Can the model operate normally if there are nodata values (on national boundaries) in the land use data within the study area? Due to the high spatial resolution and large research area, I am asking this question to avoid unnecessary trial and error. Thank you:)
Welcome to the forum @siyuliu -
Anywhere there are NoData values in the input rasters, there will be NoData values output by the model, it cannot operate normally without all of the data being present.
If you’re concerned about boundaries, you can always add a small buffer to your watershed area of interest, and use that to clip your input data, to be sure that the input data covers the boundaries too.
~ Stacie
1 Like
Thank you for your detailed reply!
After we input the data as required, the model always stops running at a similar stage. This situation has occurred many times and we have also changed computers to operate it.
It can be ensured that in our input data, the DEM, LUCC and pre raster data all have the same projection, range, number of rows and columns, and resolution. Additionally, it should be noted that since the study area includes part of the national boundary range, the land use is buffered outward by 5km according to the DEM, and the parts without data have been filled with 0. In the input data, we can no longer detect any errors, but the model always fails to run successfully.
Furthermore, the raster data is aligned with the DEM resolution of 30m, so a tif is approximately 5G. Has it reached the limit of model operation?
Attached is the work log of our last run (the others all stopped at this place in a similar way). We are very curious about this and look forward to your answer. Thank u so much:)
InVEST-natcap.invest.ndr.ndr-log-2025-09-03–23_18_36.txt (139.2 KB)
Hello @siyuliu , I’m sorry about the delay in following up here. Were you able to get the model to continue to execute?
Looking through your logfile (thank you for including it!), your output files are large, but certainly not the largest number of pixels that I have seen InVEST produce. We have put a lot of time and engineering into making sure InVEST can handle very large landscapes, and so it is more likely that there is some low-level library issue causing this crash than a limit of the model’s operation.
Although I can’t tell from the logfile alone exactly what the issue might be, the first thing I would suggest trying is that you try running the latest version of InVEST, which can be downloaded from InVEST Downloads, User Guides, and Data Sources | Natural Capital Project to make sure that you have the latest fixes and features.
If the issue persists, could you share your input data with us over something like google drive so that we can try to reproduce and fix the issue?
Thank you!
James
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.