I am Hayk Igityan from National Academy of Sciences of the RA. I want to use InVEST program. I need to find 2050-2100 climate change impact on crop. Is it possible to solve this issue with InVEST crop Percentile or Regression models?
Great question. As far as I know, the crop production models now rely on crop maps known as the “Manfreda dataset” (1), which represent the (potential?) crop distributions at a static point in time. To model the impacts of climate change, one would need updated crop distribution maps, as well as landcover maps representing possible landcover in 2050 or 2100.
@jandrewjohnson do I have this somewhat correct?
- “Farming the planet: 2. Geographic distribution of crop areas, yields, physiological types, and net primary production in the year 2000.”” C. Monfreda N. Ramankutty, J. Foley. Biogeochemical Cycles, Vol.22, GB1022, doi:10.1029/2007GB002947. 2008)
A similar question here. If I understand correctly, the current regression model is rely on climate bin in 2000. Could I update the climate data under “InVEST_3.7.0_x86\CropProduction\model_data”, and then model the crop production in different years, for example, 2005, 2010, 2015.
also, I did not find any detailed information on how this model work. Does it follow exact the same methods in “Closing yield gaps through nutrient and water management”?
I notice there are several datasets under “InVEST_3.7.0_x86\CropProduction\model_data”, but not sure which are input the model. for example, there are two similar data: "extended_climate_bin_maps“, ”climate_bin_maps“， which is the input for the regression model? @dave
btw, do you find a way for your qs? @hayk
Hi @yingjie, sorry for the delay. It appears that the Crop Production models are only using the “extended_climate_bin_maps”, not the “climate_bin_maps”. But this is not simply climate data, rather I believe these data are the product of the methods in Monfreda paper referenced above.
I think the best reference for how this model works is probably the User’s Guide, if you haven’t already read it. http://releases.naturalcapitalproject.org/invest-userguide/latest/crop_production.html
I am working on the Crop Production( Percentile) model, but I have a problem and its error is related to input land cove and CSV table of the crop. the number of each crop in CSV is related to land use? I attach my error and I need to see a sample of input land use map.
I am working on crop production (percentile) model, but I have a value error that I attached it, could you please send me an example of land cover and CSV file related to crop production?
Hi @Adele -
We provide sample data for each model that you can use to help format your own inputs. When you installed InVEST, there is an option to also install the sample data for each model. Then you can find the sample data in the InVEST installation folder/CropProduction.
Also, the Data Needs section of the User Guide provides details on how the input CSV table must be formatted so its data can be joined to the land use/land cover map through the “lucode” field.
Hi @Adele -
Were you able to use the sample data and User Guide to help format your data and move past this error?
Unfortunately not yet, I am trying but I couldn’t yet
I run the model with the sample data but I have the same error with my data.
I am confused
@Adele, I see this error when the incorrect folder is provided to the
Directory to model data input. The correct folder within the InVEST sample data is
I am working with the crop production model(percentile) and I need to find the related equations and how can be calculated… Model mathematical equations. I couldn’t find them in papers. Please introduce me some related documents.
Thanks for all your help.
Hi @Adele -
The User Guide provides a description of how the percentile model works, which is pretty simple, mostly mapping crops to the percentile table provided with InVEST. The one thing it does not include enough detail on is step 3, “Calculate nutrient valuation per crop”. Is this the step that you would like equations for?
Otherwise, I’m not very familiar with this model, so will let other NatCappers respond further.