Basic steps for processing source apportionment 

1) Use combine and hr2day to make annual (seasonal) average files of bulk + tags

2) Regrid to the common grid for PCAPS/SR matrix which is 12US2 for now (matchy)
	Puts -9 in cells without information so places off the domain do not get 0's in the final normalized
	surface average

3) generate individual patterns in text format with emissions and stack height info (peakdist.pcaps)
	No limitations on area to write out to the text file - write out entire domain

4) make a single file with all the individual extracted patterns (mrgpattern)

 	This is important because it creates a group of files that will become the underlying database input to PCAPS

5) Make location specific normalized surfaces for different precursors/stack height combinations (normpattern)

 normpattern does not need to move the gradient to a generic box around the source 
 PCAPS needs to be changed to know that all patterns on 12US2 domain and do the needed remapping internally

6) run PCAPS


