed during winter months can be about 50% higher than during summer, so if you would include two winter and one summer in your data you will end up with a large bias in your final AEP calculation. Regards Rogier Floors