We used secondary data in the form of an aerial photograph and a map 1:20000 to direct us to our site. We also used information that we found on an interpretation board to help us identify Spartina. We had no key to identify plants, instead our instructor pointed them out to us, as we were unconfident of our ability to remember the names of so many plants we drew rough sketches.
Collection of Data
- Describe what took place and why. Take photographs and notes.
We chose to use transects as we were looking for an environmental gradient across the salt marsh. Sampling was conducted in a systematic way, there were four pairs with each pair sampled one transect. The transects were spaced evenly with 2m in between them as shown in Fig ? We combined the data from all the transects as the larger the size of the sample, the greater is the probability that it accurately reflects the distribution of the parent population and therefore basing our evaluation on one transect would be unreliable.
Analysis, Interpretation and Evaluation
Our scatter graph shows a positive correlation, there is one piece of anomalous data. The soil moisture content is high as the marsh is flooded and waterlogged by the tides; the relative flatness means drainage is slow after flooding.
Our scatter graph shows a positive correlation meaning the soil has a very high salt content which decreases towards the upper shore, there are however two anomalies. When taking the soil sample from 12 m we noted that it was in a depression, subtle changes in the topography could explain some of the anomalies.
Spearman’s Rank
Soil Moisture
From the scatter graphs a trend was identifiable so we used Spearman’s Rank Correlation Coefficient to measure the strength of the relationship between; soil moisture and distance along the salt marsh, and soil conductivity and distance along the salt marsh. It was an appropriate test as our data was recorded in intervals making it easy to rank, we also had 12 pairs, which is greater than the minimum of 10 required. There are limitations with Spearman’s rank, as it ignores the magnitude of the differences in observed values, it is also problematic trying to rank numbers which are the same. It is also important to note that a correlation does not imply causation, a third unmeasured variable may be responsible for the relationship.
∑d² = 112
rs = 1 –
rs = 1 –
Calculated Value = 0.608
Critical Value = 5% 0.506
The critical value is less than the calculated value therefore we can reject our null hypothesis. However there is a clear anomaly in the data and to provide conclusive proof I will remove the anomalous and redo the calculations.
Without anomaly
∑d² = 56
rs = 1 –
rs = 1 –
Calculated Value= 0.745
Critical Value = 5% 0.506
= 1 % 0.777
The critical value is significantly less than the calculated value therefore we can reject our null hypothesis, that there is no change across the salt marsh. We can conclude to a 99% of accuracy that there is change along the salt marsh.
Spearman’s Rank
Soil Conductivity
∑d² = 66
rs = 1 –
rs = 1 –
Calculated Value= 0.769
Critical Value = 5% 0.506
The critical value is less than the calculated value therefore we can reject our null hypothesis. However there are two clear anomalies in the data and to provide conclusive proof I will remove the anomalous and redo the calculations.
Without anomalies
∑d² = 16
rs = 1 –
rs = 1 –
Calculated Value= 0.903
Critical Value = 5% 0.648
= 1% 0.794
The critical value is significantly less than the calculated value therefore we can reject our null hypothesis, that there is no change across the salt marsh. We can conclude to a 99% of accuracy that there is change along the salt marsh.
Pie charts, annotated photographs.
- Graphs put onto located map.
- Why have you chosen that specific method of data representation e.g. Scatter graph for comparison or bar chart to see variations between components.
-
Interpretation: Linking results to original hypothesis. sashman
Salt marsh plants must also be tolerant of continual submergence and the resulting low oxygen level. Many salt marsh plants are equipped with hollow passages through which air can pass, connecting stomata on the leaf surfaces with roots and providing essential oxygen to root cells. In addition, a chemical reaction between the roots and the surrounding soil produces iron oxide and ferrous sulfate, important nutrients which are particularly essential to cordgrass. The light brown color around the roots reveals this oxidation process.
Presentation of Summary
There were limitations with our soil moisture test, which may have caused the anomaly. A variety people collected the soil samples and they may have been getting them from different depths and horizons. The anomaly could have been explained by subtle changes in the topography that were not picked up by our calculation of the gradient.
There were also limitations with our conductivity test and the result will be influenced by contamination. We should have washed the utensils with a mild detergent and rinsed thoroughly in deionised water. It is also important to note that the electrical conductivity varies not only to the concentration of salts present, but also to the chemical composition of the nutrient solution. Some fertilizer salts conduct electric current better than others.
In conclusion our results for soil moisture and conductivity are, despite the limitations, reliable and after the use of the spearman’s rank correlation coefficient we can reject our null hypothesis.
Our identification of plants for the vegetation zonation hypothesis may have been inaccurate as we had no key, however when we were unsure we photographed the plant and then were able to identify it when back at the centre. Many plants that were not in flower in September, which may have led to some being wrongly identified. Having chosen to combine results from all four transects we had plenty of data, and therefore there is a great probability that our results accurately reflects the distribution of the parent population. It is clear from the results that we can reject our null hypothesis that there is no evidence of zonation.
When calculating the gradient we had only foresight measurement, a backsight measurement would have made the results more reliable. The abney did not pick up subtle changes in the topography and this was a key limitation.
Overall we conducted a very successful investigation and could reject all of our null hypothesises with confidence.
Further work:
It would help confirm our findings to redo the same investigation on another saltmarsh. It would however also be interesting to investigate the change in abiotic factors on a grazed saltmarsh.