Gridded computations

[cneyens]

We're building a pywatershed model for northern Belgium to compute transient groundwater recharge. HRU's follow a rectangular grid based on a user-specified cellsize. This is mainly done for easier pre- and postprocessing of gridded input and results. Cascading will not be active. I saw on your website that "gridded configurations" are planned for a future release. In the absence of cascading flow, what would this entail in terms of new features or perfomance? Can you not make a gridded configuration already by simply evenly spacing HRU's of equal size? I'm asking this to evaluate if it would be beneficial to hold off some of the work until those features become available.

[jmccreight]

Hello @cneyens!
Thanks for reaching out. I'm not 100% percent sure, but I think the key question is if you have "inactive" grid cells or not. Inactive cells are in the domain but are not included in the computation. If the answer is "no", then I think you can use pywatershed as-is. I would have to double check this to be sure. If you have inactive grid cells (i.e. the answer is "yes"), then you could wait for the upcoming release or you could use the branch where I've already implemented the skip of inactive cells. I believe it is all working and tested with cascades off. For the release, I just have a few mass-balance issues in cascades that I need to resolve and I've been putting it off. I've been looking for some motivation to get that all wrapped, so this could be it. Cheers,

[cneyens]

Thanks for the reply @jmccreight! We don't have inactive HRU's (inactive areas are simply omitted from the whole domain and don't have a corresponding HRU).

Thanks for all the work you put into pywatershed!

[jmccreight]

Sounds good. Let me get back to you to confirm that the 2.0.1 release handles this situation correctly. I believe it does, but i think I can check that quantitatively sometime next week.

…

------------------------------------------------------- James L. McCreight Research Hydrologist ~ Software Engineer USGS Enterprise Capacity UCAR CPAESS cell: 831-261-5149

On Fri, Mar 7, 2025 at 1:40 AM Cas Neyens ***@***.***> wrote: Thanks for the reply @jmccreight <https://github.com/jmccreight>! We don't have inactive HRU's (inactive areas are simply omitted from the whole domain and don't have a corresponding HRU). Thanks for all the work you put into pywatershed! — Reply to this email directly, view it on GitHub <#329 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC7DIYEGKZ6TJ3NH74FVBB32TFLQJAVCNFSM6AAAAABYPBAXYSVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTENBSGM3TMNY> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[jmccreight]

I just did some acrobatics to run a gridded domain (sagehen 5yr) using both my WIP feat_cascades branch and main after setting all cells to active (hru_type ≠ 0). The results were identical for all output variables I had turned on, which included recharge. That just confirmed what makes sense and I'm not thinking of something.

I put out a patch release 2.0.2 but it will match 2.0.1.

[cneyens]

Great, thanks @jmccreight