The climate change results are calculated by analyzing hourly historical weather records for weather stations near to the location requested.
To produce the clearest historical weather picture possible, we combine the historical weather records for the nearby stations weighting closer weather stations more than stations that are further away.
If a station does not report the weather for a particular hour (or longer), the dashboard will look at other stations as necessary.
Historical temperature change
For temperature analysis, we use the hourly weather records to produce annualized statistics for the mean temperature across all hours, the mean daily high temperature and the mean daily low temperature.
These values are plotted on the chart to ilustrate the values graphically. The dashboard then applies a linear regression analysis on the data to identify any trend within the data.
This regression analysis provides the estimate for the annual temperature increase - we multiply the resulting temperature trend by ten to highlight the ten year change in temperature.
Historical rainfall and precipitation change
For precipitation analysis, we use the hourly weather records to produce annualized statistics for the annual preciptation total and also the number of days with more than one inch (25.4mm) of rainfall.
These are then plotted on the precipitation chart. Linear regression is used to identify any trends within the data.
Weather station changes
By analyzing the past weather history for all the weather stations surrounding the location requested the dashboard is able to extend the available time of analysis.
This is because many weather stations have not be available for more than 20 to 30 years. Weather stations locations change over time.
Caution should be applied when intrepreting the results for very long term climate trends as different weather stations wil be chosen as stations will be added or removed from the network.
If the local area experiences different climates, such as one local weather station being located near a river or alternatively being located higher in the mountains,
the choice of weather station may cause the results to be modified incorrectly.
Feedback or questions?
We would love to hear from you if you questions, comments or suggestions. Please