Conduct WBEEP research for water use

[abriggs-usgs]

Conduct WBEEP research for water use

Completion Criteria - spend a bit of time thinking about water use. Similar to how we did background research for road salt, we want to be prepared to talk about water use at the Friday meeting so it would be useful to come with some idea of what it is. Also include any questions you have about it - likely, your questions would be similar to someone who doesn't know anything about the topic and would be helpful to inform what shape a visual would take.

Research should be no more than 10 hours.

Add notes on research here . . .

• [x] Aaron
• [x] Lindsay
• [x] Megan
• [x] Alicia (?)

[lindsayplatt]

In this example, they chose to not show "thermo" because most of the water is returned (though at a diff temp I think). Also, displayed it per capita to show "stress".

https://thewatercrisis.us/

[lindsayplatt]

Thoughts:

• Daily water use based on 2015 data doesn't seem super interesting by itself. Especially because I'm not sure how much it would change day-to-day. Seems like you need to make water use relative to something.
• One way to show soomething would be to do daily data at each HRU relative to CONUS median or CONUS average. It's either below average WU, average WU, or above average WU. That would likely highlight regions with tons of WU and those that don't use much at all which could be cool. BUT how much would it change day-to-day ...
• I like the idea from the Water Crisis viz above to remove thermoelectric values. Most of thermoelectric withdrawals are put back into the stream, so if our main goal is to communicate how much water we suck up without returning, then that category doesn't make much sense.
• Scaling symbols (as we have done with both the state and county-based USGS visuals) prevent larger regions from dominating visual space and focuses more on the values assigned. I could see somthing like this being useful for the HRUs since they are of varying size.
• What if we did water in (precip) - water out (withdrawals) ? I know that's missing a lot of nuance but may be OK at current stage of this map as a concept.

Two ideas about how to visually represent WU:

• water use makes me think of "removing water" so what if we created a 3D space where the holes meant greater water use. I'm thinking it may end up looking kind of like a topographic map, though it may put too much emphasis on the low (tall) and not the high (sunken) water use. Example kind of like this: https://www.reddit.com/r/dataisbeautiful/comments/bprbqs/3d_visualization_of_us_population_density_per/ or http://www.moonshadowmobile.com/portfolio-items/3d-map-u-s-voters-party-registration-bubbles/
• water use also makes me think of the term "drying up". With that idea in mind, I was thinking of those images of cracked, dry earth. We could represent the high water use as the cracks (so increasing darkness), avg water use as the tan earth, and low water use as puddles on the earth (so some kind of light blue scale). Like this (photo from this site)

---

If we were to combine this with availability, I think the availability model would need to be much more mature. However, we could related the two into a "stress" factor by dividing storage by water used (if they were the same units). If the stars aligned and we could do that, we would end up with stress = 1 means we use just as much as there is stored/replenished, <1 = we use more than we store, and >1 means we use less than we store. That kind of metric may show mismatches in areas that use more water than they have because the water is sources from a different region. That nuance may be too much for this kind of visual though.

[abriggs-usgs]

Water Use. Interesting graphics for fresh water withdrawal and increases in use per person https://www.epa.gov/watersense/how-we-use-water

Aquifers of the US https://water.usgs.gov/ogw/aquifer/map.html

The ‘Water Watches’ have some related information: https://groundwaterwatch.usgs.gov/LWLHome.asp

Seems like there is an important distinction between areas with confined aquifers that may take thousands of years to cycle water compared to unconfined aquifers that take only hours to days. Additionally, it seems relevant to the idea that water is a bit like energy, in that it is rarely created or destroyed, just stored in different forms/locations. Thus, drawing water from unconfined aquifers is less significant than drawing water from a confined aquifer.

Here is a Mapbox example of interest. My thought is to have the entire map ‘extruded’ to a base level for the normative values, then proceed to extrude or recess from there based on value. This is assuming that the values will have a diverging pattern as those in WBEEP Natural Water storage. https://github.com/mapbox/mapboxgl-jupyter/issues/83

I would think having a togglable overlay of indication impervious aquifers of significance would add to user’s ability to focus on area of importance where withdrawals will not be quickly replenished. The areas that have unconfined aquifers may be interesting when overlaid onto regions of Natural Water storage as those regions would be likely replenished or diminished by the amount of Natural Storage water available.