Child pages
  • Home
Skip to end of metadata
Go to start of metadata

NCK summerschool case Fine Sediment Wadden Sea

Wadden Sea SPM datasets

You can download a cache of these data with the Windows wget equivalent BITSadmin (which will be phased out after Windows 7). This is useful in palces where internet access is low. In this example I download 2 vaklodingen netCDF files.

set from=
set into=         d:\\thredds\dodsC\opendap\rijkswaterstaat\vaklodingen\

bitsadmin /transfer Job001 /download /priority normal

bitsadmin /transfer Job001/download /priority normal

Creating Google Earth polygons

In Google earth you can easily draw polygons to delineate characteristic features, such a mud patches or mussel beds or gullies. For the Dutch Wadden Sea turn on Historical imagery in the Google Earth View menu, to get the optimally bright original images, and not the darker blended ones.

Figure: how to start drawing a polygon

After selecting this button a pop-up will appear and you can start drawing a polygon. Left mouse clicking will either the active point (blue), generate a new point (green and then blue). The position of the new point is always next to the active blue point, but be aware of the sense of rotation of the entire polygon (clockwise or counterclockwise) that determines at which side of the blue point it will appear. The second-lat point (red point next to the blue point) can be deleted by right-mouse clicking, also this depend on the sense of orientation of the entire polygon.

You cannot pa or zoom with the mouse as long as you draw a polygon. Use the arrow keys on the key board to pan, and use page-up/page-down to zoom.

You can also click ok and stop editing the polygon to go back to mouse pan/zoom. You can resume editing the polygon afterwards. To extend or modify an exiting polygon, right-click in the menu on the left and the polygon-data points will appear again.

Figure: how to edit a polygon

Figure: how to continue editing a polygon

After saving this polygon to kml, you can read it into a NaN-seperated polygon into Matlab with lat,lon = KML2Coordinates('doc.kml'). You can split the NaN-seperated polygon into individual patches, with {lon,lat = poly_split(lon,lat)}} and calculate their area (sign gives sense of rotation), with {{for i=1:length(lon);area(i)= poly_area(x{i},y{i});end
}} after coordinate conversion to local system with convertCoordinates. the fill code then becomes

kmlname = 'doc.kml';
[,p.lon] = KML2Coordinates(kmlname);
[p.x,p.y] = convertCoordinates(p.lon,,'CS1.code',4326,'CS2.code',28992);
[p.lon,] = poly_split(p.lon,;
[p.x  ,p.y  ] = poly_split(p.x  ,p.y  );
for i=1:length(x)
p.area(i)= poly_area(p.x{i},p.y{i});
p.lonmean(i)= mean(p.lon{i});
p.latmean(i)= mean({i});
'CBcolorTitle','area [m2]')

This is the kml we collectively made with 4 students and 1 lectures during a 1.5 hour getting-to-known-the-google--earth waddensea session: crowdsourcing. You can process it with this script. We learned that individual interpretation of what us mud and what are mussels vary quite a lot. This means that automated image processing procedures will be difficult: how to program into a machine something that we humans can't agree on?

Recently Updated

Navigate space
  • No labels
Write a comment…