I am doing a project at work looking at rooftop solar resources. I have a set of building footprints for a few small cities, which "become" a set of raster building footprints when I clip our solar isolation raster to the building footprints.

I have completed the whole process, but now my supervisor would like me to go back and find the top 10 buildings (rooftops) in terms of their potential solar resources for each city. The problem is that the raster footprints, obviously, do not have attribute tables and (despite being clipped to the shape of the footprints) are just a collection of individual data points.

I need to somehow aggregate/match the 350,000 unique raster data points (yes, I converted the raster to integer and created an attribute table) back to their respective building footprints, so I can then find the top 10 buildings in terms of their solar resources.

Anyone have an idea how to go about this?


It would be best if you imported your buildings into a personal or file geodatabase if they are not already so. This is to make the OID (Object ID) of each feature static; unlike shapefiles geodatabase features unique ids' don't change, so if you are using a shapefile and perform any edits the linking IDs will no longer be valid. Repair geometry is also a good thing to do as you buildings might contain some geometry errors that will affect the next step.

As you've performed an Extract by Mask I have assumed you have the spatial analyst extension available.

Using the spatial analyst tool Zonal Statistics as Table you can create a table with the SUM or AVERAGE of your solar reflectance raster for each building polygon. Then you can either add a field to contain your reflectance values and join by attributes followed by calculate field or using the tool Join Field bring then statistics value table into your building polygons, removing unnecessary fields for the sake of neatness.


Your Answer

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

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