|
     |
1998 Fall Assessment of Peavíne CreekHNE 385Ellee Austin, Amelia Ravin, and Aubrey Mescher with Data Collection Efforts of HNE 120Submitted to: Dr. WegnerCONTENTS
|
|||||||||||||||||||||||||||||||||||||||
The concept and necessity of sustainability on Earth is being continually approached by ecologists and economists alike; it is of universal concern. There is no doubt that human influence has a potentially devastating impact on the global environment. With human populations continually increasing, the resulting environmental stress is of paramount importance. The use of an environmental assessment function to determine the health and condition of a specific ecosystem. Through the study of an ecosystem's present health, it is possible to assess the effects of on-going human activity and predict future environmental impacts.
Environmental assessments address sustainability concerns. Urban sprawl and development, industrial pollution, and agricultural management are just a few factors that influence sustainability on a large scale. By using environmental assessments to determine options for sustainable development, environmental conservation and management can be approached from a precautionary standpoint, as opposed to current "emergency room" procedures. The application of environmental assessments, particularly on a large scale, will also generate public awareness, a key factor in any movement as massive and influential as environmental protection.
The Georgia Adopt-A-Stream (AAS) program is one example of an environmental assessment. Because ecosystems are extremely complex and interrelated, there are many factors that support normal functioning and can be focused on during an assessment. Unless a researcher follows predetermined guidelines as in the AAS manual, he/she must determine how to approach an assessment through the specific elements that he/she will study. The AAS program does allow for variation in the extent to which an assessment may be conducted, in that it distinguishes between Level I, II, and III monitoring procedures. Level I involves regular visual assessments, basic clean-up, and public outreach activities (AAS manual). Levels II and III are more comprehensive than Level I. They require training sessions for biological monitoring, chemical testing and habitat enhancement procedures (AAS manual). The AAS program is designed to focus on relatively small areas of land: it provides the information needed to conduct a small-scale environmental assessment.
Operating on the same basic principals as AAS, the Environmental Monitoring and Assessment Program (EMAP) is an example of large-scale environmental assessment. A division of the Environmental Protection Agency (EPA), EMAP is an organization that monitors the distribution of ecological resources on a national level and continually evaluates their condition (Vogt et al. 252). This large-scale, long-term monitoring reveals trends in environmental change that may be unnoticeable at the local level, but devastating over a period of years. For example, atmospheric monitoring has revealed a global warming trend, commonly described as the greenhouse effect. Also using procedures consistent with more extensive environmental assessments, scientists have been able to track direct consequences of the greenhouse effect. Such as rising sea levels, melting glaciers, and increasingly severe weather patterns.
Environmental assessments are highly valuable on a long-term scale. Consistent monitoring reveals the normal characteristics of an ecosystem, thus providing a basis of comparison for trends that appear to be abnormal. In addition. the interpretation of long-term monitoring results often identifies environmental change that represents damage on a more comprehensive level. For example, the gradual climbing of global atmospheric temperature indicates a problem with the ozone layer, which normally functions to stabilize temperature.
Assuming that our work on Peavine Creek is the first of many assessments in a long‑term program. It may be used to reveal trends specific to that ecosystem, as well as a more comprehensive analysis of environmental damage. Our work was done mainly during the months of October and November. In future years, classes or research groups can follow the same procedures during the same months and using our data and interpretations as comparison, determine how and why the stream's condition has changed, if at all. It will then be effective to develop management procedures and guidelines that allow for the gradual rehabilitation of the existing ecosystem while also avoiding actions that are known or thought to be harmful.
Ideally, the goal of an environmental assessment is to assess the ecological integrity or ecological health of an ecosystem or region. In addition, determining the type and degree of disturbance and anthropogenic impacts are usually a part of an environmental assessment. Unfortunately, it is nearly impossible to achieve such measurements. The main problem is defining "ecological integrity" as both a concept and as something that can be measured concretely.
Ecological integrity is a complex concept, one that can be defined and measured in several ways. This ambiguity can create conflicts for environmental management and assessment. The phrase itself implies that ecosystem structures and functions are unimpaired by human caused stresses (Woodley 1993a). Virtually all ecosystems are impaired or impacted to some degree by human activity. Therefore, deciding what is healthy compared to disturbed, and assessing the degree of disturbance can become an issue of values. Placing value on the function of an ecosystem for the output of resources for exploitation often yields different ideas about the integrity of the ecosystem than focusing value on the natural processes and biodiversity of the system. In addition. there is also a different interpretation of integrity when assessing the functional (nutrient cycling, energy flow) aspects versus the structural (biodiversity and trophic levels) aspects of the system (De Leo and Levin 1997).
Measuring ecosystem integrity and overall ecosystem health is also a complicated issue. There is no universal simple measure of either ecosystem health or of a good ecosystem state. Characterizing integrity is problematic because ecosystems are not static; they change through natural processes as well as a result of human impact, and these changes are often erratic and unpredictable (Noss 1995). It is difficult to determine how far the current state is from the non-human-impacted ecosystem state (Vogt et al. 1997). It is also difficult to detect the resilience or resistance of ecosystems to various types of disturbances (Vogt et al. 1997).
Instead of measuring ecosystem health, it is only possible to measure certain variables for comparison to what is expected or desired as healthy (Vogt et al. 1997). There are several factors and variables that can actually measured and have proved useful in detecting the state of ecosystems and their response to disturbances. These include measuring energy flow or primary production, monitoring macroscale properties such as tree size or the number of large carnivores, and monitoring indicator species.
Indicator species are an effective alternative to whole ecosystem environmental assessment and monitoring. It is very difficult to design experiments to assess the functional and overall health of an ecosystem. To do this one would be inclined to measure natural processes on large scales, something that not feasible for practical purposes. Instead, it is sometimes possible to isolate one or two species that indicate the state of the ecosystem and the level of disturbance. Indicator species are organisms whose characteristics, either absence, presence, abundance, dispersion, or reproductive success, can be used as an index of attributes of the environment which are too difficult to measure directly (Vogt et al. 1997). For example, these species can reflect the presence or effects of environmental contaminants or the habitat quality for other species or communities. However. there must be a clear relationship between the indicator species and the ecosystem in question (Vogt et al. 1997). One example of such a relationship is that between the abundance of macroinvertebrates and water quality given by Georgia Adopt-A-Stream program.
This fall, Emory University’s Human and Natural Ecology (HNE) department adopted a one mile stretch of Peavine Creek as part of Georgia's Adopt-A-Stream program (AAS). This program was developed by the Environmental Protection Division to "promote citizen involvement in learning about and protecting streams, rivers, and lakes" through a hands-on approach aimed at involving volunteers, citizens, business, and government. Part of this program included biological and chemical monitoring of the stream habitat that was conducted by students in the HNE 120 classes. Students also assessed the stream and vegetation visually. Although not part of the AAS program's protocol, some students provided a very thorough history of Peavine Creek and the impacts that humans have subjected it to in the past.
The objective of our environmental assessment was to provide information to add to the data collected by the HNE 120 students about the ecological health arid human impact on this one mile section of Peavine Creek. The biological arid chemical data that was collected was intended to indicate the, water quality. We wanted to examine other possible indicators of the ecological state of the creek and the possible effects of urbanization.
When deciding what to measure for our environmental assessment we were confronted by some limitations. First, we were limited in the staffing department. The HNE 385 class this fall consisted of only three students, not including Professor John Wegner. We were also limited by the amount of time we could spend surveying, given the duration of the semester course. In addition, we were at a slight disadvantage because of the fall season, and the difficulties this presented in the course of our work. However, these limitations are not unlike those faced in environmental assessments in general. One of the most limiting factors when conducting all assessment is the scale, both spatial and temporal. When trying to determine the state of an entire ecosystem it is often difficult, because of a lack in resources, to adequately survey the region on a large enough scale or over a long enough time period to be able to make conclusions about the state of the system. Since our assessment was on such a small scale, we were mainly limited by the time and people available to conduct our survey.
We chose to conduct three different studies as part of our environmental assessment of Peavine Creek. There were two main objectives to our study. The first was to determine the overall state of the stream ecosystem and to provide baseline data for future monitoring purposes. The second objective was to assess the possible impacts of urbanization on Peavine Creek by comparing different aspects of the stream environment behind Kroger's in Emory Village and at a downstream section of the creek behind the Asbury House on Emory's campus.
We chose to survey the tree species composition and size because of the visible difference in forested vegetation between the two sites. We also surveyed bird species composition and fish species composition and relative population density at both sites for comparison. Birds are effective for long-term monitoring purposes, and are relatively easy to identify by both sight and sound. We chose fish species as another indicator in part because of the great amount of background information on fish populations in Peavine Creek available from projects conducted by students Biology Department at Emory University. Fish can also be good indicators of water quality.
The two sites we surveyed for comparison are approximately one mile apart from each other along the length of Peavine Creek. Peavine Creek runs through Emory Village and is a major avenue of drainage for Emory University and the surrounding Druid Hills community (Thom and Moseley 1996). Land use patterns in the area illustrate a long history of human disturbances to the stream environment. Up until the early 1900's the surrounding land was used mainly for agriculture and mill factories (Yerkes and Yin 1998). Emory Village construction around North Decatur Road began in the 1920's and has only grown since.
Peavine Creek runs directly through Emory Village, bordered by parking lots. Directly upstream lies the Druid Hills Golf Course (see MAP I ), the green consuming the entire stream bank on one side. Runoff pipes line the banks of the creek through the village, which are several feet high to accommodate the amount of water that floods the creek during heavy rains. We compared the stream directly behind the parking lot at Kroger's to a section downstream, behind the Asbury house and the baseball field on Emory's campus (see MAP 2). Although part of this section is bordered on one side by the road, the overall appearance is very different compared to the stream at Emory Village. At the Asbury House the creek is surrounded by vegetation and forested area, not parking lots and golf green.
Our research of vegetation at Peavine Creek focused exclusively on trees. The procedures used to sample vegetation at Kroger and at Asbury are as follows:
1. At each site, establish random plots measuring 10m x 15m.
2. Identify every tree >2.5cm dbh within each plot.
3. Record dbh of every tree.
The randomization of sample plots is important because plots that are arranged in close proximity to each other or in areas with identical characteristics may not be representative of the entire site of study. To eliminate complications involved if differentiating between young trees and other shrubs, we established that any plant with a diameter greater than 2.5 cm at chest height qualifies as a tree, while all others are shrubs. This method simplifies sampling procedures and also provides an accurate analysis of tree growth. The dbh of each tree is useful by providing a size comparison between sample sites.
Species identification within each plot provides a means of analysis and comparison between sites. The health and structure of a stream is directly influenced by nearby vegetation. Stream bank and habitat stability are largely determined by vegetation health, structure, and diversity. Therefore, it is essential to determine the diversity and density of trees of each plot as well as between them, as this information is indicative of the stream's general characteristics.
When environmental monitoring was in its Infancy, only chemical testing was used to determine water quality. Ecologists learned this was not enough so they added fish and invertebrate biodiversity indexes to come tip with more accurate conclusions. There is another problem. Current programs are forgetting to include species that drink water and use its resources to survive such as birds. They are as much of the system as those species who live in Peavine Creek. "Until an integrative perspective dominates our collective conscious, the conditions of the rivers will continue to decline" (Karr 1991).
There are multiple advantages for using birds to monitor water quality. First, birds are very easy to identify. Differences in color, size of body, length of tail and beak, create a clear classification key. Not only can birds be recognized by sight, but also by their calls (Furness et al. 1993). Over the years birds have become more and more favorable in the eye of the public. With a much larger appeal than other species used for monitoring, the warm fuzzy factor comes into play. For example, a dead bird will create much more of a outcry than a dead caddis fly.(Miller 1996). It is important to have public backing if changes are to be made. In terms of monitoring, the popularity of birds will not only provoke bird watchers to help out, but more information can be obtained from residents of the surrounding area of interest.
Birds have been studied for a long time relative to other organisms. Researchers have discovered many things about their behavior which will explain changes in distribution or abundance. Using birds as monitors decreases the risk of misinterpreting the ecology of birds as environmental disturbances (Furness et al. 1993). Mobility and eating habits are useful bird attributes to add to the monitoring equation. Since birds move through their territories daily, scientists can record the quality of the entire riparian area instead of just a small stretch of stream. Birds use water for a variety of different reasons. Some only use the water to drink and to bathe in, while others cat tile fish and other organisms. Knowing the birds eating habits allows for a better indication of where the pollution is coming from and its concentration.
Being higher on the food chain than plants, macroinvertebrates and fish, birds indicate chemical disturbances fairly well. Any compounds in overabundance will accumulate as they are moved tip the food chain (Miller 1996). This monitoring technique is advanced for an amateur water monitoring program, but dead birds could be collected and sent to labs to check the concentration of compounds in their fat reserves. Another good indicator of resource quality s the strength of the bird egg shell (i.e. thickness) and tire chemical contents of the yolk (Furness et al. 1993). Testing these components would also require more expertise.
To follow this point, birds are very useful for monitoring environmental stress over a long period of time (Furness et al. 1993). They have a relatively long life span in comparison to the fish and invertebrates being studied. Birds can be banded and followed throughout their life. The effectiveness depends on the length of the monitoring and background history of the local birds to compare with present data. Acute disturbances will not be measurable unless drastic, but chemical and visual tests will be able to detect these variations. As long as Adopt-A-Stream maintains its monitoring fervor, using birds as bio-indicators will be very useful.
First, pick bird species which should be members of the local community. For this assessment, the University of Georgia Bird web site was consulted. This site listed common Georgian birds that are easy to learn how to identify in a short period of time, At the end of the study, this list was narrowed down to just include those species which require riparian habitat. Sensitive species were not used because they are already lacking from the urban landscape.
Second, the bird monitor must learn to recognize the birds by sight and calls. I used Peterson's North America Birds Multimedia Guide for identification. Next, I went out into the field and completed a test trial at each location to find the best censusing location for determining presence and absence. At Kroger, I observed birds along the stone fence at the apartment complex across from North Druid Hills Golf Course overlooking Peavine Creek. At the Asbury location, I watched for birds in a clearing about 500 meters from the edge of the creek. It was at the end of the path parallel to Peavine Creek and was filled with a variety of microhabitats (i.e. scrubs, fallen trees, full canopy, in the background, and a few long-standing coniferous trees). I picked these two spots for their locations next to the creek, for the ease with which I could identify the species, and the variety of birds seen in comparison to other spots in those areas (Map I & 2).
The best time I found to observe the birds was after 10:00 a.m. It is recommended not to go out when it is cloudy or rainy. After 10:00am the birds were active until dusk. There might be a difference during the breeding season, but this has not been tested at these sites yet.
Each site should be visited at least five times during the assessing period, although ten times would be preferable. At each site sit silently for 20-25 minutes. Then walk along a stretch of the creek for five minutes to see what other birds pop out of the brush. Try to walk as silently as possible. In the fall, the leaves make it more difficult. If possible, do not walk down the middle of the creek because the birds hear the splashing long before you can see or hear them. Record presence and absence at each site. Total the findings to determine the frequency of sightings.
Aquatic resources are vulnerable to the effects of human activities, and many of the landscape changes humans induce can cause irreversible damage resulting in long-term biological consequences (Frissell et al. 1996). Urbanized watersheds, such as Peavine Creek, provide a system in which to study the effects of anthropogenic disturbances on fish populations. Seasonal and weather related flooding, sewage and other runoff, and other non-point sources of pollution in Peavine Creek can have drastic effects on fish species diversity and population size.
We have been provided with a historical perspective of land use in the area of the Peavine Creek Watershed by students in the HNE 120 classes (see Yerkes and Yin 1998). In addition, we are also fortunate to have information on the fish populations in Peavine Creek dating back to 1992, from undergraduate student projects in the Vertebrate Population Biology class taught by Dr. Shure in the Biology Department at Emory University. These studies of the Peavine Creek Watershed all found the greatest species diversity and abundance in the section of Peavine Creek in Emory Village. The students suggested the diverse aquatic habitat as a possible reason for this finding. We considered the data from these student projects when designing our study and also for analysis and comparison with our results.
We used seine nets to collect fish for identification and counting. There were two main reasons for this choice of methodology. First, we wanted to be able to compare our results with previous studies that had also used seine nets. Second, we wanted to design an experiment that could be easily repeated by future classes. Although one past study had included the use of an electrofisher, we decided that because seine nets are available through the biology department at Emory, these would be more accessible for future use. We surveyed approximately the same length of stream in Emory Village and at Asbury House for comparison.
Two people are needed to use the seine, one to hold each end. To collect fish we walked upstream with the net rolled out to the length appropriate for the sampling pool, and then pulled in the ends to place the fish in buckets. In order to seine, the depth of the pool must be at least 2-3 feet. This requirement and the presence of fish were the two factors influencing where, or in which pools, we could collect specimens. In some cases, we seined more than once in one location to increase our efficiency at catching fish. In areas where there was substantial vegetation overhang, a preferred stream habitat for fish, we used the kick seine net to collect fish because these areas are inaccessible to the large seine net. Our sampling was done during midday hours and was spread out over a period of 2-3 weeks.
Our objectives were to identify the species present and to sort them according to relative abundance. To do so, we came up with an arbitrary scale for abundance: 1-5 fish were categorized as RARE, 5-20 fish as FEW, 20-50 fish were common, and 50 or more were ABUNDANT. After were finished counting and identifying the fish we returned them to the stream.
MACROINVERTEBRATES/CHEMICAL
See Adopt-A-Stream
Packet
Data from these vegetation sampling procedures show that there is a greater species diversity at Asbury, where 11 different species were found, as opposed to Kroger, where 8 species were found. At the Asbury site, the average number of different species in each plot was eight. At Kroger. the average number was six. Several species present at each site were absent at the other, and most species were present in both areas (Table 4.1.1).
The most noticeable difference between these sites is seen in tree density. At Asbury, the average number of trees per plot was 43, whereas at Kroger the average number of trees was 14. Assuming that these results are representative of the two separate ecosystems, I converted the number of trees found per plot (150 square meters) into the number of trees found per hectare (10,000 square meters). This eliminates any discrepancy that may have occurred due to the fact that there was one more sample plot at Asbury than at Kroger. The conversion reveals that Kroger's ecosystem sustains less than one third the number of trees per hectare that Asbury's ecosystem does (Table 4.1.2).
There are several notable differences between the characteristics of Asbury's tree population and those of Kroger's population. In every Asbury site, there is a clear layering of plant growth. In some of the more dense areas, branches and leaves in the upper canopy are so tangled that it is difficult to distinguish which branches belong to which trees. There is also a fair amount of diversity in the amount of young trees versus older trees. This is indicated differences in size and species-specific characteristics that change as the tree grows such as bark color or texture. The Asbury plots can also be characterized as containing a lot of biomass and woody debris, which is providing nutrients to the developing forest.
In comparison with Asbury, the trees at Kroger are not as tall and more spread out. After a preliminary visual assessment of each site, I predicted that the trees at Kroger would generally have smaller dbh than the trees at Asbury. However, Figure 3 indicates that there is not a significant difference in dbh between Asbury and Kroger. For several species. including the tulip poplar, sycamore, ash leaf maple, and elm, Kroger's average measurements are slightly larger than Asbury's. I can only explain this variance between my results and my prediction by say ing that visual assessments are often misleading. Perhaps Kroger's trees appear to be smaller in diameter because most are much smaller in stature than Asbury's trees (Table 4.1.3).
One of the largest trees found at Kroger, in diameter at breast height, was the magnolia which was not found at all at Asbury. This is probably due to the high degree of variance in the environments surrounding Kroger and Asbury. Vegetation near Kroger is obviously constrained by surrounding developments. On both sides of the stream there are businesses parking lots, fencing, and other developments. These factors influence the amount of vegetation that can grow in the area and to some extent, they also influence the types of trees suitable for the area (pollution tolerant plants will be more successful).
In some areas on the Kroger side, the distance remaining between the stream's edge and the nearest development is often reduced to a steep riparian zone. Vegetation in this zone is essential for a variety of reasons. Riparian vegetation prevents erosion, maintains the stream's course (especially during periods of flooding), regulates water temperature through the presence or absence of shade, and provides a regular source of organic nutrients and materials for the stream. At Asbury, the nearest road runs parallel to the stream, almost always with a distance of at least 20 meters. Although its presence is an imposition on the stream's natural ecosystem, its distance allows for vegetation beyond the riparian zone.
The difference in types of trees, species dominance, and density found at each site in combination with the surrounding levels of urbanization indicate significant variation in habitat between Kroger and Asbury. For instance, the tulip poplar is by far the most abundant species present at Asbury, whereas the maple is dominant at Kroger, and a less substantial margin. The absence of very tall trees at Kroger, like the pines found at Asbury, changes habitat composition considerably and allows for different kinds of plants and animals to occupy each site. For example, birds and other animals that prefer Asbury's large pine trees may not be found near Kroger, where most of the trees are not very tall, and still other plants and animals thrive in ecosystems specifically like Kroger's. This leads me to the conclusion that human activities significantly influence the diversity of species and characteristics of trees and thus, the types of habitat they provide. The first step in ecosystem management, especially areas like Kroger and Asbury, is public awareness and cooperation. Due to impinging urbanization, humans play a central role in ecosystem functions arid management (Christensen et al. 1995). In order to save Peavine Creek and the surrounding area from further pollution, nearby residents and local businesses must be willing to modify their behavior.
As seen in the varied ecosystem composition between Kroger and Asbury, human impact is different on each site. At Kroger, nearby parking lots drain run-off into the stream. This run-off contains oil and grease from the pavement, as well as any chemicals that may have leaked from areas such as the Kroger garbage collection site or air conditioning units. By diverting this run-off and finding a different means of disposal. the stream will remain much cleaner.
At both sites, development is reaching closer to the stream. Management policies that limit this development and stipulate pollution guidelines will help to protect the surrounding ecosystems. Consistent monitoring of the stream is also essential to its future health. This will allow for further predictions of human impact and enable the implementation of ecosystem management strategies before further damage is done.
Out of twenty‑two species, nineteen were spotted at Asbury and sixteen at Kroger. All species missing at both sites have methodological and behavioral characteristics to explain their absence. These reasons are cited in Table 4.2.1. The only difference between Kroger and Asbury (when expertise problems ignored) when comparing the composition of species are the Northern Flicker and the Eastern Towhee. Both were absent from Kroger because these species do not like cities, although they will reside in urban areas. Thus, comparing, these two sites, there is minimal difference in the composition of birds species and no difference as a result of environmental disturbances.
Analyzing the frequency of sightings, fifteen out of the twenty‑two species turned out to have a 20% or less difference between the number of times the species were seen at Asbury compared to Kroger (Table 4.2.2). The seven species which were observed 40% or more of the time over the other site can be explained by the expertise level of the bird monitor and behavior of the birds. For example, mockingbirds are hard to distinguish by call from other birds because they mock their songs. Another example is the tufted titmouse. This species likes to hide in thick brush and is not very vocal in the winter. The frequency of the red-tailed hawk was low, in both sites. According to a golf employee, the golf course is home to a couple of red-tailed hawks, but I never saw any. Typically, this bird is much more abundant in the winter than any other time of the year. Why such a low frequency? More research must be completed in this area
The longer I spent out in the field, the more I discovered about the behavior of the local birds and the more apt I became in recognizing them. For instance, I only began to really be able to identify the difference between some of the smaller species towards the end of the study. I also discovered that certain birds are only seen during certain periods of the day, while others are out during all times of the day. For example, the red-headed and downy woodpeckers were only seen late in the afternoon moving through the trees next to the golf course.
Most of the discrepancies and the low frequency of sightings can be explained by the [low] expertise level of the monitor. Low frequencies can also be explained by the decreased amount of calling in the non-breeding season compared to the breeding season. Both areas were dominated by dense shrubbery which provided good places for the birds to hide. For example, the Brown Thrasher, which is the state bird of Georgia was very hard to detect because after August it is less vocal and prefers to seclude itself from humanity. The birds were also very wary of my presence. Once they detected me, they would move before I could identify them. There were also species, such as the Catbird, that were just concluding their migration to the more southern regions of the state (Burleigh 1958). Therefore, due to the inconsistency in data collecting and behavior of the birds, one cannot assume the absences and low frequency sightings are due to the effects of environmental disturbances.
After completing this study, I would recommend the following birds to be used in further assessments of Peavine Creek.
At the section of stream through Emory Village there were five pools suitable for seining (see Map1), and we found fish in all of them. However, at Asbury House (see Map 2), only two of the four pools that were deep enough to use the seine had any fish. This may indicate something about the relative abundance of fish between the two sites.
We ran into some problems when trying to identify the fish we collected. Essentially, we were lacking in expertise in advanced fish identification skills. While we were able to distinguish between minnow and sunfish. We had difficulty positively identifying three species of minnows. Two of the three unidentified species, the common minnow and the striped minnow, were also the two most abundant species found. For future study we recommend contacting an expert in local fish populations, possibly someone from Fernbank or the Department of Natural Resources.
Our results show no significant difference between the two sites. At Emory Village we found seven fish species (Table 4.3.1). Five of these species, tile bluegill (Lipomis macrochirus), red shiner (Cyprinella lutrensis), unidentified striped minnow, unidentified clear minnow, and the smallfin redhorse sucker (Moxostoma robustum), were also found at the Asbury House in similar densities (Table 4.3.2). Two rare species, the pumpkinseed sunfish (Lepomis gibbosus) and the snail bullhead catfish (Ameiurus brunneus), were not found at Asbury House. Similarly, two species found at Asbury House were not found at Emory Village. One of these species., the redbreasted sunfish (Lepomis auritus), was only found in rare abundance. However, the other species not found at Emory Village was the unidentified common minnow, that was found in relative abundance. This constitutes the only substantial difference we found between the two sites.
However, when comparing‑ our results to past surveys of fish populations in Peavine Creek we found some significant differences. Of the 4 studies completed in the years 1992, 1993, 1995, and 1996, the survey done in 1996 (Thom and Moseley 1996) yielded results most similar to ours. This finding supports the proposal that the variance found in these different years is a result of environmental change due to human impact on the creek, instead of merely natural environmental or seasonal variation.
We identified two species that were not previously documented in Peavine Creek: the snail bullhead catfish and the smallfin redhorse sticker. All of our other identified species were found in the. past four studies. All four of these studies cited the red shiner as the most abundant species, whereas we categorized it as rare or few in relation to the other species found. The survey done in 1995 (Regar and Beasley 1995) showed the greatest species diversity in the creek, identifying 15 species compared to the 9 species we found in relatively the same stretch of the creek.
Overall, rhere seems to have been a shift in species composition to lower levels of diversity, since the start of these studies on fish populations in Peavine Creek. A possible and probable cause of this trend may be the increased human impact and resultant degradation of the urban stream habitat. Increased construction in the surrounding area, especially at Emory, storm water runoff and flooding, and chemical runoff from the Druid Hills golf course all have the potential to affect negatively the stream habitat and water quality. The fact that we did not find any difference between our two study sites may indicate that the effects cannot be isolated on such a small scale. We have already identified possible trends in temporal scale through the use of past studies. We recommend the continuance of these annual studies to monitor temporal trend in fish populations, as well as expanding the spatial scale of the study to include a larger proportion of the watershed.
The data for the macroinvertebrate analysis of water quality was collected by groups of HNE 120 students. All of the data was compiled together for the Asbury and Kroger sites. Three groups collected data for Kroger, and four groups collected data for Asbury. The goal of the study was to determine the consistency of data collection between groups. Although no conclusions can be made on whether or not to accept this data as accurate, it has been recommended that there needs to be more practice identifying and collecting species before going out into the field. One must also take into account that there were only a few samples to use to gauge water quality. Finally. the collections were taken on different days and could have been altered by variable weather patterns.
First we will question the similarity in the species which each group collected. At both Kroger and Asbury, there was only one species that all groups caught i.e. Aquatic worms (Kroger) and Caddisfly (Asbury). All of the other species varied in the frequency of collection among the groups (Table 4.4.1) The number of each of these species collected ranged one to nine.
Another questionable area is the abundance data for each of the species. At Asbury, 50% of the groups found over 100 caddisfly during their time in the water, while 25% found nine or fewer caddisflies in one setting. This inconsistency also occurred at Kroger with the aquatic worms and midge larvae (Table 4.4.1).
The inconsistency between groups could affect the final assessment of the water quality, but fortunately it does not have to. The implication it does have is that Asbury has five more species than the Kroger site, which suggests that Asbury is more biologically diverse than Kroger. But if one crosses off all of the species on the list that were only found by one group, the total number of species per site changes to six at Asbury and four at Kroger, which is not as big of a difference. The water quality still ends up being poor either way, but for the future this might not be the case (Table 4.4.2).
The chemical data was also collected by groups of HNE 120 students. The sample sizes were small and the level of expertise must be questioned as well. The weather and other external factors could have played a large role in the radical disparity of some of the data, but we cannot comment without proof.
First, we questioned the most extreme data points. For example, although the average temperature of the water was 17°C, it ranged from 15°C to 26°C at Asbury. According to John Wegner, "Water at 26°C is hotter than bath water arid would kill the fish." (1998 pers. comm.). There were also very low numbers in the range of dissolved oxygen. If the extremely high water temperature and low dissolved oxygen content were collected on the same day, these numbers would make more sense. As the temperature rises, the fewer the number of oxygen molecules that can remain in solution. But the low amount of dissolved oxygen was collected on a day when the water temperature was measured to be 16°C. Therefore, this data should be thrown out of the calculation. All these extreme numbers were left in the data sets for future understanding of possible problems with the collection of chemical data. Overall, the averages for air and water temperature, pH, dissolved oxygen and settable solids did not inply anything unusual about the stream quality. If the data was consistently over/underestimated, the assessment of stream quality might change.
The chemicals that did indicate the stream to be unhealthy were the nitrate and ortho-phosphate levels. Both of these areas had large measurements in their ranges. An unpolluted stream is considered to have nitrate levels below 1 mg/L (Georgia EPD 1995). The average nitrate level for the Peavine Creek was 2.6 mg/L. Even if the high amounts (i.e. 5 mg/L) were omitted, all of the readings were still at least at a level to indicate polluted waters. The ortho-phosphate data followed the same trend. There were values up to .5 mg/L, although the average amount of ortho-phosphate was .1 mg/L. Levels above .1 mg/L may stimulate plant growth to increase beyond natural eutrophication levels (Georgia EPD 1995).
| Average number of species | Average number of trees | Number of trees per hectare | |
| Asbury |
8
|
43
|
3831
|
| Kroger |
6
|
14
|
1133
|
After comparing the results of the chemical, invertebrate, fish, and bird data, we have concluded that there is no difference between the Asbury and Kroger sites. When beginning this assessment, we chose the two sites because there were observable differences in the amount of natural habitat remaining. Our findings agreed with the null hypothesis and suggest "that potential water quality impacts are not spread evenly throughout the watershed, nor are they a simple decreasing function of the distance from the city" (Wear et al.. 1998). There were differences in the vegetation seen at Kroger versus Asbury, but these differences are due to human alterations in the riparian terrain, not differences in water quality. The chemical and invertebrate tests indicated poor water quality. The fish data suggested there has been a slight decrease in water quality. The vegetation and bird samplings do not suggest anything about the water quality of Peavine Creek, but they do give more information about the overall quality of the entire ecosystem. These two areas of study will become important in long term studies of the area. When interpreting these results, one must keep in mind that the number of samples for each area of study was small and the expertise level of the monitors varied greatly. Nevertheless, the majority of the groups found the water quality to be poor and found there to be no difference between the Asbury and Kroger sites.
Although we found no difference between these two sites, comparing our data with a theoretical, undisturbed stream illustrates the degree of human impact on Peavine Creek. Vegetation in an undisturbed watershed would be more diverse in species composition, allowing for species that are more sensitive to pollution. In addition, vegetation would extend beyond the riparian zone, creating a more extensive and thorough support system for the stream. When vegetation surrounding a stream is restricted or sparse, as we found at both sample sites near Peavine Creek, the stream is vulnerable to many threatening factors, including flooding, toxic runoff, unstable temperature, loss of nutrients, and ecosystem degradation.
Fish found in an undisturbed stream would be different from those found near Kroger and Asbury for several reasons. When compared with studies conducted in previous years, it is apparent that the degree of disturbance in Peavine Creek is rising, as indicated in a loss of fish species diversity. Studies conducted in an undisturbed stream would reveal greater fish species diversity than found in Peavine Creek. In addition, the small parasites present on most fish in Peavine would be absent from fish populations in an undisturbed stream. Parasites can indicate a disruption in the natural habitat.
All of the birds found near Kroger and Asbury were highly tolerant species, usually found in urban environments. In all undisturbed habitat, the bird population would be more diverse. Birds that are sensitive to pollution and urbanization require a more isolated habitat than is available near Peavine Creek. The main difference we would find at an undisturbed section of stream is that species composition would include more sensitive species.
In an undisturbed situation, the biological and chemical tests would have shown much higher water quality than was found in Peavine. Ideally, macroinvertebrates would have been found in higher numbers and with greater diversity. In a normal stream the amount of nitrates and phosphates found would be lower than was found in Peavine. In addition, water temperature would be more consistent, thus ensuring even distribution of dissolved oxygen in the water.
After completing this assessment, we recommend some additions and alterations to our work. With the understanding that the next set of students who embark on this will be better prepared in their individual areas of study, we suggest that this study be done on a much larger scale to see if there really is a difference between upstream and downstream water quality. The next place to test is above the golf course to assess the effects of grass maintenance on water quality. To gather more information about the daily events in the area of the creek, local residents should be contacted. They will be able to fill in the gaps for the time when researchers are not out in the field. In addition to residents, local experts should be questioned about ecosystem trends. This would reduce the risk of misinterpreting data.. Finally, we believe that in order to fully understand how to interpret the results, the next group of students should complete an in-depth study of how to define ecosystem health, good water quality versus bad water quality, disturbance, and stability. These terms are often too ambiguous to make judgments and determine management schemes.
Monitoring will eventually lead to management, but it is necessary to use this report as a guide on where to start. Once more data has been collected and verified, management of Peavine Creek will be feasible. Although students will be responsible for continuing our assessment, they will benefit from involving residents in the community who also hold an interest in protecting the environment. This combined effort will help make the management and protection of Peavine Creek possible.
Bryce, Sandra A. and Sharon E. Clarke. "Landscape-Level Ecological Regions: Linking State-Level Ecoregion Frameworks with Stream Habitat Classifications". Environmental Management 20.3 (1996): 297-311.
Burleigh, Thomas D. 1958. Georgia Birds. University of OK Press: Norman, OK.
Christensen. Norman L. et al.. The Report of the Ecological Society of America Committee on the Scientific Basis for Ecosystem Management". (1994/1995). http://esa.sdsc.edu/execsum.htm
De Leo, Giulio and Simon Levin. The Multifaceted Aspects of Ecosystem Integrity. Conservation Ecology. (online) 1( 1): 3 at http://www.consecol.org/vol1/iss1/art3
Fisher, Stuart G. et al.. "Material Spiraling in Stream Corridors: A Telescoping Ecosystem Model". Ecosystems I ( 1998): 19-34.
Frissell, Christopher A. and David Bayles. "Ecosystem Management and the Conservation of Aquatic Biodiversity and Ecological Integrity." Water Resources Bulletin. vol. 32, no. 2, April 1996.
Furness, R.W., J.J.D. Greenwood and P.J. Jarvis. 1993. "Can birds be used to monitor the environment?" In R.W. Furness and J.J.D. Greenwood (eds.) Birds as Monitors Environmental Change pp. 1-35. Chapman & Hall: London.
Georgia EPD. 1995. Georgia Adopt‑A‑Stream.
Karr, James R. 1994. Monitoring Biological Integrity. In Gary K. Meffe and C. Ronald Carroll (eds.), Principles of Conservation Biology, pp. 415-16. Sinauer Associates: Sunderland, MA.
Miller, G. Tyler. 1996. Living in the Environment, 6th ed. pp. 484 & 638. Wadsworth: USA.
Naiman. Robert J., Monica G. Turner, David N. Wear. "Land Cover Along an Urban-Rural Gradient: Implications for Water Quality". Ecological Application 8.3 (1998): 619-630.
Noss, Reed, "A World Wildlife Fund Canada/World Wildlife Fund US Discussion Paper: Maintaining Ecological Integrity in Representative Reserve Networks." January 1995, WWF.
Regar, Allen and Cyrus Beasley. "A comparative study of fish populations in the Peavine Creek area." Biology 445, Dr. Donald Shure, May 2, 1995.
Thom, Rosemary, and Suzie Moseley. "Fish populations and Ecology of Peavine Creek." Atlanta, GA. Biology 445, Dr. Donald Shure, May 3, 1996.
Vogt, Gordon, Wargo, Vogt et al. Ecosystems: Balancing Science with Management. Springer Press, New York, 1997.
Woodley, S. "Monitoring and Measuring Ecosystem Integrity in Canadian National Parks." Pp. 155-176 in S. Woodley, J. Kay. and G. Francis, Ecological Integrity and the Management of Ecosytems. St. Lucie Press, Ottawa, Canada, 1993.
Yerkes, Sarah and Hang Yin. Adopt-A-Stream History. HNE 120, Dr. John Wegner, Fall 1998
|
© 2009 Emory University |