GNSS Satellite (GIOVE-A)

GNSS Applications

Using GPS for Numerical Weather Predictions

WeatherA company we work with GPS Solutions has developed a really cool application which I would like to introduce to you here.

GPS solutions specialises in high accuracy GPS solutions and applications. Besides accurate positioning there are many other things one can do with GPS. One of these things is using GPS based tropospheric delay estimates to improve numerical weather predictions. GPS solutions is generating weather predictions based on the more or less standard models and data but in addition includes GPS estimates to improve these weather predictions. The really fun part of this is that they have coupled the weather predictions with Google Earth. Thus you can "click" on any location on the world and get the weather forecast for that location for the next couple of days. I find that a really cool application!

Just think about you want to go out sometime in the next days. You just go to any point on the Earth using Google Earth, click on the location and up comes the weather forcast for the selected location.

Want to try it out, just click this link. This links shows you the location of the PosiTim head quarters and the weather forecast for the next couple of days!

GPS and Weather?!

You may wonder "how does this work". What does GPS, or in general GNSS, have to do with the weather. Well that is actually very simple. In high accuracy GNSS one of the major disturbances comes from the lower part of the Earth's atmosphere, the part we call the "troposphere". To overcome these disturbances we have to estimate parameters to model these signal delays. These parameters are typically called tropospheric zenith delays. For GNSS people these parameters are a real nuisance as we have to estimate at least 1 such parameter every 2 hours per station. With station networks sizes of 100 or more this amounts to at least 1200 parameters per day. However, as it turns out for weather prediction people these parameters are very interesting. The contain information regarding the humidity of the air through which the GNSS signals have travelled. And humidity is one of the not so well observed quantities in our Earth system and at the same time it represents a significant amount of energy. So if in GNSS analysis we model what atmosphere scientists call the "dry" or "hydrostatic" part of the Earth atmosphere properly our estimates give a significant amount of information about the "wet" part of the atmosphere which if included in the numerical prediction models does improve the quality of the numerical weather predictions.

So what used to be a nuisance for GNSS specialists has turned out to be a really valuable piece of information. Because besides being very useful for numerical weather predictions it is also of value for climate studies. However, for climate we need long time series so there we have to be a bit patient.