Consider allowing users to specify column names corresponding to dt and flw input in function call

Standardize output plotting for a standard set of graphics but also be clear on export of that data used to make the plots
This will mean a new function and clearer export of results tables

Need to just duplicate the code so that it's clear to run this for both Depth and Velocity prediction

Change function such that it takes a date data type rather than a character vector.
Change function such that it checks for and gracefully fails if days are missing.

Particularly for the hsa functions
The utils_flowformat should define the columns that it creates. Each of the outputs should have a description of what they are beyond a name

Should the tidyr read_csv be used instead of read.csv?

It appears that all of the outputs (csvs, tables, lists, etc) from the previous 2 functions are created only to be used by the "utils_flowstopredict" function. Seems to me like these three functions should be one or two functions that produces a single output data frame. I think utils_floodid and utils_flowstopredict should be one function especially since I can't really interpret the outputs of utils_floodid and I don't know how those connect with the inputs of utils_flowstopredict (without just using the script you wrote). I think you could keep utils_flowformat as one function with an output as long as those output variables are defined and the input to the utils_floodid + utils_flowstopredict function is also defined.

Address this via a function

Need to make headers parallel with text in user guide