I’m using this model to understand how carbon storage has changed in Brazil. I’m only considering mangroves and only aboveground biomass with my field data. I’m using 4 LULC maps representing the landscape in 1986, 2000, 2005 and 2020.
After running the model on inVEST 3.9.0 I used Arcgis 10.5 to understand the total outputs. It seems it is working flawlessly nevertheless it appears that are some rasters missing: the raster with carbon emissions between the first period 1986-2000 and the net present value for 1986.
I also noticed that there might be some rasters with wrong information, the files with carbon accumulation (1986-2000; 2000-2005; 2005-2020 and 2020-2040), the first period (1986-2000) demonstrate accumulation for all mangrove areas but the next periods this carbon accumulation maps don’t show areas that keeping being mangroves over the years as the imagens attached on the google drive link bellow as mangrove accumulation areas.
I would like some help to understand it. Is it right or am I doing something wrong?
Thanks for posting, and that’s great to hear that the model is running well!
Although I agree that it’s a little confusing, this is actually the model working as it’s designed to. The model assumes that the period before the first transition event is accumulation-only (no emissions). Since NPV is calculated from emissions, it won’t be produced either until after the first transition.
You could work around this limitation by creating another raster for the year 1985 representing a baseline of sorts, and then have 1986 be a transition year. Then you’d end up with emissions being calculated between 1986 and 2000.
I also noticed that there might be some rasters with wrong information, the files with carbon accumulation (1986-2000; 2000-2005; 2005-2020 and 2020-2040), the first period (1986-2000) demonstrate accumulation for all mangrove areas but the next periods this carbon accumulation maps don’t show areas that keeping being mangroves over the years
Oh that’s interesting … so just to make sure I understand, you’re seeing accumulation between 1986-2000, but then no accumulation after that, even though there’s no change in landcover?
If that’s the case, could you share your inputs with me so I can try to take a look? Feel free to just link a file-sharing folder here, or else email them to me at jdouglass@stanford.edu. Could you include your logfile from your recent CBC model run as well?
Thanks for sharing your inputs and outputs @joaquimor !
The key issue here is in your transitions table. When I change the “transition” pair “mangue” → “mangue” from NCC to accum, we have more reasonable-looking outputs:
They key here is the use of NCC values in the transitions table. The model understands NCC to represent no changes in carbon whatsoever … neither accumulation nor emissions due to disturbance. So when the model saw the transition from mangue → mangue resulting in no carbon changes, it set it to 0, which is what you were seeing in the model outputs. The solution is to clarify that when we have mangue → mangue, we want carbon to accumulate.
Here’s the updated table for you to use: transition4.csv (726 Bytes)
There’s a related question here around what the model should be doing in this case … when a landcover class doesn’t change during a transition year, should the transition table be respected? We’ll have to give some thought to that, and I don’t have an immediate answer for you.
In any case, could you try running the model with the table I included above and make sure that you’re getting results that look reasonable to you? Please do let us know if they aren’t looking reasonable.
