Stream Morphology

By Special Contributing Author: Dennis Jackson, Hydrologist

During a rainstorm, the stream channel network begins in small rills on hillslopes. At a certain point on the hillslope, there is a sufficient depth of flow to initiate the formation of a small channel. After the storm stops, the point of flow initiation may retreat downhill as the near-surface ground water table declines. This process continues through the dry season until the point of perennial flow is reached. In exceptionally dry years, surface flow may cease in various parts of the channel network.

In the Santa Cruz Mountains, it is possible to find streams that maintain surface flow in their headwater region but have one or more downstream reaches that go dry almost every year.

The shape of a stream channel at any point in the channel network is a result of a balance between the erosive force of moving water and the material over or through which the water is moving. The following paragraph from Leopold (1994) describes the process of shaping the stream channel.

The shape of the cross section of any river channel is a function of the flow, the quantity and character of the sediment in motion through the section, and the character or composition of the materials (including the vegetation) that make up the bed and banks of the channel. Because the flow exerts an eroding force per unit area, or shear stress, on the bed and banks, the stable form the channel can assume is one in which the shear stress at every point on the perimeter of the channel is approximately balanced by the resisting stress of the bed and bank.

The concept of bankfull is central to understanding the morphology of stream channels. Bankfull discharge, the flow at which the active stream channel is just full, is the discharge that shapes the channel. Leopold et al. (1964) have pointed out that, over an extended period, moderate flood flows move the most sediment. On the one hand, while large flood events move great amounts of material, they are very rare; on the other hand, small floods occur frequently, but do little work. E. D. Andrews demonstrated quantitatively that the "effective discharge," or the channel forming flow, was very close to bankfull discharge (Leopold 1994). For many rivers, the bankfull discharge occurs about every 1.5 years to 2 years.

Leopold et al. (1964) observed that the initiation of movement of median size gravel in stream riffles requires a depth of flow equal to about 75 percent of the bankfull depth. A flow of this depth has a recurrence interval of about one year. Leopold also noted that the depth of the average annual discharge is approximately 33 percent of the bankfull depth.

Stream channels are formed, maintained and altered by the water and sediment they carry. Channel size and streamflow are two related attributes of the stream channel. Channel size is determined by four basic factors: sediment discharge; sediment particle size; stream flow; and stream slope. Channel equilibrium occurs when all four variables are in balance. Increases or decreases in flow or sediment disrupt equilibrium and result in changes to the stream channel. While this is a natural, on-going process, human disturbance can severely impact the balance. For example, increased delivery of sediment to the stream channel can result in significant deposition on the streambed, which can lead to channel widening and an increase in over-bank flooding. A decrease in flood discharge flow due to on-stream reservoirs can also result in significant streambed deposition. Significant streambed deposition can lead to a fundamental change in the morphology of the stream over the long term. Significant streambed deposition is usually accompanied by an increase in the proportion of fine sediments on the streambed. Excessive fine sediments on the streambed have significant biological impacts on the biological community and salmonids in particular. Fine sediments in the bed can reduce the availability and quality of spawning and rearing habitat and can diminish the vigor of the macroinvertebrate community or change its composition.

The slope of the streambed and the type of bed and bank material play a central role in determining the shape of a channel. The slope of the streambed and the depth of the flowing water determine the amount of force that is exerted on the bed and bank. Montgomery and Buffington (1993) classified channel morphology (channel form) on the basis of slope for streams in the Cascades. Their work, which classifies channel forms as colluvial, cascade, step-pool, plane-bed, pool-riffle, and regime, appears to be applicable to the Santa Cruz Mountains.

The colluvial channel form is one in which the hillslope processes dominate over the fluvial (moving water) processes in shaping the "channel." The regime channel form refers to a low energy channel that is similar to a canal.

There is a general tendency for channel slope to decline as the distance from the watershed divide increases. Thus, there is a general correspondence to channel form and position in the watershed; however, exceptions to the trend are common.

The straighter sections of stream channels in the Santa Cruz Mountains tend to be roughly trapezoidal in cross section. That is, their bottoms are roughly flat and the banks slope away from the water's edge. In some locations, the channel bottom is very flat while in other locations there may be a low gravel bar next to a shallow pool. Deeper pools tend to occur at bends or are associated with large woody debris.

The pool-riffle channel form tends to occur when the channel slope is in the range of 0.1 percent to 3 percent. Riffles are deposits of bed material. The riffle crest is higher relative to streambed elevation immediately upstream or downstream. Riffles often form the downstream edge of a pool. During times of low to moderate discharge, water velocities decrease in pools and increase on riffles due to the increased bed slope on the face of the riffle. Salmonids depend upon riffles and pools to create habitat for all life stages. Deep pools supply resting areas for migrating adults, and rearing habitat to juveniles and refuges during low flow periods. Riffles create salmon (Oncorhynchus sp.) spawning grounds and juvenile steelhead trout (O. mykiss) rearing habitat. They also provide macroinvertebrate prey resources.

Sinuosity

Natural stream channels are rarely straight. Sinuosity is a term indicating the amount of curvature in the channel and is computed by dividing the channel centerline length by the length of the valley centerline (FISRWG 1998). The value is useful for comparisons of habitat conditions among or within streams. In general, low sinuosity indicates steeper channel gradient, limited bank cutting, and limited pools. High sinuosity is associated with flatter gradients, overhanging streambanks, and pools on the outside of bends.

In a meandering reach, the stream channel moves across a flood plain, destroying old stream banks and building new ones as it re-establishes a main channel. This process creates healthy instream and riparian conditions, but creates problems for people living and working within the floodplain.

Many streams are artificially straightened and armored to prevent the natural meandering process and thereby protect property. Salmonids use all habitat types for different life stages. Areas with lower sinuosity may provide spawning habitat, while areas with high sinuosity may provide rearing habitat.

In the Santa Cruz Mountains, sinuosity tends to occur in low gradient reaches with broad alluvial valleys such as portions of Pescadero and Aptos Creeks. In general, most of the streams in the Santa Cruz Mountains tend to have a low value of sinuosity.

The Hyporheic Zone

A stream cuts its own channel across the surface of the Earth. The stream channel is surrounded by bedrock, colluvium, or alluvium deposited by earlier streams. The stream deposits a mixture of clay, sand, gravel, and boulders in the trench it cuts through the surrounding material. The size of the trench that the stream cuts is scaled to the bankfull discharge. This process has been in continuous operation over geologic time, so the actual width of stream deposited material may by wider than the present bankfull width.

The stream deposited material tends to be porous, so water from the stream can flow into the deposited material. The subsurface region surrounding the stream that exchanges water with the stream is called the hyporheic zone. The chemical and biological character of the hyporheic zone can differ from both the water in the stream and the water in the surrounding ground water table. For example, the dissolved oxygen content of the hyporheic zone may be lower than the water in the stream but may be much higher than in the ground water table. The water temperature of the hyporheic zone may tend to be cooler and less variable than the stream water temperature, but may be higher and more variable than the temperature of the ground water. In low rainfall years with hot summer temperatures, streams with little shading may become too warm to support salmonids in general. However, local refugia may develop in areas where cooler hyporheic water reenters the stream channel such as at the downstream end of gravel bars.

Along the stream course, ground water periodically interacts with the stream, creating a series of "gaining" or "losing" reaches. Gaining reaches receive water from the ground water table or from the hyporheic zone, which increases the flow in the stream channel. Losing reaches are sections of the stream where water leaves the stream channel and enters either the hyporheic zone or the ground water table.

The complex nature of streamflow presents challenges to resource managers, but understanding the interconnection of surface water and ground water resources may be a key to restoring and managing salmonid habitat. For more information about the interaction of surface water and ground water see United States Geological Survey's (USGS) on-line circular 1139: Ground Water and Surface Water: A Single Resource.

Streamflow

Precipitation is the ultimate source of streamflow, and its quantity, quality, and timing will affect many aspects of streamflow. Baseflow (precipitation that percolates to the ground water and moves slowly through substrate before reaching the channel) sustains streamflow during periods of little or no precipitation. Reduction to baseflow, which may result from ground water pumping and water diversions, can result in serious reductions in available water resources. Pools may become isolated with no connecting flow or reaches may go completely dry if insufficient baseflow maintains discharge during dry months.

Stormflow (precipitation that reaches the channel over a short time frame through overland or underground routes) impacts stream form by moving sediment, scouring stream bottoms, eroding banks, and over-bank flooding. Although these impacts are necessary to maintain salmonid habitat, extreme stormflow may create undesirable conditions by flushing juveniles downstream, preventing upstream spawning migration, and excavating developing eggs and embryos from their gravel nest. While large floods are natural events, human modifications within the watershed can create higher flows and larger flood volumes than might naturally occur. Larger floods may also deposit significant amounts of fine sediment in the stream channel with potentially deleterious effects on salmonids.

The Santa Cruz Mountains have a Mediterranean climate with a distinct winter wet season from November through April and a dry summer. Frequent and often heavy precipitation from November to March leads to highly variable stream discharge over time, with multiple peaks that closely correspond to precipitation events. At the beginning of winter rains, soil moisture is low, causing most of the rainwater to replenish dry soils. At the same time, evapotranspiration rates begin to decrease as temperatures drop. Consequently, precipitation events of similar intensity will result in higher peak flows in the winter, when soils are more fully saturated and transpiration demands are lower than in the fall.

Streamflows are lowest during the summer when precipitation is low, evapotranspiration demands are high, and soil moisture is depleted (Spence et al. 1996). The long summer months with higher air temperatures and no precipitation often result in a very low volume of stream flow. Some stream reaches become a series of unconnected pools or even dry out completely.

The Central Coast precipitation patterns create a succession of natural environmental hazards for salmonids. These hazards are often made worse by human activities within the watershed. For example, high flows may scour stream bottoms causing redds of eggs and embryos to be lost. Channel modification such as straightening a stream course for road development can increase flow through stream reaches. During low water summer months, ground water pumping may result in even lower water levels.

Hydrographs for the wettest, driest and median discharge years for the Pescadero Creek near Pescadero, CA (from USGS gauging stations).

The streamflow in the Santa Cruz Mountains can vary by a factor of almost one thousand between summer and winter. The USGS has maintained a stream gauge on Pescadero Creek near Pescadero since 1951. The watershed area above the gauge is 45.9 square miles. The figure above shows the daily discharge for 1983 (the wettest year), 2002 (a median year), and 1977 (the driest year). In 2002, the median runoff year, daily average flows ranged from about 1.4 cubic foot per second (cfs) in September and October to about 1,050 cfs during a November storm. So the daily average stormflow in 2002 was about 750 times the fall discharge. The difference between stormflow and the fall discharge is probably greater at locations with smaller drainage area.

In 1977, the streamflow at the gauging station dried up for a period of 34 days during August and September. There may still have been flow through the hyporheic zone or even isolated pools. The USGS discharge record would not distinguish between the case of no discharge and no pools from the case of no discharge and the existence of isolated pools without discharge between pools. The USGS gauge on Pescadero Creek is down on the coastal floodplain where the depth of the alluvium may be deep. Upstream of the station, Pescadero Creek flows through a canyon. The depth of alluvium in the canyon is probably significantly less than it is at the gauging station. Therefore, it is possible that there was flow in the upper reaches of Pescadero Creek or its tributaries during 1977, the driest year on record.

The maximum daily average discharge during 1977 was only 45 cfs. In contrast, the minimum daily average discharge in 1983 was about 4.3 cfs and the maximum daily average discharge was 2,400 cfs.

Streams are categorized based on the balance and timing of the stormflow and baseflow. Ephemeral streams flow only during or immediately after storms; intermittent streams flow only during certain times of the year; and perennial streams flow continuously during wet and dry times, with baseflow dependent upon ground water movement into the channel. Although salmonids are generally considered to be residents of only perennial streams, intermittent streams, especially backwater sloughs developed during high water events, provide critical habitat for sheltering. Even ephemeral streams play an important role by influencing the quantity and timing of flow and the movement of inorganic sediments and organic debris (Reid and Ziemer 1994).

Floods Shape the Channel

Floods are the process that allows a stream to create its own channel. In the context of stream morphology, a flood is any high water event whether or not water flows over the streambanks. Floods are characterized by their magnitude and frequency. The magnitude of a flood is its peak instantaneous discharge in cubic feet per second (cfs). The frequency is measured by the probability of a given magnitude flood event occurring in any given year, at a particular location. Usually, flood frequency is assigned only to the largest flood that occurs during a year. The list of all the annual maximum flood events for a gauging station is called the annual series. The inverse of the probability of occurrence is the return period, which is the average number of years between flood peaks of the same magnitude. For example, a flood that has a 4 percent probability of occurring in any given year has a 25-year return period.

Flood frequency is a useful concept since it allows events on different rivers to be compared or for different locations within the same river system to be compared. For example, a flood event with a 2-year return period on a small tributary can be compared to a 2-year event on the main channel even though the event on the main channel will have a greater discharge. Another use of flood frequency is to assign risk such as the Federal Emergency Management Agency's (FEMA) use of the 100-year (1 percent probability of occurring in any given year) storm to define the areas subject to flood damage.

The channel-forming discharge (bankfull discharge), for most rivers, is considered to have a return period between 1.5 years and 2 years. The mean annual flood has a return period of about 2 years. When the mean annual flood for 24 USGS gauging stations in the Santa Cruz Mountains on streams that drain directly to the Pacific Ocean are graphed, the magnitude of the mean annual flood shows a strong correlation with watershed area since the correlation coefficient (R²) is 0.929, indicating that watershed area accounts for 92.9 percent of the variability of the mean annual flood.

The mean annual flood for 24 USGS gauging stations in the Santa Cruz Mountains show a strong correlation with watershed area (R2 = 0.929).

The equation for the line of best fit is Mean Annual Flood = 99.698 x (Watershed Area) 0.897 where the "Watershed Area" is the drainage area upstream of the gauging station, measured in square miles. The exponent for the "Watershed Area" is less than 1.0 indicating that the contribution of storm runoff, per square mile, to the mean annual flood diminishes with an increase in watershed area. For example the predicted mean annual flood for a location with 1.0 square mile of watershed area is 99.7 cfs, but the predicted mean annual flood for a 10 square-mile watershed is 787 cfs. So a tenfold increase in watershed area only results in a 7.9 fold increase in the predicted mean annual flood.

Sediment Transport

Sediment is characterized by its size. There are six major size classes arranged in a geometric series with a ratio of two. There are six major sediment size classes, based on the Subcommittee on Sediment Terminology of the American Geophysical Union.

From the perspective of biology and stream morphology there are two basic size classes of sediment. The two basic types of sediment are fine sediment and coarse sediment. Fine sediment can be carried in the water column and is also known as suspended sediment. Coarse sediment is carried along or very near the bed of the stream and is known as bedload. Coarse sediment or bedload shapes the streambed. Gravel bars and riffles are deposits of bedload.

The division between fine sediment and coarse sediment typically occurs somewhere in the sand size class. The division is dynamic because it depends on the sediment transport power of the stream. The sediment transport power of a stream at a particular location along the channel depends on the discharge and the slope of the water surface. So, in larger magnitude floods larger sediment sizes can be carried in suspension. In highly energetic streams even gravel can be carried in suspension.

Dunne and Leopold (1978) provide a good discussion of the basics of sediment transport. They note that the sediment transport process in rivers has four basic features:

• The motion of the grains is in the form of a shearing motion in which various layers of solid grains slide over one another.
• An impelling force or tractive force exerted by the fluid in the direction of motion is required to maintain motion.
• The solids are heavier than the fluid, and therefore tend to sink or are pulled downward by gravity.
• If the motion is persistent and approaches a steady state, the forces acting on each layer of the moving solids must be in equilibrium in a statistical or time-averaged sense.

Motion of gravel on the streambed is initiated when the flowing water has enough force to slightly lift the particles in the upper layer over their neighbors. Only two mechanisms can supply the upward force needed to move particles of sediment: the first is by either continuous or intermittent grain-to-grain contact, and the second is the transfer of momentum from the water to the particles. These two mechanisms define the difference between bed load and suspended load. Bed load transport requires at least intermittent grain-to-grain contact, because the particles are too heavy to be suspended in the water column. Suspended load is light enough to be continuously supported by the upward transfer of momentum in turbulent eddies and so is carried totally by the water column. The sediment transport rate is the product of the immersed weight of the sediment times the mean forward velocity of the grains. The immersed weight of a sediment particle is the dry weight of the particle minus the weight of the water the particle displaces. A stream transporting a sediment load can be characterized as a machine doing work. The rate of work done by a machine is:

Rate of Work = Available Power x Efficiency

The available power for a river is the release of energy as the water flows downhill. Most of the energy released is dissipated as friction from the motion of the water. The friction of motion heats the water. The heat produced is lost by radiation and convection to the surrounding air. A small amount of the energy released by the flowing water is used to transport sediment and other debris and to shape the channel.

The total amount of energy available to the river is from gravitational potential being converted to kinetic energy from the drop in elevation. So, a river flowing down a channel with a constant slope does not accelerate; that is, the water velocity is a constant. The physics of flow water is such that, during flood events, the water velocity is nearly constant anywhere along the stream channel network. As the discharge increases, at a point the roughness caused by the bed and banks has less effect so the mean velocity increases. In general, the velocity of the bankfull discharge is somewhere in the range of four to six feet per second.

The available power of a river is proportional to the discharge times the slope of the water surface. These quantities can be directly measured in a stream. However, the efficiency of the river in moving sediment is very difficult to evaluate, so, at the present time, there is no single formula to compute the sediment transport rate for a river. There are several formulas that work reasonably well for certain ranges of conditions. These formulas have to be fitted to data to produce results and they can be in error by a significant amount.

The important concept is that, with respect to the morphology of a stream, the difference between fine and coarse sediment is that fine sediments are transported in the water column and so can move rapidly downstream, whereas coarse sediments are so heavy that they move by being pushed along the bed and take much longer to move downstream. Coarse sediment forms the bed of most streams. The coarse sediment provides spawning beds for salmonids and the living surface for macroinvertebrates.

Fine sediment tends to be moved rapidly down the stream channel to the ocean during floods. However, if fine sediment makes up a large fraction of the total sediment load, a portion of it will be deposited on the streambed when the discharge decreases. The presence of a substantial amount of fine sediment on the streambed indicates that the stream carries a relatively large load of suspended sediment.

Excessive amounts of fine sediment on the streambed are deleterious to salmonids and other members of the stream community. Fine sediments may clog spawning gravels, fill pools and negatively impact the macroinvertebrate community.

Geology and Fish Habitat

Portions of the Santa Cruz Mountains are underlain by a sedimentary geology marked by weakly consolidated sandstones such as the Santa Margarita or the Butano units. These formations range from deposits of virtually unconsolidated beach sand to units that are sufficiently consolidated to produce cobble and gravel size sediments. Streams in watersheds dominated by these types of sandstone formations tend to carry high loads of sand.

Formations of siltstones and mudstones also cover a significant portion of the Santa Cruz Mountains. These types of formations can produce boulder, cobble, and gravel sized sediments of low density that fracture easily. Siltstones and mudstones have a noticeably lower density than igneous or metamorphic rocks, so the same size siltstone cobble is lighter than an igneous one. A single person can easily lift small boulders of siltstone or mudstone, whereas a small boulder of igneous or metamorphic rock can pose a challenge for a person to lift. The low density of siltstone and mudstone rocks means that they can be mobilized by smaller discharges than similar igneous or metamorphic rocks. Therefore, floods are more likely to wash salmon eggs from a spawning bed composed of siltstone and mudstone than from a spawning bed composed of igneous or metamorphic gravels.

Siltstone and mudstone rocks fracture easily, so they tend to break apart during the sediment transport process. The breakup of siltstones and mudstones releases fine sediment into the stream. These types of sedimentary rock have a lower value as a spawning substrate than igneous or metamorphic rocks, and they are a source of fine sediment, which can be detrimental to salmonids. Therefore, even undisturbed watersheds dominated by sedimentary geology tend to produce lower quality salmonid habitat than a similar watershed dominated by either igneous or metamorphic geology.

The Pescadero-Butano Watershed Assessment (Environmental Science Associates 2004) makes the following conclusion about fish habitat in the Pescadero Creek watershed.

Overall, the condition of the watershed's fishery is good: adequate, but generally not excellent habitat exists throughout a relatively large area of the watershed that is accessible to anadromous fish.

Most of the tributaries of Pescadero Creek are dominated by siltstones, mudstones, and weakly consolidated sandstones. In contrast, San Vicente Creek is widely considered to have high quality salmonid habitat and its watershed has a high percentage of igneous and metamorphic geology. Thus, watersheds with a siltstone, mudstone, or weakly consolidated sandstone can provide adequate salmonid habitat, but are less likely to provide excellent salmonid habitat. On the other hand, watersheds with denser, more durable rocks from igneous and metamorphic formations tend to have a higher percentage of excellent salmonid habitat than sedimentary watersheds.

In addition, land management practices are more likely to lead to higher erosion rates in watersheds with siltstone, mudstone, and weakly consolidated sandstone geology than watersheds underlain by igneous or metamorphic formations.

References

Dunne, T., and L.B. Leopold. 1978. Water in Environmental Planning. San Francisco, CA: W. H. Freeman and Company.

Environmental Science Associates. 2004. "Pescadero-Butano Watershed Assessment." Final Report, 248 pp. View document (PDF).

Federal Interagency Stream Restoration Working Group (FISRWG). 1998. "Stream Corridor Restoration: Principles, Processes, and Practices." Federal Interagency Stream Restoration Working Group (FISRWG). GPO Item No. 0120-A; SuDocs No. A 57.6/2:EN 3/PT.653. ISBN-0-934213-59-3. View on-line document.

Leopold, L.B. 1994. A View of the River. Cambridge, MA: Harvard University Press.

Leopold, L.B., M.G. Wolman, and J.P. Miller. 1964. Fluvial Processes in Geomorphology. San Francisco, CA: W.H. Freeman and Company.

Montgomery, D.R., and J.M. Buffington. 1993. "Channel Classification, Prediction of Channel Response and Assessment of Channel Condition, Washington State." TFW-SH10-93-002.

Reid, L.M., and R.R. Ziemer. 1994. Evaluating the Biological Significance of Intermittent Streams. In Issues in Watershed Analysis. Discussions at Interdisciplinary and Interagency Workshops Held at the Humboldt Interagency Watershed Analysis Center in McKinleyville, California, edited by L. M. Reid: USDA Forest Service, Pacific Southwest Research Station. View on-line source.

Spence, B.C., G.A. Lomnicky, R.M. Hughes, and R.P. Novitski. 1996. "An Ecosystem Approach to Salmonid Conservation." ManTech Environmental Research Services Corp. TR-4501-96-6057. (Available from the National Marine Fisheries Service, Portland, OR.)

Back to top