SKILLS

Spearmans (strength of relationship) and Standard deviation (is the data close to the mean?):

Spearmans: Using spearmans was good as it allowed for the relationship between sites downstream and and the mean and sites downstream and standard deviation pebble size. Also means that the strength and direction of trends can be easitly identified.  

Using only six sites is very limited to prove a correlation. Using 4 degrees of freedom a spearmans of 0.9 would be needed, which is virtually impossible. To improve, sample more sites so a higher degrees of freedom is needed so that the spearmans is not so unrealistic.

Spearmans ranks mean data and standard deviation data. 

If we had used shreve order on the scattergraphs instead, then a peasons ccorrelation coeffiecient could have been used instead. Better than spearmans because it doesn't rank the data- more accurate and powerful analysis 

Located bar graphs or pie charts were used to show how bedload roundness changed. 

Chi- squared used to see whether there was a difference sites downstream in the river calder in terms of size.

Side by side dispersion diagrams (box and whisker) were used to show how bedload size and uniformity in size changed at the different sites. 

Scattergraphs were also used to show sites vs mean pebble size and sites vs standard deviation. 

Conclusions:

Data shows that the hypothesis that bedload gets rounder downstream is correct.

Shown by 95% confidence limit reached on chi-squared. The patterns of over andunder representation confirm expectations of how the river behaves.

Disproved Hypotheses that bedload gets smaller and more uniform in size downstream though.

This section of the channel moves in the opposite way to what was expected.

Evaluation:

Due to the secondary data (previous invesitgations) our findings were considered as valid as in previous years, the same trends were found.

(Evaluation of each question is found throughout)

To measure the bedload, pebbles were picked randomly (representational), as this avoided human bias, as if you had to choose pebbles, then hand sized, non- slimy pebbles would have been selected. 

The pebbles were measured in a pebble box that allowed for easy measuring of the a, b and c axis of the pebble.

The pebbles were also categorised in terms of roundness, using powers index of roundness. This is good as it provides images to use to compare the pebbles to but it is subjective, and a way to improve this is to have one person judging the roundness of all of the pebbles, from every site so that the results were consisitent. 

Bedload roundness:

Located bar graphs on a base map are a simple and effective way to present data such as this. They are clear and it is easy to see where each graph is referring to. 

Chi- squared:Total- 27.9

Positives: Accurate, as the sample size is very large. 
Negatives: Lack of data in Very rounded and Very angular categories, so they were combined with the rounded and angular category. To improve, more sampes could be collected so that the Very angular and rounded categories will have more data. 

Results: Bomery 1 and 2 over represented in angular pebbles and under represented in angular pebbles. This is expected. Worm Gill and Stakes bridge show over representation of rounded material and under representation of angular material. This is expected. Monks Bridge sites show observed values close to the expected values. Expected as there shouldn't be any extreme results as these sites are in the middle course of the river. 

Overall: Shows that pebbles are getting more rounded downstream due to attrition and abrasion. 

Bedload gets smaller and less variable in size:

Side by side dispersion diagrams: easily comparable. Log paper used which is hard to interpret. Using normal scale paper would have allowed for easier comparison of standars deviation.

Scattergraphs: the x axis of scattergraphs suggests equidistant sites (Systematic) so shreve order could be used instead (explain what shreve is in exam). Allow us to see the direction of the trend. 

Mean pebble size was calculated. a+b+c axis/3 -Done on excel, as this makes the calculations much easier, quicker and more accurate.

Due to large sample sizes, extremes are unlikely to distort the mean. Extremes won't distort the measures of central tendency such as the median or using interquartile ranges. 

Problems/ Improvements of data collection:

Lower course of river inaccessible due to Sellafield nuclear reprocessing site. A different river (of a similar size) could have been chosen so data could have been collected from the upper, middle ad lower course of the river. 

The sites that were selected were too close together, (bomery 1 and 2, monks bridge 1 and 2 were only 50m apart) as there were limited amounts of staff to supervise. If more staff were available then more sites (around 10) could have been chosen in order to establish clearer trends. This would have allowed for the sites to be more systematic and henceforth giving a more accurate idea of how the river behaves as it moves downstream.

Although 40 pebbles were sampled at each site, which totals 240 pebbles in total, a larger sample could have been taken which would have been even more representative of how the bedload behaves. 

River selection: River Calder is accessible in the upper, lower and middle course. 

Site Selection: Six sites chosen pragmatically and with stratified sampling (2 in upper, middle and lower course)-

Bomery 1 and Bomery 2 before and after a significant tributary (Shreve index of 5) joins the river.

Monks Bridge 1 and 2 before and after another tributary. This was done so the impact of increased discharge of the river can be identified.

Stakes bridge in the lower course. 

40 pebbles from each site were chosen at random 

P - Practical- The investigation needs to be do-able

R - Risk- There should be no or very few risks involved- any risks posed should be minimised and managed

A - Access- The location should be accessible. The use of maps as a secondary resource is important here. 

T - Time- Should be able to be completed in a sensible timescale

S - Scale- Should be reasonable for a sixth form student (or group of)  to carry out. 

E - Equipment- The investigation shouldn't invole much complicated equipment. Should be available to sixth form students and be able to be used on site (and easily transported there)

C - Contrast- Needs to involve differences between places to be observed. 

Bedload in the river Calder gets smaller, more uniform in size and rounder downstream.

POSITIVES OF HYPOTHESES:

• Clearly focusses on bedload
• Takes Bradshaws model into account

WEAKNESSES OF HYPOTHESES:

• Very narrow

POSSIBLE IMPROVEMENTS:

• If width and depth were measured then the hydraulic radius could be worked out and link bedload data to channel efficiency 

We collected data on the siza and shape of bedload- 40 pebbles from each of the six sites

Primary data:

Collected simultaneously with a member of staff supervising. 

Secondary data:

OS maps- allow to choose a suitable river that is accessible, and suitably sized.

Previous groups of students have done the same investigation too so the results obtained in this investigation can be compared with the results from other investigations. 

Bradshaws Model- gave us an idea of what kind of results to expect.

Hypotheses: Bedload in the River Calder gets smaller, more uniform in size and rounder downstream. 

MUST INCLUDE AT THE START OF EVERY SKILLS ESSAY

If appropriate, include a sketch of the river calder basin 

Include: -Scale

-Shreve Values

-North arrow

-Names of sites and tributaries

-Upper, middle and lower course labelled.

• 15Km long
• South Westerly direction
• Sellafield nuclear nuclear reprocessing site restricts access at the lower course/ mouth of the river
• 40 pebbles from each or the 6 sites sampled

• Shreve total= 59
• Bomery 1= 5
• Bomery 2= 8
• Monks Bridge 1= 15
• Monks Bridge 2= 18
• Worm Gill= 41
• Stakes Bridge=47