The Problem
Domestic livestock grazing on public lands is the most damaging land-use to streams and rivers in the west. While the various types of mining may be more destructive on a per-acre basis, livestock grazing occurs on virtually every acre of Federal land from the west coast to the Great Plains, and Canada to Mexico. That’s tens of thousands of miles of streams that are susceptible, every year, to the impacts of cows.
When I was hydrologist on the Beaverhead National Forest in southwest Montana (1984 to 2007) I measured over 600 stream channels. After discarding the ones with multiple impacts (timber harvest, mining, roads) I found that of the ones with grazing as the only land-use within the watershed, 45% were either non-functioning or functioning-at-risk as a direct effect of grazing (Bengeyfield 2008).
Non-functioning stream on public lands in Idaho (photo courtesy of Greg Bevenger).
I think that when most people consider the effects of cows on streams, they immediately think of fisheries. That’s understandable, and certainly the effects of livestock on fish are significant and prolonged. Livestock have affected fish populations throughout the west by adding sediment, trampling redds, increasing stream temperatures, and generally degrading habitat. These are real effects, but they pale in comparison to the larger scale impacts of grazing on how the watershed works – its function.
Streams and rivers are not on the landscape to look pretty or to provide water for human needs. They’re there to do a job. They must move water and sediment through the watershed in the most physically efficient manner so that the capability and productivity of the watershed can be maintained. Only then can human needs come into play – and only if they do not prevent the streams from functioning. For if the function is impaired, the stream will find a different way to do its job, and that way is not likely to be friendly to humans and their needs.
Streams are self-forming and self-maintaining. They are physical systems first. Integrating physical factors such as precipitation and geology, the efficiency requirements are met through the adjustment of channel morphology.
The components of channel morphology that respond to changing inputs are: width, depth, entrenchment (relation to floodplain), sinuosity, and slope. If one of those factors changes, through either changing inputs or direct action, the channel will self-adjust to create a new efficiency by adjusting one or more of the other components. With all the combinations of slope, geology and precipitation, there is a wide variety of possible channel configurations, or stream types, across the landscape (Rosgen 1996). When considering the impacts of grazing we’ll consider the stream type that is most susceptible to livestock impacts.
Picture a typical watershed on public lands in the west. The upper slopes are likely to be steep and rocky and the resulting stream channels will mimic those conditions. Here streams are likely narrow, entrenched (minimal floodplain), and held together by rock. Although cattle will get into these streams, the channels are pretty impervious to cow damage.
Farther down in the watershed, conditions are much different. Here the valley widens, the slope of the stream flattens, it begins to meander, and its stream banks are held together by vegetation (mainly willows and sedges). It’s most likely flowing through a small to large meadow, many of which have a high (close to the surface) water table. Soils in the valley bottom are deep. This is the riparian area, where water, soil, and vegetation interact to form the most productive part of western watersheds.
The stream type that has evolved in these meadows is fairly flat, highly meandering, has a wide floodplain, and a narrow, deep active channel (where the water is). This is the configuration that is the most efficient mover of water and sediment through these meadow systems. When floods happen, water leaves the active channel and spreads out on the floodplain, reducing energy that could cause bank erosion and recharging the water table.
The high-water table allows the establishment and maintenance of water-loving plants such as willows and sedges, both of which have extensive root systems (there can be up to a mile of roots in a cubic yard of sedge). These plants line the stream banks and are the dominant factor in maintaining the narrow, deep configuration of the active channel.
Basically, it’s vegetation that holds the channel together. This is the classic western riparian area, where the integration of soils, water, and vegetation form a functioning and productive whole. It is also the area of the watershed that is most sensitive to livestock impacts, and where the cows spend much of their time.
Functioning stream on public lands in Utah (photo courtesy of Greg Bevenger).
When most people think of how cows affect streams, they focus on the grazing and browsing of streamside plants. This is a factor in the process, but the real culprit is the trampling of stream banks by the cows themselves. The root systems of sedges and willows, so resistant to the lateral stresses put on them by flowing water in the channel, are very susceptible to concentrated downward pressure of cow hooves. As this pressure is applied, stream banks collapse into the active channel, and the stream is immediately widened. The widening of the stream initiates a series of physical responses within the channel that lead to an overall loss of function for the entire riparian area.
Remember that the stream channels that have evolved in these meadow systems are narrow and deep – a kind of steep-sided U. As such, the velocity of the water is concentrated across a relatively short distance, making it easy for the stream to carry its sediment load. As the stream widens, the water velocity decreases and the stream loses its ability to carry its sediment load, dropping it in the channel. Additionally, the cross-sectional area of the channel increases. When this happens, it takes more water to fill the channel before spilling over onto the floodplain.
Let’s say the U-shaped channel would spill over onto the floodplain at a streamflow of 5 cfs (cubic feet/second), but with the enlarged cross-sectional area, it might take 10 cfs to reach the floodplain. That means there would not only be 5 cfs of extra water in the stream, but also an extra 5 cfs of energy to erode the bed and banks of the channel.
Now let’s look at what happens to the riparian vegetation because of these changes in the shape of the channel. Under the U-shaped configuration the channel is full when water begins flowing onto the floodplain. This is a result of the annual recharging of the ground water from flood flows spreading out from the channel across the valley.
As the flood flows recede, much of the water makes its way back to the channel. The “water table” is defined by the level of water at the surface of the stream and extends across the valley at roughly that same level. With the widened cross-section channel scenario, the water only fills the lower portion of the cross section, leaving a vertical gap between its surface and the point of spillover onto the floodplain. Unfortunately, this gap coincides with the root zone of the plants.
With reduced flood-flows to the floodplain (remember, it now takes 5 more cfs to reach the floodplain), the groundwater isn’t recharged as often, and the water table drops. The sedges and willows are put under stress, and if the situation persists, they eventually die. Without roots holding them in place, the stream banks become more susceptible to erosion. Now that additional 5cfs of streamflow really comes into play as bank erosion increases and more channel widening takes place, further exacerbating the situation.
Instead of a steep-sided U with a water table close to the surface, the channel evolves to a shallow V with a water table well below the surface. The vegetation changes from willows and sedges to species that belong on drier sites like sagebrush and bunchgrass. The stream is still doing the same job of moving water and sediment through the system in the most efficient way possible given the conditions, but the widening of the stream has changed conditions enough so that the different efficiency imposed on the system results in a different stream type and thus produces a much different riparian area than was there previously.
This scenario has played out on thousands of riparian areas throughout the west over the past 130 years as ranchers and agencies have virtually ignored the management of riparian areas within livestock allotments. When I started with the Forest Service in the 1970’s, riparian areas were considered “sacrifice areas” in the Range Management section of the Forest Service manual.
By observing the change in the stream channel as it goes through the scenario described above, the changes to other riparian values become evident. Fisheries habitat disappears entirely, as the shallow V-shaped channel contains none of the undercut banks, overhanging vegetation, or deep pools that characterize the original U-shaped channel. Additionally, the shallow, open channels are more susceptible to temperature changes and are more likely to freeze in the winter and heat up in the summer. The replacement of sedges and willows by sagebrush and bunchgrass eliminates an important habitat component for a variety of wildlife.
The Solution
So, what can be done? Does this scenario mean that livestock must be eliminated from Public Lands? No, of course not.
The key to maintaining riparian function is to maintain the width of channel that results in the U-shaped cross section. That means limiting stream bank trampling to a level that will achieve those results. On the Beaverhead National Forest, we developed a methodology that predicts the annual allowable bank trampling for a variety of stream types. For the meadow systems described above, that level is 18% (Bengeyfield and Svoboda 1996). We would commonly measure bank trampling levels of 40%-60% for a single grazing season.
Now that we’ve determined what to do (keep bank trampling below 18%), we need to determine how to do it, and what measurements to take to prove that we’ve done it. On the U-shaped channel, the measurement of stream width starts at the spillover point at the edge of the floodplain.
This is called the “bankfull width” and is the point at which the measurement of width must be taken. Then, at the same point on the stream take a measurement of stream depth at its deepest point. Combine these two numbers into a “width/depth ratio.” For these meadow systems, data from ungrazed and properly grazed areas indicates this number should be under 12.
Now, for the actual measurement of streambank trampling. The critical point of measurement is that spillover point where the bankfull width measurement was taken. Extend this line along the streambank, and at any point where a cow’s hoof has caused bankfull width to increase, count it as a foot. In some places it will be a single cow track, and in others there will be 3-4 feet (or more) of continuous trampling.
Do this for 100 feet, measuring both sides of the stream, then average them. I used to use a 3 ft. piece of rebar, marked in 1 ft. increments as a guide. It takes about 10 minutes to do a transect. When bank trampling reaches 18%, it’s time to move the cows.
The question always arises of how many transects to do. The answer: as many as it takes for you to feel comfortable you’ve characterized the stream. Like all monitoring, there has to be some level of honesty inherent in the process.
In some places in the west “stubble height” of sedge along the stream bank at the end of the grazing season is used as a trigger to move cows. This is fine in theory, longer stems will trap more sediment, but it falls apart in practice. When the cows are along the banks creating the stubble, they are necessarily trampling the stream banks and widening the channel at the same time. You have to measure the disturbance that’s causing the problem. Using stubble height as a trigger promotes the very thing we’re trying to control.
Because of the presence of water, riparian areas are pretty resilient. In the case of damaged systems, sediment becomes a plus as it moves through the system and rebuilds stream banks to their former configuration. With a rising water table, riparian plants can become re-established, fulfilling their former role providing bank stability. After all, the same watershed evolved to a U-shaped channel to begin with. There’s no reason it can’t do so again. It’s the annual pressure of bank trampling that prevents recovery from getting started.
On the Beaverhead National Forest, we had 164 livestock allotments. On one of them we successfully implemented an Allotment Plan using these principles (Bengeyfield 2006). From the first year of implementing the bank trampling standards the streams began showing an upward trend and by nine years enough of the damaged stream reaches had recovered to a point where we considered it a success. The permittees lost no time or numbers in the process.
This is how they did it: we provided the bank trampling standards, generally 18%, for the streams, and they figured out how to meet them.
We didn’t tell them how to manage the cows. They discovered that riding up and down the bottom and shooing the cows uphill didn’t work. Often the cows would beat them back to the stream. They finally settled on gathering small bunches of cows and moving them far enough away from the stream where it would take days for them to drift back. In the mean time they would move other bunches of cows a similar distance away. By the time all the cows had been moved, the first group was ready to move again.
They adhered to their rest-rotation grazing system and we developed off-site water where they recommended. The riders were responsible for the “trigger” monitoring. This has to be done by people who are on the ground every day and can assess current conditions. The agencies generally don’t have the personnel to do that. We would periodically do oversight monitoring.
The permittees biggest advantage was their attitude. In the beginning they were certainly skeptical, but as they saw the results they became proactive in the process. As far as I know, the allotment still operates on these principles. Similar management changes have occurred in other areas of the West with similar results.
So, we know it’s possible. And we know it works.
By recognizing the problem (stream widening), identifying a cause (stream bank trampling), determining a standard (18%), matching a solution to the problem (within pasture riding), and monitoring (check width/depth ratio), we have a logical way to approach and solve an issue that has been plaguing land managers for decades.
One more thing. Climate change is real and it’s occurring in real time. In the future the West will be both warmer and drier. Water will become scarcer and even more valuable, if that’s even possible. All these riparian areas in the headwaters streams (most often on Public Lands) that are most vulnerable to livestock damage are, in reality, large sponges that hold water in the upper reaches of a watershed and release it slowly throughout the growing season in a downstream direction. That’s their function. If that function is lost, which is exactly what happens when streams widen and water tables drop, then the late season flows that provide irrigation water downstream will be lost as well.
LITERATURE CITED
Bengeyfield and Svoboda, 1996, Developing Allowable Use Levels for Livestock Management in Riparian Areas, Specialty Conference on Riparian Areas, AWAR Specialty Conference, Reno.
Bengeyfield, 2006, Managing Cows With Streams In Mind. Rangelands, Vol 28, No. 1
Bengeyfield, 2008, Quantifying the Forest-Wide Effects of Grazing on E Stream Types. Proceedings, AWRA Specialty Conference, Seattle.
Rosgen, Dave, 1996. Applied River Morphology. Self-published. Wildland Hydrology.
Pete Bengeyfield grew up on Long Island, got a BS in Forest Management, and an MS in Forest Hydrology from West Virginia University, and moved west as soon as possible. He has worked for private industry in Washington State as well as State and Federal Government. Starting out as a hydrologist for the State of Montana, he moved to the Forest Service on the Wasatch, Idaho Panhandle and the Beaverhead National Forests. During his 26 years on the Beaverhead he was heavily involved in grazing and riparian issues. He is now retired and living in Dillon, Mt.
Pete has made it his mission to photograph as much of the west as he can. See his work at: http://www.bengeyphotos.com.