I finally remember a disk and it 'may not be formatted'. Bollox it isn't formatted! I was using it last night!! Bloody computers..
The theme of how moisture content, organic content and pH affect species diversity across a sand dune seemed to form an appropriate investigation as it could be related to, and compared with, the Waugh and AS textbook examples of Morfa Harlech and Braunton Burrows respectively. Due to the difficulties encountered during the execution of the tests, the site and certain aspects of the methodology may now be considered less appropriate.
The results gathered during the course of this investigation suggest that, rather than another factor/set of factors being responsible, it may be down to the limitations of the investigation itself. The pH readings offer an example of this: published examples - textbooks (secondary data) suggest that the pH change normally ranges from 8 at the embryo dune to between 5 and 6 on the scrub community while the results of this investigation offer a range of 6.5-3.9. A universal indicator solution and a series of colour charts were used to analyse the pH of the collected samples, misreading the charts would not account so such a dramatic misjudgement on all pH data so it can only be assumed that there’s another explanation; perhaps the solution was contaminated or the sample bags weren’t sterile - this could be overcome by using a pH probe on site, cleaning it thoroughly between testing samples. The time delay between sample collection and sample testing could also have an effect on results. Every effort was made to ensure that sample bags were sealed, kept away from direct sunlight and handled with clean equipment. Human error may have led to contamination of the sample or damage to the sample containers.
The method used to collect the samples may also have contributed to the confused results - a random sampling technique may land near rabbit burrows, away from vegetation or on a footpath, for instance, though this is difficult to overcome as altering the position from which a sample should be taken causes bias within the experiment so an alternative method should be considered for future experiments or replications. A stratified sampling method would have ensured an equal number of samples were taken from each dune:
This is also an easier method to replicate.
As samples were collected in late July and, therefore, at the height of the tourist season, it must be assumed that the area - and therefore the samples, were affected by trampling, litter and noise: frequent use of paths and surrounding areas for recreation by tourists causes compaction of the soil, reducing pore space; litter, including food scraps, add artificially to the organic content of the soil, it also encourages more animals to the area such as birds and rabbits, again, adding to the organic content of the soil, it could also be argues that this increase in animal population will combat the compaction caused by tourists; noise will discourage animals from inhabiting the area. For replications, an alternative site should be considered, a site that has all aspects of a dune system available for testing.
Although the data’s reliability and validity is questionable, there was enough collected to carry out statistical analysis using Spearman Ranking Correlation Coefficient - eight pairs of data are needed and seventeen were collected.
Throughout the investigation health and safety issues were addressed and preventative measures were considered and enforced. Tide times were consulted before a date for data collection was chosen and travel times were organised around this. Protective clothing was used when heating the samples (when testing for moisture content) and using U.I. (for pH testing). A pilot study may have helped to prevent many of the problems experienced during the investigation - a preliminary study of the proposed area/dune system would have made it clear that a site clear of human influence would be beneficial. Given that the data was easily affected by human influence, two dune systems could be considered and compared to a control such as Braunton Burrows. Any differences between the two sites should then also be investigated. The sites could also be assessed in two distinct seasons (collecting two sets of data from each site: one in July and one in November, for instance), as results are affected by climate also.
The conclusions drawn in this investigation were not clear due to problems with equipment, the area chosen and the methodology. Even though the study focused on three factors’ (moisture content, organic content, pH) affect on species diversity, other factors were considered, such as land height, wind speed and light intensity for example. As there were no discernable links found between these factors and species diversity, the reasons for this could also be investigated if the study was to be repeated and extended.