Advancing Water Resources Research and Management
|
|
| Proceedings: Specialty Conference on Rangeland Management and Water Resources |
Selection of a Reach to Evaluate
Calculating the Riparian Health Rating
ABSTRACT: Public awareness of the need for healthy ecosystems has
driven an effort to formulate an assessment index for natural
riparian function. This assessment quantifies variables that indicate
the current ability of a riparian system to function. These
quantities are weighted to yield a comparative rating. A manager can
use the evaluation to: 1) assess present functional health of a
stream segment, 2) identify factors needing remedial attention,
and/or 3) monitor effectiveness of management change. This assessment
is intended for use by professionals as well as landowners. It can be
used to evaluate a 200m stream segment in about 2 hours. The
procedure does not require precise measurements, but relies on visual
estimations of nine readily observable features. With minimal
training in the procedure, evaluators with diverse backgrounds have
obtained consistent and repeatable results.
KEY TERMS: Riparian management; riparian health; monitoring;
trend.
Public and private land managers are being asked to improve or
maintain riparian habitat and stream water quality on lands
throughout the West. Those who live and work on the land can usually
tell which riparian sites support diverse, vigorous plant and animal
communities, which sites have lost their capacity to retain spring
season waters long into the summer dry season, and which sites are
biologically depauperate. While it may be easy for an astute observer
to see that a site has been degraded by human use, it is often
difficult to quantify such changes. Presented here is a method for
rapidly assessing riparian health. It provides an indexed site rating
useful for setting management priorities and stratifying stream
segments for remedial or more rigorous analytical attention. It is
intended to serve as a first approximation, or "coarse filter," by
which to identify stream segments in need of closer attention so that
the manager can more efficiently concentrate effort.
We use the term "riparian health" to mean the ability of a stream
(including the channel and its riparian zone) to perform certain
functions. These functions include sediment trapping, bank building
and maintenance, water storage, aquifer recharge, flow energy
dissipation, maintenance of biotic diversity, and primary
production.
In some cases management steps may have already been taken to remedy
a functionally degraded riparian area. In many such cases, however,
it is unclear how the results of those changes can be assessed.
Moreover, if allotment management will be affected, the permittee is
right to ask for an evenly applied and dependably repeatable
assessment. How, for example, can the health of a riparian site on
one allotment be compared to that on a neighboring allotment? Or, how
can we stratify sites on a large management unit among those
functioning well, those functioning with slight impairment, those
having lost much of their functional capacity, and those so severely
impaired that restoration would be too costly and difficult?
Some more rigorous methods to determine status of a stream's channel
morphology are Dunne and Leopold 1978, Pfankuch 1975, and Rosgen
1996. These relate their ratings to degree of channel degradation but
do not integrate other riparian functions into the rating. Other
methods are available for determining condition from perspectives
that also include vegetation, most notably the USDI Bureau of Land
Management (BLM) proper functioning condition (PFC) (1995). The PFC
method relies on a team of diverse expertise and is not designed to
yield an indexed rating useful for comparative or monitoring
purposes.
We propose this method for rating ecosystem health in terms of site
potential. It is based on assessment of nine channel and riparian
vegetation factors. The procedure has been tested in Montana,
surrounding states, and western Canada since 1992. Some potential
uses for this rating are: 1) for stratifying streams or stream
reaches by degree of ecologic dysfunction, 2) for identifying
ecologic problems, and 3) when repeated over time, for monitoring to
detect functional change. This method is not designed for an in-depth
and comprehensive analysis of ecologic processes. Such analysis may
be warranted on a site and can be done after this evaluation has
identified areas of concern. Nor does this rating yield an absolute
rating to be used in comparison with streams in other areas or of
other types. Comparisons using this rating with streams of different
types (Rosgen 1996), different orders (size class), or from outside
the immediate locality should be avoided. Appropriate comparisons
using this rating can be made between segments of one stream, between
neighboring streams of similar size and type, and between subsequent
assessments of the same site.
A single evaluation provides a rating at only one point in time. Due
to the range of variation possible on a riparian site, a single
evaluation cannot define absolute status of site health or reliably
indicate trend (whether the site is improving, degrading, or stable).
To measure trend subsequent evaluations must be conducted on a reach
over a number of years.
This assessment attempts to balance the need for a simple,
quick index of health against the reality of an infinite variety of
riparian situations. There are some visible changes to site health
for which we have no simple way to measure. An obvious and commonly
encountered example is excess entrained sediment. This may indicate
serious degradation, but we leave it out of the assessment due to
difficulty in knowing how much is normal. Instead, we address on-site
causes of sediment production: bare ground, banks with poor root mass
protection, and human-caused structural damage to the banks. Another
serious degrading factor for which we have no simple measurement yet
is dewatering of the system by irrigation diversion/pumping and by
upper drainage retention dams. Although this approach will not always
work perfectly, we believe that in most cases it will yield a
usefully accurate index of riparian health.
A less direct, but also important, value of an environmental
assessment of this kind is its educational potential. By getting land
managers to focus on individual riparian functions and ecologic
processes, they may come to better understand how the parts work
together and are affected by human activities.
The first step in the assessment is to map the site at a scale
appropriate for management application. This map should show the
stream channel and lateral extent of the riparian zone. Many riparian
areas are easy to recognize and clearly distinguished from adjacent
uplands. The evaluator should sketch general position and extent of
important riparian plant communities.
Identification of plant communities by vegetation type (such as by
Hansen and others 1995, Kovalchik 1987, or Padgett and others 1985)
is useful in determining appropriate management. These may be in a
mosaic difficult to map. An area may have a mix of herbaceous
communities, shrubs, and forest. These communities have diverse
resource values and may respond differently to a management action,
but it is seldom practical to manage such communities separately.
Community composition can be described as percentages of component
types. Management actions can then be keyed to the higher priority
types present.
Having mapped and identified the major plant communities, we now
need to select the specific reach (segment) to evaluate. If time is
available, or length of stream is short, the entire stream can be
rated. If not, then one or more reaches may represent the whole. The
evaluator may choose either a critical
reach (an especially sensitive spot) or one
representing (typical of) the larger
area. It may be wise to assess both critical and representative
reaches.
We recommend the length of reach be at least one channel
meander cycle, though two is preferable. Bank cutting will be
overestimated if the reach is located mostly on an outside curve and
underestimated if it is mostly on an inside curve. A complete meander
cycle has equal inside and outside curvature (Figure 1).
Figure 1. A schematic example of meander cycle delineation showing two cycles
Scale should be considered in determining reach length. Whereas a
200m reach length may include two meander cycles on smaller streams,
such a length would be inadequate on a river 30m wide. If the reach
to be assessed must be shorter than a full meander cycle, the
evaluator should look beyond the delineated reach to include a full
meander cycle when rating channel morphology and streambank factors.
If it is impractical to assess a full meander cycle, we recommend a
200m minimum length.
In addition to reach length, riparian zone width must be considered.
The evaluation should include the riparian zone on both sides of the
stream if both are under the same management. Along a large stream,
the same operator may not manage both sides. The stream may be so
large that livestock (or evaluators) cannot easily cross. In such
cases it may not be feasible to evaluate both sides at once.
The riparian zone is that generally green and relatively flat area
influenced by water from a stream and its floodplain. The contrast
between a riparian zone and adjacent upland is most notable in late
summer when much of the upland herbaceous plants have gone dormant.
The area to be assessed includes any terraces dominated by
facultative wetland and wetter plant species (Reed 1988), the active
floodplain, streambanks, and areas in the channel with emergent
vegetation (Figure 2). Reference to Reed's list of plants found in
wetlands should not be necessary to determine the area for
evaluation. The evaluator should simply focus on that area which is
obviously more lush, dense, or greener by virtue of proximity to the
stream. On very wide riparian areas that extend far back from the
stream, it may be necessary to limit width of the zone assessed to
some reasonable distance. Five channel widths is a good arbitrary
distance. In such a case one should document that distance for
subsequent repeatability.
Figure 2. A schematic example of a typical riparian zone cross section showing near-channel landform features. Note: FAC (facultative), OBL (obligate), UPL (upland), etc. refer to categories of frequency a species is found on wetland (Reed 1988).
If the stream to be rated crosses more than one management
unit, at least one reach should be assessed in each unit. Fences
exert a strong influence on livestock movement and grazing patterns;
therefore, assessed reaches should be located at least 75m from any
fence.
Some factors on the evaluation will not apply on all sites. Sites
without potential for woody species are not rated for factors
concerning trees and shrubs. Vegetative site potential can be
determined by using a key to site type (e.g., Hansen and others 1995,
Kovalchik 1987, Padgett and others 1985). On severely disturbed
sites, vegetation potential can be difficult to determine. On such
sites clues to potential may be sought on nearby sites with similar
landscape position.
To monitor trend, health assessment should be repeated in subsequent
years during the same time of year. Evaluation should be conducted
when most plants can be field identified and when hydrologic
conditions are most nearly normal (e.g., not during peak spring
runoff or immediately after a major storm). Management regime should
influence assessment timing. For example, in assessing trend on
rotational grazing systems, avoid comparing a rating after a season
of use one year to a rating another year after a season of rest.
Most of the factors rated in this evaluation are based on ocular
estimations. Such estimation may be difficult on large, brushy sites
where visibility is limited, but extreme precision is not necessary.
While the rating categories are broad, the evaluator needs to
calibrate his eye with practice. It is important to remember that a
health rating is not an absolute value. The factor breakout groupings
and point weighting in the evaluation are somewhat subjective and are
not grounded in quantitative science so much as in the collective
experience of an array of riparian scientists, range professionals,
and land managers.
Each factor below will be rated according to conditions observed on
the reach. The evaluator will estimate the scoring category and enter
that value on the score sheet.
1. Amount of the floodplain and streambanks covered
by plants. Vegetation cover helps to stabilize banks,
control nutrient cycling, reduce water velocity, provide fish cover
and food, trap sediments, reduce erosion, and reduce the rate of
evaporation (Platts and others 1987). Vegetation cover is ocularly
estimated using the canopy cover method (Daubenmire 1959).
Scoring:
6 = More than 95% of the reach soil surface is covered by plant growth
4 = 85% to 95% of the reach soil surface is covered by plant growth
2 = 75% to 85% of the reach soil surface is covered by plant growth
0 = Less than 75% of the reach soil surface is covered by plant growth
2. Percent of streambank with a deep, binding root
mass. Streamside vegetation stabilizes the soil to the
extent that it provides deep, binding roots. All tree and shrub
species provide such roots. Herbaceous annuals lack
this quality. Perennial herbs provide it in varying degree. Some
rhizomatous species, such as sedges (Carex spp.), are
excellent streambank stabilizers. Other rhizomatous species, such as
Kentucky bluegrass (Poa pratensis ), have shallow roots
and are poor streambank stabilizers. In all cases
greater plant density and vigor means greater
stability of the bank. For this item consider the near-stream area to
be the active floodplain out to a maximum of 10m from the channel on
large streams.
Scoring:
6 = More than 85% of the near-stream area has a deep, binding root mass
4 = 65% to 85% of the near-stream area has a deep, binding root mass
2 = 35% to 65% of the near-stream area has a deep, binding root mass
0 = Less than 35% of the near-stream area has a deep, binding root mass
3. Percent of the riparian zone covered by noxious weeds.
The presence of noxious weeds indicates a degrading
ecosystem. While some of these species may contribute to riparian
function, their negative impacts reduce overall site health. Consider
the aggregate area covered by all species of
noxious weeds listed by your state or county extension service. The
evaluator must use a weed list that is standard for the locality, and
be sure to note any other species additionally considered as noxious
weeds for this item.
Scoring:
6 = No noxious weeds on the site
4 = Less than 5% of site covered by noxious weeds
2 = 5% to 25% of site covered by noxious weeds
0 = More than 25% of site covered by noxious weeds
4. Percent of the site covered by disturbance-induced
undesirable herbaceous species. A large cover of
disturbance-induced species, native or exotic, indicates displacement
from the potential natural community (PNC) and a reduction in
riparian health. These species generally are less productive, have
shallow roots, and poorly perform most riparian functions. They
usually result from some disturbance which removes more desirable
species. Noxious weeds, considered in the previous item are not
reconsidered here. As in the previous item, the evaluator should
state the list of species considered. A list of undesirable
herbaceous species appropriate for use in Montana includes:
- cheatgrass (Bromus tectorum )
- clovers (Trifolium spp.)
- dandelion (Taraxacum spp.)
- Kentucky bluegrass (Poa pratensis )
- mustards (Brassicaceae )
- strawberry (Fragaria spp.)
- Japanese brome (Bromus japonicus )
- plantains (Plantago spp.)
- pussy-toes (Antennaria spp.)
- common cocklebur (Xanthium strumarium )
Scoring:
3 = Less than 5% of the site covered by disturbance-increased undesirable herbaceous species
2 = 5% to 25% of the site covered by disturbance-increased undesirable herbaceous species
1 = 25% to 50% of the site covered by disturbance-increased undesirable herbaceous species
0 = More than 50% of the site covered by disturbance-increased undesirable herbaceous species
5. Degree of browse utilization of trees and shrubs.
(Skip this item if the site lacks trees or shrubs; for
example: the site is a herbaceous wet meadow or cattail marsh. Some
sites with potential for trees or shrubs may have had them eliminated
by human-caused disturbance, but this is not always easy to
determine.) Many riparian woody species are browsed by livestock
and/or wildlife. Heavy browsing can prevent establishment or
regeneration of these important species. Excessive browse can
eliminate them from the community and result in their replacement by
undesirable invaders.
When estimating degree of utilization, count browsed second year and
older leaders on representative plants of woody species normally
browsed by ungulates. Do not count current year's use since this may
not accurately reflect actual use because significant browse can
occur late in the season. Determine percentage by comparing the
number of leaders browsed with the total number of leaders available
(those within animal reach) on a representative sample (at least
three plants) of each tree and shrub species present.
Scoring:
3 = None (0% to 5% of available second year and older leaders are browsed)
2 = Light (5% to 25% of available second year and older leaders are browsed)
1 = Moderate (25% to 50% of available second year and older leaders are browsed)
0 = Heavy (More than 50% of available second year and older leaders are browsed)
6. Woody species establishment and regeneration.
(Skip this item if the site lacks potential for trees or
shrubs.) Woody species potential can be determined by using a key to
site type (Hansen and others 1995, Kovalchik 1987, Padgett and others
1985). On severely disturbed sites seek clues to potential on nearby
sites with similar landscape position. Vegetation potential is
commonly underestimated on sites long disturbed.
The presence of seedling, sapling, pole, and mature age classes of
late seral or climax species indicate long term stability.
Cottonwoods (Populus spp.) are considered equally
desirable, but cottonwood regeneration cannot be expected under the
canopy of older stands. Replacement stands of cottonwoods are to be
found on the recent alluvial deposits near the channel. Cottonwood
regeneration must be sought on such deposits. Their lack may be a
function of scale of the reach or of ecological degradation. On a
site where channel incisement is preventing deposition of suitable
sediment, the site would not be rated for cottonwood regeneration due
to having lost potential for the species. Site degradation would be
assessed in the channel incisement item. Note:
Presence of a woody species age class is indicated
by plants of that class in density of at least ten individuals per
acre. Climax and late seral species are listed below for Montana and
surrounding region.
Scoring: (See below for determination of age classes)
6 = Seedling, sapling, and pole ages of late-seral/climax woody species or cottonwoods present.
4 = One of the three younger age classes of late-seral/climax woody species, or cottonwoods, is absent; OR the site is dominated by early or mid seral shrub species
2 = Two of the three younger age classes of late-seral/climax woody species, or cottonwoods, are absent; OR the understory is dominated by disturbance-increaser shrubs (hawthorn [Crataegus spp.], snowberry [Symphoricarpos spp.], and wild rose [Rosa woodsii ])
0 = Only mature, decadent, and dead trees or shrubs remain. The understory is dominated by herbaceous species; OR Russian olive (Elaeagnus angustifolia ) and/or salt cedar (Tamarix spp.) have at least 5 percent cover on the site
Age classes of trees are based on species and size as follows (dbh is diameter at breast height):
1Rocky Mountain juniper (Juniperus scopulorum ) is an exception to the specifications given, as it does not have typical (or consistent) coniferous size, age, and growth form relationships. The evaluator should subjectively assign age classes to individuals of this species based on size, reproductive ability, and overall appearance.
2Other Broadleaf Species refers to green ash (Fraxinus pennsylvanica ), box elder (Acer negundo ), peach-leaf willow (Salix amygdaloides ), quaking aspen (Populus tremuloides ), and American elm (Ulmus americana ).
Age classes of shrubs are based on relative height and stem size
by species. Shrubs are in three age classes: seedling/sapling,
mature, and dead/decadent. Generally, those plants with stems up to
one inch (2.5 cm) thick and/or no more than half as tall as the
tallest individuals of that species on the site, are considered
seedling/saplings. Mature plants have stems thicker than one inch
(2.5 cm) or those having reproductive structures. Dead/decadent
criteria are same as for trees.
7. Percent of site with human-caused bare ground.
Bare ground is soil not covered by plants, litter or duff, downed
wood, or rocks larger than 2.5 in. Bare ground caused by human
activity indicates a deterioration of riparian health. Sediment
deposits and other natural bare ground are excluded as normal and
probably beyond management control. Human land uses causing bare
ground include livestock grazing, recreation, roads, and industrial
activities. The evaluator should consider the causes of all bare
ground observed and estimate the fraction that is human-caused.
Scoring:
6 = 1% or less of the site is human-caused bare ground
4 = More than 1% to 5% of the site is human-caused bare ground
2 = More than 5% to 15% of the site is human-caused bare ground
0 = 15% or more of the site is human-caused bare ground
8. Percent of streambank structurally impaired (altered) by
human causes. Streambank structural integrity is vital to
good channel configuration and bank shape. Impaired structure can
mobilize channel and bank materials, cause loss of fishery and
wildlife habitat, lower the water table, etc. Bank alteration can
result from such causes as livestock hoof shear, recreation, and
resource extraction. In rating this item consider the bank area from
the water's edge up to 18 inches beyond the top of the bank. The bank
top is that point where the upper bank levels off to the relatively
flat surface of a floodplain or terrace.
Scoring:
6 = Less than 5% of the bank is structurally altered by human use
4 = 5% to 15% of the bank is structurally altered by human use
2 = 15% to 35% of the bank is structurally altered by human use
0 = More than 35% of the bank is structurally altered by human use
9. Channel incisement (vertical stability).
A stream is incised when downcutting has lowered the
channel bed so that two-year flood events cannot overflow the
banks. Incisement can lower the water table enough
to change current vegetation and site potential. Four typical
downcutting indicators are: a) headcuts; b) exposed cultural features
[pipelines, bridge footings, culverts, etc.]; c) lack of
sediment and exposed bedrock; and d) a low, vertical scarp at the
bank toe on the inside of a channel bend.
Channel incisement can occur in any of several stages. A severe
disturbance can initiate downcutting, transforming the system from a
steady state of high water table, wide floodplain, and high
productivity to one of degraded water table, narrow [or no]
floodplain, and low productivity. These stages of incisement can be
categorized in terms of Rosgen Level I channel types (Rosgen
1996).
A top rating goes to either a Rosgen E or C-type un-incised channel
from which 2-5 year floods can access a wide floodplain (not
entrenched). The lowest rating goes to Rosgen F or G-type entrenched
channels where even greater floods cannot overtop the high banks.
Intermediate stages can be improving or degrading. They can represent
those slightly incised channels not yet so incised that intermediate
floods cannot access the floodplain; or they can be old incisement
that is healing and rebuilding floodplain at a new, lower elevation.
Figure 3 illustrates the stages of channel incisement.
Scoring:
6 = Channel is vertically stable and not entrenched (1-2 year floods access a wide floodplain). Active downcutting not evident. Any old incisement is characterized by broad floodplain inside with perennial riparian plant communities well established. (1A or 1B of Figure 3).
4 = Either of two incisement phases: 1) An early phase where the channel is just beginning to downcut. There may be small headcuts, but bankfull flows still have access to the floodplain. (Look for cutting in channel bottoms). 2) An old incisement in which the channel may still show limited active downcutting. A new floodplain is well formed at the lower level, although much narrower than it may become. Lateral bank cutting is likely still widening the incised system on outside curves. Perennial riparian plants are becoming well established. (2 of Figure 3).
2 = Two phases of incisement also fit this rating: 1) An intermediate phase, with downcutting and headcuts probable. The channel is not yet so deeply incised that medium (5-10 year) floods cannot escape the banks. 2) A deep incisement that is starting to heal. In this phase new floodplain development, though very limited, is key. Look for widening of the incised system and for early establishment of pioneer perennial plants on the new depositional surfaces. (3 of Figure 3).
0 = The channel is deeply incised to resemble a ditch or a gully. No floodplain development has begun. Only extreme floods overtop the banks. Downcutting is likely ongoing. (4A or 4B of Figure 3).
The scores are totaled for all the factors rated, and that total is divided by the possible perfect score. Figure 4 represents an example scoresheet.
Figure 4. A sample scoresheet of a site with no
apparent potential for trees or shrubs
Rating = (Total Actual) / (Total Possible) * 100%
Rating = (26) / (39) * 100% = 67%
The manager should realize that a less than perfect score is not
necessarily cause for concern. Ratings of individual factors can be
useful in detecting strengths or weaknesses of a site. A low score on
any factor may warrant management focus. For example, the sample
scoresheet in Figure 4 has low scores for noxious weeds and bare
ground (items #3 and #7). These are factors that management might
improve that would show up in a subsequent assessment.
Daubenmire, R. D. 1959. A canopy-coverage method of vegetation
analysis. Northwest Science 33:43-66.
Dunne, T. and L. B. Leopold. 1978. Water in environmental planning.
W. H. Freeman & Company, San Francisco, CA. 818 pp.
Hansen, Paul L., Robert D. Pfister, Keith Boggs, Bradley J. Cook,
John Joy, and Dan K. Hinckley. 1995. Classification and management of
Montana's riparian and wetland sites. Miscellaneous Publication No
54. Montana Forest and Conservation Experiment Station, School of
Forestry, The University of Montana, Missoula, MT. 646 pp.
Kovalchik, Bernard L. 1987. Riparian zone associations: Deschutes,
Ochoco, Fremont, and Winema National Forests. USDA Forest Service
Region 6 Ecology Technical Paper 279-87. Pacific Northwest Region,
Portland, OR. 171 pp.
Padgett, Wayne G., Andrew P. Youngblood, and Alma H. Winward. 1989.
Riparian community type classification of Utah and southeastern
Idaho. USDA Forest Service Reg. 4 Ecology 89-01. Intermountain Res.
Sta., Ogden, UT. 191 pp.
Pfankuch, D. J. 1975. Stream reach inventory and channel stability
evaluation. USDA Forest Service, RI-75-002. Government Printing
Office #696-260/200, Washington, DC. 26 pp.
Platts, W. S., C. Armour, G. D. Booth, M. Bryant, J. L. Bufford, P.
Cuplin, S. Jensen, G. W. Lienkaemper, G. W. Minshall, S. B. Monsen,
R. L. Nelson, J. R. Sedell, and J. S. Tuhy. 1987. Methods for
evaluating riparian habitats with applications to management. USDA
Forest Service General Technical Report INT-221. Intermountain
Research Station, Ogden, UT. 187 pp.
Reed, Porter B., Jr. 1988b. National list of plant species that occur
in wetlands: Northwest (Region 9). US Fish and Wildlife Service
Biological Report 88 (26.9). USDI Fish and Wildlife Service, Research
and Development, Washington, DC. 89 pp.
Rosgen, D. L. 1996. Applied river morphology. Wildland Hydrology,
Pagosa Springs, CO. 246 pp.
USDI Bureau of Land Management. 1993, revised 1995. Riparian area
management: process for assessing proper functioning condition, TR
1737-9, revised 1995. Bureau of Land Management Service Center,
Denver, CO. 51 pp.
William H. Thompson, Research Specialist, Riparian and Wetland Research Program, The University of Montana School of Forestry, Missoula, MT 59812.
Robert C. Ehrhart, Research Specialist, Riparian and Wetland Research Program, The University of Montana School of Forestry, Missoula, MT 59812.
Paul L. Hansen, Associate Research Professor, Riparian and Wetland Research Program, The University of Montana School of Forestry, Missoula, MT 59812.
Thomas G. Parker, Wetland Ecologist, Bitterroot Restoration Inc., 445 Quast Ln., Corvallis, MT 59828
William C. Haglan, Wildlife Biologist, United States Fish and Wildlife Service, Charles M. Russell National Wildlife Refuge, Lewistown, MT 59457.
home page 
