As new data centers are headed to Eastern Oregon, opponents share their concerns

Earlier this year, Amazon purchased more than 1,000 acres near Boardman, Oregon. As reported by The Oregonian/OregonLive.com, it could be the site for the state’s first “exascale” data center.

Kaleb Lay, director of policy and research at Oregon Rural Action, says that an uptick in data centers brings a number of concerns to the group, including water contamination, water use and growing demand for electricity. He shares more on their concerns.

Note: The following transcript was transcribed using AI and validated for accuracy, readability and formatting by an OPB volunteer.

Dave Miller: This is Think Out Loud on OPB. I’m Dave Miller, coming to you this week from Hermiston. As you drive around parts of Umatilla and Morrow Counties, it is very hard not to notice the data centers. They are huge and common, and more are on the way. This has meant an infusion of jobs and public money in the region, and tomorrow we’ll talk to the Hermiston assistant city manager about what he sees as the benefits of this industry. But Kaleb Lay is worried. Kaleb is the director of policy and research at Oregon Rural Action. We heard from him earlier this week about nitrate contamination in drinking water. He is also focused on data centers. I asked him to give us a sense for the scale of these buildings here in Northeastern Oregon.

Kaleb Lay: So a lot of folks will have heard about data centers in Hillsboro and Washington County, and currently there are quite a few. In Morrow and Umatilla Counties of the Lower Umatilla Basin, I believe we currently have more and have thousands more acres on the way. These are predominantly hyperscale AI data centers, 100 acres or more. Almost all are Amazon and Amazon Web Services.

And that includes what would be, to my knowledge, the largest data center in the Pacific Northwest in what Amazon calls its Exascale Project. The 1,300 acres would use a gigawatt of electricity, enough to power hundreds of thousands of homes, and is going to consume as much land as roughly the entire city of Boardman. It’s a massive footprint, and that is just one of, I believe, 20 different data center clusters or campuses that are either already built, under construction, or in the planning phase.

Miller: When we talked before, our big focus was on nitrate contamination. Amazon was one of the entities that was sued in this class action suit, including the Port of Morrow, and then various large farming operations. Amazon settled about four months ago and said, OK, we’ll give about $20, $21 million. They said, we did nothing wrong here, but we want to be good folks and be done with this, and so here’s this money. As I said, they said that we did not contribute to this problem. What’s your contention about the connection between data centers and nitrate pollution?

Lay: And what Amazon will say is we didn’t add any nitrate to our water. Here’s the thing, though. There was already nitrate in that water when it went into their data centers. They use massive, massive amounts of water to keep their servers cool because running all those chips and answering all those AI prompts, that takes a lot of energy, creates a lot of heat. They need that water to keep them cool.

What happens is, for folks who’ve ever made Jell-O on their stovetop or a simple syrup or something like that, basically, you start with water that has a certain amount of sugar in it, and you boil off a bunch of the water, so the concentration increases. The same thing happens in these Amazon data centers. Water goes in with nitrate in it. They also add things like biocides and descaling agents to prevent things from growing on the inside of those pipes.

And in that process, because they use what’s called evaporative cooling, a lot of the water evaporates off into the atmosphere. All of the contamination stays in the water. So as they run it through over and over and over again, their contamination levels actually go up. That’s a result of what they’re doing in those data centers. And then many of those data centers here in the lower Umatilla Basin send their wastewater to entities like the Port of Morrow, that then take that wastewater and dump it on farmland. Back to the same problem.

Miller: How much data do you have about the water going in and the water going out?

Lay: It’s hit and miss. We don’t know what’s going in and out of every data center. We do know what’s happening at some of them. One of the best examples is PDX90. It’s a large data center, basically smack dab in the middle of Boardman.

Miller: Called PDX90, but it’s here.

Lay: Yes, exactly. Yeah, Amazon loves their weird terminology. PDX90 gets its water from the city of Boardman, and because it’s city drinking water supply, it’s tested pretty regularly for things like nitrate. In July 2025, the water going into PDX90 was, I think, 4.26 mg. per liter for nitrate. A month later, the water coming out of PDX90 in August 2025 was 5 times higher. So now, it’s not a perfect 1 to 1. We don’t know if that’s the same water that went in as went out. It is a quagmire trying to understand the operations of these data centers because they’re the farthest thing from transparent.

But what we know is that the water coming out of them is significantly higher, in terms of contamination levels, than what went in, and then one of two things is happening at these data centers. Both cause problems. Either they’re sending it out to farmland and land-applying it, dumping it as a means of disposal, which threatens our groundwater, poses a risk to our groundwater, or they evaporate off the water completely, which is highly water consumptive, and we do live in a high desert. We have water supply issues across the board here.

Miller: So, water contamination and water use and electricity use are two of the big issues that we’ve just been talking about. What are your other concerns when it comes to the existing level of data centers and the likelihood or plans for many, many more?

Lay: Yeah, how much time we got?

Miller: Only a few minutes, I think.

Lay: OK, I’ll do the quick version. I mean, there are many problems. At the top of the list is transparency. What we have seen is that a lot of the public officials in this area, who are engaged in these negotiations with Amazon, they signed non-disclosure agreements that are still active today. That applies to both data centers that are being built, have been built, and the ones that are going to be built in the future. I believe, if I’m not mistaken, most of the Boardman City Council has NDAs with Amazon, and that goes into other counties in the area as well. It’s by no means just a one-off.

Also, the loss of agricultural land is a very serious, very real problem here. And in some of the rezoning that we’ve seen being done for this exascale project, the big one, there was this supposition that the land where they want to put this thing that was sold to Amazon and rezoned to support a data center had no agricultural value. But if you go out there and actually look around right across the street is almost 2,000 acres of grapes. That’s one of the highest value crops imaginable, especially in an area like this where we get tons and tons of sun. So there are issues with agricultural land loss, there are issues with transparency, and then at the end of the day, people don’t know these things are even going in until they’re already built.

The public engagement process is about as bad as I can imagine it being. Even I, as somebody who does this for a living and is engaged on this every day, even when I’m on my day off because I can’t stop thinking about it, it’s hard for me to keep up with how fast these things are getting approved, getting built, and coming online. And one of the worries that I personally have is that a lot of the rhetoric coming out from Amazon and data center proponents is using estimates and metrics that are based on way outdated data. If you have one data center that’s using a couple million gallons a year, well, what about the other 19? And what about the scalability? How much more water are the new big, giant AI data centers using than the old ones?

Miller: The other big arguments that the companies themselves or their civic boosters often talk about are related, but they’re jobs and then money often in lieu of property taxes, that often they’ll get property tax breaks, but then there’ll be annual fees, franchise fees. Then they can, in the case of Hermiston, they’re talking about millions or tens of millions of dollars which they say, it built our new city hall, for example. So I’m curious how you reckon with, in some cases, large job numbers and this money that can go to public coffers.

Lay: Yeah, I mean, the money is great, we need the money. You know how else we could be paying for all those things is if they would just pay their taxes. Tax breaks are a problem. I would rather there be money in public funds that the public has some control over than just taking big tech’s word that they’re gonna do the right thing and give the money where it ought to go. The other thing too is we have a lot of resource issues here, and we’ve just talked about thousands of people can’t drink the water in their homes because of nitrate, and we need big amounts of funding to create the infrastructure and pay for the long term upkeep, maintenance, and water bill costs for everybody. Why should we not be taking more money from Amazon and other big corporations and big industries around here and requiring them to pay into these funds? I mean, the class action lawsuit is a step in that direction. I think it would be great to see government take a step in that direction as well.

Miller: Kaleb, thanks very much.

Lay: Thank you, Dave.

Miller: That was Kaleb Lay, the director of policy and research at Oregon Rural Action.

“Think Out Loud®” broadcasts live at noon every day and rebroadcasts at 8 p.m.

If you’d like to comment on any of the topics in this show or suggest a topic of your own, please get in touch with us on Facebook, send an email to thinkoutloud@opb.org, or you can leave a voicemail for us at 503-293-1983.