AN ASSESSMENT OF THE BIOLOGICAL INTEGRITY OF THE
WESTERN CANADIAN RIVER BASIN IN TEXAS

May 1998

This paper presents the results of a bioassessment study of the western Canadian River Basin in Texas. The study was performed under a monitoring plan developed under the Texas Clean Rivers Program (CRP) during July of 1997. Under this plan, biological monitoring plays a major role in supplementing chemical and physical data to provide a complete assessment of water quality in the Canadian River Basin.

The primary objective of this study was to assess the water quality of streams within the western Canadian River drainage in Texas. Collections of fish (Appendix A) and aquatic macroinvertebrates (Appendix B) were collected at 8 sampling stations, which encompass the entire Canadian River Basin at and west of Lake Meredith. These biological collections were then used to calculate a biological integrity score using Rapid Bioassessment Protocols (Plafkin et al., 1989). By performing habitat assessments at these stations, it is possible to determine biological impairment due to water quality, as opposed to natural habitat characteristics.

Table 1 provides a description of these stations and Figure 1 shows a map of the stations. These stations are exposed to several different land uses having potential to impair water quality.

These include urban activities, such as wastewater effluent discharges and stormwater runoff carrying excess nutrients and pollutants. Secondly, these stations ( to a varying degree) encounter the potential degradation from farming activities, including sedimentation and excess runoff carrying various pollutants. Lastly, these stations are exposed to ranching activities, which results in sedimentation, introduction of fecal pathogens, and deterioration of the riparian zone/stream banks. Table 1 displays which of these activities are predominant at each of the monitoring stations.

The western portion of the Canadian River watershed is from the Texas-New Mexico state-line downstream to the Sanford dam of Lake Meredith (Oldham County to Hutchinson County). There are two subwatersheds in the reach totaling 3,760 square miles of contributing drainage area in Texas and New Mexico (3,108 square miles in Texas). Average annual rainfall for this region is 15 to 20 inches a year. Mean streamflow of the main stem measures 287 cubic feet per second.

The streams of this region are generally low to moderate gradient and possess clay/mud/sand substrates. Vegetation presence is minimal, dominated by grasses. Overall physical habitat in this region is rather inadequate for biological communities to thrive. Limiting habitat characteristics for aquatic life in this region include the lack of suitable habitat, poor bank stability, limited vegetation, and low to no flow conditions. Chemical properties of the waters in this region is characteristic of moderately high total dissolved solids. This is common for the mainstem of the Canadian, however, some tributaries generally possess lower concentrations of total dissolved solids. These total dissolved solids originate from brine inflow below Ute Lake, New Mexico.

Pollution tolerance values for macroinvertebrates were assigned based on values established by the Texas Natural Resource Conservation Commission (personal communication - Bill Harrison), the EPA (Plafkin, 1989) or from Lenat, 1993. Functional feeding groups for macroinvertebrate families were assigned based on classifications obtained from Merritt and Cummins (1996). Several insect families have been assigned multiple functional feeding group classifications by Merritt and Cummins (1996). In order to clarify these classifications, organisms were further identified to genus, and when possible, to species (Merritt and Cummins, 1996, Parrish R.K., 1975, and Thorp and Covich, 1991). The data thus recorded was used to score seven metrics for a slightly modified Rapid Bioassessment Protocol II (RBP II) (Plafkin et al., 1989 and Barbour et al., 1992).

The following metrics were scored in the study:

1. Taxa Richness = total number of taxa collected at the site.

2. EPT Index = number of genera belonging to the Orders Ephemeroptera, Plecoptera, and Trichoptera which were collected.

3. Community Loss Index = (taxa richness at the reference site - taxa common to reference and sampling sites)/taxa richness at the sampling site.

4. Family Biotic Index (modified) = S (xi ti)/n, where:

xi = number of individuals within a genus,

ti = tolerance value for the taxa,

n = total number of organisms in  the sample.

5. Percent Contribution of Dominant Taxa = percent contribution of the dominant taxa to the total number of organisms collected.

6. Ratio of Scrapers/Filtering Collectors = number of scrapers/(number of scrapers + number of filtering collectors).

7. Quantitative Similarity Index = (Functional Feeding Groups or FFG’s) - compares two communities in terms of presence or absence of FFG’s, also taking relative abundance into account.

Each metric value obtained was given a Biological Condition Score of 0, 3, or 6, based on its percent comparison to the metric value obtained from reference station data. Scoring criteria for the Percent Contribution of Dominant Family was expressed as the actual percent contribution, not percent comparability to the reference station. The Community Loss Index and Quantitative Similarity Index (FFGs) were not given a percent comparability to the reference station, because a comparison to the reference station is incorporated into these indices. The metric scores for each sampling site were totaled, and compared to the total metric score for the reference site to obtain the Percent the reference site to obtain the Percent Comparability to Reference Site value. The reference site in this study was Coetas Creek, due to it’s superior biological and physical health, as well as minimal environmental disturbances.

A habitat assessment matrix was completed for each sampling site using forms taken from Plafkin et al., 1989. The matrices were evaluated to determine percent comparability of habitat between the sampling sites and the reference site. The percent comparability was used to judge the potential for each sampling site to support a similar level of biological health compared to its reference site.

The sampling methodology for this study consisted of using a D-framed kick net to sweep woody habitat, vegetation, and to perform kick samples within riffles. Due to inconsistencies in the amounts and types of habitat between the sampling sites, a perfectly consistent sampling method could not be implemented. Therefore, the sampling method consisted of performing 5 minute sweeps, partitioning the 5 minutes between the most suitable macroinvertebrate habitats. The different habitats are sampled approximately the same percentage as they are present. For example, if the habitat consisted of approximately 60 percent bank woody habitat and 40 percent channel woody habitat, then the bank woody habitat would be sampled for 3 minutes and the channel woody habitat would be sampled for 2 minutes. If less than 100 macroinvertebrates are collected after the 5 minutes, then another 5 minutes of sampling was conducted.

Fish species were identified and assigned origin groups, tolerance values, and trophic levels (Plafkin et al, 1989; Hubbs, et al, 1991). This information was used to score 8 RBP V (or IBI) metrics (Plafkin et al, 1989). RBP V allows for some discretion in selecting individual metrics for analysis. The metrics used to calculate an IBI score for these stations was a modified version developed for fish in the Western High Plains and Southwestern Tablelands (Ecoregions 25 and 26) (Linam, personal communication).

The 8 metrics which were evaluated in this study were:

All metrics, except "total number of individuals" were scored according to the previously cited references. The "total number of individuals" metric was changed due to different seine mesh sizes being used in this study as compared to mesh sizes used at stations at which the metric scoring criteria were calculated. Smaller mesh sizes were used at stations used in calculating the metrics, resulting in a greater number of fish to be collected. A score of 1 would have been obtained for this metric at all stations, therefore, an alternate methodology was used. The total number of individuals from each of the 10 stations were ranked and the value at the 90^th percentile was deduced. This number was split into thirds, giving the three ranges for the metric scoring. This is consistent with the methodology used for the calculation of other metrics.

The metric scoring consists of each metric receiving a 1, 3, or a 5, based on its numeric value. The 8 metric scores for each sampling site were totaled to obtain an Index of Biotic Integrity score. Each sampling site was classified as being in Limited (<24), Intermediate (24-33), High (34-35), or Exceptional (>36).

Fish were sampled at each site, using a backpack electroshocking unit and/or seining techniques. Due to differing conductivity levels and habitat, a consistent fish sampling methodology could not be implemented. The normal protocol for fish sampling calls for 15 minutes of seining and 7 seine hauls. However, if electroshocking could not be conducted, then 10 seine hauls were performed. Additional seine hauls were performed if a new species was collected in the final seine haul, until a seine haul with no new species occurred. Sampling gear consisted of a Model 12-B Backpack Electrofisher, a 10 foot seine with ¼ inch mesh, and a 25 foot bag seine with ¼ inch mesh.

All habitats, such as snags, rootwads, riffles, and undercut banks were sampled if present, in order to maximize the capture of different fish species. A representative of each species was preserved in 10% Formalin solution and returned to the lab for identification. All other fish collected were released, unless identification could not be performed in the field. All fish were identified to species and the number of each species was recorded

The overall biological integrity for this geographical region falls into moderately impaired category. The results are shown in Table 2. Appendix A contains fish data and Appendix B shows macroinvertebrate date for each station. RBA (invertebrates) scores range from 38.5 to 107.7, with the majority showing moderate impairment. RBA (fish) scores ranged from 20 to 38, with most scores falling in the Intermediate category. Results for this region indicate that physical factors are the prime causes of biological degradation at these stations. These physical limitations include inadequate substrates, lack of vegetation, sedimentation, and harsh weather conditions. Another major factor in limiting biological community health is the presence of elevated levels of total dissolved solids in this region (i.e. main stem of Canadian River).

It is interesting that the only station greatly affected by anthropogenic activities scored in the highest categories for both macroinvertebrate and fish methodologies. The flow from this station is comprised of runoff from the city of Amarillo and monthly effluent discharges from Amarillo’s River Road Wastewater Treatment Plant. The reason for the high biological integrity scoring can be attributed to the presence of relatively good physical habitat and additional flows, which aids during the hot, dry summers in this region. These hot, dry summers are a major factor for biological communities, as seen at station Punta de Agua @ Hwy 767. This station generally has running water all year, except in mid to late summer. During these periods of no flow, all aquatic biota must survive in pools spread out along the creek channel. The physical habitat at this

station consists of abundant fish cover, high-quality clear water, mud substrate, and lacks stream banks. This habitat should support biological communities, however, due to the yearly "no flow" conditions, it is believed this negatively impacts the health of biological communities. Additionally, the results indicate that the macroinvertebrate community endure the conditions presented at this station.

The station, Big Blue Creek @ FM 1913, is characterized by shallow clear water running over a sandy substrate. In performing a habitat assessment for this station, it is evident that this station has a non-supporting habitat for sustaining healthy biological communites. The RBA scoring for the fish results contradict this assumption, in that the score (34) fell in the "High" category. The fish collection at this station consisted of 65 plains killifish, 5 bluegill, and 1 green sunfish. This collection did not appear impressive, however, scored well due to the relatively high number of fish (i.e. plains killifish), the presence of 2 sunfish species, and the dominant percentage of insectivores. The macroinvertebrate collection produced completely opposite results. Only 14 organisms were collected after 10 minutes of collection time. This illustrates a very low abundance of macroinvertebrates, which would be expected with the lack of suitable habitat. An RBA score was derived for this station, even though 14 macroinvertebrates are not adequate for this protocol. Interestingly, 8 taxa were identified within the 14 organisms collected, as the score (46.2) fell in the moderately impaired category. Although results for this station were conspicuous, it is evident that this station is physically limited and biological impairment would mainly be due to these limitations.

The three stations, Coetas Ck on Alibates Ranch, Chicken Ck on LX Ranch, and Bonita Ck on LX Ranch are three small aquifer fed creeks in close proximity to one another. Coetas Ck and Chicken Ck showed similar biological results, as these streams are physically similar. Macroinvertebrate collections from both of these stations scored the highest in the study (non-impairment) and RBA (fish) scores for these stations showed "Intermediate" health. The scores were lowered due to a low number of species, no Cyprinid species collected, and only one sunfish species (green sunfish) collected at each station. Overall, the biological health at these two stations are excellent for this region. Bonita Ck on LX Ranch, however, shows more impairment in its biological communities. The RBA score (fish) shows Limited health, as the macroinvertebrates showed Moderate Impairment. The physical habitat for this creek was somewhat different from the above two stations. There were beaver dams along the stream, which pooled up much of the stream. Sedimentation was much greater at this station, affecting the suitability of the substrate (as compared to the cobble substrate of the other two stations). This station lies in very close proximity to a home and picnic area, which can cause degradation by certain anthropogenic activities.

The two stations on the Canadian River, differ from the other stations in the study, due to it’s water quality characteristics. The Canadian River is characterized by highly turbid, moderately saline water. The river is also characterized by a low quality of physical habitat for the support of biological communities. Macroinvertebrate collections were very limited in number for both of these stations, as only 15 to 16 organisms were collected in 10 minutes. Fish communities for both stations showed Intermediate health. Impairment is most likely caused by a combination of elevated total dissolved solids and less than optimal physical habitat.

Overall biological integrity for the stations sampled in the western region of the Canadian River in Texas appears to be Intermediate. Detected impairment obtained through the calculation of Rapid Bioassessments can be attributed mostly to natural conditions (i.e. poor natural habitat and harsh weather patterns). Anthropogenic activities play minor roles in water quality and quantity in this area, with the exception of the station East Amarillo Ck @ US 287, where man’s activities actually augment flows needed for healthy biological communities to flourish. It is evident that all streams and rivers in this region are exposed to harsh weather patterns, such as wide temperature fluctuations and very dry conditions. In combination with poor physical habitat and natural chemical chcaracteristics, it is evident biological communities are at a disadvantage, as compared to other regions of the state.

These stations will be revisited at a later date, in order to detect long-term trends in quality for this region of the Canadian River Basin.

Barbour, M.T., J.L. Plafkin, B.P. Bradley, C.G. Graves, and R.W. Wisseman. 1992. Evaluation of EPA’s Rapid Bioassessment Benthic  Metrics: Metric Redundancy and Variability Among Reference Stream Sites. Environmental Toxicology and Chemistry. 2:437-449.

Harrison, Bill. 1997. Personal communication through handout. Texas Natural Resources Conservation Commission.

Hubbs, Clark, Robert J. Edwards and Gary P. Garrett. 1991. An Annottated Checklist of the Freshwater Fishes of Texas, with Keys to Identification of Species. The Texas Journal of Science. 43(4): Special Supplement.

Lenat, David R. 1993. A Biotic Index for the Southeastern United States: derivation and list of tolerance values, with criteria for assigning water-quality ratings. Journal of North American Benthological Society. 12(3): 279-290.

Linam, Gordon. 1998. Personal communication. Texas Parks and Wildlife Department.

Merritt, R.W. and K.W. Cummins. 1996. An Introduction to the Aquatic Insects of North America. Second Edition. Kendall/Hunt Publishing Company. Dubuque, Iowa.   722 pp.

Parrish, Fred K. 1975. Keys to Water Quality Indicative Organisms of the Southeastern United States. United States Environmental Protection Agency. 2^nd Ed.   Cincinnati.

Plafkin, J.L., M.T. Barbour, K.D. Porter, S.K. Gross and R.M. Hughes. 1989. Rapid Bioassessment Protocols for use in Streams and Rivers: Benthic Macroinvertebrates and Fish. EPA/444/4-89-001.

Thorp, James H. and Alan P. Covich. 1991. Ecology and Classification of North American Freshwater Invertebrates. Academic Press, Inc. San Diego. 911 pp.

Go To Beginning

Home