Suggestion for the make_covariates function

[agruss2]

Just an idea for future development/enhancement:
The nearest neighbour procedure in the make_covariates function is not ideal for stream network models. So, in the future, it may be good to adopt an interpolation procedure that is more suited to stream networks in the make_covariates function when the VAST model of interest is a stream network model.
I reckon this issue may be easier to address with tinyVAST than with VAST?

[James-Thorson]

Yeah I doubt I'll do that level of development in VAST. But I'm open to ideas about developing helper functions for tinyVAST, perhaps in an optional tinyVASTextras package

…

On Sun, Feb 25, 2024, 1:49 PM Arnaud Grüss ***@***.***> wrote: Just an idea for future development/enhancement: The nearest neighbour procedure in the make_covariates function is not ideal for stream network models. So, in the future, it may be good to adopt an interpolation procedure that is more suited to stream networks in the make_covariates function when the VAST model of interest is a stream network model. I reckon this issue may be easier to address with tinyVAST than with VAST? — Reply to this email directly, view it on GitHub <#386>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB46UTP2ZQIHS4C3MF6GJETYVOWWVAVCNFSM6AAAAABDZFXR5SVHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2TEOJVGYYTEOA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>