I am currently running the Coastal Blue Carbon model, and I am facing some issues with the output results. I have run the model twice from the beginning to ensure accuracy.
In the first run, I generated carbon stock output maps for the years 2015, 2020, 2025, and 2050. However, the carbon stock maps for 2015 and 2020 appear exactly the same, with no visible changes between them. Additionally, for the years 2020 and 2025, the “low” value is showing as -inf, and the entire output map appears completely black, with no clear visualization.
To resolve this, I re-ran the model again from the initial stage. However, the issue persists — all four maps are showing similar low-to-high ranges, and the maps look almost identical.
I kindly request your guidance on how to resolve this issue. Please let me know if there could be any problem with the input data, LULC layers, transition matrix, or parameter settings.
However, the carbon stock maps for 2015 and 2020 appear exactly the same, with no visible changes between them.
Have you opened the maps in a GIS and inspected the values to confirm they are identical? Or better, computed a difference map to check if they are actually the same or just similar?
Additionally, for the years 2020 and 2025, the “low” value is showing as -inf, and the entire output map appears completely black, with no clear visualization.
Which version of InVEST are you using? This could certainly be a bug in how NoData is calculated. If possible, could you share your data?
However, the issue persists — all four maps are showing similar low-to-high ranges, and the maps look almost identical.
What are you expecting to see instead? For example, if there are no LULC changes (post-alignment), we wouldn’t expect to see a difference in the carbon maps.
But there is still one issue. The legend on the side shows low to high values, and it indicates that carbon stock is increasing over time. However, the images for all the years still look the same. The baseline map from 2015 is still being displayed, and there is no visible change in the image.
First, have you inspected the values in the rasters (or created a difference map)? Symbologies can hide differences in rasters. If the raster values are close and the GIS is stretching with the same minimum and maximum values, the maps could look the same, even if they’re not.
I would also recommend you take a look at the User’s Guide, which describes the model and outputs in detail: Coastal Blue Carbon — InVEST® documentation . It provides some useful info which could help diagnose your issue, e.g.,
The carbon stocks for year t represent the carbon stocks at the very beginning of year t…In practice, if accumulation or emission happens, for example, in the year 2030, they will start appearing in the carbon stock output in year 2031, not 2030, after there has been a year for carbon to be accumulated or emitted.
So in your case, if the first “real” change occurs in 2020, it wouldn’t show up until the next year.
You might take a look at total-net-carbon-sequestion-between-[year]-and-[year][Suffix].tif and if this is 0, the stocks will match.
In addition, please look at your transitions table and ensure this looks good (see: Coastal Blue Carbon — InVEST® documentation ). Are the blue carbon to blue carbon transitions set to accum? If common transitions are set to NCC, you will not see accumulation and the stocks will look unchanged.
If these points do not provide clarity on why you’re seeing the results you’re seeing, again, please share your logfile.
This is a bit unclear - do the carbon maps look the same or do they look different? Or are you referring to different maps in these two sections? Are you expecting the carbon stock to decrease over time? Are you using the default symbology or have you tried to display the maps on the same scale (maybe with more limited ranges than default) to better show any differences? It would be helpful if you could share (at minimum) images of your carbon maps and detail what you are expecting to see. It would be even better if you could please share your datastack.