Subject: Validity of using historical climate data (WorldClim 1970-2000) with a 2025 LULC map

Science, Research & Data
, ,
1

Hello everyone,

I would like to raise a question regarding temporal consistency in the application of the InVEST Annual Water Yield (AWY), Sediment Delivery Ratio (SDR), and Nutrient Delivery Ratio (NDR) models.

I am currently conducting a study aimed at mapping the spatial distribution of ecosystem services related to water yield, soil conservation, and water purification, using the AWY, SDR, and NDR modules of InVEST.

My concern relates to the climatic input data. Specifically, I am considering using datasets available through the IEEM platform (IDB/IBM), which provide long-term climatic averages derived from WorldClim 2.1 (1970–2000). This choice is mainly driven by the difficulty of obtaining reliable, locally calibrated datasets for key variables such as reference evapotranspiration (ET₀).

However, the land use/land cover (LULC) map used in my analysis represents a baseline scenario for the year 2025.

Given this situation, my question is:
Is it methodologically acceptable to combine a recent LULC scenario (2025) with historical multiyear average climate data (1970–2000) when applying the AWY, NDR, and SDR models in InVEST?
I would greatly appreciate any guidance or insights on how to handle this temporal mismatch in a way that is consistent with the assumptions of the InVEST framework.

Thank you very much in advance.

2

Is it methodologically acceptable to combine a recent LULC scenario (2025) with historical multiyear average climate data (1970–2000) when applying the AWY, NDR, and SDR models in InVEST?

Yes, this is what we often do. While we strive to get the “best available” data, practically it doesn’t always (in fact, often doesn’t) line up exactly. This is just the reality of the data available to work with. So we are very transparent in citing our sources, and the limitations of the data.

~ Stacie

1 Like
3

Thank you very much, Swolny. It is very reassuring to know that this is a common practice, especially since I have been quite concerned about this aspect of my master’s thesis. I will continue making my best effort to obtain the most locally representative datasets possible, and when this is not feasible, I will clearly document and justify the use of these available data sources.

Thank you again for your guidance and reassurance — it is truly appreciated.

4

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.