Wind data is relatively difficult to display and analyse in detail. This is because both its direction and speed change frequently, so any graphical display must also have a time or frequency dimension. Unfortunately, the most available form of wind data from any meteorological station is simply statistical tables - which are very laborious to analyse and understand. However, most stations also keep hourly records of speed and direction. Whilst such long lists of meaningless numbers are even more difficult to understand than summary tables, they are perfect for computer-based analysis. The WeatherTool available at this site provides a unique system for displaying and analysing hourly wind data, as shown in Figure 1 below.

Figure 1 - Annual hourly wind direction data for New York, USA.

Select Wind Data: :: Annual Frequency :: Annual Avg.Temperature :: Annual Rel.Humidity :: Annual Rainfall :: Frequency - Summer :: Frequency - Autumn :: Frequency - Winter :: Frequency - Spring :: Annual Comparison :: Monthly Comparison

This system involves analysing wind data into 16 different directions, each centred on a multiple of 22.5° (360/16). The speed of the wind can then be represented as a distance from the centre of a polar chart.

This is done in 10 concentric rings for winds up to 50km/h. This gives a grid onto which various properties of the wind each hour can be plotted using a colour scale to indicate relative frequency. The benefit of this system is that average temperature, humidity and rainfall can be displayed as well as wind frequency. The charts can also be generated for any range of dates and times, making it possible to compare summer afternoons with winter mornings, etc.

As an example, by looking at the various plots available in Figure 1 above, it is quite clear that New York experiences regular southerly and south westerly winds throughout most of the year, except in winter where wind tend to be more from the north west. The southerly winds also tend to be the warmest and the most humid. The rainfall graph also shows driving rains from the south accompanied by up to 55km/h winds.

Such information is vital to the designer when considering rain protection, the best location for outdoor areas and natural ventilation pathways. Click here for an example of how climate data can be used at the pre-design stage to determine the viability of a night purge system.

Wind Roses

Another way of displaying wind data is to show its relative frequency as radiating lines from an octagonal chart. Each direction has 12 lines, one for each month of the year running in a clockwise direction around the chart. Each line on the chart represents the percentage of time per month that winds came from each of the resulting 8 directions. Thus, based on the dotted line representing 12.5%, if you added up the lengths of each month's 8 lines, they would add up to 100%.

Figure 2 - The octagonal wind rose, showing data for Perth, Western Australia.

The result is a diagram which clearly shows the dominant wind directions throughout the different months of the year. Such diagrams do not clearly show diurnal changes in wind direction. Thus two diagrams are normally constructed, one for 9am to show morning patterns and one at 3pm to show afternoons.