windNavigator FAQ

The maps delivered through windNavigator were created with AWS Truewind's MesoMap® system, well known in the wind industry as the technology behind ground-breaking wind maps published for over 30 US states as well as many other countries.

MesoMap is a combination of mesoscale and microscale atmospheric models. The mesoscale model simulates weather conditions for a representative meteorological year (366 days sampled from 15 years) on a horizontal grid of 2.5 km. Starting from an initial condition established by regional weather data, and using the complete set of physical equations governing the atmosphere, the model simulates the evolution of weather conditions from the start to end of each day in the year. The microscale model then refines the wind fields from the mesoscale model to capture the influence of fine-scale topography and surface roughness changes at a high resolution of 200 m.

The maps have subsequently been fine-tuned through an error-correction procedure developed by AWS Truewind, which uses data from over 1000 wind-monitoring stations from public and authorized private sources (the windObs Database™).

Incorporated in the initial conditions for the mesoscale simulations are weather observations from many thousands of platforms, including surface stations, instrumented balloons, satellites, aircraft, and others. These data have been assimilated into the NCAR/NCEP Global Reanalysis (NNGR) database, which provide a snapshot of weather conditions every 6 hours on a 2.5 degree resolution grid for nearly the past 60 years. Data from rawinsonde stations (instrumented balloons), as well as sea-surface temperatures, land cover, topography, and other geophysical data, also drive the simulations.

The error-correction procedure relies on data from over 1000 wind-monitoring stations in AWS Truewind's windObs Database. The data have all been carefully screened, adjusted to represent historical average wind conditions where appropriate, and projected to the map height using a measured or estimated wind shear factor.

In creating this database, AWS Truewind drew on a wide array of both public and authorized private sources. Included are hundreds of Automated Surface Observing System (ASOS) weather stations, which are 10 m (33 feet) in height, as well as hundreds of tall towers ranging from 20 m (66 feet) to 100 m (328 feet) in height and instrumented expressly for wind resource assessment. Two hundred and seventy-five towers in the data base are taller than 30 m. There is an average of about one station every 100 km. However, the spatial distribution is uneven. The greatest concentrations of towers are in the Northeast, upper Midwest, California, and the Northwest. The lowest concentrations are in the Southwest and Southeast.

We wish to thank our clients for donating their data to this endeavor in the interest of improving the accuracy of the wind map for all users. AWS Truewind invites others to donate data as well for future map updates.

Thorough verification of wind maps with reliable observations is a hallmark of AWS Truewind's approach to wind resource mapping. The same philosophy has been applied to windNavigator. To produce a true estimate of the map error margin, each station in AWS Truewind's windObs Database was withheld from the error-correction procedure and the map error (difference between the map speed and the observed speed) at that station was determined.

Through this objective procedure, the mean bias of the high-resolution wind maps available for sale through windNavigator is found to be virtually zero, while the standard error (after accounting for uncertainty in the data) is about 0.35 m/s. See the chart below for a scatter plot comparing the predicted mean wind speeds with the observed after projection to the same map height.

The free, 2.5 km resolution wind maps have a much greater error margin because of their coarse resolution. The estimated standard error for these maps is 0.75 m/s.

The standard error is the standard deviation of the map errors at the wind-monitoring stations in AWS Truewind's windObs Database. Assuming these stations are representative of all locations in the wind map, and that the errors are normally distributed, the true mean speed should be within the standard error of the map speed in 68% of the mapped area. The error margin may vary, however, depending on the complexity of the terrain and surface conditions and on the density and quality of wind-monitoring stations in the surrounding region.

The windNavigator maps differ from AWS Truewind's previous state maps in two key ways.

First, they are more accurate because they have been subjected to a sophisticated error-correction procedure employing a data base of over 1000 wind-monitoring stations in the windObs Database. According to our best estimates, this procedure has eliminated almost all bias between the map and observations and has reduced the standard error from a typical range of 0.5-0.7 m/s to 0.35 m/s, or less than 5% for sites with a mean speed greater than 7 m/s. Overall, the estimated mean wind speeds in the Northeast, the Mid-Atlantic States, and the Northwest have decreased compared to the previously published maps, while the speeds in the Plains states and the Southwest have increased.

Second, windNavigator presents maps of 15 states that have never before been made available for public sale or distribution by AWS Truewind. The states are Texas, Oklahoma, Kansas, North and South Dakota, Minnesota, Wisconsin, Illinois, Iowa, Kentucky, Tennessee, Louisiana, Alabama, Mississippi, and Florida.

The result is the first seamless, high-resolution map of the conterminous United States backed by AWS Truewind's proven MesoMap technology and extensive validation.

AWS Truewind has made available via windNavigator a free, low-resolution wind resource map of the country. This map, which is what you see when you first open the windNavigator home page, was created by averaging the wind resource estimates from the seamless 200 m high-resolution map over a 2.5 km square area around each point. The 2.5 km Wind Speed Value is the annual average speed.

The 200 m Wind Speed Range is the range of mean annual speeds within each 2.5 km square area extracted from the underlying high-resolution map. It gives an indication of the spatial variation in the wind resource at that scale. In some regions of mountainous terrain, the spatial variation can be very large; in others, such as the Great Plains, it is smaller. The 200 m Wind Speed Range is provided free for any point on the 2.5 km resolution wind map. The 200 m maps can be purchased as PDF or GIS products for any project area you define in windNavigator.

We're glad you like it. For many casual users, the free, low-resolution wind resource map may be perfectly adequate to obtain a general picture of the wind resource in a given area. Have fun!

More demanding users, however, want the greater accuracy and finer detail available in the 200 m maps (see comparison below). This scale is the same as that favored by AWS Truewind when mapping regions and selecting project sites for its private developer clients. In addition, the high-resolution map data are available in GIS format, allowing users to overlay other data layers (roads, transmission lines, land parcels) and design preliminary project layouts, among other useful tasks.

However, even the error-corrected, high-resolution wind maps must be confirmed by on-site measurements before substantial investments should be considered.

Wind roses, speed distributions, monthly, and diurnal data are available for purchase at this time. We are currently in the process of automating such purchases by fall. In the mean time, please contact us directly to acquire information for your project area.

Alaska and Hawaii are next on the list. Then we plan on introducing new countries as fast as they can be completed. Stay tuned, and visit the site often. We also encourage you to join our e-mail list, to be notified of the latest map updates and new regions available through windNavigator.

Pricing is based on your area of interest in kilometers squared. For PDF purchases, a minimum fee of $950.00 for an area of 100 km² or less is required. An incremental fee is applied for each additional square kilometer past the initial 100.

GIS purchases require a minimum purchase of $2,300 for an area of 100 km² or less. An incremental fee is applied for each additional square kilometer past the initial 100.