Data was collected from over 25 sites along a 1.5km stretch of river ending at the confluence of the River Alderbrook with the River Blythe. Systematic line sampling was implemented, as this is the best form of sampling to measure change along downstream. The data we collected in the river was primary data – we physically measured data ourselves and obtained original results in order to have first hand evidence for our fieldwork. For a piece of secondary research, the class briefly looked at the Environment Agency’s website to give some background knowledge and previous statistics. The data collected was supposed to be evenly spaced along a transect line, however we often measured the riffles for ease of access into the shallower water – therefore our results may not be so reliable. It is important to collect accurate and reliable data so the results can be stated as true. As a result, all the data from each group was collected on the same day. At times the accuracy of the sampling strategy was questioned as it was not always possible measure the intended point of the river due to the safety of accessing water too deep to work in. It was also important to ensure the data is written down accurately in the field and then transferred effectively onto the database, in some cases there was a confusion with where the decimal point should go with the values!

The sampling strategy we used was systematic line sampling. We sampled in groups of 4-5 people, as we believed that a larger group would increase personal safety and resources. We did a large number of sampling sites (over 25) because a higher number of samples make the validity of the results more reliable. In addition, the repetition of the systematic sampling technique improves the reliability. A huge disadvantage is that some groups could distort the data, causing bias and therefore affecting the results. In order to avoid this, team leaders could check on the groups more often out in the field in future investigations and repeating the exercise a few times would improve the results further. Also, a larger amount of samples or area sampled would give a more accurate representation of how the river should be.

We wanted to analyse the variations in stream characteristics. The results we obtained as a class were much better than originally expected as we used the most appropriate techniques to represent our data. Not all results were accurate because different groups used different techniques and some may be biased as some groups may have distorted the results. Of the sites sampled at least 5 sites became useless due to inaccuracy in technique and the actual river itself – some groups measured just in the pool areas. All groups used the correct equipment to ensure fair and accurate results however, a lot of the equipment may not have been calibrated – the hydro-prop propellers did not spin at the same speeds as some were clogged with silt. This leads to areas with no results and others where the results were questionable.

From the basic results of velocity, (width + depth=cross-sectional area) discharge was calculated

From the basic results of cross-sectional area and wetted perimeter, hydraulic radius (or stream efficiency) was calculated.

In order to analyse the data we had collected, we used simple graphs to show width and depth – it was also possible to show average (mean) and maximum depths at each site. One of the averages taken was the mean of the sediment size at each site, this made the data statistically more reliable. Modal and median results were also depicted. Several scattergraphs with best-fit lines were drawn to see if some of the data – such as width and depth would give a positive correlation. Once a clear pattern was established, the data was than calculated using the Spearman’s Rank Correlation Coefficient were a correlation of over +0.7 established that the Bradshaw Model could, in part, be applied to the AlderbrookRiver. We believed that scattergraphs are the best way to analyse our data collection because they clearly show correlation patterns and can handle a lot of data.

The data we collected shows that generally, all the categories, discharge, velocity, bank-full width, wetted width, wetted perimeter, depth and sediment size follow the trend of the Bradshaw Model. However the results may not be accurate because the AlderbrookRiver has previously been partly channelised (during the 1930s to prevent flooding in the gardens of the housing adjoining BruetonPark), however the Bradshaw Model is based on a river’s natural course. Therefore the channelisation does not fit the model due to human intervention and influence, and so, the data cannot be fully reliable. We feel we have answered all the initial questions and aims because the river is so similar to the Bradshaw Model. For example, one of our original aims was to find out how river discharge varied and from our results we are able to state that as the river approached the confluence the discharge (Q) increased.

There were many strengths in our investigation including a good sample size, good weather and clean water and limited health and safety issues that the task was made available to everyone. Yet these factors do not make it reliable. It was felt that the investigation was somewhat unreliable as there was uncalibrated equipment (e.g. the hydro-props), the data collection was different for each group, the planning and consistency were poor, the river had been partly channnelised and the vegetation obstructed the equipment. Also some groups found erroneous data on their database due to a confusion between centimetres and metres, this needed to be addressed as velocity, discharge and stream efficiency calculations could go wrong. These factors all outweigh the positive factors that included the shear number of sites were investigated (over 25). If we were to repeat the investigation, we should have fewer people investigating a longer stretch of river (hopefully finding the source) over a longer period of time. Also there was no measure of the environmental parameters of the river including water temperature, fish populations and the presence of plant life (flora) and aquatic invertebrates (insects). The lack of this type of data also needs to be addressed.