I am working on two large datasets with buidling height (300000 buildings) and the 500*500m buffer intersection (90000 buffer). I want to get the average height of buidlings in each buffer. Two approaches were formulated but not successful at all.
1. I pairwise intersect both layers, then dissolve the shapefile with the buffer no., but it returns 999999 error at intersection (which is probably due to such large size). This methods work fine in my 20*20m buffer. Are there any ways in handling that?
2. I try on spatial join on 500*500m buffer and its gives the result. However when checking the application using the 20*20m buffer, it did come up with slightly off results, with some large errors due to the Clementini rules (it counted buildings touched the outer boundary of buffers). Are there any suggestions on the minimal search distances difference between intersect and spaital analysis to get the exact same function as intersection tool above?
Thanks!
You may have better success with the source and destination data as featureclass in a locally stored file geodatabase. Shapefiles are not the way to go because of its limits on file size and the fact that buffers are not represented as true circles.
so try featureclasses first, then see what the results are