3 ms·

It would be nice to be able to look more closely at the data here. What makes up this scalar 'water stress'? Maybe those areas using the available water are ef

It would be nice to be able to look more closely at the data here. What makes up this scalar 'water stress'? Maybe those areas using the available water are efficient? As other commenters point out, there's a 2nd order effect: if you live in a low stress area, near (in a transportation sense, not Euclidean distance) a high stress area, you're bound to have problems unless border control comes to rescue. I should have paid more attention in geopolitics class..

theandrewbailey 7y ago
I'm wondering why areas along the Mississippi River and near the Great Lakes are considered stressed.
- Bartweiss 7y ago
  That's a good question, I had to go poke at the original publication and tool to get some idea. This looks like a very good database, but the NYT is being misleading when it cites "high stress" and then only gives examples that are actively running out of water. The actual definition of water stress used is: > Baseline water stress measures the ratio of total water withdrawals to available renewable water supplies... Available renewable water supplies include surface and groundwater supplies and considers the impact of upstream consumptive water users and large dams on downstream water availability. Higher values indicate more competition among users. So the first note is that this is measuring sustainability, not risk of water shortage; it doesn't factor in that Lake Superior has ~800x the current volume of Lake Powell. And it's meant to study competition for water rather than actual deficit, so 'high' stress is 40%-80% usage, and 'very high' stress is >80% usage. (In fairness, >80% is likely unsustainable because we don't capture water as it arrives. Drawing 100% doesn't mean a state-wide rainbarrel, so water will still reach surface plants as wells run dry.) Second is that it's not clear how rainfall onto the lakes is being assigned. Michigan is "high" in the middle and "low" around the edges, so maybe it's being assigned to the nearest land area without accounting for piping and the usage compacts that govern where it goes? That wouldn't matter in most places, but Lake Superior in particular would distort things hugely. Third, these are watershed-level colorations. If you zoom in on the dark red along the Mississippi, it turns out to be Jonesboro, Arkansas. That's a hilly/mountain region with flatter land on either side, and a bunch of manufacturing. As soon as it's streams hit the White River watershed, it drops back down to yellow. If they're drawing hard without pumping water uphill they might get into trouble, but things would still be ok downstream because no one on the Mississippi is relying on that area for water. https://www.wri.org/aqueduct/ https://www.wri.org/aqueduct/

theandrewbailey 7y ago

I'm wondering why areas along the Mississippi River and near the Great Lakes are considered stressed.

Bartweiss 7y ago
That's a good question, I had to go poke at the original publication and tool to get some idea. This looks like a very good database, but the NYT is being misleading when it cites "high stress" and then only gives examples that are actively running out of water. The actual definition of water stress used is: > Baseline water stress measures the ratio of total water withdrawals to available renewable water supplies... Available renewable water supplies include surface and groundwater supplies and considers the impact of upstream consumptive water users and large dams on downstream water availability. Higher values indicate more competition among users. So the first note is that this is measuring sustainability, not risk of water shortage; it doesn't factor in that Lake Superior has ~800x the current volume of Lake Powell. And it's meant to study competition for water rather than actual deficit, so 'high' stress is 40%-80% usage, and 'very high' stress is >80% usage. (In fairness, >80% is likely unsustainable because we don't capture water as it arrives. Drawing 100% doesn't mean a state-wide rainbarrel, so water will still reach surface plants as wells run dry.) Second is that it's not clear how rainfall onto the lakes is being assigned. Michigan is "high" in the middle and "low" around the edges, so maybe it's being assigned to the nearest land area without accounting for piping and the usage compacts that govern where it goes? That wouldn't matter in most places, but Lake Superior in particular would distort things hugely. Third, these are watershed-level colorations. If you zoom in on the dark red along the Mississippi, it turns out to be Jonesboro, Arkansas. That's a hilly/mountain region with flatter land on either side, and a bunch of manufacturing. As soon as it's streams hit the White River watershed, it drops back down to yellow. If they're drawing hard without pumping water uphill they might get into trouble, but things would still be ok downstream because no one on the Mississippi is relying on that area for water. https://www.wri.org/aqueduct/ https://www.wri.org/aqueduct/