This investigation is based on the river Horner just north of Exmoor national park. I am going to investigate how the river changes downstream.
On the trip we were split into two field groups with leaders and teachers. We then split down into groups of three to collect our results. We went to various sites on the river, at each site we took our measurements and recorded them. We also took photos and made field sketches.
We chose to go to the river Horner partially because it is reasonably small. This makes taking readings easier and much quicker as we could use basic equipment. It also meant that as a class we could get all of our data in one day.
In my investigation I collected information on the following parameters:
Cross sectional area
These parameters cover many of the aspects and key characteristics that a river has. They also tie into each other; this is useful for finding patterns and explanations for my results.
Erosion is the breaking up of material.
As the river moves through its course it changes from vertical to lateral erosion:
There are four types of erosion
> Hydraulic action – the river wears away the river bank from underneath.
> Attrition – the rocks being carried by the river smash together and break into smaller particles.
> Abrasion – the rocks carried along by the river wear down the river bed and banks.
> Corrosion – the smaller particles are dissolved by the acidic river water.
As a river develops its order of stream increases. This tells us how many tributaries have joined the river. A 2nd order stream has two tributaries, a 3rd order stream has four and a 4th order stream has 8 tributaries and so fourth. So when two 3rd order streams meet at a confluence a 4th order stream is created.
Predictions and hypotheses
With my knowledge of river basics I have made the following predictions on how the channel parameters will change downstream.
Predicted changes downstream
Reason / theory for prediction
Increase – bigger increase when river order increases
These parameters increase because the river is going to get bigger downstream.
Cross sectional area
As the river gets bigger it should have less of a wetted perimeter to water ratio meaning the hydraulic radius increases.
Typically a hill becomes less steep as you go down hill.
Decrease – I think that the gradient decreasing will mean the velocity decreases.
Where the river is steeper the water should run faster
(due to gravity).
As the river gets bigger and holds more water it should have a greater capacity to move the water (discharge).
Attrition (explanation in river basics)
Become more rounded
To find the changes in channel parameters down the river Horner I am going to collect a series of data at six different sites along the river’s course. I am collecting my data with a group of three friends from my class.
I am going to measure the width of the river at each site so that I can calculate the cross sectional area (CSA) and the hydraulic radius. The width will also show me how the river progresses.
I will measure the width of the river along the waters surface to the point where the water meets the bank. I will measure the width using a tape measure because a metre stick simply isn’t long enough and therefore wouldn’t be accurate. When measuring this I will need to keep tension on the tape otherwise it will go downstream as the water pushes on it. This would be inaccurate as it needs to be a straight measure from bank to bank.
I am going to measure the wetted perimeter so that I can calculate the hydraulic radius.
The wetted perimeter is a measurement of the river bed starting and finishing where the river meets the bank to get this measurement I will use a tape measure as it is flexible, this will enable me to measure the bed accurately.
I am measuring the depth to get an average depth of the river at each site. Using the average depth I can calculate the cross sectional area. And also produce cross section diagrams.
I will measure the depth of the river at five evenly spread places across the river. Five depths should give me a good average.
To make the measurements an equal distance apart I will divide the width by six, the answer is the distance the measurements will be apart. There are limitations as I am using a meter stick I could not get an accurate reading for a depth of more than 1 metre.
Cross sectional area
Finding the cross sectional area will mean I can calculate the rivers discharge and hydraulic radius. To calculate the CSA I multiply the width and the average depth. I will measure it in metresï¿½.
The hydraulic radius will tell me the efficiency of the river at each site.
You find the hydraulic radius by dividing the CSA by the wetted perimeter. The bigger the hydraulic radius the more efficient the river is.
If the CSA of the two streams is equal then which is more efficient? The one that has the bigger efficiency is the one with more water touching on a smaller area (wetted perimeter). In this case it’s the river on the left as the wetted perimeter is smaller and thus the hydraulic radius is bigger. The measurement of hydraulic radius should show me how the river is changing as it flows downstream.
I will measure the gradient at each site so I can see if it affects the velocity and any other river parameters. I will measure the gradient using a clinometer. Before starting I must calibrate the clinometer. I will do this by standing on a flat surface three meters from a partner I will look through the clinometers sights and line it up so that it reads 0. I will note where on the body the sights are pointed at. When I want to find the gradient I will aim the sights at the spot and take the clinometer’s reading.
I am measuring the velocity of the water at each site so that I can calculate the discharge. And see how the velocity changes downstream. Also the velocity should have an influence on other parameters and I want to explore its impact on them.
I am going to obtain velocity using a hydro-prop flow meter (above). This measures the flow rate of the water. To gain a fair reading from each site
I will take three readings. As I did with the depth I will divide the width by four and the answer will be the distance between each of the readings.
I decided that when I was to take a reading I would place the impeller one turn from the top (far left on the above picture) this technique prevents it being too tight and not moving. When someone timing starts you release the impeller allowing it to turn. When it reaches the end you stop the timer.
To calculate the velocity you use this sum:
I am going to find the discharge because it will show me the changes in the amount of water the river moves every second at each site. I am going to measure the discharge in meters cubed per second or cumecs (mï¿½/s).
I’ll calculate this by multiplying the CSA and the average velocity.
I am going to investigate the pebble size and shape to see how they change downstream. I will measure pebble size using a pebbleometre; I will measure the A axis (the longest length) of the pebble.
I am going to use the pebbleometre (above) to do this as it is more accurate than a ruler as I can be sure where to measure to.
I will also be looking at pebble shape the pebbles I pick up I will put into the following categories:
Very angular (VA)
Sub angular (SA)
Sub rounded (SR)
Very rounded (VR)
Putting them into categories will make it easier to identify the pebbles into groups which will be very beneficial when I am presenting my results.
Data presentation and analysis
I have collected results for all of my river parameters; I am going to present them each with graphs and analyse each one. I will then compare parameters and see how they are interlinked and hopefully come up with explanations for my results.
The graph shows that as I predicted the wetted perimeter increases downstream. I think this is because the depth increases as well as the width downstream.
As you can see the width also increases downstream. This is because as the river goes downstream it carries water and so the river gets bigger it does this by increasing its width and depth. There are some exceptions the graph does not show a perfect increase – site three seems to break the pattern. I think this maybe because the river isn’t as deep at site three in proportion to the other sites. This may be due to a hard rock type, which would mean that it couldn’t erode and become deeper so easily, so the river became wider. In my field notes I had written “big boulders – possibility they fell from cliff on the east bank”.
This may suggest there was a particularly hard rock type at site three. If there was a particularly hard rock type at site 3 then the pebble size should be very big at this site (see average Pebble size graph). Another explanation for the width at site 3 being particularly big is that the erosion has changed from vertical to lateral. This means that the erosion doesn’t concentrate on the rivers bed but erodes the rivers banks more creating a wider but shallower cross section. If this is true then it may be because the gradient has decreased. (See gradient graph)
The depth also increases downstream i.e. the next river site is deeper. Site 6 is an exception to this its depth is smaller than site 5’s. I believe this is because the erosion has changed from being vertical to being lateral. I think this because there is a big change in gradient between site 5 and 6 and this is what changes the focus of the erosion – where the river is descending steeper the river erodes its bed more than its banks, this is the other way around when the river reaches a flatter part of its course (normally and in this case further downstream). When the erosion is more lateral, as I believe site 6 is, the river will get wider and not be as deep. This is evident in both the depth graph above and on the width graph on the previous page.
In most rivers this changes is gradual and starts in the middle course of the river, according to my results the change happens much closer to the mouth (the lower course of the river) on the river Horner. I think this is because the river Horner is particularly small and comes of the hill right by the coast. From the map I can see that the land is steep until very close to the sea. As you can see from my map (introduction) there is a bigger distance gap between site 5 and 6 than between the other sites. So as I did not collect data between the sites I don’t know how quickly these changes happened or where exactly they happened.
The cross sectional area increase as the river develops towards its mouth. This is hardly surprising as both the average depth and width have increased (the CSA=average depth x width). I have already explained why these two parameters have increased.
This graph shows me that the rivers efficiency in shifting water increases downstream. It shows me this because the hydraulic radius is a measure of river efficiency. I have already explained this in my methodology, just to clarify – a bigger hydraulic radius must have a bigger / equal amount of water touching a smaller area (wetted perimeter).
The graph tells me that downstream the river discharge increases. Sites 1 and 2 are very small in comparison to the other sites. This is because their cross sectional areas were much smaller than the rest. Between sites 2 and three the river becomes a 3rd order stream meaning that two more streams have joined the river. This explains why there is such a gap between site 2 and 3 because at site 3 the river will be carrying more than two streams worth of water more than at site 2. Once again site 6 is smaller than site 5.
I have found out that once again another parameter (velocity) is increasing downstream. There appears to be a jump in velocity between sites 2 and 3. This is because at site two the river is a 2nd order stream by site 3 it’s a 3rd order steam meaning two more tributaries have joined the river. These extra tributaries will have dramatically increased the amount of water flowing down the river (evident in discharge graph), and therefore increased the velocity. The two biggest increases in velocity were at sites 3 and 5 where major tributaries had joined the river.
I am however uncertain as to how the river is speeding up downstream. I had originally thought in my predictions that the gradient would be the factor that determines the velocity. It is not, this graph shows why more clearly.
The graph shows a strong negative correlation, this means that as the gradient increases the velocity decreases. Clearly the velocity does not increase because the ground is becoming flatter; so on the whole the gradient is not the main factor influencing the velocity.
If it is not the gradient then this increase surely must be due to an increasing CSA.
The graph shows a very strong positive correlation i.e. as the CSA increases the velocity increases. I think that the cross sectional area is the factor making the velocity increase because the more water in the river the more mass it holds therefore the faster it moves.
There is one other factor that I think may influence the velocity. That is the hydraulic radius, where the river is most efficient will be where the river is fastest. For example I think the river is most efficient at site five. This is shown on the hydraulic radius graph, as site five is the biggest bar and fastest site by some way.
This scatter graph shows a strong correlation. It shows that as the river becomes more efficient it gets faster. This proves that it is certainly a factor influencing the velocity. So I think that the velocity increases due to an increasing CSA and hydraulic radius.
The graph shows that as the river goes downstream the pebble shape becomes more rounded. The graph shows this as the bars that represent the more angular pebbles peak earlier, and the more rounded later. I think that the pebble shape increases downstream for a couple of reasons.
Firstly, in the upper course of the river on the moors the rock type is quartzite, one of the hardest rock types. This would make the rock hard to erode and therefore a majority of pebbles would be angular. Secondly, I think that the pebbles get rounded downstream due to velocity. The faster the water is moving then surely the more attrition will happen in the river. When attrition happens the pebbles are knocked into each other and slowly become more rounded. Also pebbles are always moving downstream and therefore the ones that are further downstream tend to have been eroded longer and therefore will be smaller and more rounded.
This graph compares the pebbles at site 1 and demonstrates that they become much more rounded downstream (for reasons already specified).
I have been using the same explanation for pebble shape getting rounder as I have for pebble size getting smaller downstream (pebble size explanation is 2 pages on). This graph shows the relationship with pebble size and shape, that the bigger pebbles are the most angular.
As the graph shows you the pebble size is decreasing downstream. The graph also supports my theory about the width and depth for site 3, as the pebble representing site 3 is bigger than all of the others apart from site 1. This shows that there is a chance that site 3 did contain a hard rock type as the harder the rock the harder it is for the river to erode the pebbles, so on average the pebbles are bigger. Also when I looked at the width graph I said that the big width at site 3 could also be due to a change in erosion from lateral to vertical, this would be due to a decrease in the gradient.
As this graph shows there is a dip in the gradient at site three. So I think that the river is wider and shallower at site three due to a change from vertical to lateral erosion and due to a possible harder rock type. This is the same explanation for why the pebble size is so great at site 3. The gradient dips at the two sites where I said that the erosion was more lateral than vertical.
This graph does show a correlation, however there are two anomalies (circled). I think there is a relationship between velocity and pebble size, but that other factors, such as a hard rock type, also influence the river’s pebble size and shape.
There is a clear pattern in all of the graphs shown so far, other than width and wetted perimeter. They all show that the various parameters are decreasing between sites 5 and 6 to. This is for the reasons that I specified under the depth graph. I am going to draw a table to show the difference between site six and five for the various parameters and explain the differences.
Difference between site 5 and 6
Due to changing to lateral erosion by site six the width increases, as it is the banks that are eroded.
Due to such a large increase in width the wetted perimeter has also increased, despite the decrease in depth.
Due to changing to lateral erosion by site 6, the bed of the river is eroded less and the river isn’t as deep.
Cross sectional area
Cross sectional area has almost remained the same as site 5. As the depth has decreased rapidly the width increased making it almost the same as site 5.
The river has become less efficient due to its increase in width and decrease in depth. This is explained clearly in the methodology section, where the left hand diagram is like site 5 and the right hand is like site 6.
The gradient has been constantly decreasing downstream.
For the same reason as the erosion changing to lateral there is a big decrease in the gradient between the two sites and therefore the water is flowing more slowly.
The river is discharging less water due to the decrease in velocity.
The pebble size has constantly been decreasing due to the factors explained above. The same factors mean the pebble shape has also constantly been rounding.
The data and analyse broadly supports the hypotheses set out above. However there were some important exceptions, in particular velocity appears to be dependant on the volume of water and the efficiency of the channel rather than the gradient. In addition, there are some important changes as the river approach its mouth.
From my results and analysis I have generally discovered that the following changes occur downstream on the river Horner:
* The river gets deeper
* The river gets wider
* The rivers wetted perimeter increases
* The rivers cross sectional area increases
* The hydraulic radius increases so the river becomes more efficient
* The river discharges more water
* The gradient of the river decreases
* The velocity increases
* The rivers pebbles get smaller as well as rounder
* The erosion changes from vertical to lateral
I have found possible explanations for all of the above in my data analysis.
I have also found that by site six some channel parameters have started to change in the following ways:
* Depth decreases
* Hydraulic radius decreases
* Discharge decreases
* Velocity decreases
And I have explained them all in a table in my analysis.
On the whole I think my results were fair and within a reasonable degree of human error. However there are some readings I think I could have got more accurately. One of these was the readings was depth. When I was taking the readings I noticed deep areas followed by shallow areas these are called pools and riffles when I took my readings for depth I took them regardless of weather that particular cross section was a pool or a rifle. This means my data may not be totally representative of the river. If I were to repeat the experiment I would make sure to either measure all sites in a pool or all the sites in a rifle. An even more accurate way to do this would be to measure the river at each site in both a pool and a rifle and average the results to get the data. This is the main reason I used class data for most of my representation, so that it would balance out factors such as pools and riffles and the data should represent each site of the river.
The other reading that I felt was inaccurate was the pebble shape. It was quite easy to judge the shape of the pebble incorrectly, when looking at six diagrams to try to identify which label to give to it. This reading was highly subjective and could be inaccurate. To improve this method I would.
When measuring the width it wasn’t always easy to determine the edge of the river due to large rocks and overhanging vegetation. To keep my results fair I always measured under vegetation as far as the water went.
Vegetation and sediment caused the impeller to stop several times when measuring velocity. The impeller also needed a certain force to get going meaning that sometimes it wouldn’t start for a long time. A solution to this could be to use a float and time it to get to a certain distance.
Measuring the gradient was inaccurate at times as the person was never at a specified point from the other person measuring. Also I question how accurately the clinometer was calibrated.
So overall I am pleased with my investigation, although I felt my method could have been improved in some areas I have managed to conduct a fair experiment, and I feel that my analysis and explanation of the rivers characteristics was accurate.