Final Report: Evaluation of Rangeland Stream Condition and Recovery Using Physical and Biological Assessments of Nonpoint Source Pollution

EPA Grant Number: R823487
Title: Evaluation of Rangeland Stream Condition and Recovery Using Physical and Biological Assessments of Nonpoint Source Pollution
Investigators: Herbst, David B. , Knapp, Roland A.
Institution: University of California - Santa Barbara
EPA Project Officer: Hahn, Intaek
Project Period: September 1, 1995 through August 31, 1998
Project Amount: $249,647
RFA: Exploratory Research - Environmental Biology (1995) RFA Text |  Recipients Lists
Research Category: Biology/Life Sciences , Health , Ecosystems


Arid climate and limited water resources in drainages of the Great Basin often give rise only to small streams with narrow riparian zones. These sensitive riparian areas of Great Basin streams are particularly vulnerable to impacts caused by livestock, and grazing has severely impacted stream and riparian habitats in this region (Minshall et al., 1989). In many areas, overgrazing has eliminated riparian vegetative cover, resulting in increased soil erosion and sedimentation. These increases in nonpoint source pollution and loss of channel stability have resulted in deteriorating water quality and diminished ability of these ecosystems to maintain healthy aquatic communities (Platts, 1991).

A recent meta-analysis reviewing the effects of livestock grazing on stream and riparian habitats found negative impacts on water quality, channel structure, and riparian and aquatic life occurring across local and watershed scales (Belsky et al., 1999). Though impacts to rangeland stream riparian areas have been well documented, the effect of grazing on the aquatic life use of streams has received less attention. The goal of this study was to improve understanding of within-stream ecological impacts and quantify the biological response of aquatic invertebrates and fish to grazing-related habitat alterations.

There is a distinctive syndrome of degraded stream habitat conditions that characterizes overgrazing by livestock, which often includes unstable and eroded banks, sedimentation, burial or embedding of rock substrates, loss of riparian vegetation cover and associated organic matter inputs, increased width-to-depth ratio, reduced current in shallow water, nutrient enrichment, and increased algae growth. Lack of bank cover, burial of spawning gravels, and warming of shallow water may limit fish habitat. Potential ecological impacts of overgrazing on benthic invertebrate communities include suspended sediment releases, loss of habitat through choked and silted substrate conditions, algae-microbial mats covering substrates, reduced dissolved oxygen, higher temperatures, and reduced leaf and wood litter as food and habitat structure. The usual approach to monitoring grazing impacts has employed measures of channel morphology and riparian habitat, but these fail to provide direct indications of in-stream biological integrity. The objectives of our studies were to: (1) develop relationships between physical habitat degradation and ecological impairment using benthic invertebrates and fish as biological indicators of grazing effects; and (2) contrast seasonal changes in habitat and invertebrate community indicators before and after the annual grazing period. Invertebrate indicator metrics were used as the primary basis for documenting ecological differences among sites related to habitat degradation.

To reduce the impact to water quality caused by nonpoint source pollution, including livestock grazing, Section 319 of the Clean Water Act requires that states assess nonpoint sources and develop Best Management Practices to improve water quality. The EPA has further recommended that states develop biological criteria through resource inventory and identify reference areas to which waterbodies may be compared for impact assessment. The paucity of relevant scientific information on how nonpoint source pollution from livestock grazing can best be detected and ameliorated has hindered efforts to improve grazing-impacted water quality and stream and riparian ecosystems.

Many studies of livestock influences on streams have been plagued by similar problems relating to methods and interpretation. Grazing studies have often emphasized comparisons between ungrazed and overgrazed systems inside and outside fenced enclosures, and have seldom included long-term data collection or monitoring of benthic stream communities. To evaluate stream responses to grazing, it is essential to have appropriate indicator variables and a gradient of grazing exposure that allows description of the range of response to environmental stress. The results presented here constitute 3 to 4 years of seasonal monitoring that integrates habitat, benthic invertebrate, and fish responses over a gradient of grazing-impacted stream sites.

This study was conducted from 1992 to1995, on nine stream reaches in the upper Owens River watershed on the east slope of the Sierra Nevada in California. These reaches included one ungrazed control and eight grazed sites that have been exposed to varied levels of cattle grazing. Benthic invertebrate sampling and habitat surveys were conducted in the spring of 1993?1995 and fall of 1992?1995, framing conditions before and after the cattle-grazing season each year. Fish population censuses were made in the fall of 1993?1994. This approach allowed us to examine both differences among sites with similar environmental settings, and changes within sites before and after grazing season.

Stream surveys involved measuring habitat features known to be affected by livestock grazing impacts, along with regular sampling of benthic invertebrates, and periodic censusing of fish. Benthic invertebrate data were evaluated using bioassessment analysis, which partitions community structure into taxonomic diversity, stress tolerance, and trophic components. Analysis involved progressive screening of physical attributes of habitat to enable future predictions of the expected biological response to grazing stress. Based on the significance of correlation coefficients, 7 of 10 habitat variables and 10 of 20 invertebrate community metrics were selected. Multiple regression models developed for each metric identified another subset of habitat variables that would form an integrated measure of benthic habitat quality (used to relate biological responses to the effects of grazing). Ten invertebrate metrics also were combined into a standardized multimetric index (BICI), used as an indicator of biological integrity to assess differences between sites (ungrazed and grazed), and seasonal changes before and after grazing within each site.

Summary/Accomplishments (Outputs/Outcomes):

The habitat changes observed between sites and seasonally were consistent with spatial gradients of grazing impact over the study area, and with declining seasonal habitat quality during exposure to grazing. Though site quality correlated year-to-year and seasonally, poor quality sites were more varied between years and more degraded in the fall than sites of higher quality. The ungrazed reach had consistently superior and more constant habitat conditions. Increased spring-to-fall embeddedness and algae-aquatic vegetation cover were the most pronounced substrate features affecting benthic invertebrates. Metric regression models showed these to be the most significant variables, but also that embeddedness (along with percent fines) was most important in spring, while algae-aquatic vegetation cover dominated in fall. The primary determinants of habitat impacts on benthic invertebrates in the rangeland streams in this study appear to be substrate quantities buried from below by sediments (embedded) and covered from above by algae (and aquatic vegetation) growth. These results validate the intuitive appeal of this portrayal of habitat alteration and, together with the related features of substrate size and percent fines, comprise the composite substrate quality index and are critical habitat attributes in assessing potential grazing impacts on benthic stream communities.

The invertebrate metrics most responsive to habitat quality differences were similar to those reported in other studies, but changed seasonally. In general, richness metrics (total taxa, EPT) were most effective in spring, and tolerance metrics (intolerant, tolerant taxa) in the fall. Pregrazing disturbance metrics may respond to the overall range of habitat quality and availability (richness), while the most effective postgrazing metrics may be those that distinguish effects of stress or disturbance (tolerance values). This group of metrics showed that when substrates were embedded, composed of fines, covered by algae and aquatic vegetation, or small, there were significant losses in taxa richness and sensitive indicator groups, and/or increases in tolerant indicator groups.

The relationship between composite substrate quality and the multimetric BICI was significant in spring but not in fall, when sites with otherwise good habitat showed declines in BICI. The loss of biological integrity seen on these stream reaches may be due to the spring-to-fall increase in embeddedness and algae cover. Other grazed reaches also showed these seasonal increases, but without diminished biological integrity. Those reaches were either in a preexisting state of poor habitat quality and lower BICI, or were resistant, possibly due to large substrates and less cumulative upstream grazing.

Trout in the streams studied here attained very high biomass on the ungrazed reach and were correlated with a composite index of fish habitat quality over all sites. Where stream habitat had little riparian vegetation or bank overhang pool cover, trout biomass was reduced and more tolerant native fish increased in abundance. Physical and biological assessments of habitat condition indicated that while some grazed reaches were unimpaired, others were in a marginally impaired state year-around or only during the fall at the end of the grazing season. Under the most degraded habitat conditions, another site continued to show significant impairment at all times.

Ample evidence shows that extreme overgrazing degrades hydrologic features, riparian habitat, and wildlife resources of streams (Ohmart 1996, Platts 1991). Although, less clear is the extent of ecological degradation caused by lower grazing intensities, and methods to detect the magnitude and timing of impairment to biological integrity of in-stream ecosystems. Study results suggest that bioassessment can detect varied degrees of impairment on rangeland streams relating mainly to benthic substrate quality, and that monitoring should include a seasonal component to track changes before and after grazing on pastures with cyclic annual grazing.

Though the grazing disturbance regime documented in this study undoubtedly represents a narrow range of the habitat conditions found on rangeland streams, the patterns of biological response may be generalized to other situations. The benthic invertebrate community responded most to changes in substrate quality, although both the indicator metrics and the dominant substrate features changed with season. Taxonomic richness measures were most sensitive in the spring when habitat quality was dominated by embeddedness and fine sediment deposition, while tolerance measures were most reliable in the fall when algae cover of open canopy streams was extensive. Seasonal changes found before and after grazing suggest that spring and fall monitoring provides useful assessment information. Using seasonal and cumulative comparisons of biological integrity among sites, designations that are intermediate between impaired and unimpaired can be made to distinguish perennial marginal habitats from seasonally marginal.

The results of this study could be applied as: (1) a screening tool for evaluating habitat factors contributing to ecological degradation by grazing; (2) biological measures for comparing the level and timing of impairment between stream reaches located in different pastures; and (3) a method for monitoring recovery of biological integrity in degraded streams under various grazing management treatments.

Supplemental Keywords:

RFA, Scientific Discipline, Waste, Water, Ecosystem Protection/Environmental Exposure & Risk, Water & Watershed, Hydrology, Ecosystem Protection, exploratory research environmental biology, Chemical Mixtures - Environmental Exposure & Risk, Contaminated Sediments, Environmental Chemistry, Ecosystem/Assessment/Indicators, Chemistry, Ecological Effects - Environmental Exposure & Risk, Ecological Effects - Human Health, Biology, Watersheds, Ecological Indicators, ecological exposure, rangeland, aquatic biota , ecosystem assessment, landscape indicator, erosion, stream ecosystems, contaminated sediment, land use effects, runoff, ecological impacts, aquatic ecosystems, assessment methods, bioassay, environmental stress, water quality, vegetation

Progress and Final Reports:

Original Abstract
  • 1996
  • 1997 Progress Report