Error: No wavewatch data was found near the aoi (Coastal Vulnerability model)

Dear Team,
While running the coastal vulnerability model for the coastal states of India individually, I encountered an error stating No wavewatch data was found near the aoi. I have been using the global wavewatch sample data provided by InVEST.
Is there anyway to tackle the issue? Also I would appreciate if any suggestions/recommendations for wind and wave data (point shapefile ready to use in the model as I have no experience of working with wind and wave data nor such knowledge of generating my own data set.

I really appreciate any help you can provide.

Hello @pravinyab,

One other thing I noticed just from your screenshot is that it looks like you’re using an older version of InVEST. We always recommend using the latest version of InVEST available whenever possible, as we are constantly updating and improving these models to improve their stability and performance, and it is possible that updating to the latest version of InVEST might address this issue.

If updating to the most recent version of InVEST doesn’t fix the issue, it’s possible that an update to your AOI might address it. The InVEST Coastal Vulnerability model looks for points in the wavewatch 3 sample dataset that are within approximately 3 degrees of your area of interest. So, you should be able to work around this by extending your AOI out towards the nearest WW3 points in the dataset. Could you try this and let us know how it goes?

James

There looks to be good coverage of WW3 data all around the coastline of India. This makes me wonder if there is an incorrectly defined coordinate system, or some other issue with the AOI or landmass inputs. Despite the wording in the error message, extending the AOI further offshore should not make a difference because the model actually searches in proximity to the shore points, not the AOI boundaries.

Updating invest is a great first step, though. @pravinyab please also upload your entire logfile, it has other useful details in it. Thank you,

Thank you Dave for the reply.
I have been using a aoi shapefile with meters as units but all other inputs are in degrees (raster and vector). It ran successfully for Goa; however, when I run the same for other states it failed due to no wind data near the aoi. Attaching log file for Goa (ran successfully) for your reference.
InVEST-Coastal-Vulnerability-log-2024-07-10–15_43_56.txt (353.5 KB)

Also as suggested by James, I am now using the most recent version of the software. However, before running the model I am now encountering issue with a shapefile for natural habitat. The error says the coral reef shapefile is not utf-8 encoded but when checked it is. All the natural habitat shapefiles are in the geographic coordinate system (Degrees).

Uploading two log files of Gujarat and Odisha where the model did not run successfully.

Also, just to get clarity… earlier while I was using the older version, I did not face issues with the natural habitat shapefile geometry, but now in the newer version the model requires a polygon or multipolygon. How can I rectify this issue?
InVEST-Coastal-Vulnerability-log-2024-07-10–20_11_27.txt (4.8 MB)
InVEST-Coastal-Vulnerability-log-2024-07-10–16_50_10.txt (7.5 MB)

Interesting, this might be a flaw in our validation system. I think in order to resolve this issue, and the issue of the WW3 error, we will need to look at your input data. Do you mind sharing your data?

If the files are large, it might work best to upload to a file-sharing site and then share a link. You may send me a private message with the data if you prefer.

Thanks,

I currently use www3 and continental shelf contour provided by InVEST, DEM from Asterdem, Bathymetry from GEBCO. For Aoi and landmass, I am presently using the LULC maps converted into shapefile. I successfully ran for Goa, Kerala, and Tamil Nadu. ‘Save as’ in the workbench is unable to save the parameters and data to share you with the input data.

That’s interesting. What happened? Was there any indication of an error?

Perhaps you could manually package your data and upload to a file-sharing service? We won’t be able to reproduce the problem without getting our hands on the exact files you are using.

Thanks again,

I can’t say without seeing the data, but this may be problematic. Vectors converted from rasters often have lots of unintended polygons in them. It could be useful to review the User’s Guide advice on the Area of Interest: Coastal Vulnerability Model — InVEST® documentation