I'm working with county level data where I'm using ordinary kriging in QGIS 2.14.2 Essen (SAGA plug-in for reference) to characterize a point data in each county. The problem is that I don't want data points from another county influencing calculations near the border of an individual county.

The simplest but brute force way would be to separate my point data layer (x and y coordinates, response variable, and county name) by county and then, run ordinary kriging as a batch process for each county, by clipping each raster by the county polygon layer, and merging each county raster into one overall raster file. This seems like a lot of manual work each time I might run this data or add new counties though.

In the parameters for the Ordinary Kriging function, I can specify a maximum search distance for points to use in the calculation. What would be really nice would be to add on another condition that not only would each point used need to be < x distance away, but also fall within the the same county polygon. Is there a simple way to functionally do this, or would that come down to modifying the Ordinary Kriging function itself?

I haven't worked with Python yet (hence checking here for other potential options before I dig into it), but the other option I'm considering is to look into setting up a do loop for the list of counties using the work flow I mentioned in my second paragraph. Is this latter option the simplest way to move forward at this point, or is there some way to add this additional condition?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.