Wind velocity warning + SLR + projections

Hi I am using the coastal vulnerability model but for every run I get: “WARNING Found wind velocity of 0.00, using 1m/s in wave height calculation instead” or 0.92. I was wondering what this means, if it’s a problem, and how to resolve it.

I am also trying to create SLR scenarios for the year 2050 and 2100 for the whole area, without variations in the area. The SLR intermediate results are all the same (0.575) but do generate some differences in the exposure index. Why is this and is there a way around this other than to calculate the exposure myself?

Lastly, some of my projections are ESPG:4326 - WGS84 and others ESPG: 32621 - UTM 21n. Is this a problem? Many thanks!

InVEST-natcap.invest.coastal_vulnerability-log-2023-08-18–16_42_41.txt (162.8 KB)

Hi @EmilievdHM , thanks for your questions,

The model uses wind velocities in an equation that calculates wave heights. This equation only makes sense for wind velocities >= 1 meters/second. The warnings suggest that many velocities < 1 were found in the WaveWatch3 dataset. I would not be concerned by these, except that there are quite a lot of them. I would check one of the intermediate outputs:
intermediate/wind_wave/wave_energies.gpkg. The maxH_local and minH_local fields describe the max and min wave heights that were calculated for each point. If these heights look reasonable, then there’s probably nothing to worry about.

If SLR does not vary across space within your area, then it may not make sense to use this parameter. As with all the components of the exposure index, SLR data is converted to a rank to describe the exposure that each shore point experiences, relative to the other shore points. So that’s why all the intermediate values for SLR are the same at each point.

That’s okay so long as you are okay with the assumptions the model will make for you. These are captured by log messages like,
C:\Users\emili\OneDrive\Documents\Imperial\InVEST\CoastalVulnerability\Guyana data\guy_ppp_2020.tif is unprojected and will be warped to match the AOI and resampled to a pixel size of 250.0 meters

If you prefer the raster data inputs not be warped & resampled with these assumptions, you could project them yourself ahead of time.


This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.