Fieldwork: Studland Sand Dunes

Location: East coast of the Isle of Purbeck in Dorset, nearest village is Studland, close to Poole and Bournemouth

Aim: To investigate environmental and vegetation changes across Studland Bay Sand Dunes

Hypothesis: Percentage of vegetation cover and number of plant species will increase as you move inland from the embryo dunes

Why were Studland Sand Dunes Chosen?

• Fully developed 600 year old dune system as otherwise if it was not fully developed there would be data missing leading to inaccurate conclusions
• They are a National Nature Reserve and a SSSI owned by the National trust so they are protected and have been able to develop with  a minimum of human interference - any changes we observe in the natural environment should be linked to natural changes in the immediate physical environment
• The dunes are accessible at all tidal times and there is no risk of being cut off at high tide - full access to the sand dune system and kept fieldwork risks to a minimum
• The beach can have up to 25,000 visitors on it in the summer, and as the dunes are not heavily protected or patrolled we could also observe the human impact on sand dunes
• Very accessible as close to a B road and good parking

There are:

Embryo Dunes followed by

Grassy Dunes

Then a Heather Slack

Then a main dune ridge

Then a tree line

Fieldwork: Risk Assessment

There was a pre-visit to the dunes by staff.

We discussed possible risks in groups and as a class did a risk assessment, looking at the site on google earth to help us.

Risks:

• Sharp Dune grass causing cuts on fingers
• Sand blowing into eyes
• Eating poisonous plants
• Being cut off by high tide
• Snake bite, more serious with a venomous bite than a dry bite
• Verbal abuse and possible physical harm from members of the public

We assigned the risks with a Severity Score and a Likelihood score each out of 5, then used these to determine an overall risk score.

eg  Sand blowing into eyes = Likelihood score: 4   Severity Score: 1   Total risk score:   4
How we would combat this: Student briefing and carrying a first aid kit with eye wash.

Snake bite   Likelihood Score: 2    Severity Score: 4    Total Risk score: 8
How we would combat this: Student brief, wearing appropriate clothing like walking boots, checking where putting quadrat down, having mobile phones to call 999

Fieldwork: Spearman's Rank Correlation Coefficient

A simple statistical test examining the relationship between 2 sets of data.

95% significance: 0.377       99% significance: 0.534

For % of vegetation cover: +0.41 so there is a 95% certainty that it and distance from the sea are related.

For number of plant species: +0.47 so there is a 95% certainty but is more than vegetation cover

There is a moderate positive correlation.

Given that we had had 20 sites we can say with certainty that there is a 95% correlation between the variables. We can't be 99% sure.

Advantages:

• Tells whether direction of relationship is positive or negative
• Tells exactly the strength of the relationship unlike the scattergraph which only gives a general indication

Disadvantages:

• Correlation does not equal causation
• Simplistic technique
• Need at least 10 sets of data

Fieldwork: Scattergraph

A scattergraph of distance from sea against percentage plant cover.

Dependent Variable: % of Vegetation cover
Independent variable: Distance from the sea (the one we changed)

Positives:

• Shows the 'direction' of a relationship between 2 sets of variables through the line of best fit
• Indicates strength of correlation by how close the points are to the line of best fit
• Simple and visual representation of correlation
• Shows anomalies

Negatives:

• Have to subjectively assess the strength and correlation of the relationship
• Individual sites are hard to identify

Ideally we would use it in conjunction with a spearmans rank.

Our results so show a positive relationship so vegetation does seem to increase but it is a fairly weak relationship as there are lots of anomalies.

Fieldwork: Improving Reliability

In order to improve the overall reliability of our results, reduce the impact of anomalies and strengthen our conclusions we could have:

1. Repeat our experiments at another time - the environment changes throughout the year
2. Other samples would mean a more valid conclusion
3. We could have visited other sand dunes to compare - otherwise our conclusions are just about Studland Bay Sand Dunes and they might be anomalous
4. Do more transects or spread them further apart to come to genuine conclusions about the sand dunes

Braunton Burrows, Devon:

would reduce the impact of anomalies and increase our results validity.

Fieldwork: Interrupted Line Transect

We did an interrupted line transect across the sand dunes every 10m.

Appropriate?

Yes

• We were looking for changes between the sea and trees, so it was appropriate for our hypothesis.
• If we had done random sampling all 20 sites might have been in the same place and we wouldn't have observed change with distance

Fieldwork: Systematic Sampling

We sampled slope angle, vegetation, wind speed and temperature every 10 metres.

Advantages:

• It meant every area of the dune was sampled - we could see that trend changes with distance

Disadvantages:

• Most of the sand dunes are missed out because we only sampled every 10m
• We may have picked out a trend in nature - eg underlying limestone might cause chalky soils which only allow some plants to grow - This could give us a false impression

Appropriate Method?

Could have used random sampling and put numbers 1 - 200 in a hat to decide where to do our measurements.

BUT in order for it to be truly random we would have to put the numbers back, and all the places we picked out might be the same place 20 times which wouldn't give us an appropriate picture

Fieldwork: Quadrat

To assess vegetation changes we used a Quadrat. It had 25 squares (each square was 4%), and we placed it on the ground every 10 metres (systematic sampling). We estimated the total vegetation cover as well as the number of species, which was vital to our hypothesis.

Advantages:

• Gives consistent/closed area to analyse
• Helps estimate percentages

Problems:

• Can crush/squash larger vegetation like marram grass
• Percentages were a crude guess
• Some species might have been misidentified or obscured
• Might have missed out some vegetation that grows in clumps
• Mights have missed out some species eg trees