A Physical, Chemical, and Biological Assessment of Buffalo Creek

A Report by:

Lycoming College Clean Water Institute Interns

Laura Lockard

Brad Musser

Dr. Mel Zimmerman

December 7, 2004

During the summer of 2004, Lycoming College Clean Water Institute interns began work on a two-year project aimed at identifying and monitoring the condition of Buffalo Creek and its five major tributaries. The main branch of Buffalo Creek stretches a total of 28 miles and drains an almost 134 square-mile watershed (Versar). The creek begins along Kessler Trail in the heart of Bald Eagle State Forest, passes through the outskirts of Mifflinburg and Lewisburg, until finally converging with the Susquehanna River. The first year of this project was predominantly devoted to documenting the frequency and severity of erosion along the banks of the main branch of Buffalo Creek. Preliminary water chemistry data and macroinvertebrates were also collected from the main branch of the creek during this time.

The potential for bank erosion was determined by a combination of bank height, bank angle, the density of roots present, and the particle size of the bank substrate. Each of these factors was rated for high, moderate, or low erosion potential. Banks less than 6-feet high were considered to have low erosion potential. Banks between 6 and 9-feet high were considered to have moderate erosion potential, and banks over 9-feet high had high erosion potential. Erosion potentials based on bank height along Buffalo Creek are summarized in Table 2 (Chart 2.1 thru Chart 2.3). Bank angles of less than 45 degrees were said to present low erosion potential, bank angles of 45 to 90 degrees presented moderate erosion potential, and bank angles greater than 90 degrees (undercut banks) presented high erosion potential. Erosion potentials based on bank angle along Buffalo Creek are summarized in Table 3 (Chart 3.1 thru Chart 3.3). The concentration of roots throughout a bank is also a prime indicator of the erosion potential of that bank. Roots fortify the bank substrate, preventing erosion. Bank surfaces more than 60% covered with well-rooted vegetation are considered to have low erosion potential based on root density. Bank surfaces 30% to 60% covered by well-rooted vegetation were considered to have moderate erosion potential, while banks less than 30% covered by well-rooted vegetation had high erosion potential. Erosion potentials based on bank root density along Buffalo Creek are summarized in Table 4 (Chart 4.1 thru Chart 4.3). The particle size of bank substrate also helps determine erosion potential. Large particles, such as bedrock or boulders, were considered to present low bank erosion potential, while basketball-size rocks to pebbles created moderate bank erosion potential, and fine particles like sand and clay produced a high potential for bank erosion. Erosion potentials based on bank particle size along Buffalo Creek are summarized in Table 5 (Chart 5.1 thru Chart 5.3). As shown in Table 1 (Chart 1), erosion made up 59.6% (366 total) of the 614 total disturbances along Buffalo Creek. Of these 366 erosion sites, 170 (46.4% of 366) were on the left bank, while 196 (53.6% of 366) were on the right bank.

In addition to erosion, other disturbances, possibly caused by human influence, were identified and documented along the entire stretch of the creek (see Table 1). There were 614 total disturbances recorded throughout the creek. Of these disturbances, 45 (7.33%) were gravel bars. A gravel bar is a disturbance usually formed when severe upstream erosion or sedimentation causes the formation of a downstream gravel deposit that protrudes above the water surface. Similarly, mid-channel bars are often formed on well-developed gravel bars and characterized by the existence of permanent vegetation. In all, 68 of the 614 (11.1%) disturbances were mid-channel bars. Point sources made up 8.14% (50 out of 614) of the total disturbances. A point source is, by definition, a specific location where pollution, wastewater, or run-off water enters a stream in measurable quantities. The term is used in this paper to denote all pipes and drainage ditches. Rip rap is a general term used to describe any material, most often limestone boulders, that has been placed over a bank in hopes of preventing further bank erosion. Although this method of bank preservation seems to succeed in preventing erosion of the particular bank on which it is placed, fortifying the bank in this way causes water pressure to build up against the material and surge with greater force against banks downstream, causing them to become eroded as well. There were 32 instances (5.21%) of rip rap along Buffalo Creek. The term ‘debris deposit’ is used to describe any substantial build-up of natural material, such as logs, sticks, and leaves, which obstructs the width of the water pathway and hinders normal water flow within the creek. Only the most severe debris deposits are documented in this report since nearly all can be naturally flushed downstream during high-flow periods. Only 2 (0.33%) debris deposits were severe enough to be included in this data. There were 18 (2.93%) tributaries documented along the creek. Here, the term ‘tributaries’ encompasses the five major tributaries of Buffalo Creek (Rapid Run, Spruce Run, Stony Run, North Branch, and Little Buffalo Creek) as well as all minor tributaries identified during this study. Bridges (those large enough to support an automobile) and footbridges (those only suitable for pedestrian crossings) were also included as disturbances because they can cause stream channelization and divert the natural water pathway. There were 25 (4.07%) bridges and 8 (1.30%) footbridges along Buffalo Creek.

The strengths of the creek’s in-stream habitat and riparian buffer systems were also investigated in the field. Two standardized forms were completed at sites approximately every mile and a half along the length of the creek. These forms detailed specific properties used to determine the overall habitat quality in and around the creek bed. The first of these sheets, The EPA Habitat Assessment Field Data Sheet (Figure 1), focused mainly on riffle/run prevalence and the presence of adequate aquatic habitat to estimate the ecological fitness of each site. Each factor was evaluated based on the category ranges (0-20) provided on the data sheet. The sum of the category scores for the sites are presented in Table 9. The Riparian Assessment Data Sheet (Figure 2) was the other form completed at each of these sites in order to record the health of the riparian buffer surrounding the creek. The parameters listed on this data sheet were evaluated using the accompanying Riparian Assessment Guide (Figure 3). Scores could range from 1-10 for each parameter, included bank and riparian vegetative width, thickness, and type. Land use outside the buffer was also recorded. Once all parameters were evaluated, the sum was compared to the highest total possible to produce a percentage. This percentage fell into a range from 0% to 100% (0% being the worst possible and 100% being the healthiest possible) and was used to estimate the condition of the riparian buffer. The percentages and riparian evaluation score for each site is presented in Table 10.

Buffalo Creek was at one time considered an exceptional trout stream, however, the combined pressures of increased development, irresponsible agricultural practices, and airborne acid deposition have perhaps contributed the most to the degradation of the creek’s aquatic habitat (Versar). In an effort to at least quantify the degree of chemical impairment of Buffalo Creek, water samples were collected and analyzed for the months of June and July from six sites along the length of the water body. These sampling sites were chosen based on their proximity to possible sources of chemical influx and the ease at which they could be accessed by car. Table 6 shows the location of each water chemistry testing site.

Two 1000mL plastic bottles were filled with water at each site. Each bottle was rinsed twice with creek water before being filled a third time to ensure that no residue would remain in the container and possibly impact the results of any chemical analyses. Dissolved oxygen readings were recorded, as well as site length, width, depth, and stream velocity. The amount of nitrates, nitrites, phosphates and orthophosphates in each bottled sample were determined using spectrophotometric techniques within the lab. The pH and alkalinity of each sample were also ascertained using an in-lab pH meter. Average results of all chemistry data was calculated for each site and the values are displayed in Table 7 and Table 8.

Macroinvertebrates were collected in July at each of the water chemistry testing sites. Samples were collected using the kick-net technique, in which a flat seining net is held at an incline facing upstream. The substrate one square-meter directly upstream from the net was kicked or disturbed in some manner so the dislodged organisms would wash into the net. The contents of each kick-net (debris and macroinvertebrates) were placed in plastic, screw-cap jars and preserved in dilute 70% ethanol. Once in the lab, all macroinvertebrates were separated from debris collected with the original samples and stored in 25mL glass vials of 70% ethanol. A random sample of 200 macroinvertebrates will be identified from each site at a later date. The results will be tabulated using an EPA Rapid Bioassessment Protocol and the prevalence of certain indicator species will help estimate the pollution level of Buffalo Creek.

Works Cited:

Versar Inc., Biological and Hydraulic & Hydrological Investigations of Buffalo Creek

Watershed, PA.