1

Is it possible to interpolate the specific value at a point using the Kriging tool in the spatial analyst toolbox? If not, then, how can this be done in Arcgis?

To my knowledge, Arcgis only produces a raster. This isn't applicable for me because I need to interpolate something quickly and it has to be at a specific point (geographic coordinates). I'm trying to interpolate weather data (temperature, humidity etc).

Improve this question
3
  • Yes? I think we'll need a little more detail on what data you have and exactly what you're trying to do to provide any meaningful answers.
    – Chris W
    Mar 13, 2015 at 19:33
  • Hi Chris, thanks for the reply. I'm trying to interpolate weather data (temperature,humidity etc) at a specific point using Kriging. To my knowledge, Arcgis only produces a raster. This isn't applicable for me because I need to intepolate something quickly and it has to be at a specific point (geographic co-ordinate).
    – user1361488
    Mar 14, 2015 at 19:44
  • Most interpolation methods I'm aware of (which isn't many) have software implementations that produce rasters (well, besides TINs). Regardless of whether or not one interpolation method is more appropriate for your use than another (we have several questions on this topic), you can always use the point you have to extract the value from the resulting interpolation raster. Otherwise maybe you're looking at a pure statistics/math solution rather than a GIS one.
    – Chris W
    Mar 15, 2015 at 20:14

1 Answer 1

Reset to default
2

In case someone is interesting to accomplish this task, without needing it to be in ArcGIS for Desktop.

In R, there is the automap package, function autoKrige. The new_data argument, which indicates where the interpolation will happen, accepts points (besides grid and polygons).

The manual says:

new_data: a sp object containing the prediction locations, new_data can be points set, a grid or polygon. Must not contain NA's. If this object is not provided a default is calculated. This is done by taking the convex hull of input_data argument and placing around 5000 gridcells in that convex hull.

If there are multiple points of interest, the interpolation can be done in just one step.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.