Category Archives: science practice

Renewable energy from a Devon river – the new Totnes Weir Hydro

About a mile upstream of the south Devon town of Totnes, the tree-lined tranquillity of the river Dart is interrupted by a weir. Water cascades over this concrete barrier and after heavy rain there is a spectacular display of power with swirling whirlpools and foamy white water. Slate-grey herons and sparkling white egrets stand sentinel by the weir and the occasional grey seal lurks below, waiting to feast on fish that linger too long. There has been a weir at this bend in the Dart since the 16th century, built originally to harness the power of the river; the present rather bland construction dates largely from the 20th century.

Totnes weir
The Totnes Weir viewed from the upstream pool. The picture shows the concrete weir after the installation of the new Hydro (off the picture to the right) and after a period of low rainfall so that water flow across the weir is quite low. The gulls are enjoying the calm conditions.


The weir is a downwards-sloping concrete barrier that interrupts the flow of the river so that a large pool of water accumulates upstream, isolated from the tidal downstream water about three metres below. This pool of water is a store of potential energy that was used in the past to drive several water mills in the town a mile away. A channel, the leat, ran from the pool all the way in to Totnes and providing the leat stayed above the level of the river it contained the energy to drive a water wheel. Only one mill building now survives: the Town Mill dating from the 17th century but with 19th century additions. This was used as a water mill until 1945 and currently houses the Tourist Information Centre. The leat is still intact and can be viewed along much of its path, through an industrial estate, under the main railway line and passing near the front of Morrisons superstore. The leat is celebrated in the name of the town’s large medical centre, the Leatside Surgery.

Water Mill
A water mill at Dartington in Devon showing the principle of the leat. The leat takes water from the stream and providing the leat stays above the stream it can drive the mill wheel.


Turbine Building exterior
The turbine building of the Totnes Weir Hydro. The Archimedes Screws can be seen on the right.


Hydro and weir
The two Archimedes Screws alongside the weir


Over the past year, a neat, stone-clad, turf-roofed building has materialised by the side of the weir. This is the turbine house of the new Totnes Hydro which once again harnesses the power of the Dart. On the downstream side of the building are two tube-like structures roughly aligned with the descending surface of the weir, each tube containing an Archimedes Screw. Water from the pool behind the weir passes under the turbine building to enter the tubes, pressing on the blades of each Archimedes Screw causing them to turn and driving the turbines. The Archimedes Screws can be viewed from the downstream side and I find them mesmerising – turning steadily, water splashing, feeling almost alive – as they transform the potential energy of the water in to kinetic energy and subsequently electrical energy.

turbine in action
Renewable energy in action: a close up of water emerging from one Archimedes Screw.


When there is a good head of water, the turbines generate about 250 kW of power.  Output will depend on flow in the river (higher power after heavy rain) and the head across the weir (typically about 3m but reduced by spring tides).   Generation may cease altogether for about two weeks in a dry summer when water flow in the Dart is low.   Currently, the electricity generated is powering the local comprehensive school and an aluminium foundry on the nearby industrial estate and any surplus enters the grid.  To put this in to perspective the overall energy produced is enough to power the equivalent of about 300 homes. In time, the hydro will also provide electricity for the new ATMOS project.  This is a community-led development of homes and businesses on the former Dairy Crest site in Totnes.

The river Dart is an important route for migrating fish and the weir already contained a fish pass to help sea trout and salmon overcome the barrier. The pass was, however, in poor condition so that fish were having difficulty moving up the weir leading to losses to hungry herons and seals. The new Hydro project includes renovating the existing fish pass and building an additional modern fish pass alongside the turbine building. These should help migrating fish so that, in time, the piscine population on the Dart increases; new fish counters have also been installed to help monitor traffic.


So far so good, but if equipment is installed to capture the energy of the river it is bound to alter the flow over and around the weir.   You can see this clearly on the upstream side of the turbine house where water in the pool flows towards the new building to enter the Archimedes Screws, eventually discharging in to the river below.   Although water flow through the turbines is carefully regulated by sluices to make sure that the weir does not dry out, less water now flows across the weir than before.  This redistribution of water has remodelled islands in the river downstream and night fishermen have had to relearn safety on the river.    We should not forget, however, that when the weir was first built and water was directed down the leat to power the Totnes mills some 500 years ago, water flow in the river must have been changed to a much greater extent.    There is also the question of noise.  The new Archimedes Screw turbines do emit noise as they turn and there is some splashing of water.  The turbine building is insulated and the current level of noise from the new installation is no more than I can remember coming from the weir on a full flood.

Water enters hydro
The pool of water behind the weir showing water preferentially entering the turbine building.


The Weir Hydro project was developed by the owners of the weir, Dart Renewables, working closely with the Totnes Renewable Energy Society (TRESOC). TRESOC was set up by local residents to enable the community to develop renewable energy and to retain control of the resources. On a practical level TRESOC aims to supply local homes and businesses with “local” energy. If everything works to plan, the Totnes Weir Hydro should generate 1.35 GWh of electricity each year, saving 550 tonnes of carbon dioxide. The majority of this electricity will be used to power local enterprises.

Disclaimer: I am a member of TRESOC and have invested in some of their projects.

Nurofen – a popular pain killer, a case of misleading marketing and the power of suggestion

Sounds great doesn’t it? Pain killers targeted at different kinds of pain. You can pick the pill for your particular pain.

Consumers in Australia were, until recently, offered this choice. The pain killer, Nurofen was sold in different brightly coloured packets each labelled to target one of four kinds of pain (Tension Headache, Migraine Pain, Period Pain, Back Pain). The only problem was that, despite the specific labelling, the drugs inside the four packets were identical and behave the same whatever the labelling says. Following a long campaign by the media, the Australian Federal Court last week ordered the makers, Reckitt-Benckiser to stop selling the drug in this misleading manner.

Nurofen and the over-the-counter market

In many parts of the world, Nurofen is the brand name for the drug ibuprofen, one of the leading over-the-counter (non-prescription) analgesics. This market is very lucrative for pharmaceutical companies but it is also very competitive. Here in the UK, the over-the-counter market for all medicines and supplements nets £2.5 billion a year, corresponding to sales in pharmacies, drug stores, supermarkets, corner shops and filling stations. Pain killers, including ibuprofen, are the largest individual sector in this market, worth about £550 million. With such large sums of money and in a competitive market, companies work hard to maximise their share and this perhaps explains Reckitt-Benckiser’s behaviour. Should you want to know about other tricks used by companies to boost sales, take a look at Ben Goldacre’s book, Bad Pharma.

Targeted marketing in the UK

Given the Nurofen marketing scam in Australia, I wanted to find out if the same was happening in this country so I went to our local branch of Superdrug. This is not a pharmacy but they sell a range of permitted medicines alongside all kinds of beauty products, perfumes, baby supplies etc. The over-the-counter drug section is not hard to find and there is a dizzying array of brightly coloured products especially in the pain section. Nurofen is the principal branded form of ibuprofen on sale but it’s not the only one and generic versions of ibuprofen are also available.

When I looked carefully I found that Nurofen is indeed being sold in the UK in the same targeted manner as in Australia. I didn’t find all four targeted forms but I did find packets of Nurofen labelled Tension Headache and Migraine Pain, both containing the identical drug. It’s not just Reckitt-Benckiser who play this game: Feminax Express (made by Bayer) is marketed for period pain relief but inside the packet it’s just ibuprofen. To be fair to Feminax, they do say that the drug targets other kinds of pain but you can’t avoid the connotations of the brand name. Superdrug have their own generic versions of ibuprofen, ibuprofen labelled for period pain and ibuprofen labelled for migraine pain and again the drug inside each packet is the same.

Should we be concerned about targeted marketing in the UK?

The Australians believe this targeted labelling is misleading but should we also worry in the UK? This turns out to be a complex question.

Labelling packets to imply that they treat different kinds of pain is deliberate deception when the drug inside is the same. It’s not that the ibuprofen doesn’t work, it does work but it cannot target a specific kind of pain. On this basis, the targeted packets should be withdrawn.

There is, however, another side to this story and to understand this we need to go back to 1981 to an experiment conducted by Branthwaite and Cooper who studied 835 women with headaches. They gave the women aspirin or dummy pills (placebo) and the two treatments were further subdivided by packaging. Half were supplied in neutral boxes, half were supplied in flashy, branded packets. Neither the experimenters nor the subjects knew who was receiving aspirin or placebo. As expected, aspirin was better than placebo for headache but, surprisingly, both aspirin and placebo did better when supplied in branded boxes. If we can extrapolate these findings to the Nurofen story, then perhaps the targeted packets of Nurofen do perform better against specific kinds of pain, not because of pharmacological effects but because the packet says so and the consumer believes this.

This borders on quackery and makes me feel very uneasy. Surely we should not be using misleading marketing to achieve a specific effect, and make a tidy profit along the way. Can I suggest a compromise? Ditch the targeted labelling and have a single Nurofen or generic product but put all four indications on the packet. That way there would be no deception but people would be reassured that the drug worked on the pain they wanted to treat.

Dorset’s big farming experiment

The village of Briantspuddle lies near tranquil water meadows in the valley of the River Piddle, some eight miles to the east of Dorchester in the south-west of the UK. Nowadays, Briantspuddle is all pretty cottages, thatched roofs and peace and quiet. The village was far from quiet in the first half of the 20th century when Briantspuddle became a centre of agricultural and social innovation.

River Piddle at Briantspuddle
The river Piddle near Briantspuddle (doesn’t look like a piddle to me)


The Bladen Estate: Ernest Debenham’s vision for a new agriculture

Ernest Debenham was an educated and practical man, an innovator, always keen to try new ideas. He was, after all, the grandson of the founder of the Debenhams drapery and department store empire. By 1900 he was in charge of the company and became very wealthy. Around this time, he decided that agriculture would benefit from being organised as a business. He developed the idea of self-sufficient farming where centralised processing and selling directly from the farm would “cut out the middle man”, reduce costs and boost rural employment. In 1914, he purchased several farms in the Piddle Valley around Briantspuddle where he intended to test these ideas. This land became the Bladen Estate, named after the old form of Blagdon or Blackdown, the hill above Briantspuddle.

Houses for workers

Cruck Cottage Briantspuddle
Cruck Cottage Briantspuddle – one of the houses present before Debenham started building. It shows the “Dorset” style he tried to emulate.


At the beginning of the 20th century, Briantspuddle was a sleepy hamlet of about a dozen houses. To realise his vision of self-sufficient farming, Debenham planned a substantial expansion of the village although, because of the outbreak of war, new building did not start until 1919. He believed that good housing led to good work, so his first priority was to provide new houses for the estate workers. These were built in the traditional, Arts and Crafts style with thatched roofs, designed to blend sympathetically with the local environment. The new cottages were equipped with a bathroom and inside lavatory, and with self-sufficiency in mind, a quarter acre garden and a pig pen. Debenham also encouraged tree planting as a means of harmonising the new development with the surrounding countryside.

Cottages in Briantspuddle's Bladen Valley
Some of the “new” houses built by Debenham for estate workers in Briantspuddle’s Bladen Valley


Scientific agriculture

Dairy Ring 2
Part of the semi-circular dairy complex, now private housing


The Bladen Estate was established as an experiment to test how centralised processing and the application of recent scientific discoveries in agriculture might improve food production. Many aspects of farming were examined but perhaps the most innovative development was the central dairy in Briantspuddle. This was a purpose-built facility for collecting, testing and packaging milk from Estate farms. The new buildings were intended to be functional, the semi-circular design allowing easy access for transport. They were also meant to be aesthetically pleasing, imparting a special character to the area and, of course, they had thatched roofs. A unique aspect of the dairy was a bacteriological laboratory capable of testing milk for bacteria as well as fat content. Bonuses were paid to workers from farms supplying milk with the lowest bacterial count, so encouraging cleanliness in the milking parlour. The central dairy processed 1000 gallons of milk each day in to Grade A milk, butter, cheese and pig food. Milk was transported in covered motor wagons to a depot in Parkstone where it could be on sale within an hour of leaving Briantspuddle.

Old dairy buildings Briantspuddle
Some buildings I found at the back of the dairy complex, now private housing but looking like they once had an important function on the farm.


Animal husbandry was also approached systematically and scientifically. For each cow, detailed records were kept of weight, health, food consumed etc. Twice a year, the estate Veterinary Service examined animals for tuberculosis; cows testing positive were vaccinated. Similar intensive approaches were tried in relation to sheep, pigs and poultry and there were 70 bee colonies. Livestock were fed arable crops grown on the Estate; also balanced rations supplied by a company established by Debenham. The Estate had dedicated power and water supplies and its own transport depot, contributing to self sufficiency.

The end of the experiment

At its peak in 1929, the Bladen Estate farmed 10,000 acres of land in and around the Piddle Valley, including many individual farms, providing employment for 600 people. These were difficult times for business, especially for farming and the Estate required continual financial input to stay afloat. Eventually the funds required to subsidise the venture ran out, the Estate went in to decline and the individual farms were sold.

Ford over the river Piddle at Turner's Puddle
One of the constituent farms of the Estate at Turner’s Puddle, seen across the ford on the river Piddle


Despite the apparent failure of his experiment, Debenham should be seen as one of agriculture’s pioneers. His ideas for self-sufficient farming were ahead of his time. Many “modern” farming practices were tested on the Bladen Estate but at the time the tools to make them work were unavailable e.g. antibiotics to control disease under intensive conditions. Debenham was, sadly, wrong in one of his beliefs: increased production and centralisation have not allowed more people to live on the land; in fact the opposite has proven to be true.

21st century Briantspuddle

The contemporary visitor to Briantspuddle will encounter an attractive village with a remarkably consistent architectural style, a legacy of Ernest Debenham’s experiment and vision. The best place to experience this is the Bladen Valley, a small coombe populated by substantial, white-washed, thatched cottages originally built for estate workers, most still retaining their original look.

Bladen Valley 2
Cottages in the Bladen Valley, Briantspuddle – originally built for estate workers from 1919 onwards


At the foot of this valley lies the unusual War Memorial commissioned by Debenham to commemorate seven local men who died in the Great War (six names of WW2 fatalities have since been added). The memorial, sculpted in Portland Stone by the controversial artist Eric Gill, was dedicated in 1918, one day after the armistice had been signed.

War Memorial Briantspuddle 2
War Memorial


Detail from War Memorial Briantspuddle 2
Madonna and child on War Memorial


Detail from War Memorial Briantspuddle
Commemorating the names of the WW1 dead


In the main part of the village, there is the semicircular former dairy complex, now private housing, and the fine thatched Village Hall based on a converted 18th century barn which, together with the Social Club, provides a focus for village activities. Next to the Hall is the Village Shop and Post Office. This was once a granary but in 2002 became a community shop run by volunteers, “by the village, for the village”. It seems that in Briantspuddle, social experiments continue to the present day.

Dairy Ring 1
Another view of the semi-circular former dairy complex


Village Hall and shop Briantspuddle
The Village Hall (left) and Community Shop (right) in Briantspuddle


This article appeared in the February 2015 edition of the Dorset-based Marshwood Vale Magazine.


When I visited the area, I came across the now-disused church at Turner’s Puddle a little way along the Piddle from Briantspuddle and was surprised to see snowdrops in profusion at this time of year (January 13th).

Turner's Puddle church with snowdrops
The disused church at Turner’s Puddle – to the left of the steps you may be able to see snowdrops in flower in January


National Science and Engineering Week – Science in Cornwall

The UK’s “widest grassroots celebration of all things science and engineering” is currently in full swing.  This is National Science and Engineering Week and the 2013 theme is Inventions and Discoveries.  There should be something going on near you and you can find out using the searchable web site.

As part of this celebration, I wrote an article for Cornwall Today a popular local magazine in the South West of England.  The editor asked me to pick out six Discoveries and Inventions made by people from Cornwall or made in the county of Cornwall. It was an interesting piece to write and it made me reflect on how the scientific landscape of one part of a country is generated.  In the case of Cornwall, the geography and natural resources have strongly influenced how science has impinged on the county and you can read the article here.

John Snow: the man who understood cholera

2013 is the bicentenary of the birth of John Snow, a pioneer of anaesthetics and epidemiology.  In the 19th century he studied the distribution of cases of cholera in London and inferred that the disease was spread by contaminated water.  This was the first study using the of the distribution of a disease to highlight its cause: we now call this epidemiology and Snow is considered to be the father of this discipline.  Here is what I wrote about him for the Dorset-based Marshwood Vale Magazine:

Cholera is a deadly disease killing more than 100,000 people each year, mostly in the developing world.  Although cholera no longer affects the UK, the mid 19th century saw Britain ravaged by repeated waves of the disease with many thousands dying.  There was little understanding of the disease until John Snow, the Victorian medical pioneer showed how cholera could be contained and prevented; 2013 is the bicentenary of Snow’s birth.

In 1849, a new wave of cholera struck the UK.  Dorset was not spared and Fordington, then a village adjacent to Dorchester, suffered a serious outbreak of the disease.  Fordington was, at that time, a place of great deprivation.  There were many small cottages, mostly without gardens so that all waste, including sewage ended up in the open drain or the Mill Pond.  The Mill Pond was also where most people took water for washing and sometimes for cooking.

Five years later, in 1854, cholera was again on the rampage in the UK.  Millbank prison in London was badly affected and the authorities decided to move seven hundred “uninfected” convicts to a place of safety. Their chosen refuge was a disused cavalry barracks in Fordington.  The Mayor and townsfolk of Dorchester tried hard to resist the move, fearing the consequences, but the convicts duly arrived.  Although attempts were made to minimise contact with the new jail, two women in Fordington contracted to wash bedding from the prisoners. Within days cholera appeared in Fordington and thirty or more people eventually died.  The energetic and intelligent Vicar of Fordington, Henry Moule, worked with the sick and dying, burning contaminated clothing and bed linen, spreading help and sympathy.  He became convinced that poor sanitation was at the root of the spread of the disease.  He recounted how when he sat at the bedside of a dying man, the overflow of one privy shared by thirteen families trickled between him and the bed.  More than 100 miles away, in London, another energetic man, John Snow, had also been thinking about the role of poor sanitation in the spread of cholera.
John Snow

John Snow was born in York in 1813, the eldest of nine children from a poor family.  Snow was determined to be a physician and held three apprenticeships with medical practitioners in the North East where he saw his first cases of cholera.  At the age of 23 he moved to London to continue his medical education, qualifying as a practising physician in 1844.  He became an expert in anaesthetics showing how they could be used more safely and effectively during surgery.  One of his notable patients was Queen Victoria to whom he administered chloroform for pain relief during the birth of two of her children.

Snow was also very interested in infectious diseases.  At the time, the prevailing theory of infection was the “miasma” theory whereby infectious diseases were thought to be spread by poisonous gases emanating from sewers, swamps and decaying organic matter.   Snow did not believe these ideas and wondered if infections might be spread by invisibly tiny organisms.  He applied himself to understanding cholera and during the 1849 outbreak, proposed that the disease was caused by contaminated water.  At the time, no infectious organisms had been identified, so his ideas were not well received.

His greatest insights came during the 1853/4 outbreak.   The west London district of Soho, near where Snow lived, was densely populated; living conditions were poor and sanitation even poorer, reminiscent of Fordington but on a greater scale.  Soho was very badly affected by cholera at this time and many died early in the outbreak.  Snow noticed that many of the deaths occurred in the vicinity of a pump in Broad Street from which local residents obtained water.  He went to the Register Office, checked the records of the victims and found that of the 73 victims who lived close to the pump, 61 drank the water.  The pump became the prime suspect in the search for the culprit which Snow believed was a microbe in the water.  In support of his ideas, he noticed that incidence of cholera was much lower among people in the neighbourhood who used alternative water supplies such as residents of the local workhouse and employees of the local brewery. He convinced the authorities to remove the Broad Street pump handle until the outbreak had subsided.  It was eventually found that a leaking cess pool had been contaminating the water from the Broad Street pump, supporting Snow.  However, despite Snow’s efforts, most people still continued to believe in the miasma theory of infection.

There was, nevertheless, a gradual acceptance that the poor state of sanitation in the capital was a public health risk.   By the 1850s much of London’s sewage was ending up in the Thames which was essentially a very smelly open sewer as well as being the source of drinking water for many.  This was an obvious health hazard whether or not you believed the miasma theory.  Eventually, a vast network of new sewers was built so that the sewage could be diverted downstream of London in to the sea and diseases like cholera disappeared.

Snow died in 1858 and was gradually shown to have been correct about cholera:  in the 1860s, Louis Pasteur showed that diseases could be caused by microorganisms and in 1884 the cholera bacterium was discovered.  Most importantly, Snow was the first person to show that careful mapping of the incidence of a disease, epidemiology, could provide important clues to its cause.  Snow was the father of epidemiology and his work paved the way for other landmark studies such as the demonstration in the 1950s by Richard Doll and Austin Bradford Hill of the link between smoking and lung cancer.

Bloody British weather or a wake-up call?

Here is an article I wrote for the December edition of the Dorset-based Marshwood Vale Magazine considering whether the recent awful summer weather could be a reflection of climate change.

Summer this year was the wettest for a century and Dorset experienced some particularly heavy downpours early in the season.  Rainfall at Portland and at Hurn in the first ten days of July was nearly three times the normal monthly average.   Several rivers burst their banks and towns were flooded including Bridport and Burton Bradstock.  Devon was also badly affected and Modbury, Ottery St Mary and Yealmpton were inundated.  The torrential rain in Dorset caused a landslip at the Beaminster Tunnel where two people died.  The rain also undermined the cliffs on the Jurassic Coast leading to a 400 ton rock fall that killed one person near Hive Beach.

Flooding in Burton Bradstock, West Dorset, July 2012

Our notoriously changeable weather depends on many factors but this summer it has been dominated by the Jet Stream.   This is a band of fast moving winds running eastwards across the Atlantic high up in the atmosphere separating cooler northerly air and warmer southerly air.  The Jet Stream is important for our climate as it guides Atlantic weather systems bringing rain and unsettled conditions.  In a typical summer the Jet Stream lies to the north of the UK shielding us from these weather systems and bringing milder, settled conditions.   This summer, the Jet Stream lay below the UK so the weather was cooler and wetter and because it did not move, this pattern persisted for several months.

The position of the Jet Stream could simply be a result of natural variation and most of us will shrug our shoulders and blame the awful summer of 2012 on “bloody British weather”.  We do, however, live in a world where changes in conditions in one part of the globe may affect the weather in another part.  Some climate scientists are beginning to wonder if the weather in Europe is being influenced by changes in sea ice many miles away in the Arctic.

At the North Pole there is a large mass of pack ice formed from sea water in this very cold part of the world; this is the Arctic sea ice.   The size of this ice pack varies by season, partly melting in the warmer months and then refreezing as the cold returns.  There has been a long term trend towards less ice in the warmer months and this reached a record low in September 2007.  Slightly more summer ice was seen in subsequent years but in 2012 the record was broken again; the level of Arctic summer ice was the lowest since records began.  This had not been expected and scientists were surprised and shocked by the extent and speed of Arctic sea ice loss.  If, as seems likely, the loss continues, summer sea ice will disappear altogether from the Arctic some time this century.  This represents a major change to the planet that should concern us all.

One of the principal drivers of this major change in the North Polar ice cap seems to be global warming; as the sea temperature increases so the ice melts more.  Much of this effect on the polar ice can be attributed to human activity; burning of fossil fuels (e.g. coal, gas, petroleum) raises levels of the greenhouse gas, carbon dioxide, causing warming.  Loss of such a large amount of sea ice has major effects on the planet.  Most importantly, loss of the ice accelerates warming of the sea; the sun’s rays are no longer reflected away by the ice and the heat is absorbed by the sea.  This is very serious for the stability of the climate.

There is also an indication that loss of the sea ice affects the position of the Jet Stream and makes it stay in one place longer.    In consequence, weather patterns are maintained for longer; periods of wet weather may lead to floods and periods of dry weather may generate drought.  These changes are compounded in that a warmer atmosphere holds more moisture leading to heavier rain.  The loss of Arctic sea ice may, therefore, have contributed to our poor summer and we may now be experiencing the effects of climate change in the UK.

The unexpectedly large and continuing loss of the polar ice cap should be a major wake-up call to governments that the planet is changing.  We are all beginning to feel the effects.     The solution is clear, we must reduce use of fossil fuels; this will not reverse the changes but may slow them down. Unfortunately, governments still look on with disbelief and do nothing.   It is a disgrace that climate change did not feature at all in the recent US presidential debates, especially when the US is one of the major users of fossil fuels.  Perhaps the devastation wrought by Hurricane Sandy will make the US think again about extreme weather events.

But it gets worse.  The response of industry to the melting of the Arctic ice has been to see opportunities for trade.  Oil, gas, mining and shipping companies are rushing to expand operations in the Arctic as the ice recedes and new sea routes open up.   This has the potential for environmental catastrophe as well as leading to greater emission of carbon dioxide and acceleration of climate change.

Let’s finish by quoting Bill McKibben, the environmental campaigner, on the loss of Arctic sea ice:  “Our response (so far) has not been alarm, panic or a sense of emergency.  It has been “Let’s go up there and drill for oil”.  There is no more perfect indictment of our failure to get to grips with the greatest problem we’ve ever faced”.

The Higgs Boson and faster-than-light neutrinos – what should we believe and can statistics help us?

I’ve been pondering this topic ever since I read Philip Ball’s column in the Guardian on Christmas Eve.  The recent announcement of tentative evidence for the Higgs Boson and the report of neutrinos travelling faster than the speed of light have created both interest and confusion.   In his article, Philip Ball proposed some guidelines on how we should respond to these new “discoveries”.   In his view, statistical measures associated with the experiments should be ignored but rather we should be lead by the prevailing beliefs of scientists in the field.  Taking this approach, we could conclude that the Higgs boson might have been found but that neutrinos might not travel faster than light.

I had always thought that scientists should follow the statistics and that following opinion might run the risk of being tainted by bias so I wanted to write something about this apparent paradox.   I am also not really sure why we need to second guess the outcome of these experiments.  It might be better to wait until more data are available and, in the case of the neutrinos, to wait until the experiments have been repeated by other groups.  But scientists are intrinsically gossipy and will always want to speculate on the science done by others so here we go.

If you delve a little, you find that Philip Ball’s piece was triggered by an article by Italian particle Physicist Giulio D’Agostini.  Now this is a difficult read but it makes some good points.  D’Agostini is concerned about the misuse of statistics by journalists, by lay people and by scientists.    It’s easier to see his argument with an example.  He uses data from a study at Fermilab in the US that were interpreted by some as evidence of a “new physics”.  Scientists at Fermilab had seen an unexpected “bump” in their data and wanted to evaluate whether this was a real difference (implying a “new physics”) or just part of the variation seen in any experiment (no “new physics”).  Here they used a p-value which is a statistical measure of the probability that the results they had observed would be seen even if there were no “new physics”.  

The Fermilab scientists came up with a p-value of 0.00076 for the experiment and this tells us that with no “new physics” the probability of seeing the “bump” would be 0.076% or about 0.1%.  Putting this in a different way, the results seen by the Fermilab scientists could have occurred with standard physics although the probability is quite low.   What worries D’Agostini is that this probability value is then misused.  The misuse is to turn the p-value into a predictor of the hypothesis, in this case “new physics”, and he quotes several media reports that made this error. These reports suggested that because the p-value was very low there was a very high chance that the hypothesis was correct.  As D’Agostini points out, the p-value tells us about the experimental data (i.e. what is the probability of observing the data in the absence of new ideas) but not about what the experiment means i.e. the hypothesis.   

If the statistics can’t help then how do we understand this experiment?  Here D’Agostini suggests we look at what is already known and ask what other scientists in the field believe.  Do they believe the hypothesis that a “new physics” is required to explain the results or do they not?  Here we have to look at the way science works.   Scientists perform experiments to test hypotheses and gradually, as the data emerge, networks of beliefs about science arise.  These networks of beliefs can be called theories.  These theories will be believed by many scientists in the community although there may be dissenters.  If a prevailing theory, based on extensive prior information, is challenged by new data indicating an alternative idea then we will require substantial further experimental work to make us change our beliefs.  As D’Agostini points out, this is why many people did not believe the Fermilab results.  The faster-than-light neutrinos also fit this category, being a completely new observation that challenges existing ideas.  If new data can be interpreted within existing theories then they are more likely to be believed.  This seems to be the case for the data obtained on the Higgs Boson at the Large Hadron Collider. 

This scheme, proposed by D’Agostini and endorsed by Ball does not, however, completely fit the way progress is made in other branches of science.  In the life sciences, for example, p-values are a mandatory aspect of reporting the results of experiments and they are used extensively to assess outcomes of experiments.  In examining the possible effect of a perturbation on a system, typically the experiment is set up with two groups, control and treated.  The data are analysed and mean values calculated for the two groups.  The two means may differ and the p-value is then the probability that the mean for the treated group (different from the control group mean) will be seen even if there is no real effect of the treatment.  A low p-value implies a low probability and, as we have seen above, this should not be taken to infer the magnitude of the probability that the treatment has had an effect.  Nevertheless, we need to be able to make progress and typically life scientists take a p-value of 0.05 (5%) as a cut off.  p-values greater than 0.05 are taken as evidence that the treatment had no effect and p-values less than 0.05 are taken as evidence that the treatment had an effect.  This is a bit different from what D’Agostini recommends but it seems to have allowed advances in the life sciences and the generation of theories of how systems function.

You may have now spotted another paradox.  The p value reported for the Fermilab experiments was 0.00076 and this is much smaller than the 0.05 figure typically used in life sciences as an indicator that data are not just a result of random fluctuation. So why were the particle physicists not more accepting of the result?  Apparently particle physicists use more stringent cut offs in their work with p-values of 4×10-7 being required for a “discovery”.   In the end, however, these figures don’t matter greatly as it is the relation between the new findings and prior knowledge that matters.  When a new finding is made, even if the experiment is supported by a strong p-value, if it goes against prior data then most scientists in the field will shrug their shoulders and wait for replication before getting too excited.

Let’s conclude by considering a potential problem in the scientific belief system.  Science proceeds by experiments that test hypotheses and gradually networks of beliefs (theories) arise.  As more data are found that are consistent with the theory, belief in the theory strengthens.  If data are found that disagree, then the theory should, in principle, be weakened.  Sometimes this weakening of belief does not work as it should because of the way people treat theories.   Within the life sciences there are large research groups and smaller research groups often working in the same field.  The larger groups often produce large numbers of papers and attain a preeminent status in the field.  Sometimes theories produced by these groups assume dominance and are awkward to dislodge.  It can even become difficult to publish data that disagree with the prevailing theory and there may be a reluctance to interpret data in terms of other frameworks.  Sometimes data that disagree with the dominant theory don’t get published at all.

Strong theories can also have insidious effects on the way experiments are performed and interpreted.  Sometimes when analysing data there can be a tendency for a researcher to reject data points that “don’t fit” and it is essential to avoid such bias.  It is for this reason that, where drugs are tested on humans, the randomised controlled trial is the only acceptable method.  This includes randomisation of subjects between, for example, control (placebo) and treatment groups.  Ideally, there should also be “blinding” where neither subject nor observer knows which group a subject is in and there should be enough subjects enrolled on the trial to eliminate chance effects.  This design provides a way to reduce bias to a minimum so that a fair appraisal of the effects of a drug can be obtained. 

The success of the randomised controlled trial system also depends on full disclosure of data.  So, for example if a pharmaceutical company performs ten trials on the effects of a new drug in different parts of the country, all of those trial data should be made public.  An editorial and associated papers in the British Medical Journal this week show that full disclosure of trial data is not occurring.  The results of many trials are not disclosed leading to deficits in knowledge.  This is a form of bias and in some cases leads to misleading conclusions (positive and negative) about the effects of a drug.