Habitat Quality Model Best Practices for Threats

Dear Natural Capital Project Forum,

I am trying to identify the best strategy for incorporating variable hydrology into the InVEST Habitat Quality model. I have used the model before, but am working on a different group of organisms that require more complexity.

For the new model, we have 6 different levels of surface water frequency estimates for our landscape and would like to use this information to incorporate one or more hydroperiod related threats. As I understand it, we have three strategies to accomplish this:

  1. Build one raster layer with the water frequencies as the value for each grid cell, which becomes the only hydroperiod threat raster layer. Here I worry that the units of this raster layer will not align with other threat layers (e.g., presence/absence of roads). I also don’t completely understand where this threat raster cell value is incorporated into the habitat quality metric. The benefit of this approach is that there is only one hydroperiod threat layer and therefore it does not dilute the impact of other threats.
  2. Incorporate each of the six surface water frequencies as separate threat layers with different sensitivities based on their anticipated impact. Here I worry that having six threat layers just to deal with hydroperiod will dilute the impact of other threat layers. When I say dilution, I’m referencing the calculation of habitat quality that incorporates a multiplier based on the weight of the specific threat relative to the total threat weights in the model.
  3. Incorporate each of the six surface water frequencies as a separate threat layer with the same sensitivities but different threat weights. As with the above option, I worry that having six threat layers just to deal with hydroperiod will potentially dilute the impact of other threat layers.

Any thoughts on best practices for the InVEST Habitat Quality model?

I think this is a valid concern because I read in the User Guide: “All threats should be measured in the same scale and units (i.e., all measured in density terms or all measured in presence/absence terms) and not some combination of metrics.” Where the water frequency metric you describe would be considered a “density” metric in this context.

That said, this still sounds like a more intuitive approach than the 2nd and 3rd options that would treat each surface water layer as a different threat. Is it reasonable to treat your other threats as density metrics rather than simple presence/absence?

Alternatively, have you considered a method like the Habitat Risk Assessment model? Habitat Risk Assessment — InVEST® documentation

It probably requires quite a bit more work to parameterize that model, but it might be better suited to this use-case.

1 Like