Hallo Assami,

In ArcGIS you have a stack of filter technicks (e.g. Spatial Analyst). With focal statistics you can even build a entire clean-up chain (using Model builder or python scripts). Which you'll take depends on the type and size of the anomaly. What is it that you want to clean-up?

My experience is that ASTER GDEM v2 is much better than the v1. But, of course, you still can have some artifacts.

Best regards,

Jewgenij

Hella Assami

You can use GRASS software with function r.neighbors (http://grass.osgeo.org/grass64/manuals/r.neighbors.html). Their is a lot of filter useful to smooth topography depending your goal.

Some tools to analyse depression in hydrological analysis (http://grass.osgeo.org/grass65/manuals/r.watershed.html) are also useful.

Best regards

Frédéric

Hi,

I have the same problem. I am using ASTER GDEM data from the Lesser Antilles region. Unfortunately there are spikes over the islands which can be off by over 1000 m. I wonder what these are caused by, perhaps cloud cover? I am trying to use grdfiter (a GMT routine) to remove them, but it looks like it might not be easy. If there where only a few isolated spikes a median filter might work, but in my case the are clusters of spikes. I will see if it works.

Cheers

Michele

Hi again,

After all I ended up using the SRTM90 dem instead. It has a lower resolution (~90 m instead of 30 m), but is free of artifacts, at least in my region of interest. I didn't need the higher resolution so it works better for me. The SRTM90 grid can be downloaded here:

http://srtm.csi.cgiar.org

The ASTER GDEM is still quite raw but it will certainly be improved in the future. For the meantime I suggest using SRTM90 instead, unless you need the higher resolution and have the time and resources to check and correct artifacts.

Cheers

Michele

Hi!

We found that SRTM performed better than GDEM2 for tropical forest area. The resolution can be increased (e.g. 1m, 2m and 5m) by resampling technique while maintaining the accuracy (SD=9.0m-10.4m; RMSE=9.3m-10.6m).

Conference Paper Accuracy assessment of global topographic data (SRTM & ASTER...