inmap source receptor matrices gridding

[Anil Gogebakan]

Hello everyone,

I have been trying to figure out why we ended up with 52412 (52411+1) geographic cells (I assume they are the product of the gridding while simulating?) for the data provided in InMAP Source-Receptor Matrix (ISRM) dataset (zenodo.org). I am using s3fs (amazon s3) along with a jupyter notebook to explore what is in this file. When I check the dimensions of the columns stored in there I see (596444,) for the most columns, and (3,) for layer and layers, for (3, 52411, 52411) pNH4 PNO3 PSO4 etc. I have no clue why these are the resulting dimensions. I suspect the column called AOrgPartitioning is responsible for these dimensions but again I have no clue what this column contains (I can see numbers, but I don't know the meaning). Can someone help me understand what this "mystery" (mystery to me) file is exactly providing and how the simulations have been run to obtain that sr matrices? I couldn't find much detailed explanation/documentation explaining what these dimensions are or how this file has been created. Also, can we create our own column inside this sr matrices (I believe it is not permitted), it would be great if there is a workaround so I can use my own new column. Any guidance on this would be greatly appreciated.

Ps. I would guess the grid number would be around 3,000 if we were thinking of US counties and I am guessing the grid number would be around 220,000 if we were thinking of block group levels. But when it is 52,000, I cannot think of a geographic level that would give this number for US territory.

Thanks in advance!

Anil Gogebakan