Safety Code of American Whitewater

1998 revision, related articles

Upgrading The American Version of The International Scale of River Difficulty

By Lee Belknap, Chairman, AWA Safety Committee

Two and a half years ago American Whitewater introduced a project to update the international scale of river difficulty. Frankly, we were hoping that someone else would step out from under a waterfall somewhere and take on this task, but nobody threw that rope and the American Whitewater safety committee was swept into the bottomless canyon that the project became. Mind you, the results are not ours. More than 100 expert paddlers from across the country responded to this project and almost 80 of them contributed directly to the ratings that will result. The remaining responders added comments that were helpful in determining the best way to carry out and present these changes. To everyone who responded, regardless of the nature of your response, the safety committee sincerely thanks you for your assistance. We would like to apologize at this time for not being able to respond to some of you who wrote involved and carefully thought out letters. The massive amount of computer time this project required prevented us from providing the level of response that these letters deserved.

Why not leave well enough alone? Because “well enough” was falling apart. While many believed that the rating definitions were very specific and were doing the job, a step back to look at the overall picture found some very large discrepancies. There were even rapids being rated from Class II to class V depending on your information source (and if you had an old enough guidebook). Groups of experts were forced to downgrade the whole system as they pushed for higher difficulty. Folks just learning were having a hard time figuring out the conflicting signals they were getting between the experts, old timers, government agencies, guidebooks, etc. It was getting hard for an entry level Class III paddler to enjoy a run on the other end of the Class 3 scale, and that’s not to mention what was happening in the Class IV range. At one point in time this writer started wondering if the dry joke he frequently repeated was coming true: when asked what to rate unfamiliar rapids the response was: “It’s all Class IV”!

For those who have concerns about the following modifications, please show some mercy. There are a million equally valid ways to have done this and the majority of the last 2 1/2 years was spent distilling the best consensus possible under the circumstances. One might guess that about half of you will find the results satisfactory and the other half will have reservations, some very serious and well thought out. The rating system is used by a wide variety of groups including experts, beginners, all skills in-between, guidebook writers who will have many thousands of changes to keep up, and government agencies, including some that reference the rating system in the code of law. In many cases these references will be in conflict with this updated scale of whitewater difficulty.

We received many well-thought-out letters describing various reservations about any proposed system.

One common reservation about the rating system in general, and one that sometimes resulted in a refusal to take part any further, was the idea that the rating system caused a “tremendous problem” in the sport due to ”…paddlers motivated by ego, image, being cool, etc., who are not in touch with their abilities and can’t identify a safe rapid from a dangerous one.” would this “problem” (if it is in fact a problem) go away if the rating system were abolished? Unlikely! We would also have to abolish river names, rapids names, and all campfire war stories. I don’t know about you, but I’d probably resort to pictographs to make up for these lost communication methods – what would a skull-and-bones symbol rate? Pretty gnarly, I suppose.

The goal of any whitewater rating system is to improve communication between whitewater boaters around the country. As an important communication tool that boaters have relied on for decades, the rating system was falling apart. The following modifications are an attempt to patch up the worst holes, and are were some major ones. There may be much more that can be added to this effort, and those ideas can be implemented later if there is enough interest in the boating community. For now, these changes should put us off on a good start. We will try to explain as carefully as possible why things were done the way they were. If you find this effort totally unsatisfactory, please discuss it with your friends and if necessary American Whitewater. Experience has shown us that it is very difficult to communicate these modifications and after some discussion, we usually find that we are not as far off as first thought.

When discussing rivers or rapids, paddlers regularly discuss a number of factors. The questions asked may include: how dangerous is it, where is it, how big or small, how remote, how easy are the portages…(see inset). All of these factors can be described in very specific terms and generally we choose to do so. For example, hazards can be discussed in very specific terms that describe the exact nature of the undercut, strainer, or whatever. But there is one major factor that cannot be described in such specific terms and it is a factor that is perhaps the most talked about. That factor is difficulty, or “how hard is it.” The only way to answer a question like this is via comparison to another more familiar run. A standard discussion of a rapids may go like this: “Lumpy Falls is almost as hard as Heavy Falls on the Harden River with a little bit of Twister Rapids thrown in and a nasty hole next to an undercut wall on the bottom left.” The narrator may go on to describe how close to the road this place is, the water level, how constricted, any other significant factors, and, if you’re lucky, a good war story. But what if the listener(s) aren’t familiar with the Harden River? This is where a rating system can most effectively fill in the gaps.

The purpose of the scale of river difficulty is to compare the difficulty of rivers and rapids. When we specify a rating for a rapids, we are comparing it to other rapids that are about as difficult.

In the past this was done with definitions that were based on the paddler with a scale that had no room to grow. This left the scale with many human variables including individual perceptual differences, drifting average skill levels (as skills increased the ratings decreased), variations between groups, individual group dynamics, and even individual ego.

The best compromise solution to this basic problem requires 3 basic changes. First, focus more on the variable hardest to describe, specifically “difficulty.” Second, open the scale for future growth as more difficult runs are made and provide more graduations within the scale to patch up the damage that has already been done by downgrading, and third, anchor the system in the physical world instead of the mental by creating a list of benchmark rapids that the paddling community can use to compare all other rapids and rivers to. The list of benchmarks will be introduced in the next issue of American Whitewater (more on this later in this article). It is important to note that rapids were chosen for the benchmarks instead of rivers because rapid difficulty is the largest factor used to determine the difficulty of a river and not generally the other way around. (Having written this, I’m sure someone will find an exception.)

By focusing more on difficulty and less on danger we are actually following a trend that has evolved over time within the paddling community. As long as we are careful to specify hazards when we describe a rapids, this seems to work well. Proposals have been made to add an extra designation to the classification for particularly hazardous locations, or where fatalities may have occurred. These proposals are good ones and could be added later if there is enough demand for this. For now, the focus of today’s revisions will be the difficulty component. Again, if there is enough dissatisfaction with the new level of emphasis, then an additional designation to note danger can be added. Comments on this are encouraged.

The scale will be revised to open the top end and to add increments within the lower classes. For the lower classes, Class II+, III-, III+, IV-, and IV+ have been added. In general these work out to be the top or bottom quarter of a classification range.

Of the many excellent possibilities to open the top end of the scale, the most popular method has been to mimic the climbing scale with an open ended class 5.x scale. The decimal range has no top end. In a century or so there could be a class 5.306 (but I doubt it) which would be way higher than a class 5.4 or a class 5.99. With this scale class VI is problematic. Countless discussions regarding what to do with class VI resulted in changing it to an extreme and exploratory class. Once an extreme rapids has been run more than a few times, those familiar with it should be able to use the 5.x scale to classify it, something that would be difficult without the initial experience of actually running it.

The final and most complicated task to upgrade the rating scale is to anchor the system by basing it on a list of benchmark rapids. As mentioned earlier, this list will be introduced in the next issue of American Whitewater. In order for this to work, each benchmark will apply to very specific conditions. These will include name, location, and water level range. The latter is very important because of the large effect of water level on difficulty. Other variables will be assumed as follows:

  • Climatic conditions to be normal for the season during which the river is most likely to be runnable.
  • The rapids are rated for the paddler on a first run who is following another paddler who is familiar with the rapids.
  • It is always important to specify hazards in words when describing a rapids. There is no way a rating system can do this in any but the most minimal way. As a result, the contribution of danger to the difficulty rating can be reduced to match what most paddlers seem to be doing already. Having said this, danger is still a factor because participating paddlers were asked to review the definitions as defined in the American Whitewater safety code. As a result, danger still plays some part of the rating.
  • While rapids are made of solid rock, they do occasionally change. When this happens to a rapids on the list, the AW Safety Committee will somehow need to be notified of any change in difficulty so that the rating can be kept up to date.

The benchmarks were created by polling almost 80 boaters and generating nearly 3000 data points. Each data point was a specific rapids at a specific level rated by one of our 80 participants. To make a complicated story short, averages were generated for each rapids and the rapids were placed in order. Countless filters were required for the results to make any sense at all including a final manual elimination of a few that just didn’t look right for various reasons.

The list of benchmarks is still being proofread by a large number of boaters. This is the reason why they will not be published until the next issue of American Whitewater. In the meantime feel free to contact us with comments. After reviewing any comments received, American Whitewater will make any last-minute corrections necessary, and then make the changes in all future publications of the American Whitewater Safety Code.

(copy of revised rating definitions omitted here, see the International Scale of River Difficulty in the Safety Code of American Whitewater for that part of this article.)


Inset 1.
What’s New?: three major changes are being made to upgrade the International Scale of River Difficulty.

  • Focus more on the variable hardest to describe: “difficulty.”
  • Open the scale for future growth as more difficult runs are made while providing more graduations within the scale. The system will now include pluses and minuses as well as a decimal system within class 5.
  • Anchor the system in the physical world instead of the mental by creating a list of benchmarks at specified flows that the paddling community can use to compare all other rapids and rivers to.

Inset 2.
List of potential dimensions that should be considered when discussing a river or rapids.

• Location
• Hazards
• Flow
• Size
• Partners
• Remoteness
• Length/time required
• Equipment (which boat, which paddle, camping gear, etc…)
• Portagability
• Air and water temperature
• Available time

Inset 3.
The classes: a list of class I through 5.x


Join AW and support river stewardship nationwide!