CyrusOne News and Blog

CyrusOne’s Path to Zero Water

March 15, 2022

At CyrusOne, we understand that reducing our onsite water consumption is an important part in our effort to minimize our impact on regional water stress and protect our business from risks associated with water shortages (read more about our impact in this blog post). However, no matter how much we reduce our onsite water consumption, as long as CyrusOne facilities are reliant on grid electricity we are indirectly responsible for the consumption of large amounts of water through traditional thermoelectric electricity generation—our energy supply chain water consumption.

We have developed efforts to quantify this energy supply chain water consumption, or “embodied water of electricity,” in order to understand both our full impact on water resources and the risk of electrical supply volatility. With many of our facilities in regions that are currently or predicted to be  high in water stress, we face a risk of electrical supply volatility in the event of a short or long-term water shortage.

The water consumed in electricity production, sometimes referred to as “virtual water,” is often used to justify employing less expensive data center cooling, such as water-consuming evaporative cooling, in order to save electricity. The rationale is that water expended onsite is simply replacing water that would have been used in electrical generation, and that it all evens out in the end.

There was some truth to this analysis, especially when the electricity consumed came from thermoelectric sources (like fossil fuel or nuclear generation). However, solar and wind generation consume a negligible amount of water in comparison. As both electrical grids and individual consumers like CyrusOne replace thermoelectric sources with wind and solar generation, the water embodied in the electricity we consume decreases dramatically. When we at CyrusOne reach our net zero carbon target through the use of renewable energy, we will consume effectively no water for cooling at the vast majority of our facilities, since we will neither evaporate water for cooling nor use water-consuming electricity.

But in the meantime, over the last three years we’ve had an opportunity to conduct a case study in grappling with the embodied water of electricity and the tradeoffs with onsite water consumption for cooling.

In 2019, our largest data center in Carrollton, Texas withdrew 13.3 M gallons of water onsite for its hybrid air- and water-cooled system. In 2020, our assessment of current and future water stress indicated that North Texas is a high water-stress region and predicted to remain water scarce into the future. Also in 2020, we upgraded the facility to a 100% water-free cooling design, protecting this facility from water risk and reducing our impact on regional water stress. This had the impact of slightly raising the facility’s power usage, but reduced the onsite water withdrawal by 65% down to 4.6 M gallons (used for landscape irrigation, fire system maintenance, and domestic water for the office space). Only a portion of the water use is actually consumed through irrigation and the rest is discharged to the water treatment works; however, for our case study, we counted all 4.6 M gallons as water consumption to be conservative.

This 65% decrease looks great in theory, but we wondered if the supply chain water for the extra electricity would mean that the total water consumed by the facility actually stayed the same or even increased due to the upgrade. Luckily, the World Resource Institute published the first working guide for organizations on how to calculate embodied water electricity. Using this guide, we estimated the total water consumed (onsite and for electricity generation) by Carrollton in 2019 and 2020.

While we discovered that supply chain water greatly outstripped the water used onsite, Carrollton’s overall water use still decreased by more than 5 million gallons between 2019 and 2020 due to its switch to a water-free cooling design. This result challenges the conventional wisdom that consuming water for cooling saves total water, at least in today’s supply chain.

But we went further.  In 2020 we also purchased Water Restoration Credits to offset our onsite water use at Carrollton, restoring 20% more water than we withdrew in order to achieve our net positive water designation (Read more about our net positive water methodology in this blog post).

While our path to zero water is not perfect, it’s heading in the right direction and constantly under evaluation. Climate change is a serious threat to the world, and at some facilities we are likely emitting more carbon today than we would if we used water-consuming evaporative cooling. Some may argue that it is better in this moment to consume water than electricity.

This is a worthwhile discussion, but we don’t design facilities only for current conditions; rather, when we build a new data center, we design it to operate for decades to come. Global water stress is predicted to increase dramatically in that timeframe, and we want our facilities to be resilient. We do not ignore our carbon footprint and are aggressively pursuing low-carbon electricity. Our facilities are designed for a future where they will neither consume large amounts of water nor emit large amounts of carbon.

Therefore, in 2020 we invested in a new renewable electricity source to cover an estimated 70% of Carrollton’s power needs with renewable solar electricity beginning in 2021. Based on the WRI’s tool, solar electricity has a water consumption factor of zero, thus reducing our energy supply chain water consumption by 70%. The total water consumed at Carrollton in 2021 will be less than a third of the consumption in 2019, demonstrating the promise of our onsite water-free cooling enabling a truly water-free cooling future for this facility. By investing in both water-free cooling and renewables, we are progressing toward both our water-free and net zero carbon goals.  Why choose “either or” when you can have both?