Geography - Ivestigation of the River Colne, Buckinghamshire
25 Pages
Introduction Pages 3-6
Hypotheses (at top of page): Page 3
Maps of Location of River: Page 4
3D map of river: Page 6
Methodology: Pages 6-8
Data Presentation: Page 8-19
Raw Data Table: Pages 8-9
Photos: Pages 10-14
Volume of flow: Proportional Line map: Page 15
Width, Depth, Velocity and Volume of Flow Graphs: Pages 16-17
Vegetation and Height Above Sea Level Graphs/Charts: Page 18
Differences in Height Above Sea Level Graph: Page 19
Data Analysis: Page 19-22
Hypothesis 1: Pages 19-20
Hypothesis 2: Pages 20-21
Hypothesis 3: Pages 21-22
Conclusion: Page 22
Evaluation: Pages 22-25
Final Conclusion: Page 25
Bibliography: Page 25
I formulated my hypotheses based on my current knowledge:
* I expect the river to get wider and deeper as it flows downstream because it will be joined by tributaries, and other water e.g. from rainfall, so the volume of water in the river will increase, therefore the width and depth of the river must increase to accommodate this extra water. The extra water will also entail more hydraulic action, so the banks of the river will be eroded more, and more water also means more sediment. This sediment will also erode the banks more; rocks carried by saltation or traction will erode the bed, making the river deeper, and alluvium in suspension will erode the bank. In the upper course of most rivers, due to the bedload, there is usually more vertical erosion than lateral (as the river tries to reach its base level), so the river is deep, but narrow, with steep banks. In the middle and lower course, there is progressively less vertical erosion and more lateral erosion, so the river should get a lot wider and a bit deeper as it flows downstream.
Human intervention could change the shape of the river, and indeed the nature of the water itself, and plants and fishes can build up the riverbed with decaying material, or deepen it by foraging for food or growing roots. Finally, the banks themselves are likely to be eroded more easily, because they are less likely to be made of rock, as they are in the upper course, and more likely to be made of soil or formerly deposited sediment, which are easier to erode. It is by this common sense and geographic theory that I know that rivers widen and deepen as they flow downstream.
* I expect the rate of flow will increase because there are fewer obstructions in the river channel. Whereas in the Upper Course of most rivers, the riverbed is uneven and contains many loose stones, e.g. Figure 1, whereas in the Middle and Lower Course of the river, most of this bedload will have been broken down by attrition, abrasion and corrosion, and it is now sediment in suspension or solution, e.g. Figure 2. Therefore, the river will be losing less energy through friction between the bedload and the riverbed, so it will have more kinetic energy and will flow faster. Conversely, because the river is no longer in mountains, the gradient is shallower, which may mean that the river flows slower. On the whole, however, even the steep incline in a river's Upper Course will not compensate for the energy loss through friction and erosion (both erosion of the banks and of the bedload). Furthermore, for the section of river I will be studying, there will be no mountains (I have included a 3D map of the area just to prove this [The River Colne is highlighted in pink]), so the gradient should decrease slightly over the length of my study as it normally would in the middle course of a river.
* I expect the gradient of the river will get shallower because the land itself will get flatter, as the river is no longer near its source in hills or mountains. Also, as the river flows through its Middle Course, it will begin to meander, partly as a result of the decreased gradient in the first place, which means that, in every kilometre the river flows, it will only drop a fraction of the height it would drop if it was flowing in a straight line. Furthermore, as the river flows into its Lower Course, I would expect it to develop a floodplain, so the gradient is naturally flatter than it was upstream
In this case, the Colne doesn't actually have a floodplain as such. Instead, a succession of man-made gravel pits, now lakes, and a canal run alongside it; so the land is naturally flatter anyway. In addition, I have procured a 3D map of the area (next page), so I know that the immediate surroundings of the river are reasonably flat, and that the river's gradient decreases gradually over the 10km I have chosen to study.
Methodology:
Unfortunately (in retrospection), I did not do a pilot study or investigate the River Colne before I did my main data collection day. If I had, I would have changed my preference or rivers and studied another, as the Colne turned out to be too wide and too deep to measure whilst standing in it, leaving me no choice but to measure from bridges, which was no problem in practise because there are many bridges on the river, although it may have affected my readings. However, I did plan where I would collect my data using Memory Map, an Ordnance Survey application, and I further prepared myself by printing off a small scale local map to help me navigate the river Colne, its byways and its access routes on the day I collected my data. Furthermore, because I was unable to borrow a hydro-prop from school, I had to design and make a velocity measuring device myself from a plastic container with a length of string, knotted at 5cm intervals, through its lid. Then I experimented with varying amounts of weight (I used 1 penny coins as weights) in it to make it float in the water rather than on it.
On the 26th of August when I arrived at the river with an assistant, who helped me take readings, I was surprised by its width and depth, as I had not noticed the width while I was planning, nor had I remembered the width from previous visits or made a pilot study to check. Unperturbed, I decided I would study the Colne anyway, at ten sites roughly a kilometre apart so as to gain a fair sense of how the river changed as it flowed downstream. However, I needed to make sure each site had a bridge so that I could take readings there, as the river was too deep to stand in. A good way to check, I found, was to pick sites, looking at the map I had printed out, where roads, tracks or paths crossed the river, whilst still keeping the prospective sites as close to a kilometre apart as possible. Thus, I decided upon my ten sites, and marked them on the printout. I have since marked them on the OS application, as below. Then I proceeded to measure the river where I was, at my first site.
First of all, I measured the width. Unfortunately, unprepared as I was, my tape measure was only 3 metres long. So I measured 3 metres horizontally along the sides of the bridge, from vertically above the bank. When I reached the end of the tape measure, I marked the endpoint, pivoted about it, and then measured another 3 metres. I repeated this process across the bridge to get the entire width in metres. Then I had to measure depth. I thought of attaching a weight, a stone, to the bottom of the tape measure, but then I thought that I could take all the weights out of my velocity measurer and fill it with stones, then measure how far up the string the river came, and hence find the depth. So I dropped it into the river, making sure the string was as taut and vertical as possible, and noted the knot it was at. I counted the knots to the bottom as I pulled it out, and then used the tape measure to find the depth in metres. I repeated this process at three more places equidistant from each other, and then used the tape measure from the bank to get readings for the depth at each bank, giving me six depth readings an equal distance apart. Unfortunately, the current of the river made the string belly a little downstream, so my readings were not, perhaps, as accurate as they could have been. It was important for me to attain measurements of the river's width and depth in order to work out the river's cross-sectional area, and the
Next, I measured the river's velocity. I took the stones out of the depth measuring device, and put the weights back in and dropped the device into the water at the near bank. I made a mark on the bank level with where it started and timed a minute. Then I made a mark on the bank level with where it was when the minute was up, and measured the distance between the two marks. Then I repeated the procedure at the other bank, and then at the four sites in the middle of the river where I ...
This is a preview of the whole essay
Next, I measured the river's velocity. I took the stones out of the depth measuring device, and put the weights back in and dropped the device into the water at the near bank. I made a mark on the bank level with where it started and timed a minute. Then I made a mark on the bank level with where it was when the minute was up, and measured the distance between the two marks. Then I repeated the procedure at the other bank, and then at the four sites in the middle of the river where I had measured the depth. When I did it in the middle of the river, I stood on the bridge and dropped the weight in, then fed the string out, but my assistant timed the minute and marked the bank level with the start and finish. So I had the speed in metres per minute, so I divided it by 60 to get metres per second, then multiplied it by 100 to get the speed in cm/s. It was important for me to attain measurements of the river's width and depth in order to work out the river's cross-sectional area, and then, with its velocity, the river's volume of flow. I needed this data in order to prove or disprove hypothesis two, that as the river flows downstream it will get wider and deeper. I also needed the width and depth measurements to analyse hypothesis 1, that as the river flowed downstream it would get wider and deeper.
Then, I took a general look at the river and estimated the vegetation coverage, and noted it in percentage. From my existing data, I then worked out an approximation of the cross-sectional area of the river by multiplying the average depth by the width. Then I took the speeds at each of the six sites, and averaged them, and divided this value by 100 to get the average velocity of flow in m/s (so that all the units were the same for the next calculation). I worked out the river's approximate volume of flow by multiplying this value by the cross-sectional area I had previously worked out. The penultimate thing I did was to look at my OS printout and attain a reading of the site's height above sea level, in order to asses how the river conformed to hypothesis three, that as it flowed downstream, its gradient got shallower. Finally, I took some photos of the site, and chose the best, the one which showed most of the river where I had taken my readings. Photos are good because they are evidence of the river I studied, and will help to identify trends when I come to analyse the data. Then I moved on to the next nine sites, where I repeated the above method.
Data Presentation
Width
Depth
Velocity
Height above sea level
Cross-sectional Area
Volume of flow
Vegetation
Other Details
SITE
Reading
(m)
(m)
(cm/sec)
(m)
(m2)
(m3/sec)
(%)
No.
22.7
0.223
.35
38
.12
0.34
30
Fisherman
2
n/a
.23
.38
n/a
n/a
n/a
n/a
n/a
3
n/a
.393
.46
n/a
n/a
n/a
n/a
n/a
4
n/a
.72
.49
n/a
n/a
n/a
n/a
n/a
5
n/a
.49
.41
n/a
n/a
n/a
n/a
n/a
6
n/a
0.34
.34
n/a
n/a
n/a
n/a
n/a
2
7.62
0.01
5
39
.12
0.11
5
Golf Club
2
n/a
0.14
1.3
n/a
n/a
n/a
n/a
n/a
3
n/a
0.195
2
n/a
n/a
n/a
n/a
n/a
4
n/a
0.295
3.1
n/a
n/a
n/a
n/a
n/a
5
n/a
0.16
1.7
n/a
n/a
n/a
n/a
n/a
6
n/a
0.08
8.6
n/a
n/a
n/a
n/a
n/a
3
6.03
0.08
3
39
.04
0.07
20
2
n/a
0.17
7
n/a
n/a
n/a
n/a
n/a
3
n/a
0.26
1
n/a
n/a
n/a
n/a
n/a
4
n/a
0.29
0
n/a
n/a
n/a
n/a
n/a
5
n/a
0.16
6
n/a
n/a
n/a
n/a
n/a
6
n/a
0.07
2
n/a
n/a
n/a
n/a
n/a
4
4.73
0.35
0.34
40
2.41
0.13
40
fish
2
n/a
0.6
6
n/a
n/a
n/a
n/a
n/a
3
n/a
0.72
0
n/a
n/a
n/a
n/a
n/a
4
n/a
0.69
9.4
n/a
n/a
n/a
n/a
n/a
5
n/a
0.37
5.8
n/a
n/a
n/a
n/a
n/a
6
n/a
0.33
0.8
n/a
n/a
n/a
n/a
n/a
5
5.74
0.11
6
41
7.40
2.21
35
lots of fish
2
n/a
0.3
20
n/a
n/a
n/a
n/a
n/a
3
n/a
0.95
33
n/a
n/a
n/a
n/a
n/a
4
n/a
0.71
50
n/a
n/a
n/a
n/a
n/a
5
n/a
0.46
37
n/a
n/a
n/a
n/a
n/a
6
n/a
0.29
23
n/a
n/a
n/a
n/a
n/a
6
8.1
0.45
22
42
6.06
.47
80
big plants
2
n/a
0.57
30
n/a
n/a
n/a
n/a
n/a
3
n/a
0.98
40
n/a
n/a
n/a
n/a
n/a
4
n/a
0.96
25
n/a
n/a
n/a
n/a
n/a
5
n/a
0.88
5
n/a
n/a
n/a
n/a
n/a
6
n/a
0.65
4
n/a
n/a
n/a
n/a
n/a
7
2.8
0.47
TMV/FALLEN TREE
45
7.91
0.54
80
weir and fallen tree
2
n/a
0.64
n/a
n/a
n/a
n/a
3
n/a
0.78
0
n/a
n/a
n/a
n/a
n/a
4
n/a
0.81
7
n/a
n/a
n/a
n/a
n/a
5
n/a
0.67
3.4
n/a
n/a
n/a
n/a
n/a
6
n/a
0.34
TMV
n/a
n/a
n/a
n/a
n/a
8
0.1
0.12
4
46
.26
0.45
70
bend
2
n/a
0.1
20
n/a
n/a
n/a
n/a
n/a
3
n/a
0.11
42
n/a
n/a
n/a
n/a
n/a
4
n/a
0.16
55
n/a
n/a
n/a
n/a
n/a
5
n/a
0.14
47
n/a
n/a
n/a
n/a
n/a
6
n/a
0.12
36
n/a
n/a
n/a
n/a
n/a
9
5.2
0.07
0
47
0.40
0.12
35
little fish fry
2
n/a
0.09
8
n/a
n/a
n/a
n/a
n/a
3
n/a
0.09
38
n/a
n/a
n/a
n/a
n/a
4
n/a
0.08
50
n/a
n/a
n/a
n/a
n/a
5
n/a
0.07
34
n/a
n/a
n/a
n/a
n/a
6
n/a
0.06
TMV
n/a
n/a
n/a
n/a
n/a
0
6.5
0.08
STATIC/TMV
49
0.62
0.01
5
shopping trolley
2
n/a
0.1
STATIC
n/a
n/a
n/a
n/a
3
n/a
0.11
2
n/a
n/a
n/a
n/a
n/a
4
n/a
0.12
3
n/a
n/a
n/a
n/a
n/a
5
n/a
0.09
2
n/a
n/a
n/a
n/a
n/a
6
n/a
0.07
STATIC
n/a
n/a
n/a
n/a
n/a
Note: The Colne merged with the Grand Union Canal in several places; My first reading was just before a confluence, and subsequent upstream readings were perhaps a little smaller than expected. The Colne then split again into the Colne Brook and the Colne proper, before merging again just downstream of another section where it was the canal; so my tenth reading appeared a little smaller than previous ones.
Note: TMV = Too much vegetation/weed/obstruction to take readings. STATIC: In some places the river was not moving, or it was eddying, so no readings, or no accurate readings could be taken.
Now, I have attained a map of where I did my study, and used a bitmap editor to adjust the size of the river, pink, to correspond to the volume of flow along the length that I studied.
Next, I plotted graphs of volume of flow against height above sea level to show what happened to the volume of flow as the river flowed downstream, which are on the attached sheet of graph paper.
I thought I should do this for width, depth and velocity as well, so I plotted four more graphs using ICT. Then I saw how uneven the line was, so I worked out a moving average for every 3 sites and plotted that on the same axes as the width, depth, velocity and volume of flow graphs.
Finally, I used ICT to plot a volume chart of the vegetation cover on the river bed (plate 3.1) and a line graph of the river's height above sea level (fig 4.1). Plate 5.1 is a graph I later plotted to help me analyse my data.
Data Analysis
My Hypotheses were:
. As the river flows downstream, it will get wider and deeper.
Firstly, my raw data may show this, as the furthest downstream site (1) is definitely deeper and wider than the furthest upstream site (10). However, there is much fluctuation within the data, and it is hard to see any real trends. So I calculated a moving average of the data by averaging results 1-4, then 2-5, then 3-6 and so on. This had the effect of smoothing the curve of and line graphs I plotted. The line graphs I plotted from the moving averages (plates 2.1, 2.2) do seem to conform to hypothesis 1. I drew an approximate best fit line for plates 2.1 to 2.4 and 4.1. The line is very approximate, and in most cases I just the line from first to last point, but moved up or down a little. The best fit lines of both plate 2.1 and 2.2 show an uphill slope as the river flows downstream, in accordance with hypothesis 1.
I thought the width and depth would increase as the river flowed downstream, because the terrain it flows through changes: it is easier to erode mod in its middle course than rock in its upper course. However, the Colne, at the point where I took my readings, was flowing over chalk not mud, as it is near the Chilterns. Chalk is a soft, porous rock, but still a rock, so it was harder to erode. However, because it is porous, water drains easily through it, so this fact may have affected my readings of volume of flow.
I also thought the width and depth would increase as the river flows downstream simply because the river is larger, which I know through experience. This means there is more water to erode the bed and banks through hydraulic action, and to carry the river's load, which will also erode the bed and banks through corrasion, attrition and possibly corrosion.
In the river where I took my readings, there was usually a bedload, moving by saltation and traction. This would have caused vertical erosion and made the river deeper. Furthermore, the abundant vegetation will gradually erode the soft chalk and soil when it dies, or if it is uprooted by strong current. Looking at plate 3.1, I can see that there was generally more vegetation upstream than downstream, which in itself is strange as I would expect the rocky upper course channel to defeat plant growth. However, I am only studying a section of the middle course of the river, not the whole river, so the bed is always mud underneath any bedload. The extra vegetation upstream, especially at site 6 and 7, could explain the out lying values 9 of width and depth at said sites. Mostly though, there was not that much suspended load apparent to the vision: the river was generally quite clear. However, sites 1 and 9 were a bit murky, and site 3 very much so, suggesting that there was quite a lot of alluvium and silt suspended in the water, which would gradually erode the banks and make the river wider. Of course, I can't see what's in solution in the river, so I don't know how it will affect the river's width and depth
It is also evident in most of my data presentation that the river varies a lot between site 1 and 10, and an average is needed to draw conclusions. This could be because the river's chalk bed affects the river's volume because it is porous and allows water to drain through it, and groundwater to rise and fill the river through it, depending on the position of the water table. It is also easily eroded, so the river's depth may be affected by it as well. Furthermore, flowing through an intensely urban area, the Colne is heavily affected by human activity, which may explain some of my outlying results. In addition, the variable weather at the end of august: alternate showers, downpours and sweltering sun, may well have had adverse effects on my data. Looking at the photos, I can see that the river gets wider and deeper as it flows downstream, because some of my narrowest and shallowest sites are 9 and 10, whereas site 1 is very wide and so deep you can't see the bottom.
Looking at photos 1, 3 and 9, we can also see that they are murky, as mentioned above, which means there is sediment in suspension in the water which could affect the river's width and depth at that point. However, this sediment is not continuous: the river is clear at most of my other sites, so the lateral erosion is also not constant. Overall, it is obvious that there is a little increase in values from upstream to downstream of the section of river I studied, but a lot of fluctuations too. So the data I took does support hypothesis 1, but is quite inaccurate, and I would need an average over a much longer section of river to be precise.
2. As the river flows downstream, the rate of flow will increase.
My raw data does seem to show this, as the only places where the river is static are upstream, and my furthest downstream site is fairly fast flowing. To analyse the river's rate of flow I will need to average out my velocity readings. Once I had calculated averages and plotted graphs, I still found it hard to see many trends.
I calculated a moving average from my velocity readings as well as my width and depth readings. This made it easier to see quite a strong trend of the river's rate of flow decreasing as it flowed downstream in plate 2.3. This is not what I expected from the theory I studied. I thought that the river's velocity would increase because, in its upper course, the river channel is smaller and there is more bedload and rocks jutting everywhere: the friction on the water will be greater so it will flow slower. I also though that, since it is very high near the source, the river will have more potential energy, but less kinetic energy than it will lower down, so it will not flow as fast.
However, it may be that, as I did not take my readings over the whole river, just over a section of its middle course, the bedload didn't have much effect because the river was already quite deep and wide, and the river channel was bigger than in the upper course, so there was fast flowing water in the middle. Looking at most of my sites, such as 2, 3, 4, 5, 6, 8, 9 and 10, I can see that the river, on the whole, does indeed flow much faster in the middle than at the banks, indicating that friction is not having much of an effect on the river's average velocity.
When I drew a line of best fit on plate 2.3, it further enforced the downward trend of river velocity. Contrary to what I expected, the river's velocity is undeniably decreasing as it flows downstream. This may be because I am only studying a section of the river, or because of the variation of vegetation coverage: more vegetation will slow the river down. Of course, the anomalous values at site 8 are most likely due to the presence of the weir. A further possible explanation of the river's velocity decrease is that the river flows over chalk, which is a soft rock and so is easily eroded, so the river will expend more energy eroding the chalk than in moving, in kinetic energy. Also, because the river is no longer in the Chiltern Hills, so its gradient is probably getting shallower all along the section I studied.
When I started my investigation, I drew my second hypothesis from theory, but because I only studied part of the river and not the whole river, so the theory I drew my hypothesis from largely didn't apply anymore. But the most obvious and most influential explanation of the anomalous results is human influence: humans take some water from the river for drinking and other uses, put chemicals and other stuff in the river and build structures like weirs and bridges in, over and beside it. So overall, my results didn't fit hypothesis 2, and that is probably due to human influence.
3. As the river flows downstream, the gradient of the river will get shallower.
My results do show this to an extent: from the line graph I plotted, plate 4.1, it is easy to see that the height of the river decreases as it flows downstream, and there may be a slight decrease in the gradient, but it is minimal. For example, upstream, the height decreases by 2, downstream by 1, as can be seen in my table of raw data. It is hard to see any real trend, and hard to be sure of any trends that I do see. It is impossible to decipher the riverbed gradient from my photos, and hard from the tabular data and graphs. If I were to redo the investigation, I would need to take a wider range of data to prove of disprove this hypothesis, and to take it over a larger length of river.
From plate 2.4 it is hard to decipher the gradient of the riverbed, as it is a graph of height rather than gradient. To find the gradient, I would have to plot a graph of the difference between heights. So I used Microsoft Excel to calculate those differences and plot that graph: plate 5.1. Even with this graph it is hard to tell, but the moving averages suggest an increase as the river flows downstream. This may be because the moving averages took away some of the data accuracy, but is more likely to be because my original data was faulty. I would need at least one more data type to help me prove or disprove hypothesis 3. Of course, some of the strangeness of the data may be due to humans again, but that doesn't explain why the Colne seems to defy the laws of physics and flow uphill. It must be a problem with my data of my data analysis. So I can't really tell whether the river's gradient gets shallower as it flows downstream, thus neither proving nor disproving hypothesis 3, but I can tell that my data shows a slight gradient decrease as the river flows downstream, but the data is probably partially incorrect anyway as it seems to show the river flowing uphill.
In conclusion, the river seems so follow hypothesis 1, but not 2 and 3. The reason for it not agreeing with hypothesis 2 is most likely to be that I took my readings over a short section of river rather than the whole river, so the theory I drew my hypotheses from did not apply. The reason hypothesis 3 is not supported by the data is because the data is likely to be partially incorrect. All of my data was affected by human influence to a large extent, and also by the setting of the Colne, the fact that it flows next to lakes and over chalk. To get a reliable answer to my hypotheses I would have to study the whole length of the Colne, and to take readings of more types of data, especially where gradient is concerned. But really, I should study a different river, as the Colne's urban setting and interaction with the Grand Union Canal makes it a bad place to study.
Evaluation:
Firstly, my method was fairly thorough, but I don't think I took a large enough sample. The results I took were fairly accurate, but there were notable fluctuations, for example when the river neared the Grand Union Canal and when it split into Colne Brook and Colne Proper. If I were to do this investigation again, I would need to take data from a larger sample. To be honest, I would probably investigate another river, or just another section of this river, because the proximity to human urban areas, the gravel pits and the Grand Union Canal have affected my readings so that, although they are reliable, the trends they show are hard to distinguish and inaccurate. Furthermore, I would liked to have taken readings of the river's bedload, but because of the river's depth this would have been impossible at more than 1 site. So if I chose a different river, I may be able to take a wider range of data from the river. I would also probably do a preliminary visit to the river if I could start the project again, as my lack of knowledge meant that it was hard to take some readings and impossible others.
My methods did have some limitations. Firstly, the fact that I had to take some of my readings from bridges due to the depth of the river. To keep my data collection techniques constant, I had to take every reading from a bridge, not just one or two, and that made it hard for me to analyse things like the river's depth: I could not hold a tape-measure vertical in the river as I had done on my practise trip with the school, and as I had planned to do for my investigation. Instead, I had to improvise with the weighted string.
This method in itself had limitations: not least at site 1, with a fast current and deep bed, the string was seen to belly a little, despite my efforts to keep it taut. Furthermore, the way I took readings was not completely reliable; I counted the number of knots that were submerged. Taking readings by inspection like this is prone to a small margin of error, and the knots themselves may not have been evenly spaced anyway, despite my efforts to tie them at regular intervals next to a metre-rule. But then, I could take this theory of tiny errors further and say that the metre-rule itself may not have been completely accurate. In further investigations, I could improve this method by putting more time and money into finding equipment: buy professional depth-measurers. Or I could simply improve upon the weighted string method by marking a pole at regular intervals to remove the problem of the string bellying in the current.
There may also have been inconsistency in some of my readings, like a surge in fiver velocity. An unlikely scenario, but it illustrates the point that I should take repeat observations and average my results to make sure they are accurate. Furthermore, the velocity measure floated at some of my sites due to the water weighting it down (with the pennies) leaking out through the hole where the string was attached. This would have affected my readings, as the rate of flow would vary between the middle of the river and the surface and bottom, due to friction with the bed and the air. The instrument also got tangled in weeds, or started trailing them, which would also have affected my readings. To counteract all of these failings, I could use a hydro-prop to measure velocity. However, it may still get tangled in weed, so I would have to devise a way of catching weed and diverting it around the impeller.
However, because my methods had limitations, the evidence for my hypotheses that I gathered may not have been wholly reliable, and yet I based my conclusions upon that evidence. There is a knock-on effect visible here: because my methods weren't accurate, nor were my results or conclusions.
My results, on the whole, were as accurate as I could make them with my limited methods and equipment, with a few notable exceptions. The depth and velocity readings, as mentioned above, may have been inaccurate, especially at the sites with a faster current. The measurement of vegetation cover was only an estimate: I could improve it by creating a scale so that I can reference it, like Power's scale of roundness. Importantly, though, some of my results do not fit my hypothesis: the river velocity, I thought, would increase as the river flowed downstream. This is what I was led to believe by common sense and the theory I have learned, but with my results it doesn't seem to be the case. This is probably because of human interference with the river.
My conclusion/analysis was fairly good, but I could only do so well with the data I collected. I used a variety of techniques in my analysis, which is good because it makes the data easy to analyse visually in several different ways. However, a good visual technique I could have used, but didn't, was a pictogram, with which I could have displayed maybe vegetation cover, or, if I had taken readings here, bedload size.
However, there are other factors that affect my investigation. I took readings over 10 km of river. Only one of my hypotheses was supported by my data, because the theory I based my hypotheses on applied to the whole river, not a section. For example, the river's rate of flow should increase along its length, but may decrease or stay constant along 10km. To counteract this, I would need to study a longer section of river in subsequent investigations, or maybe the whole river or more than one river.
Also, the river flowed through the Chilterns: chalk hills. This meant that the bed of the river was chalk: a porous rock which allows water to percolate into the groundwater flows and be carried away. However, due to man dumping some rubbish in the river, but also largely due to nature, not all of the bed was chalk; at some sites it had been covered by copious amounts of mud, or garbage (as illustrated by the shopping trolley in photo 10), which would have prevented the water from draining in these parts, while it drained well in other places. This could be another explanation, apart from the proximity of the Grand Union Canal, for the fluctuations between the size of the river at site 1 and 5, and most of the other sites.
However, the permeation of the water through chalk illustrates another point: water would only drain away if the water table was lower than the riverbed. The water table fluctuates with the seasons, and I took my results on one day, in one year, toward the end of august. The river level may change immensely over a year, and the water table certainly does. This would greatly affect my readings were I to come back in winter or spring and perform exactly the same series of tests. Furthermore, the river changes over the years as well: meanders form and shift and tributaries dry up or connect to the river. Humans change themselves and their activities, which can have indirect effects on a river that runs through a conurbation like London, but they also change the river itself, and the landscape, which has a direct effect on the river.
To be thoroughly accurate in my readings, then, I would need to make the same tests more than once: rather than finding out about the river in a snapshot of time, I would need to take readings maybe once a month every two years for ten years, or maybe that's too extreme. I could take readings once a month for a year, and use past records on the internet or in books to get a vague impression of the past. Once a month for a year would mean that my results encapsulated the effects of the season change: the spring thaw and the winter freeze (unlikely due to global warming), or the summer heat-wave and drought (more likely). In such a wet winter as we commonly experience these days, the chalk bedrock would be saturated and less water would be lost, but in summer, the chalk would be parched and in some places the river may dry up altogether. This fluctuation would definitely affect my readings: less water loss would mean that river width, depth and velocity would increase, maybe uniformly. Moreover, if I took readings over a whole year, and maybe averaged them all, it might be the case that more than just one of my hypotheses is supported; all of them could be.
The depth of the river is obviously not constant, though, unless it is manmade. The six readings that I took would not accurately describe the river's depth; I would need an average. I did work out an average for depth in my data interpretation and data analysis chapters in order to calculate volume of flow and cross sectional area, but I wouldn't get an accurate idea of either of these with six results, or with an average of the six results. Essentially, some of the methods I used in my data presentation and data analysis chapters were also not perfect, so the conclusions I drew from them were even less sound. I could do with taking more readings: rather than 6 per site, maybe 10 per site. It may also help if I did not average the depth readings at first, but calculated volume of flow and cross-sectional area from each of the six results, and then either kept it like that or averaged them after the calculations, depending on how I needed to present the data.
In addition, there was the fact that I had to find bridges to take readings from. Not only did this mean that my sites weren't exactly 1km apart, and therefore that I wasn't getting a clear picture of the river's variation, but it also meant that my readings may have been affected. For example, as mentioned on pages 22 and 23, because I took readings from bridges, it was hard to keep the string vertical/taut when measuring depth and velocity, so my readings there may not have been completely accurate. In addition, the home-made instrument I used to measure depth and velocity wasn't that accurate, so I would use the proper equipment, a hydro-prop, if I were to do the experiment again. The day that I chose to do the study on was fairly good: it was not too hot and not too cold, so I wasn't uncomfortable taking readings, and, more importantly, it had been like that for the past week, so the conditions of the river were probably average: not too little water due to drought, not too much water due to flood. However, as previously mentioned, this is still just a snapshot in time, and the river conditions could vary greatly across that week/month/year, even if there was no drought or flood that week.
All considered, I think I did just about the best investigation I could in the circumstances, but it could have been greatly improved with a lot more time, money and effort if I was a professional or had a professional need.
Conclusion:
In conclusion, the river seems so follow hypothesis 1, ("as the river flows downstream, it will get wider and deeper"), but not 2 ("as the river flows downstream, the rate of flow will increase") and 3 ("as the river flows downstream, the gradient of the river will get shallower"). It would be interesting to see whether the same results are found on a different river, or if other rivers support my hypotheses. In this investigation, I aimed to prove or disprove three hypotheses drawn from my background knowledge (refer to page 3). I wanted to investigate how the River Colne changes as it flows downstream. I have certainly completed the latter part: I have gained an insight to the Colne's treacherous nature and attained a general idea of how it changes as it flows downstream: it does not follow the theory I have learnt from my lessons, but its width, depth, velocity and volume of flow vary considerably in a very short stretch of river due to several factors. I have completed the first part of my aim too: I proved or disproved all of my hypotheses. In actual fact, I only proved one of them, but I found out that I would need to take readings over a longer time period and a larger sample to be sure of my conclusions, or to draw completely new conclusions and be sure of those. But really, I should study a different river, as the Colne's urban setting and interaction with the Grand Union Canal makes it a bad place to study.
Bibliography:
I did not actually use many external resources for this investigation, but the ones I did use are listed here, and I am very grateful to the authors.
Understanding GCSE Geography By Ann Bowen and John Pallister;
Memory Map OS 2004 Edition (copyright www.memory-map.com);
Wikipedia,(www.wikipedia.org); http://geographyfieldwork.com/GCSE.htm;
And also, I borrowed a few ideas from past GCSE Geography Students. Thank-You.
GCSE Geography Coursework: River Study James Graham, 11T
Page: 1/25 11:30:25 08 Jan 08