We have derived an hourly time series of GB-aggregated wind power generation from 1980-present using wind speed records from NASA’s “MERRA” reanalysis (please note this has now been superseded by MERRA-2). This data was used in our paper:
D. J. Cannon, D. J. Brayshaw, J. Methven, P. J. Coker and D. Lenaghan, 2015. Using reanalysis data to quantify extreme wind power generation statistics: a 33 year case study in Great Britain. Renewable Energy, 75, 767-778. doi:10.1016/j.renene.2014.10.024
The original data and models used can be downloaded freely on request here, though we’d request you cite our paper if you use them. Some additional information concerning setting up and using the model is provided below.
In the coming years the geographical distribution of wind farms in Great Britain is expected to change significantly. Following the development of the “round 3” wind zones (circa 2025), most of the installed capacity will be located in large offshore wind farms. However, the impact of this change in wind-farm distribution on the characteristics of national wind generation is largely unknown. Our wind power model has therefore been applied to determine the long term characteristics of GB-aggregated wind power for a “high wind power penetration” future scenario. To represent the future wind farm distribution, a number of assumptions have been made:
- All Round 3 zones are developed to full capacity
- All onshore wind farms under construction or with planning permission are fully commissioned.
- All existing farms remain generating at their current capacity.
The details of the model/data are discussed in:
Drew, D., Cannon, D., Brayshaw, D., Barlow, J. and Coker, P. (2015) The impact of future offshore wind farms on wind power generation in Great Britain. Resources, 4 (1). pp. 155-171. ISSN 2079-9276 doi: https://doi.org/10.3390/resources4010155
and can be downloaded here.
The model constructs an hourly time series of regional-total wind power over any specified time period since 1979, using MERRA reanalysis wind speed data. Written in Matlab, the model is run from the Master.m script, which builds a time series of regional total wind power generation from MERRA reanalysis data. It does so by calling other Matlab scripts to compute the required steps.
The model is designed so that all user-defined settings can be edited in Master.m, and the user need not edit the sub-scripts (MERRA_interp.m and MERRA_clim.m), which are called from Master.m. However, all scripts are well commented to help users understand and, if required, modify the model to their own needs.
- MERRA data containing wind speed data at 2m, 10m and 50m.
- Wind farm distribution and capacity data: “windfarms.dat”
- A data file containing a list of wind farm locations (longitude/latitude), their capacity (in MW), and a farm-average turbine hub height above ground. See the example (“windfarms.dat”).
- Store this file in the same directory as Master.m
- A wind farm power curve: A data file containing a list of wind speeds and corresponding power output (in fractional Capacity Factor: I.e., as a fraction of the total wind farm capacity). See the example (“powercurve.dat”).
- Store this file in the same directory as Master.m
Running the model
With these ingredients in place, the steps to build the time series are as follows:
- Set the user-defined settings in Master.m
- Run the Master.m script from an open Matlab session!
- If the interpolation step is computed, the model outputs MERRA wind speeds horizontally interpolated to each of the wind farm locations in windfarms.dat (stored in NETCDF format).
- The climatology step outputs:
- A plot showing the power curve used.
- A time series of regionally-aggregated capacity factor (CF). 
- A time series of showing the date and time corresponding to each CF value.
- Both the date/time and CF variables are saved to an ascii file, as well as to a “.mat” Matlab data file.
 Capacity Factor = 100 % × [Total Power Generated (MW)] ÷ [Total Capacity (MW)].
The model will loop through all days, months and years between the start and end dates specified in Master.m, extract the MERRA wind speed data and interpolate it to the desired wind farm locations using the MERRA_interp.m script. The power curve, wind farm capacity and turbine hub height data is then used to calculate the wind power output of the entire fleet in MERRA_clim.m.
Note, the MERRA_interp.m script can be slow when computing a long time series and/or a large distribution of wind farms. Therefore, the model stores the interpolated wind data so that any of the following inputs can be changed without the need to repeat the interpolation:
- Wind farm capacities
- Turbine hub heights
- Power curve
However, any change in the distribution (longitudes or latitudes) of the wind farm distribution will require the recomputation of the interpolated data . Please note the model has also been applied to a future wind farm scenario with a higher capacity (larger number of sites). This data is also available to download here.
 A useful tip: If you want to use this model to study a number of different wind farm distributions, perform the interpolation step for all wind farm locations. Then, to test a particular distribution, just set the wind farm capacity to zero for any wind farms you do not wish to include. This will avoid having to recompute the interpolation step for each distribution.
Some notes about the MERRA data
To use this model, you must have the raw MERRA data downloaded in advance. It can be downloaded from:
The model is set up to read in the MERRA data product named “IAU 2d atmospheric single-level diagnostics (tavg1_2d_slv_Nx)”, using the “Daily Data Product”. This contains hourly wind speeds which are stored in a separate file for each day. Once downloaded, the MERRA files should look like this:
where NNN is an integer (this might be 100, 200, 300 or 301) and YYYYMMDD is the date of the file.
The model uses U and V wind components from 2m, 10m, and 50m (“U2M”, “U10M”, “U50M”, “V2M”, “V10M”, “V50M”), which must be available in the MERRA data files.
The model assumes the data is stored in NETCDF format, and can be accessed at the location: pathname / year / filename, where:
- “filename”: the name of the NETCDF file
- “year”: a folder containing all NETCDF files for the year
- “pathname”: the directory where the “year” folders are stored
are all user-defined in the Master.m script.
 Website available as of 18th February 2014. If the link no longer works, try http://gmao.gsfc.nasa.gov/merra/ and navigate to the Modeling and Assimilation Data and Information Services Center (MDISC) page, and then find a link to MERRA Data Products.