I encountered an issue while running the Coastal Blue Carbon model. The spatial distribution of coastal habitat LULC in the input rasters for 2009 and 2024 is different due to transitions such as reclamation and other changes. However, when I ran the model, the output raster remained the same in distribution as the 2009 map, even though there were significant LULC changes by 2024.
How could this happen? And could you help me solve this issue?
Welcome to the forum @Raid!
Which output raster remains the same? Do they have exactly the same values for both 2009 and 2024, or just the same pattern, but with different values per pixel?
~ Stacie
The output displays the same spatial pattern as in 2009, but with different raster values. Could this be because the value I entered for soil-high-impact-disturb, following the guideline recommendation, was less than 1—thus allowing some carbon to remain stored in the sediment? However, my intention was to assume a 100% carbon loss due to reclamation for industrial development, such as the construction of jetties and shipyards within the observation area.
I have another question. Does the Coastal Blue Carbon model in InVEST not include an output report (HTML) like the one generated by the Carbon Storage and Sequestration model?
Hi @Raid -
Is “the output” you’re referring to carbon-stock-at-[year][Suffix].tif or a different model result? And how are the values different? Higher, lower, something else? Please be specific, else it’s hard to comment on the values you’re seeing, and how they may be impacted by disturbance.
No, Blue Carbon does not provide an HTML report. But you can generate those values easily by summing the model output results that you’re interested in within your area of interest.
~ Stacie
I input three LULC maps: 2009, 2024, and 2044 (the latter being a simulation result). However, the decrease in carbon storage only appears in the year 2044, even though in 2024 there is a loss of seagrass areas. Surprisingly, the carbon value in 2024 increased instead of decreasing.
Hi @Raid -
Assuming that you are talking about the carbon-stock-at-[year][Suffix].tif result, I think that the values you’re seeing are as expected. The User Guide is not clear enough about exactly how the emissions calculations are done, it says this:
Carbon emissions begin in a snapshot year where the landcover classification underlying grid cell transitions into a state of low-, med-, or high-impact disturbance. In subsequent years, emissions continue until either grid cell experiences another transition, or else the analysis year is reached.
and this:
We assume that landcover transitions happen instantaneously and completely in the first moment of the year in which the transition occurs.
This left me still wondering, if there is disturbance beginning in 2024, should that be reflected in the 2024 stock map? Or only reflected in later stock maps? So I consulted with our software team, who responded:
The model assumes that all changes in land cover happen instantaneously and completely, but we won’t actually see the changes reflected until the following timestep because carbon accumulates and/or emits over time. Let’s say that the 2024 map represents the new landscape state effective Jan 1, 2024. Then the accumulation and emissions would not actually be seen until the 2025 map (e.g. Jan 1, 2025), after we’ve had a year for carbon to accumulate or be emitted.
So one thing you can do is look at the total-carbon-stocks-[year].tif layers created in the Workspace/intermediate folder. You should see a change in carbon for the 2025 map, corresponding to the loss of seagrass area.
I will update the User Guide to be more clear about these points.
~ Stacie
“Thank you for your response. I now have a better understanding. Based on the research I am currently conducting, would it be possible for me to provide suggestions that could be forwarded to the developers to help improve the model in the future?”
Our developers monitor this forum, and we are very happy to hear suggestions from users about how to improve the models. So please post your suggestions here and we will make sure they get to the right people for consideration. Thanks @Raid.
~ Stacie