JGCRI / gcam-core

GCAM -- The Global Change Analysis Model
http://jgcri.github.io/gcam-doc/
Other
296 stars 172 forks source link

calibrate the land area #474

Open mxq517 opened 2 months ago

mxq517 commented 2 months ago

I am trying to calibrate the land area types for the base year 2015 for China in the GCAM model. Where can I modify the total land area data for 2015? In the land allocation, I only found the data for the year 1975 under land_input_1:

        <landAllocation fillout="1" year="1975">9293.14</landAllocation>

When I modify this data to reflect the 2015 land area for China, the model throws an error, stating that the land area for the USA and other countries exceeds the allocated area, causing the model to stop running. Additionally, when I only change the land area for different land types in 2015, the model reports an error saying that the area exceeds the allocated land for China. How should I properly calibrate the land area for China in 2015?

realxinzhao commented 2 months ago

Hi @mxq517, interesting that the error was for the USA even though you changed the total area for China.

But note that the total area should be equal to the sum of the land allocation to all land types across all sub-regions (basins), so only changing the total led to an imbalance.

The initial area calculation/allocation in gcamdata was probably somewhat complicated, but that's probably the place you want to check. E.g., that number in XML came from L125.LC_bm2_R which is the output of module_aglu_L125.LC_tot. And L125.LC_bm2_R was an aggregation of all land across all basins. You can probably start there by modifying areas for specific land types, but note that there might have been more adjustments before that (depending on land type). Alternatively, you might start from the source data located in extdata/aglu/LDS. They were mostly data from HYDE (certainly possible you have better sources for China) and aggregated by the package moirai.