Allan Frei

"There is no filtration on New York City water."

Scroll this
Please tell us what you do.

I’m a climatologist and a professor in the Department of Geography at Hunter College. I analyze data to see what happened in the past, and then analyze computer simulations to predict what might be happening now, and what could happen if temperatures get much warmer. How will climate change affect the water cycle and therefore the water supply? Quantitative work, like what I do, helps planners who are looking at how different extreme events can impact the infrastructure and — maybe — how we should plan for more green infrastructure in the future.

Science 101. So, can you give us some more background about the history of the water system in New York? 

Everything about the New York City water system is fascinating – the history of it, the social, political aspect of it; and the physical infrastructure of the system.

There is no filtration on New York City water. That doesn’t mean it’s not treated at all, but there is still no filtration necessary. That’s very unusual for a large water supply – among cities, we still have some of the best water quality in the world. The reason the water is so clean is because in the region where most of New York City water comes from, the watershed north of the city, natural processes maintain a clean water supply, from the time rain falls on the hills, collects in streams, and runs down to the reservoirs.

Natural filters?

Natural filters, exactly. Only about 10% of the water comes from the part of the water supply that is the oldest, and the nearest to New York City, in Westchester and Putnam counties – the Croton reservoir. That has more water quality issues because it’s more densely populated and the environment isn’t as pristine so there is a water filtration plant being built to treat that part of the water.

Also, there is minimal pumping of the New York City water flow. Water flows to New York City by gravity.

Through aqueducts.

Yes. Big tunnels, underground, that deliver the water from the reservoirs upstate.

Is that what the sandhogs do?

Exactly — the sandhogs. They dig those and other tunnels that need to be built. The oldest part of the system, Croton, started delivering water into New York City in 1842. The first water from the Catskills started being delivered in the 1920s, and the most recent of the main reservoirs was completed in the 1960s.

As the system was built, it created tension between the city and the people in the rural, upstate watershed communities. There certainly used to be tremendous distrust of the city because the city needed a vast water supply. And the city used political and economic clout to go up there and build reservoirs. Entire towns were removed — because people built towns in valleys, not up on top of the mountains. Valleys, that’s where the reservoirs end up. The local people certainly didn’t have a voice – they tried to fight it maybe, but they couldn’t.

And people probably did not get adequately compensated. Compensated for their whole life — I mean, they had to move their whole life and move someplace else. For decades, there were people who thought that New York City had hidden motives to do this. There were rumors that people in the city knew that there was oil in the ground, so they wanted to move people off — rumors like that, to explain what was happening. I don’t know how bad it is now, but there was still friction as recently as the 1990s.

As a result of the Safe Drinking Water Act, which requires all municipal water supplies to filter water unless granted permission to avoid filtration, in 1989 the EPA told New York City to come up with a plan to protect the water, or New York City would have to filter the water. Which means: build filtration plants. It’s billions of dollars to build a filtration plant, and hundreds of millions of dollars a year to run and maintain a plant. It’s a huge, huge endeavor, so obviously New York City wanted to avoid that at all costs. What followed was 8 years or so of intensive negotiations and discussions between New York City, upstate residents, and other stakeholders.

In 1997 they arrived at a Memorandum of Agreement. New York City now pays for certain things like sewage treatment plants for the local communities, and helps them with agricultural practices that will reduce runoff, so the farmers are not incurring the costs themselves, but they are protecting the environment. Across the system, there are different threats to the water quality — it depends upon the land use. The western part of the system, for example, has a lot of agriculture so there agricultural runoff is the biggest potential threat.

So the solution for clean water is coaching people and enforcing agreements?

Coaching people and working with them to adopt methods that protect the water system, even the way they grow certain plants around the fields or practice irrigation. Part of the Memorandum of Agreement is that New York City buys land if there are critical areas that are up for sale. They don’t take land by eminent domain, but if it’s up for sale, New York City will at least try to buy, if they decide it’s an important piece of land to protect for water quality.

That’s amazing. And this was in 1997?

That agreement – and there have been a few addenda to it, I think – but that agreement was signed in 1997. And that was a very, very tense time for people involved with this issue. You think of these as things in the past, but that was recent.

Did access to a supply of clean water affect the success of NYC early on?

Very early on, the city was just the southern tip of Manhattan. Over decades, there were a series of problems because they just got ground water from wells and from ponds, and gradually those got contaminated as the early city grew. And here wasn’t enough water to fight fires. Several fires raged through the city in the 18th and early 19th centuries. They didn’t have the means to get the water in fast enough to fight fires, so they decided they had to go north of the city to build some places where they could catch rain water. North of the city, at one time, was where 13th Street is now. [In 1830, the city built a storage tank on 13th Street and Broadway, to supply water for firefighting through 12” cast iron pipe laid in the ground.]

Right — it’s good to keep that in mind.

Yes. Next came plans to go up to the Croton system. In fact, they looked at different rivers up there and ended up choosing the Croton. Water started coming in from the Croton in 1842; that was a significant water supply that would provide for growth. Within a couple of decades, the city looked again, farther north, and ended up in the Catskill Mountains, where now most of our water comes from – about 90% of the water for the city comes from there.

How have events like Hurricane Irene affected the water supply in New York?

We don’t know the answer yet, in terms of water quality. But that’s certainly an issue. The communities in the watershed were affected tremendously.

Homes were destroyed.

Homes, roads, bridges, people died. Who knows what got washed into the reservoir. Parts of septic systems – I mean, who knows what got into that. And in the eastern part of the Catskills, there is certainly a lot of erosion there, that’s the problem under normal conditions. That sediment gets into the reservoir and makes the water turbid.


Muddy, and under normal conditions, that’s an issue. So whenever there’s a big storm, it’s a big issue. So when this happened, smaller streams changed their course and jumped to different places because things were eroding so fast.

And how do you anticipate that the water supply will be impacted in the future? I guess you don’t know the answer yet.

I don’t know the answer, you’re right. But it certainly is a topic of concern not only for the New York City supply, not just for the communities in the watershed, but for other communities in this region who, some of which get their water from the New York City supply. The DEP [Department of Environmental Protection] has put a significant amount of resources into a project to evaluate the vulnerability of the New York City water supply to climate change. We’re in the middle of that project now, so we don’t know the answers yet.

But there is some good news. In this region, climate models agree that there will be warming — and that affects the water supply because it impacts evaporation and transpiration from trees. People don’t realize what a big factor that is – that’s a huge, huge factor in the amount of water that stays in the system. So as summers get longer and warmer, that represents a significant drain of water from the system, because more water, whatever falls, is going right back up into the atmosphere — not staying in the ground and in the reservoirs. So just the warming, that serves to make less water available for human consumption and environmental purposes – ecological services.

Is this good news?

No, that’s the bad news. The good news is that it seems likely, although it’s less certain than the temperature prediction, that there will be a moderate increase in precipitation. That’s the other side of the budget, right? Precipitation comes in, and evaporation, or what we call evapotranspiration, goes out. And what’s left is what we use in our daily lives. And of course it’s what’s required in the creeks for the natural landscape. So, in terms of precipitation, it’s expected that there will be a moderate increase. There’s not complete agreement amongst the models. There are some models that are saying no, there will be less. But most models are saying there will be somewhere between five and fifteen or twenty percent increase in precipitation in this region.

That’s good news.

So then the question becomes, does precipitation balance out the losses, due to the warming and faster evaporation? It’s very complicated, for many reasons.  For example, you have the seasonal question: how warm is it going to be in a season? Is it going to rain more in different seasons? Rain in summer, most of it ends up getting evaporated or transpired back into the atmosphere.  On the other hand, rain in spring and fall can end up staying on the surface and ending up in the streams and reservoirs of the water system. In the winter, some of that ends up as snow, which ends up staying in the water system even more. It’s a complicated question.

Just by common sense, you think, well, there will probably be less snow. More of the precipitation will come as rain, because it’s going to be warmer, so the runoff season, the hydrological cycle in the watershed region, is going to change significantly. Everyone recognizes spring runoff. The streams are high in the spring. By the end of the century that might change significantly. You may get a lot more runoff in the winter. There won’t be as much snow in winter; there will be more rain. The spring peak will still be a snow melt, but it won’t be as much. These changes will change the ecology, and the economy, of the region, because communities upstate rely on skiing and fishing and vacationing. The economy in most of the Catskills is not that strong. They rely on these things. So this potential change is coming down the pipe.

But in terms of the amount of water coming into the system, it looks like the net result is that there’s going to be more water. That’s what we think now. So at least that is looking positive.

I don’t think we have enough results to talk about how these changes are expected to impact the water quality. That becomes more complicated. A lot of these things have to do with big events — in other words, storms. Storms affect things, like erosion, which gets the water muddy. And water quality also depends on biological processes that may be altered. The growth of phytoplankton and algae in the water can be disrupted or changed. Reservoirs or lakes may have seasonal changes in the temperature structure and how they overturn and mix in some seasons. That affects the biological processes as well. So how will all these things interact? There’s a group at DEP doing this, that I work with – they’ve got some good people working on this. We just don’t have the answers yet.

How do you think New York could better adapt to the future? As a corollary to that, can residents of New York City do anything now to make the water supply better?

Let me start with the first question, and actually answer a question you asked earlier about Hurricane Irene. In all these issues, people try to address what the vulnerabilities are. I’m going to broaden the discussion to include other communities. There has been an effort, for example, funded by NYSERDA, to identify the vulnerabilities of New York State as a whole to climate change. One of the sectors that we looked at was water. We look at the models, and on the average we wonder, will there be more water or will there be less water in different regions? But it’s also the question of extreme events. It’s expected in many parts of the world, including our parts of the world, that regardless of whether precipitation stays the same, or increases, or decreases, the frequency of large events will increase. More bigger storms. This is because in a warmer world, an expected hydrological cycle would be accelerated. You can get more extremes.

People looking at this question recently did think of this question of extremes, and now Hurricane Irene brings more attention to the issue. And it wasn’t just Hurricane Irene, it was the context in which Hurricane Irene occurred that made this so devastating to some regions. Because prior to Irene, the month of August 2011 was already a huge month for rain. In some places, record-setting. The ground was soaked. Usually when it rains, some water soaks into the ground. But when the ground is saturated, there’s no place to go but over the land surface and that’s what happened with Irene — everything was so soaked beforehand, and then you had this record-setting hurricane and flooding. Then, a week later, you had Tropical Storm Lee. It was Lee that did a lot of damage in the Catskills and other areas that had record-setting precipitation on top of what had already occurred. There were some places where the streams, after the tropical storm passed by, had really record-breaking flooding. Up and around Binghamton, New York. Record-smashing event.

So now, we have a good case study. We know it’s an unusual combination, the amount of rain, then Irene, then Lee, but is it a one out of a hundred years that we get such a series of events? Or is it one out of every thousand years? How unusual is this event? If it was extremely unusual in the past, is it going to go from once every thousand years and now happen every hundred years?  Or would what used to be a once every hundred years event become a once every twenty or thirty year event? Fifty, sixty, seventy, eighty years from now — how many should we expect? How does this event stack up in the past and in the future, and how can people prepare for that? What are the vulnerabilities for these sorts of events? Those are the important questions we’re working on.

How can New York adapt to an increase in water supply?

I don’t know that I have the answer. In terms of how can we adapt, technically, we have to make sure that our infrastructure is robust, and it seems to have held up. Some of the infrastructure, like the water supply that delivers water to New York City, held up through the storm — but there were many local infrastructure examples that failed. Bridges, local flooding, roads, tremendous number of roads were undercut because they were near creeks or rivers, and the current flowed so strong that they eroded the ground beneath the roads and the roads collapsed. People were stranded for a long time. For days, some places up to a week or two. These are going to be more common.

How can we maintain the water quality if these events are going to happen?  That’s a tough question. We have to try to figure out if we should expect more events like this. Expect the unexpected. One of the things about this climate change issue is that many people say, what if these predictions were wrong? And it’s not so bad? That’s true — but people don’t realize, what if these predictions were wrong and it’s even worse than we’re predicting?  No predictions are going to be exactly right, so if we’re looking for a ballpark in the models, some of them will under-predict, and some of them will over-predict.

There have been examples of that already, not specifically on the topic we’re talking about, but for example, the Arctic sea ice. Sea ice in the Arctic has been disappearing faster – way faster – than was predicted ten years ago. They were saying, “Well, that may happen; there could be ice-free summers in the Arctic Ocean,” and we don’t know. Maybe the last time that happened was thousands of years ago. To the best of our knowledge. They were predicting it might happen by the end of the century. People are now saying the Arctic could be ice-free in summer 10-20 years from now.

In terms of the water supply in this region, we just have to be prepared for changes. We are wealthy enough, even though the country is in an economic pickle. We have resources enough here, unlike some other places, so we can adapt more quickly. We can build the infrastructure we need and figure out how to handle the water quality issue, as well as rising sea levels. Which also actually impacts the New York City water supply in a roundabout way.

How do you think City Atlas can be most useful to New York City residents? How can it educate people about the water system that they are so lucky to have?

Again, I don’t know the answer, but that’s the important question. How can we use the Atlas, and any other tools, to help people understand? Most people certainly don’t know the history of the water supply, not even the recent history. People upstate in the communities know it better than people in the city. In the city, for most people all they know is you turn it on, and out it comes.

I think people have to be educated as to the history and how this thing functions, because it’s not free. It’s not only not-free in terms of cost or resources, but it requires a lot of thought and knowledge. People who work on these issues understand how the system works and try to understand how to keep it safe. Safe from losing our water supply or water quality. It’s a very difficult job to do that. Perhaps average citizens don’t appreciate it, and it’s worked because folks at the DEP, and others, have figured out how to make it work over years.

One of the ideas that has surfaced, a group of well known hydrologists wrote an article a few years ago saying something that hydrologists understand: stationarity is dead. In hydrology, people would assume that what we’ve measured historically — say, over the last fifty years — if you’re planning for the future that it’s going to be approximately the same. That stationary measurement is now dead. So how do we even function now?

People have to understand that just managing the water supply from day to day is a difficult job. And now the entire system is going to be less predictable. Even on calm days, decisions are made that are not that easy to make. For example, one of the things they do in one of the water systems is let the water accumulate on one side of a reservoir and then flow over the wall to the other side, so the sediment accumulates in the first side, and the clean water is on the other side. They can control the releases from both sides. But what happens is, if you release into the rivers the sediment-laden water, the one with more sediment in it, that’s what ends up in the rivers downstream of the reservoir, so the local communities aren’t happy. [This kind of maneuver can be complex even in stable weather.]

You’re juggling New York City’s needs, the local community’s needs, the ecosystem’s needs, and changing weather patterns. It’s a really hard job. Maybe I’m biased, but…I’m not praising myself. The guys who work at the DEP and manage this water system, they do a really good job. They really do. They do a hard job. You hear that, guys?

About Allan Frei:

Dr. Frei, the Deputy Director of CISC, is an associate professor in the Geography Department at Hunter College, CUNY. After receiving his Ph.D. from Department of Geography at Rutgers University in 1997, he spent four years at the National Snow and Ice Data Center (NSIDC), which is part of the Cooperative Institute for Research in Environmental Sciences (CIRES) at the University of Colorado. In 2001 Dr. Frei moved to Hunter College. Dr. Frei is a climatologist whose research interests include issues related to climate change, including links to snow cover and sea ice across the Northern Hemisphere, as well as water resources in the New York City watershed region.


Photo by Maureen Drennan
Graphic by Florian Brozek


Read more: New York’s water supply, and PlaNYC

An infographic of New York’s water from 1800 to the present