JOSS review feedback

[thurber]

Hello pygsflow team! I'm a reviewer for your submission to JOSS tracked here. I'll use this issue to track my feedback on the paper and software -- feel free to split out my comments into separate issues if that helps your workflow.

README.md

• There is a reference to installing version 1.0.0 specifically; since there is already a v1.0.1, it might make sense to generalize this sentence a bit. #13

Paper

• Lines 48-49 in the paper have a duplicated expression "binds them into a single integrated model instance" #18
• Paragraph starting on line 62: since you mention external sensitivity analysis software, it could be useful for readers to cite the calibration software typically used in GSFLOW applications

Jupyter notebooks

• Some of the image links in the notebooks appear to be broken #15
• In the 1_pygsflow_intro.ipynb notebook, there's a comment that mentions forcing the parameter type as follows, but I don't think it follows #16
• Also in that notebook, there's a couple plots of Tmax data -- the first doesn't filter out the missing values but the second does -- I think it would be good to add some commentary about that #17
• Some of the output in the notebooks is excessive (i.e. beginning of AG notebook), could be nice to trim that down so it's easier to skim through #14
• Just wanted to say that the plots in the stream vector notebook are really nice looking!

[jlarsen-usgs]

Accidentally closed wrong issue.

[jlarsen-usgs]

link to JOSS review isssue

openjournals/joss-reviews#3852

[thurber]

@jlarsen-usgs thanks for addressing these requests so quickly! A couple more small items:

• Could you elaborate a little bit more in the summary of pygsflow at the top of the README.md? This would be to satisfy the review requirement that there is a "Statement of Need" as part of the documentation (can be similar to what you have in the paper).
• Can you add a section to the README.md describing how to run the automated tests?

[jlarsen-usgs]

Closing this issue. The reference to external sensitivity analysis software was skipped as many users have their own preference and we do not want to imply bias toward USGS or other calibration tools.