Satellite altimetry is just a complicated sea level tide gauge as we need to know, among others, orbits and get auxiliary corrections for unwanted effects for instance by atmosphere. This satellite technology allows researchers to get sea level records regularly, globally, and in places where it is very hard to maintain or even acquire measurements. There are obviously some differences with in situ instrumentation. In particular, the fact that the altimeter-derived measurements are an average over a footprint and the temporal sampling is of order o days instead of minutes.

In this way, time series of sea level-changes can be constructed for a particular location along the satellite ground track, during the lifetime of the mission. A number of satellite radar altimeters have been launched over the last few decades (TOPEX/Poseidon, Jason-1, ERS-1 and 2, Envisat, Geosat Follow-On).

Google Earth represents a whole new way for scientists to connect with the world and making scientific information more accessible and useful.  With Google Earth, more people would have the chance to produce themselves time series of sea level as usually scientists do.

At present, the main problem with Google Earth is the missing of a data processing engine. Just to mention an example, it can visualize graphs generated off-line (e.g. from Matlab) but cannot produce graphs on-the-fly.   

Jon Blower and Alastair Gemmel from Environmental Systems Science Centre, University of Reading (UK) have done some work with producing graphs automatically in a web application that then you can link to these from Google Earth.  The web application contains the numerical data and produces a graph in response to a certain type of request.      

Here we provide a “simple scenario” consisting of a file containing sea level data and related times of collection and a method for generating a plot in Google Earth starting from those numerical data.  Essentially, the strategy is as follows: 

1) Generate a KML file that contains placemarks for all the locations (with their latitudes and longitudes) where we like to show sea level time series. 

2) For each placemark, the <description> field contains HTML that embeds an image, just like the "simple scenario" above.  However, in this case the image is dynamically generated on a web server by a CGI script, Java servlet or similar.  (one could also wrap the image-generating Matlab script as a CGI program.) 

3) When the user clicks on a Placemark in Google Earth, the image is loaded from the server and displayed.

Step 1 could be also refined by providing a web page that allows the user to select exactly which stations and time periods he/she is interested in.  The web page links to a CGI script or Java servlet that generates the KML dynamically.

Jon and Alastair have set up a basic example servlet. The servlet generates a KML file. When viewed in Google Earth, the KML file will display the location of the sea level data from the file as a push-pin.  Upon clicking on the push-pin, the pop-up balloon will appear with a PNG image of the sea level time series.  

The example is based on input from an ascii file with data in columns. The data file has a structure as follows:

[lon],[lat]

Year         Month       Day        Hour      Level

[year1]      [month1]    [day1]     [hour1]   [level1]

[year2]      [month2]    [day2]     [hour2]   [level2]

[year3]      [month3]    [day3]     [hour3]   [level3]

.

.

.

[yearN]      [monthN]    [dayN]     [hourN]   [levelN]

where N can be any number (as many rows of data as you like in each file) and

[lon] and [lat] are the longitude and latitude of the profile.  So the file should be like this:

The servlet is now running on a Tomcat Server handled by Pasquale Lazzareschi at CNR (Istituto di Informatica e Telematica). Pasquale re-worked the basic servlet to accept sea level data in float format and also modified the way to locate data files. 

Users with Google Earth software already installed in their computers can try the below examples by clicking on the “world” bullets to launch the geo-browser and fly to the location. Then clicking again on the placemarker the plot will be automatically generated. 

If not, you will need to install Google Earth first (available at http://earth.google.com).

 1. CASPIAN SEA 

Example of sea level data (raw) collected at Fort Schevchenko Station

2. MEDITERRANEAN SEA 

Example of sea level data (filtered) collected at Genoa Station

Example of sea level data (filtered) derived from a Jason-1 ground point near Genoa (satellite track 044)

Please note how the estimates from satellite altimetry well reproduce the seasonal signal of the sea level measured by in situ intrumentation.