We are starting a soil sampling program on the grounds of large solar farms. How can we create files for 2023-2025 (duration of our project) that would allow comparisons with the currently available files (which seem that cover up to 2019)?
Thanks
Vasilis Fthenakis, Columbia U.
Hi @Vasilisf -
I could use some clarification about what you’re trying to model. Are you only modeling soil carbon, or other carbon pools as well? What do the “currently available files” include? Are you trying to add solar farms data to a larger land cover map that includes other classes, or only considering the solar farms alone?
When talking about the Carbon model, I always note that it is extremely simple, and it is easy to do the same calculations in a GIS. So depending on what exactly you need to do, it might be more flexible to do the calculations manually.
~ Stacie
1 Like
Thank you Stacie for your reply. By “currently available files” I mean the data in Invest that to my understanding cover only up to 2019. The idea is to sample the soil in solar farms that have been operating for some years and compare such future time data with those in Invest for a year before the solar farm installation.
We will sample soil carbon only but we also want to model the change of vegetation.
We are trying to understand in what depth should the sampling be done. Since we would be looking for changes during time horizons of only a few years (e.g., 2-10 yrs) would sampling from up to 30 cm suffice or should be different depth for different soils.
Does it make sense?
Vasilis
Hi Vasilis -
If you’re referring to the sample data that is provided for the carbon model, it is specific to the Willamette Valley in Oregon, United States, and none of the data sources were documented (the sample data for that model is really old) so we do not recommend using any of the values for a real-world study. Sorry to say that you’ll need to do a literature search or something to come up with vegetation biomass values.
As for the soil sampling depth, I don’t have any experience with that, other than to say that it doesn’t matter to the model.
~ Stacie
Hi Stacie,
I see Soil Survey data for every county in the US in the Natural Resource Conservation Service website ( websoli.survey.nrcs.usdata ) with data downloadable in Microsoft access. Do you know if such data can be introduced in INVEST? Also can you give me contact info with the Invest author to see if we can develop a project together?
Thanks
Hi @Vasilisf -
I have used the NRCS data, but not for carbon. Here’s what I say about it in the User Guide for other models that have a more explicit soil data requirement:
In the United States free soil data is available from the U.S. Department of Agriculture’s NRCS gSSURGO, SSURGO and gNATSGO databases: https://www.nrcs.usda.gov/wps/portal/nrcs/main/soils/survey/geo/. They also provide ArcGIS tools (Soil Data Viewer for SSURGO and Soil Data Development Toolbox for gNATSGO) that help with processing these databases into spatial data that can be used by the model. The Soil Data Development Toolbox is easiest to use, and highly recommended if you use ArcGIS and need to process U.S. soil data.
If you don’t use ArcGIS, the database is much harder to use, since you have to process it manually, and I don’t think I’ve ever had to do that. Still, it should allow you to produce a table that maps your soil property of interest to the shapefile of soil polygons that they provide.
There isn’t a single InVEST author, we’re a small team of scientists, software engineers, data analysts and others who create the models, and often use the models in our engagements around the world. So who to talk with would depend on the nature of the project you have in mind. And I’ll be honest in saying that we are all generally very busy and unfortunately pricey to work with, which constrains the projects that we take on. If you do want to propose a project, I’ll recommend writing it up in a bit of detail, and sending it to naturalcapitalproject@stanford.edu, where it can be considered by the people who handle such things (which is not me 
).
~ Stacie
1 Like
Hi Stacey -thank you very much for your incredibly fast response and I will appreciate it if you verify or correct my understanding from our communication that I summarize below:.
In a forward looking study that will only quantify changes in soil carbon from sampling data, perhaps INVEST is not the right tool, and it may be easier to introduce and analyze new soil carbon data in the NRCS Soil Data Development Toolbox as we are comfortable with ArcGIS.
Thanks again
Yes, if you’re only looking at soil carbon, especially from sampling data, our model will not be useful. All it really does is add up the different carbon pools that you provide and map that to a land cover map. There are no dynamics or complexities related to growing or dying vegetation, soil chemistry, etc.
Great that you can use the Arc tool, it’s extremely useful, and I really wish we had a similar tool for global soil data.
~ Stacie
1 Like
Hi Stacie,
we communicated a year ago related to the applicability of Invest to estimating the impact of Land Use Change to Soil Carbon in Solar Farms, and I am now going a full circle back as I realized that detailed Soil Carbon models like CCLUB/GREET and DayCent (both based on CENTURY) do not provide the required geographical resolution, but Invest can do it with it mapping tool.
I understand that the Soil Carbon model in Invest is a simple one but I have not found any equations describing the model. I am wondering if Invest does more than just looking at the difference between Carbon Data assigned to different years.
If not could you please guide me to where the Soil Carbon model equations are listed
Thank you very much
Vasilis Fthenakis email: vmf5@columbia.edu
Founding Director, Center for Life-Cycle Analysis
Earth & Environmental Engineering
& Electrical Engineering
Columbia University
Hi @Vasilisf ,
In terms of sequestration, the InVEST Carbon model simply takes the difference in densities for any pixels that change between scenarios, and spatially maps those differences. There is nothing more sophisticated to it.
You can read details in the User Guide,
“the model assumes that none of the LULC types in the landscape are gaining or losing carbon over time. Instead it is assumed that all LULC types are at some fixed storage level equal to the average of measured storage levels within that LULC type. Under this assumption, the only changes in carbon storage over time are due to changes from one LULC type to another.”
-Jesse