Observatory for Gridded Hydrometeorology: a python toolkit -- publication management

[ChristinaB]

Week 1 Results Week 2 Discussion Week 3 Intro, method, conclusion update Week 4 References, proofreading, other issues to make readable for pre-submission editor review

• [x] Write an email to coauthors on 4 week deadline with Friday noon checkpoints

• [x] Email Dan Ames - Editor in Chief of Environmental Modeling & Software - who volunteered to review paper before submission - this makes it a REAL deadline

• [x] Invite coauthors to a results discussion zoom phone call Friday at 9:00 am

• [x] Write up and update figures for results section for Friday July 27 noon review.

• [x] Discussion for Aug 3

• [x] Intro, method, conclusion update Aug 10

• [x] Send to Dan Ames on Aug 17

• [ ] Present on HydroShare call Aug 22 (utility and feedback)

[RondaStrauch]

task 1 and 3 done as of July 20

[RondaStrauch]

• Updated HydroShare resource abstracted, key words, and authors.
• Suggest we rename the usecase notebooks without the 'Observatory' in the name...easier to locate, and capitalize Usecase.

Edits to do for Usecase #1 notebook:

1. Edit first notebook use to "Setup and prepare HydroShare"
2. Add text for 'visualize' watershed (above v1=...); explain 'multisitevisual' function; can we make map scale even numbers?; fix arrow on Upper rio salado
3. mention in Section 3 that this is a demonstration for only 1 location...change the name to pull up one of the others and even suggest that the user try it with a example line of code
4. Below section 4. Map Gridded Cell..., add text about what this is doing in more detail than title.
5. Section 5 map is awesome; add text about what this is doing in more detail than title
6. is the saving back to HydroShare section, is there anything that folks will have to change here to save it to there own space...if not, mention that they don't and that this will be saved to their own space.

[jphuong]

@RondaStrauch I'm going to respond to each of your bullets above.

• Thanks for updating those resource metadata. I'll get Sean on HydroShare to add the authorship.
• 'Observatory' is the repository name, and the use-cases would be without meaning if we remove that.

1. Done.
2. Describing the multisiteVisual function. And, the map scale reference length is controlled by the scale_ref_length parameter, which was initially set at 75. Anything below 50 might not get the black-and-white checker style.
3. To explain the visual control, I'll write a few sentences about the parameter differences.
4. The figures in section 4 use the griddedCellGradient function in loop. I'll describe the function.
5. I believe this was repeated for Section 4
6. Saving things back to HydroShare is actually a HydroShare rest client operation. I think I'll add in a few sentences in the notebook, but it's already explained in the Methods section general workflow as the stopping point for each use-case.

In general, thanks for these suggestions. I see your pattern. I will extend it into the other notebooks after I estimate the time that takes.

There seems to be some confusion about the usecases in the paper and the usecase notebooks. The Observatory_usecase1_treatgeoself notebook is only one-half of the data availability use-case mentioned in the paper (Figure 2 a-b). Observatory_usecase2_datadownload notebook is the second half. If done together in one notebook, the notebook takes too long to finish and engages too many functions and parameters. I suggest modifying the paper use-cases to describe the notebook use-cases to be clear and consistent.

[RondaStrauch]

@jphuong Nice job Jimmy. I hope this didn't take too long. I sent you an email while I'm sitting at a Dr. office for the 2nd notebook. Maybe we should call Usecase 1 & 2 notebooks Usecase #1a and #1b so they aren't confused as being 2 use cases. No problem having these being 2 notebooks; however, I just had an idea that we should reference each other in each notebook. This way folks could review independently, but also see how they are connected...just a simple sentence at the beginning and end of the notebook.

[jphuong]

@RondaStrauch I've modified the methods section in the paper so that there are four use-cases now. It corresponds with the notebooks and their outputs more consistently.

[RondaStrauch]

@jphuong Here are a few more thoughts on the notebooks (general, #2 and #3) GENERAL

• Put use case # at top of notebook
• reference the paper at the top so folks know where to go look for more info...use (in review) for yr
• reference the other notebooks in each and that they're designed to run in sequence

USECASE #2

• Describe what section 4 is doing
• Section 5 - point to the save button in the upper left corner and describe the tar’ing of the data….how would they be able to use the data afterwards - untaring? Some simple guidance.

USECASE #3

• explain what exactly a ‘mapping file’ is - remind people
• describe the last command box for section 2
• Is there any way to do a ‘double check’ “by hand” that the statistical summaries are correct?
• What is the ‘Initailize list of outputs’ and following box doing in Sect. 3 after getting Salathe ltm?
• Make ‘Visualize…’ a new section and make ‘station’ “grid”…these might get confused as being real stations rather than virtual distributed stations.
• Add a bit more explanation of what the code boxes are doing so folks don’t have to interpret….like why picked certain years
• Make "visualizations" a new section 4 with a header and then add comment boxes below that describe what is being visualized. Love how the box plots change color with values!
• Explain what old Section 4 is. Is this just a repeat of the previous section end but with a different dataset?

[RondaStrauch]

@jphuong Comment on notebook 4

Explain what the 2nd and 3rd code boxes are doing…loading data and reviewing the dictionary keys, which are like data labels.

in section 2, 2nd sentence - suggest: ‘was previously migrated into your user space on HydroShare…” Also, in next paragraph, confirm that usecase3 isn’t discussed…still usecase2?

Add come text below the ‘Summarize the file availability…’ with a bit of explanation, such as: ‘This function list the data files available for the watershed Sauk-Suiattle River.’

The output from the create dictionaries out puts text on “PRECIP dataframe reading to start: 0:00:00.175112” Is this the time it took to read these data…could you add a sentence below with this interpretation for the curious reader?

Below the print out o fall the summary data types, add interpretation sentence something like: ‘These are all the summary data statistics available for this watershed and data sources.’

For Visualize section…make this a separate section and adjust the text to the title of section 4 accordingly.

After the box plots, add another guiding comment box, something like: ‘Select a few months and show the spatial distribution of the average total monthly precipitation.’ Same thing with the next box plot set…basically expand on the first comment at the top of the code box.

The yearly box plots for average daily and total for the year look identical, os that because the average daily is just the total divided by days in year? Wondering if we need both or we could just add a comment box that states something like, ’to see the daily average precipitation by year, divide the total annual precipitation by the number of days in the year.’ This would fit between the plots with the plots switched. Interestingly, the 2003 & 2006 big storm years weren’t even the heaviest years…thus, the total annual doesn’t necessarily translate into damaging flood years. Wonder if drought would be similar or not.

Historic highs and lows needs a intro comment box. Connect this to the previous annual box plots…you selected some of the highs & lows from this and spatially displayed them. You can see the pattern I’m visioning…comment box on what we’re doing, code box, and output…repeat.

Section 4 should be a later number and needs some comment boxes to annotate what is going on and interpretation.

That's it!

[ChristinaB]

@jphuong We can close this! Almost! Do you mind if I share this on a HydroShare phone call sometime soon? Are you signed up for Emilio's Python in Geosciences seminar?

[jphuong]

@ChristinaB I'm not sure what you're sharing. A presentation? A notebook?

And, I'm not sure if I needed to sign up again. I'm on Emilio's listserv, and we had a brief discussion about scheduling to give a talk. I'm prepared. Bring it on.