Urban Cooling: minimal cooling under urban greening scenario

Science, Research & Data
1

Hi all,

I am running the InVEST Urban Cooling model to explore the urban heat island effect in a municipality under different urban greening scenarios. I am running four different scenarios, each representing a different % canopy cover across the municipality: 22% canopy cover, 33%, 38%, and 44%. The 38% scenario is the “baseline” (i.e., the municipality currently has 38% canopy cover). The way I simulated increased canopy cover in each scenario is by adjusting the shade proportion in the biophysical parameter table for specific LULC classes (focusing mostly on developed classes). I also altered some of the Kc and albedo values to account for changing impervious surface cover values with reduced or increased canopy cover. However, when I run the models, I am getting very minimal cooling with increased canopy cover. For example, an increase in canopy cover from 22% to 38% only results in about 0.1 degrees cooling (both across the urban core as a whole, and within each developed land use class). And again, an increase in canopy cover from 38% to 44% only results in about 0.1 degrees of cooling. This seems remarkably low, especially considering the large increases in canopy cover that I am simulating. I am wondering if anyone else has run similar scenarios with InVEST and what their results have been, or if anyone has any suggestions on better ways to simulate urban greening using the InVEST tool.

Thank you!!

EDIT: to clarify, I have only run daytime temperature scenarios (working on nighttime temperature input data at the moment)

2

Hi @CharG ,

Thanks for writing in to the Forum with your question about comparing the results of different scenarios using InVEST’s Urban Cooling model.

To fully assess what’s happening, it would be best if you could share your input data by giving us access to them on a shared cloud space, such as Google Drive or Dropbox. Please let us know if and when you’re able to do so.

I’m curious how your other inputs are set. Did you change the weighting of the factors from their default values at all? Please share an example of a log file (.txt) from your runs.

Typically, we recommend creating scenarios via inputting different LULC rasters representing the potential changes to the landscape. But your approach of altering parameters in the biophysical table could certainly be valid as well. Again, we’d likely need to see and touch your data to know for sure.

Lastly, a change of 0.1 °C could be significant and reasonable given the % changes in canopy cover you describe.

Thanks,
Jesse