The Roads and the Water

 On the Proposed Rescission of the Roadless Rule — and the Test It Sets for Itself

Deer Creek - Greenbank, West Virginia

In the winter of 1897, Congress wrote down why the national forests exist. The occasion was anger. Grover Cleveland had just proclaimed thirteen new forest reserves on Washington's Birthday — twenty-one million acres withdrawn from the public domain in a single stroke — and four of them sat in country I would later spend a career mapping: the Bitter Root, the Flathead, the Lewis & Clarke, the Priest River. Some nine million acres in Montana and Idaho. The western reaction was loud enough that Congress moved to suspend most of the proclamations. But in the same appropriations bill, almost as an afterthought, it did something durable. It stated, for the first time, the purposes a forest reserve would serve: to improve and protect the forest, and to secure favorable conditions of water flows.

Water was named at the creation. Not as an amenity, not as a concession to some later constituency, but as one of the two reasons these lands were held back from disposal at all. I have made that point before in this space, and I will keep making it, because it keeps being forgotten. It is being forgotten again now.

The current administration has proposed to rescind the 2001 Roadless Area Conservation Rule — the “Roadless Rule” — and open the Inventoried Roadless Areas to road construction and commercial timber harvest. The rescission would reach nearly forty-five million of the fifty-eight million acres the rule now covers. The stated rationale is twofold: wildfire prevention and timber production, with the argument that these lands need active management and that the rule stands in the way.

I want to take that rationale seriously, because parts of it are serious. Wildfire risk in the West is real and rising. The agency is genuinely, chronically underfunded. “Active management” is not a slur — I spent my early forestry years thinning stands and setting prescribed fire on the San Carlos Apache Reservation, and I do not romanticize a forest left entirely alone. If the case for rescission held up on its own terms, it would deserve a fair hearing.

It does not hold up. And the most honest way to show that is not to argue about values. It is to take the rescission's own two justifications — fire and fiscal responsibility — and check them against the record. On both, the evidence runs the other way.

First, what these areas are. A study published this year in Biological Conservation by David Mildrexler and colleagues does something I find persuasive precisely because it is spatial and measurable rather than sentimental. The authors run the Inventoried Roadless Areas through the Forest Landscape Integrity Index — a 300-meter dataset that scores how intact a forest's ecological functions remain, from observed human pressures to the loss of forest connectivity. I spent the last chapter of my federal career as geospatial program manager for the Northern Region, responsible for the spatial data across twenty-five million acres in Montana, Idaho, and the Dakotas, and I know what an integrity index can and cannot tell you. What this one tells you is unambiguous. The roadless areas score a median of 9.5 on a ten-point scale — as high as, and often higher than, congressionally protected wilderness. The unprotected national-forest lands score a full point and a half lower. The difference is the roads and the logging. The Great Burn, on the Lolo, is one of the areas the authors photograph to make the point; it sits in the region whose data crossed my desk. These are not marginal lands. They are the most intact forest the country still holds outside its wildernesses, and there is more of them than there is wilderness.

Second, the water. Because the roadless areas are overwhelmingly high, steep, headwater country, they protect the topmost links of the watershed — the cold, clean, slowly released water that everything downstream depends on. The study finds that in the western United States, roughly ninety percent of people are served by public drinking-water systems that draw from national-forest lands, and those lands contain seventy-one percent of all the roadless acreage in the country. Nationally, the Forest Service's own watershed accounting puts something on the order of a fifth of the nation's water supply on these lands, serving well over a hundred million people. This is the Organic Act reading forward into the present tense. The headwaters the roadless rule protects are the “favorable conditions of water flows” the 1897 Congress had in mind, whether or not anyone remembers the phrase. Build roads into that country and you introduce exactly what a watershed cannot filter out on its own — sediment, altered subsurface flow, culverts that fail in the heavy rain events that are becoming more common, not less.

Now the test. The rescission says: roads for fire prevention. But a 2026 assessment in the same literature, looking at three decades of wildfire across all eight contiguous-U.S. Forest Service regions, found that ignition density is highest within fifty meters of roads and lowest in wilderness and inventoried roadless areas. Human beings start most of the fires — better than eight in ten ignitions nationally — and roads are how human beings and their ignitions reach the backcountry. More road does not mean less fire. It means more of the ignitions that drive the fire problem in the first place. The rescission has the causation backward.

And the rescission says: fiscal responsibility. But here the argument turns most sharply against itself. The national forests already carry more than 386,000 miles of road — enough to circle the earth fifteen times — plus tens of thousands of miles of unauthorized road on top of that. The agency's deferred-maintenance backlog has grown from $8.4 billion in 2001 to more than $10.8 billion today. By the agency's own 2001 estimate, it was receiving less than a fifth of what it needed simply to maintain the road system it already had. You do not, as a matter of basic stewardship of public money, respond to a maintenance backlog you cannot fund by building more of the thing you cannot maintain — in the most remote, most expensive-to-reach country you administer. The fiscal case is not an argument for rescission. It is the single strongest argument for keeping the rule. An acre without a road needs no road maintenance. That is not an ecological point. It is an accounting one.

So the rescission fails the two tests it sets for itself. It would not reduce fire; it would move ignitions closer to the fuel. It would not save money; it would add liability to a ledger already in the red. That is what I mean when I say the case does not hold on its own terms. One does not have to reach for anyone's motives to see it. The argument simply does not survive contact with the record it claims to rest on.

I will add the part I know from the field rather than the literature. The roadless areas are the source country for the wide-ranging carnivores — lynx, wolverine, grizzly, fisher. In my Voyageurs years I turned radiotelemetry into wolf-pack territory polygons and supported the interagency lynx survey, and I learned that a “source population” is not an abstraction. It is a place large enough and quiet enough that animals produce a surplus that spills outward and repopulates the rest. Fragment it with road and you do not shrink the population by the acreage you took. You break the function of the whole.

And I will add the part I know from before there was a career at all. In the last years of my father's life, Interstate 80 was under construction through our hunting country in Clarion County. The cut opened a coal seam above the camp he had built on Beaver Creek, and orange sulfur water began running into the stream where we fished. I was a teenager. I did not have the words for acid drainage yet. But I watched a road change a creek, and the water still runs changed today, and the camp is still there. I have never been able to fully separate the federal career that followed from that particular piece of orange water. The roadless rule is, at bottom, a rule about what roads do to water. I learned the lesson early and by accident. The country has the chance to keep from learning it again at scale.

Of the roughly 223,700 public comments submitted on the proposed rescission, ninety-nine percent opposed it. That is not a close call, and it is not a partisan one. The authors of the Biological Conservation paper recommend the better course plainly: not to weaken the rule but to codify it — to pass the roadless country into law the way wilderness was, so it no longer depends on which way the administrative wind is blowing.

I would put it in the register of 1897. The roadless areas are the interior — the un-edged, low-noise core of the system, where the water still gathers the way it did before we arrived. Securing favorable conditions of water flows was never a slogan. It was the founding instruction. We are being asked to forget it. I would rather we didn't.

 

This piece was produced in collaboration with Claude, an Anthropic AI assistant. The argument, the sources, and the editorial judgments are the author's.