AnsweredAssumed Answered

Spatial disaggregation for large datasets

Question asked by isaacdu84 on Mar 7, 2016

I am working on a project which requires the spatial disaggregation of county-level air pollutants based on the population distribution in ArcGIS. I have the population distribution in 2.5' x 2.5' grid cells and road as vector shapefiles. What I did was to

1. divide each county into 0.25 x 0.25 degree grid

2. use "intersect" tool to generate features that contain the attributes of both counties and spatial surrogates(population)

3. summarize population in each 0.25 x 0.25 degree grid cell. The output are in the table format

4. joint the table to the grid cell so that each grid cell has the population in it

5. Here is the tricky part: I went into the attribute table and use "statistics" function by right-click on the population field and obtain the total number of population in the county since I do not know how to automatically get them.

           The principle of the spatial disaggregation is:   air pollutant concentration in one cell = county-level air pollutant concentration x (population in the grid cell/population of the county)

Then, I created another field and used the formula above to calculate the air pollutant concentration in each grid cell.

The problem is that I have 70 pollutants, 12 months, weekdays/weekends and more than 2000 counties. It seems impossible to do the spatial disaggregation using the method I described above. I am wondering whether there might be a easier way or some way to automate the process´╝č


Thanks in advance!