Coastal blue carbon model error


I’m trying to run the coastal blue carbon model (dev build that you shared in a previous thread) and am getting an error when running the main model. The preprocessor runs without an error. The log makes me think that there’s something wrong with one of my csv inputs, but I’ve looked at them all and don’t see anything different from previous runs that have worked.

Log and csvs are attached; let me know if you need any additional information.

Thank you!

carbon_pool_initial_template_upd.csv (2.6 KB)
carbon_pool_transient_template_upd.csv (3.6 KB)
transitions_upd.csv (7.3 KB)
LULCLookup_Master.csv (2.2 KB)
InVEST-Coastal-Blue-Carbon-log-2020-07-03–16_15_21.txt (3.2 KB)

Hi @kwarnell,

Could you post a link to the other thread where a dev build was shared? I’d like to make sure I’m working off the save build of InVEST.

Would it also be possible to share the inputs used in the preprocessor step? Sorry if you’ve already shared these before.


Hi @dcdenu4,

Here’s the thread with the dev build: Coastal blue carbon output questions

Here are the input rasters used for the preprocessor: (1.4 MB) (2.9 MB)


Hi Kate,

Sorry for the frustration here! The Coastal Blue Carbon model expects all of the landcover code labels to be unique, and one of the utilities we use for reading tables assigns .n suffixes to duplicate field values that it finds. So within the transition table alone, there a bunch of unique fieldnames including "accreting marsh high salinity", "accreting marsh low salinity", "accreting marsh medium salinity", and "background" (which is the KeyError you had previously mentioned). The LULC class names are used to index into the other tables, which is why they need to be unique.

I’ve made a note to improve how we handle duplicate labels in the model so that this will be much clearer in the future.

Let us know if this doesn’t solve the issue for you!

1 Like

Hello, is it possible to have the ‘‘snapshot table’’ file? Thanks!

@matteoconv the snapshot table is just a list of paths to the input rasters (as described in the users’ guide) - you’d have to update my version to reflect your paths so it’s probably not worth the time for me to dig it up. If you’re looking to run the model with pre-existing data as a test, I’d recommend using the sample data that is downloaded with the InVEST model.


This topic was automatically closed after 20 hours. New replies are no longer allowed.