Colin, much thanks for your series on DC cooling. Very informative, very helpful.
Please go a little further in re the specific chemicals you cite in this third installment: let folks know how/where to find specific information on toxcities, environmental fate, etc on these compounds including basic research tools like CAS numbers, MSDSs when available, URLs for standard user-friendly databases, etc.
A lot of this fight, not just in Marana, is going to depend on citizen science and you are making a strong contribution to that. Thanks. Onward!
Thank you Colin for your research on this topic. Once again we can see how we the people in the community are shielded from any public scrutiny into the the potential unintended consequences of the relentless building of data centers in Arizona and the the whole country for that matter.
I wonder what kind of health conditions/concerns might result from the chemicals described in your article.?
The lack of open discussion, as well as the missing levels of engagement to bring diverse expertise into these discussions to ensure residents are safe, is quite problematic.
I'm working on an addendum for this week on that exact topic. Some of the chemicals are better established, and exposures have been more broadly studied; others are newer at the levels these facilities produce and emit them, so there is less information. I'll have some more on that front coming up soon, just working my way through Google Scholar!
Thank you for your impressive research and continued activism! How can we be in touch with you? We are facing threats of data center development in our rural AZ county with severe groundwater issues.
You're welcome! These are important issues that need to be looked at, and I'm happy to help folks push back! You can find me in the No Desert Data Center group on Facebook. Just message me there, and we can talk further on how I can help!
Thanks for your research and writing on this environmental hazard of data center coolants. There is one point I need to make though -- please consider a correction on this. You state, "Closed-loop systems consume less water. That is true, and it matters." Closed loop systems do use less on-site water, but with the current 86% TEP mix of fossil fuels, cooled by evaporative water/cooling towers, data centers actually use more water -- off site. I have a detailed analysis on spreadsheet on this, if you want to look at it. Let me know.
Project Blue, next to Tucson was going to use 1910 acre feet per year (AFY), on-site. Their off-site water use by Tucson Electric Power was going to use about 7000 AFY. However, mechanically assisted (fans and convective architecture of the cooling towers) evaporative cooling on site is so much more efficient than HVAC (aka heat exchanger and heat pump) technology. In the Tucson area, evaporative cooling yields a COP/coefficient of performance of 10-40, averaging 25. That means for every kilowatt-hour of electricity (kWhe) used to cool the data centers, evaporative cooling removes heat at the rate of 25 kilowatt-hours of thermal energy (kWht).
With HVAC technology, 1kWhe only removes about 4.5 kWht. So what then happens is the electrical demand goes up dramatically. And the water use from TEP goes up to at least 14,000 AFY, probably more like 28,000, explained in my spreadsheet analysis. When PB goes with the HVAC option, the TEP cooling of their power plants increases water use dramatically.
Some would say that TEP could increase its mix of solar PV from its current 14%. However, they instead are considering nuclear energy, which has the highest water use of all major electric energy sources, at 0.77 gallons per kWhe at Palo Verde, which uses very high quality reclaimed water that has gone through primary and secondary filtration, cleaner than CAP water by a mile.
Hi Russell! Thank you for your note and engagement! I had split the topics up and had been focusing on the on-site approaches in these articles, specifically because the Marana situation is still evolving, and some of these solutions are discussed at public meetings as complete solutions to specific resident concerns without the tradeoffs.
You are absolutely right, and I fully agree that there are significant downstream differences in power needs and their effects on critical resources in the area, most specifically water. The potential power options and the final cooling selections for the facilities in the area will have dramatic effects. For the Marana facilities, dry-air-cooled (HVAC) is the primary approach in the submissions, which, as you state and I agree, will have a dramatic impact on power needs.
The other side of this, and the bias that TEP seems to have, may very well come down to the ACC rate case discussions and the cost-socialization determinations that result from them. Being able to build massive infrastructure like that could represent substantial financial gains for them at the expense of residents' utility bills and water usage.
I'm going to work on the power side of these next and would certainly be interested in any analysis you have done. The implications of these facilities across the entire area are far-reaching and will have ripple effects for generations, depending on how they ultimately settle out.
Colin, much thanks for your series on DC cooling. Very informative, very helpful.
Please go a little further in re the specific chemicals you cite in this third installment: let folks know how/where to find specific information on toxcities, environmental fate, etc on these compounds including basic research tools like CAS numbers, MSDSs when available, URLs for standard user-friendly databases, etc.
A lot of this fight, not just in Marana, is going to depend on citizen science and you are making a strong contribution to that. Thanks. Onward!
Michael Gregory
Thanks for the recommendation! I'll add an update this week with an addendum for the chemicals.
Thank you Colin for your research on this topic. Once again we can see how we the people in the community are shielded from any public scrutiny into the the potential unintended consequences of the relentless building of data centers in Arizona and the the whole country for that matter.
I wonder what kind of health conditions/concerns might result from the chemicals described in your article.?
The lack of open discussion, as well as the missing levels of engagement to bring diverse expertise into these discussions to ensure residents are safe, is quite problematic.
I'm working on an addendum for this week on that exact topic. Some of the chemicals are better established, and exposures have been more broadly studied; others are newer at the levels these facilities produce and emit them, so there is less information. I'll have some more on that front coming up soon, just working my way through Google Scholar!
Thanks again Colin I am enormously grateful for your research into all of this, we can’t change anything unless we can see it clearly!
Absolutely, and couldn't agree more!
Thank you for your impressive research and continued activism! How can we be in touch with you? We are facing threats of data center development in our rural AZ county with severe groundwater issues.
You're welcome! These are important issues that need to be looked at, and I'm happy to help folks push back! You can find me in the No Desert Data Center group on Facebook. Just message me there, and we can talk further on how I can help!
Thanks for your research and writing on this environmental hazard of data center coolants. There is one point I need to make though -- please consider a correction on this. You state, "Closed-loop systems consume less water. That is true, and it matters." Closed loop systems do use less on-site water, but with the current 86% TEP mix of fossil fuels, cooled by evaporative water/cooling towers, data centers actually use more water -- off site. I have a detailed analysis on spreadsheet on this, if you want to look at it. Let me know.
Project Blue, next to Tucson was going to use 1910 acre feet per year (AFY), on-site. Their off-site water use by Tucson Electric Power was going to use about 7000 AFY. However, mechanically assisted (fans and convective architecture of the cooling towers) evaporative cooling on site is so much more efficient than HVAC (aka heat exchanger and heat pump) technology. In the Tucson area, evaporative cooling yields a COP/coefficient of performance of 10-40, averaging 25. That means for every kilowatt-hour of electricity (kWhe) used to cool the data centers, evaporative cooling removes heat at the rate of 25 kilowatt-hours of thermal energy (kWht).
With HVAC technology, 1kWhe only removes about 4.5 kWht. So what then happens is the electrical demand goes up dramatically. And the water use from TEP goes up to at least 14,000 AFY, probably more like 28,000, explained in my spreadsheet analysis. When PB goes with the HVAC option, the TEP cooling of their power plants increases water use dramatically.
Some would say that TEP could increase its mix of solar PV from its current 14%. However, they instead are considering nuclear energy, which has the highest water use of all major electric energy sources, at 0.77 gallons per kWhe at Palo Verde, which uses very high quality reclaimed water that has gone through primary and secondary filtration, cleaner than CAP water by a mile.
Hi Russell! Thank you for your note and engagement! I had split the topics up and had been focusing on the on-site approaches in these articles, specifically because the Marana situation is still evolving, and some of these solutions are discussed at public meetings as complete solutions to specific resident concerns without the tradeoffs.
You are absolutely right, and I fully agree that there are significant downstream differences in power needs and their effects on critical resources in the area, most specifically water. The potential power options and the final cooling selections for the facilities in the area will have dramatic effects. For the Marana facilities, dry-air-cooled (HVAC) is the primary approach in the submissions, which, as you state and I agree, will have a dramatic impact on power needs.
The other side of this, and the bias that TEP seems to have, may very well come down to the ACC rate case discussions and the cost-socialization determinations that result from them. Being able to build massive infrastructure like that could represent substantial financial gains for them at the expense of residents' utility bills and water usage.
I'm going to work on the power side of these next and would certainly be interested in any analysis you have done. The implications of these facilities across the entire area are far-reaching and will have ripple effects for generations, depending on how they ultimately settle out.
Hi Colin, Thanks for the quick reply. Let me know how I can send you my Project Blue analysis. It covers the energy and water aspects.