Raster resampling in QGIS

[Fabiola Rodriguez]

Hello all,

I am resampling some environmental layers and my question is about the resampling method. The methods available to resample rasters (Im doing this via Raster-Projection-Warp) in QGIS include: nearest-neighbor, bilinear interpolation, cubic, cubic spline and Lanczos. I've found in this group that nearest neighbor to cubic could work for the environmental layers. I want to resample a land cover raster to match the 1km resolution of the environmental layers and nearest neigbor, bilinear and cubir seem to "dismiss" many of the grid cells that have little percentage of any given category. The cubic spline seems to make it a smoother final layer and includes these fragments. However in other help groups they recommend not only basing these methods on "eye-evaluations". 

Does anybody have experience with these methods in QGIS and/or know of articles relating to SDM where these methods are assessed. I found that: Phillips et al. 2006. Maximum entropy modelling of species geographic distributions. Ecological Modelling (190): 231-259.  resampled an IPCC layer using bilinear interpolation because :"...it is likely more realistic than simply using nearest neighbor...". So I just want to obtain more insight if this would be an appropriate approach for me to select a method based on what I see and my experience on-site where if I use nearest neighbor or the others I mentioned I believe it is losing potential fragments of category data.

Thanks in advance for any kind of comment.

Fabiola R.

[Dimitris Poursanidis]

There is no "appropriate" method.

Try all and see which is closer to the reality. Eye evaluation is the only way. Or if you have a similar dataset in the resolution you need (like in case of Landcover, a dataset from MODIS) and check how accurate the results of the interpolation are.

Best

Dimitris

--
You received this message because you are subscribed to the Google Groups "Maxent" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maxent+un...@googlegroups.com.
To post to this group, send email to max...@googlegroups.com.
Visit this group at http://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.

--

Dimitris Poursanidis

PhDc. Marine Ecology

My pubs @ Google Scholar

http://www.iacm.forth.gr/index.php?l=en

terraSolutions

gr.linkedin.com/in/dimitrispoursanidis