The grass I will be measuring is Wood Meadow-grass (Poa nemoralis). It forms loose tufts in woods, and other shady places, as well as forming large populations in open grassland.
After looking at possible factors affecting the grass height, I decided that it was one of two possibilities. Either trampling, which would stop the grasses being able to spread their roots, or a natural process. I then chose a site well away from any paths in order to negate the effects of trampling, which left only a natural process to affect the plant height.
Plants produce energy by photosynthesis. This is completed via a series of chemical reactions (Figure 3), which simplify to become:
Water + Light + Carbon Dioxide → Glucose + Oxygen
The carbon from the carbon dioxide is used to build up complex organic molecules, which are used to build up cell components, and enable the plant to grow.
Photosynthesis relies on the following conditions:
- Light
- Water
- Carbon Dioxide
- Suitable Temperature
- Adequate supply of magnesium and iron (to form chlorophyll)
Because the two sites were so close together, it is almost certain that the temperature, and carbon dioxide levels were not significantly different, so It would suggest that either light, water, or soil type (to alter magnesium and iron levels) was the key variable in affecting grass height.
With an Auger, I took a sample of soil 30 cm deep, from both sites. There were 3 clear layers in the sample. I found that there were no significant differences in soil type, or water levels. This meant that light was the key variable in affecting grass height.
The top 7cm was the humus layer, which was dark brown in colour, and was quite dry. The 7cm – 18cm layer was a silt loam layer, with a light brown colour, and was slightly moist. The bottom of the sample was made of silt clay loam, with a grey-brown colour, and was moist.
Clay loam soil is made from very small particles. The fact that these particles are small means that the spaces between each of the particles are also very small, which causes the soil to be largely nonporous. This is evident from the soil sample, as the soil was moister the lower, and more clay-like the soil became.
Prediction
Grass height will be greater in areas of high light intensity, where clearing has occurred, than in shaded areas, where natural succession has occurred, and trees stop light reaching the ground.
Controls and Variables
Throughout the experiment, I will try to keep as many of the factors the same, apart from those, which I am comparing. Temperature should be very similar between the two sites, as they are within 100 meters of one another. All coordinates for both sites were achieved using a random number generator, so no human bias was introduced here. The soil types are the same, which I discovered in a preliminary investigation, so this wont alter the results between the to sites. The edges of my two sites are both at least 3 meters from paths, so soil compaction should not alter either set of results.
I will be varying the canopy cover, and therefore light intensity, which should affect the amount of photosynthesis, and the growth of the grass plants.
Risk Assessment
Rushey Plain consists of a wooded area, and a cleared, grassy area. There is a foot and wheelchair path running through the wooded area, and the between the wooded and grassy areas. In the wooded area, there is the risk of low branches hitting people on the head, or poking them in the eye. This is less of an issue whilst on the path, due to maintenance by the Corporation of London; although some branches could have grown across the path again since they were last cut. In order to minimize this risk, it is necessary to watch where one is going, and use paths whenever possible.
Away from the paths (where I will be carrying out my studies) there is the possibility of fallen trees, or roots protruding from the ground. There are also some ditches and rabbit holes on the area. All of these factors present a tripping hazard. This hazard can also be minimized by watching where one is going, and staying to the paths whenever possible.
Rushey Plain used to be quite marshy, and the clay-based soil is water retentive. It is also within half a mile of a lake, so there is the possibility of mosquitoes biting, and causing irritation, or even infection. To minimize this risk, insect repellant should be worn.
The moisture retention could also cause some puddles, and areas or slippery mud to remain a problem even during the summer months. To minimize the risk of slipping, boots should be worn at all times, and muddy areas avoided as much as possible.
There are some plants at the sites, and on the journey to the sites that contain sharp thorns, or stings. In order to reduce the risk of injury on these plants, long sleeves and trousers should be worn, and one should keep to the path, where these sorts of plants have been cleared.
Whilst on the paths, there is the small risk of colliding with horse riders, or mountain bikers. This risk can be reduced, by listening out for either of these path users, and standing to one side if they approach.
There is the risk that bacteria from handling any equipment, or soil or plants could be ingested and cause illness. To reduce this risk, eating and drinking should be prohibited until after the recording is finished, and one has washed their hands thoroughly.
Method
Measure out the study site in the cleared area using two 10M tape measures, laid out fully at 90 degrees to each other. Then access the first coordinate listed, whilst minimizing effects of trampling to the rest of the study area. Once at the specified coordinate, measure the light level, by pointing the white circle upwards, and recording the value displayed. Then place the meter rule perpendicular to the ground next to the tallest grass plant, and record its height. Then place the quadrat on the study area, and move it to ground level, by threading the plants through the smaller squares. Once at ground level, count the number of squares that are fully occupied by any ground covering; then the number that are partly occupied, and estimate how many full squares this represents. Sum the totals, and record number on Vegetation Recording Sheet.
Go through the process again for the other coordinates listed, and then repeat all of this for the wooded area.
Justification Of Method
- The 10m X 10m area is big enough to get a good variety of samples from each site.
- By minimizing trampling, disturbance of later recording areas is reduced.
- Light intensity is the key variable, and therefore has to be recorded.
- Height of vegetation is affected by the quadrat being placed, so it has to be measured before that.
- The quadrat is used to measure the percentage ground cover.
- The measuring process is explained with clarity, to enable anyone to repeat the test.
- The test needs to be repeated for all types of ground cover, and in all places in order to obtain a variety of results.
Equipment
- 2 X 20m tape measures
- Lux meter
- Meter rule
- Gridded quadrat (figure 5)
- Vegetation Recording Sheet
Justification Of Equipment
- The tape measures are used to mark out the site, and then to locate recording areas.
- The lux meter gives an accurate measure of light in lux, up to 200,000 lux.
- The meter rule is used to measure the height of plants, and with a small base surface area, it has little impact on the recording area.
- The gridded quadrat allows an accurate recording of percentage ground cover to be taken.
- The Vegetation Recording Sheet is used to record all the values collected.
Evaluation
The scatter graph (figure 6) shows that there is a positive correlation between light intensity and grass height, although there are many points, which are a long way from the line-of-best-fit.
When I picked my grassland study area, I had to choose an area, which was 10m x 10m in size, and did not cross any paths. The site that met these two requirements was partially in the shadow of an oak tree. At different times in the day, these shadowed areas would have been in direct sunlight (figure 9). Similarly, some of the areas, which were in direct sunlight at the time of measurement, were shaded at other times of the day. This meant that some of my light measurements could have been atypical of the average light intensities. This could account for the weak correlation on the graph to show correlation between height of grass and light intensity.
The method used to measure ground cover uses estimation, which introduces human error. It is difficult to assess the cover when the plants have greatly different heights, such as in the cleared area of my study. Small species, are often covered by larger plants, and may not be recorded.
The measurement in height of the grass was only accurate to one cm, and human error was created, as I had to choose by sight which grass plant I thought was tallest.
Results
Rushey Plain (Cleared Area)
15-07-2005
Rushey Plain (Wooded Area)
15-07-2005
Mann-Whitney U-Test
The Mann-Whitney U-Test is used to compare two sets of data. It is useful, because it does not need the data to fit a normal distribution pattern. The test compares the medians of the two data sets, by looking at the overlap between two sets of data (Figure 7). The test needs a minimum of 6 replicates of data, but the two data sets can be unequal.
Before the test is carried out, a Null Hypothesis (H0) and an Alternative Hypothesis (H1) must be established. H0 is that there is no significant difference between the two data sets, whereas the H1 is, that there is a significant difference.
By setting out the data in a table, it is possible to rank each piece of data. Then, by a series of calculations, two Mann-Whitney U values are produced. As long as the smaller of these two values is above that needed for amount of values, H0 can be rejected.
H0 = There is no significant difference in grass heights between the wooded, and grassland areas of Rushey Plain, Epping Forest.
H1 = There is a significant difference in grass heights between the wooded, and grassland areas of Rushey Plain, Epping Forest
ΣR1 = 149
ΣR2 = 61
In the equation, the numbers n1 and n2 are needed. These are the number of replicates per data set. In my study, n1 and n2 are both 10
The final equation gives values U1 and U2. These are the Mann-Whitney values for the test.
U1 = n1 x n2 + ½ n2 (n2 + 1) – ΣR2
= 10 x 10 + 5 (10 +1) - 61
= 100 + 55 – 61
=94
U2 = n1 x n2 + ½ n1 (n1 + 1) – ΣR1
= 10 x 10 + 5 (10 + 1) – 149
= 100 + 55 – 149
= 56
U2 is the smallest value, and in order to reject H0, it must be greater than the value shown in Figure 8. The value for 10 x 10 is 23. 56 is higher than 23, and therefore, H0 can be rejected.
This shows that there is a significant difference between the two areas of Rushey Plain, which is what I set out to prove at the beginning of this investigation.
Evaluation
The scatter graph (figure 6) shows that there is a positive correlation between light intensity and grass height, although there are many points, which are a long way from the line-of-best-fit.
When I picked my grassland study area, I had to choose an area, which was 10m x 10m in size, and did not cross any paths. The site that met these two requirements was partially in the shadow of an oak tree. At different times in the day, these shadowed areas would have been in direct sunlight (figure 9). Similarly, some of the areas, which were in direct sunlight at the time of measurement, were shaded at other times of the day. This meant that some of my light measurements could have been atypical of the average light intensities. This could account for the weak correlation on the graph to show correlation between height of grass and light intensity.
The method used to measure ground cover uses estimation, which introduces human error. It is difficult to assess the cover when the plants have greatly different heights, such as in the cleared area of my study. Small species, are often covered by larger plants, and may not be recorded.
The measurement in height of the grass was only accurate to one cm, and human error was created, as I had to choose by sight which grass plant I thought was tallest.