We intend to study what effect landscape has on a stream. To accomplish this, we will survey aquatic life such as minnows and water striders, factors affecting the stream such as light, temperature, and entering debris, and elements of the stream such as width, depth, and water speed. We hypothesize that landscape influences stream habitat. In other words, streams and what lives there are effected by the landscape around them. We will measure several different physical aspects of streams in different landscapes and determine if and which of these affect the abundance of animals in the stream. In an urban landscape we expect to see the most sun and trash, and the fewest leaves of all the landscapes we survey. We also expect a high water temperature and slow water speed in an urban landscape. In a deep-woods landscape we expect to see the least amount of sun and trash and the most leaves of the landscapes in our study. We also expect to see a low water temperature and high water speed. In an edge landscape we expect the sunlight and water temperature to be evenly between the other two landscapes. The amount of leaves and trash in the edge landscape and the water speed to be not much less than the amount in the deep-woods landscape. We expect animals to be most abundant in the deep-woods, least abundant in an urban landscape, and evenly between the two in the edge landscape.
We came to this topic due to our interest in streams and aquatic life. As a class, under the excellent guidance of Chris Myers, it was decided that landscape will be included as a factor in each lab. This led to our eventual question of how a stream would differ across landscapes. Through this lab we plan to gain a greater understanding and appreciation of aquatic landscapes along with enriching our lives through discovery science.
The topic of landscape is relevant in terms of the research done by the class as a whole, and for the benefit of future students and disciples of Chris Myers. As well as in how it relates to official stream studies such as those done by the EPA. If landscape is not considered when studying a stream, then it is possible that the results may be skewed.
Isaac Schlosser has found that land-use activities near a stream result in significant changes in the population of stream fishes. He has also found that normal variations in these populations correspond to the seasonal input of leaves and other organic materials. If this schedule of input is changed, it has a huge effect on the animals in the stream. Alvarez-Cobelas et al have also found that the nitrogen and phosphorous added to streams from the decomposition of leaves and other organic materials are a measure of the aquatic health of the stream. Todd Crowl found that species interactions vary with the environmental context across different stream habitats. In the same article, Alex Flecker remarked on the importance of the environmental context when studying biological interactions. He was referring to tropical streams, but we believe it is applicable to our study as well. Luz Boyero has found that factors such as rainfall and stream flow are important in terms of nutrient cycling and foodweb dynamics. His rainfall and stream flow are similar to the depth and water speed in our study. A more complete list of literature is provided at the end of this proposal.
We intend our study to expand FleckerÌs ideas to temperate streams, and ScholsserÌs ideas to animals in addition to fishes, as well as tie together the somewhat disparate ideas in these studies.
We will visit areas of a stream and find sections in different landscapes (urban, edge, and deep-woods) ten meters long with similar width, depth, and water speed. In each of these sections we will count schools of fish, and water striders along the ten meters using a visual scan. With a hula-hoop we will mark off three areas within the ten-meter section of the stream looking for other notable organisms such as invertebrates and insects. Every two meters along the ten-meter stretch of stream we will measure the width and depth (at the middle) of the stream. We will set up traps with more hula-hoops in order to collect debris entering the stream. Other factors we will measure include light, using a spectral refractometer, and water temperature using a thermometer. However, due to season concerns and lack of technical support for the refractometers, we opted to perform a visual survey of overhang for each of our environments.
We chose these methods based on what we are capable of measuring with our limited resources. We found, in our research, that when studying streams, 150-meter stretches are usually surveyed. We did not think that with our time and resources we would be capable of surveying such a large area effectively. We chose not to study microorganisms or do water testing because they are beyond our capabilities. We also feel that minnows and water striders are good organisms to study because one lives in the water and the other on the water, giving us gauges for two different situations in a stream. This will help us distinguish between the actual stream and the habitat immediately surrounding it. We feel that our experiments are statistically sound because we asked for advice from our wonderful, omniscient teacher, Chris Myers. He is so great. (He has a neat tattoo.) We will ensure unbiased results through consistency in width, depth, and speed of the stream. We will ensure consistency in location by measuring and marking off ten meters and using the same size hoops.
The materials we plan to use include: a meter stick to measure the depth of the stream, a stopwatch and floating object (such as a cork) to measure the speed of the water, three hula hoops to mark off areas in which to observe other notable organisms. Additional hoops will also be used along with a screen, sticks, and a stapler to construct a trap to catch entering debris. We will use a thermometer for measuring the water temperature. We will use a large measuring tape to mark off the ten-meter stretch of stream and to measure width of the stream. We will also use a field guide to identify organisms. A rope will also be used to mark off the selected length of stream. We will call upon Chris Myers as a source of motivation and inspiration, with his eternal passion for science and nature.
The class will be involved in our lab by helping us to collect data. We will divide the class into three groups and take them out to three different sites. Each group will be headed by a Team Stream member. We will teach the class to help us take measurements and count organisms. The presence and guidance of a Team Stream member throughout the process will assure consistency. In case of rain, we will get wet (and we did). In case of a storm that would make it dangerous to go outside, we will have a back-up presentation of pictures of what we do and find in the stream, and explain our lab and its importance to them. We will invite them to, if their free time co-incides with ours, to come take measurements with us outside of class on a more weather-conducive day.
|Week One:||Formulated methods.|
|Week Two:||Scout out sites and pick locations.|
|Week Three:||Build and place hula-hoop traps to begin capturing debris.|
|Week Four:||Check sites and record data.|
|Week Five:||Check sites and record data. Teach NS, in place of Chris Myers.|
|Week Six:||Check sites and record data.|
|Week Seven:||THANKSGIVING no available work time|
|Week Eight:||Avoid hypothermia. Begin analyzing data. Clean up and work on final lab report.|
|Week Nine:||Analyze data and finish lab report.|
The bar graphs included above best show the relationships among landscapes as we worded them in our hypothesis and discuss them below. The above correlations are quite useful to determine what investigated items are interrelated, and possibly explaining why we got the results we did.
We hypothesized originally that an urban landscape would have the least shade of all the landscapes. However, once we went out to the streams, we realized that using a spectral refractometer would not give us a meaningful measure of leaf cover, and therefore shade. When we began surveying the streams, leaves had already begun to fall from the trees. We did, however, estimate how much leaf coverage there would be based on surrounding vegetation and branches over the stream. Our findings suggest that our predictions were accurate; the urban landscape had the least leaf coverage, and the deep woods landscape had the most. (Hi, Chris Myers.)
We hypothesized that the type and quantity of debris entering the stream was also dependent on landscape. As we predicted, in the deep woods we found more leaves than the other landscapes, as well as twigs, seeds, and berries. We also observed some bubbling brown froth that we believe is due to the nearby Petrifying Peabody Power Plant. At the edge landscape we found fewer leaves and seeds, but we found no trash. It should also be noted that our hoop trap had been moved to the side of the stream resulting in uncertain data. At the urban landscape, we found yet fewer leaves, and a hubcap. This data holds up our hypothesis. Unfortunately, we were only able to take data from the hoop traps once, as just before our second trip to the stream, a rain storm swept away our hoop traps. They continue to be Missing In Action.
We hypothesized that we would find a higher water temperature and lower water speed in the urban landscape than in the other landscapes. We based these predictions on an unstated hypothesis that the urban landscape would have the lowest water level, and the deep woods landscape would have the highest. This is upheld by our data. (Note on the correlation matrix the high correlation between depth and water speed, and depth and temperature.) However, we have since discovered that we were thinking about water speed in the wrong way. We thought that shallower water would mean slower water, when we found it is completely the opposite. Our temperature data only partially upholds our hypothesis. The urban area had the highest water temperature, which we thought would be due to the lower water level. However, the differences in water temperature were fairly slight, and not statistically significant. Our temperature measurements went up fairly significantly after the rain storm that washed away our hoop traps and this may have significantly skewed our overall data.
We found a distinctly higher number of minnows in the deep woods landscape than in the edge and urban landscapes, whose minnow populations were very similar. Perhaps this is because there were no still pools in the edge landscape. In adjacent areas, there were found pools that contained many more minnows than in our designated ten meters. We found a statistically significant difference in the water strider populations. However, we think this may also be due to the lack of still pools in the edge and urban landscapes.
We believe that our results are not necessarily representative of the conditions and populations in the stream. There were several conditions beyond our control that could have significantly altered our results, such as the rainstorm which caused deeper, higher, faster, and warmer water, and which washed away our traps in every landscape. The loss of our traps was significant because we then had to discontinue that avenue of research. Due to seasonal issues and the dangers of hypothermia, we only took data three times, which is not a large enough sample size from which to draw meaningful conclusions. The season affected the abundance of animals as the weather got colder as well as the amount of overhanging vegetation and the resulting debris entering the stream.
Within the hoops we placed in the stream (as opposed to the hoop traps) we did not find notable differences in the abundance and variation of animals among landscapes. Perhaps this means that life under rocks is not affected by the surrounding landscape. Although, once again, our data is inconclusive. It, however, may be meaningful that there was more noticeable life after the large rainstorm.
Our study, while loosely based on EPA processes, is far more limited as we have more time and resource constraints. The EPA lead us to look at the factors of human impact and the vegetation next to the stream and overhanging it. The literature we cited in our Relevance section led us to look at the particular aspects we did. Particularly notable for the purposes of this conclusion are Luz Boyero and Isaac Schlosser, because their findings help to explain some of ours. Boyero found that factors such as rainfall and stream flow have a direct impact on the stream environment. This helps to explain the large change we noticed in all our data at all our sites after the rainstorm. Schlosser discussed land-use activities near streams having significant effects on the life within the stream. In half of our bar graphs the data for the edge landscape was actually in-between the data for the other two landscapes. In the other half, however, the edge data was closer to that we had predicted for the urban landscape. Looking at the sites we had chosen, we hypothesize that this was probably due to the degree of human impact at the two sites. When we chose the landscapes we were looking at their proximity to the town of Oxford, and not so much at the human impact. On a second viewing of the sites, we realized that a ford runs adjacent to our edge landscape. On the other hand, the area we chose as the urban landscape had much higher banks and, though it was located in a public park, was actually less accessible to humans, as we discovered by climbing in and out of it repeatedly. We found that the differences we enumerated in our hypothesis are more a result of the degree of human impact and less a result of the location of the landscape. What determines the differences in the landscapes is the degree of human impact not the geographical location of the landscape. Therefore, we find that our data is inconclusive in proving or disproving our hypothesis, as only one aspect of data we collected was found to yield statistically significant differences among the landscapes. We recommend further study.
For anyone who would like to do an investigation of this sort, our advice is as follows:
Alverez-Cobelas et. al. "Hydrological and botanical man-made changes in the Spanish wetland of Las Tablas de Daimiel" Biological Conservation. Vol. 97. Issue 1. Page 89-98. January 2001.
Boyero et al. "Tropical Stream Ecology" Tree. Vol.15. Issue 10. Page 391. October 2000.
Greenwood and Metcalfe. "Minnows become nocturnal at low temperatures". Journal of Fish Biology. Vol. 53. Issue 1. Page 25-32. July 1998.
Mode et. al. "Ranked set sampling for ecological research: accounting for the total costs of sampling." Environmetrics. Vol. 10, Issue 2, page 179-194. March/April 1999.
Myers, Chris. "Science: Inspiration and Joy". Natural Systems. September 28, 2000.
Schlosser. "Stream fish ecology: A landscape perspective". Bioscience. 1991. Vol. 41. No. 10, page 704-712.
Severns, John. Student: Miami University and Employee: EPA. Email Interview. September 19, 2000.
Return to the Topic Menu
IMPORTANT: For each Response, make sure the title of the response is different than previous titles shown above!
Weather & Earth Science Resources
Tropical Ecosystem Courses
Tools & Other Stuff