Nature | News Feature

Translations

عربي

Long-term research: Slow science

The world's longest-running experiments remind us that science is a marathon, not a sprint.

Article tools

Although science is a long-term pursuit, research is often practised over short timescales: a discrete experiment or a self-contained project constrained by the length of a funding cycle. But some investigations cannot be rushed. To study human lifespans or the roiling of Earth's crust and the Sun's surface, for instance, requires decades and even centuries.

Free interview

Brian Owens talks about the efforts to continue recording sunspots and pitch drops

You may need a more recent browser or to install the latest version of the Adobe Flash Plugin.

Here, Nature takes a look at five of science's longest-running projects, some of which have been amassing data continuously for centuries. Some generate hundreds of papers a year; one produces a single data point per decade.

Experiments operating at this pace are challenged by shifting research priorities and technologies, and their existence is regularly threatened by funding droughts and changes in stewardship. But they are bound together by the foresight of the scientists who started them and the patience and dedication of those who carry the torch. If persistence predicts a long and healthy life — as one 90-year study of human longevity has suggested — then the scientists featured here could set some records themselves.

400 years: Counting spots

UNIVERSAL HISTORY ARCHIVE/UIG/BRIDGEMAN ART LIBR.

Galileo was drawing sunspots as early as 1613.

Astronomers have been recording the appearance of sunspots ever since the telescope was invented more than 400 years ago; even Galileo recorded his observations. But early observers had no knowledge of what the dark patches on the Sun's surface were, or of the magnetic fields that created them. That began to change when, in 1848, the Swiss astronomer Rudolf Wolf began making systematic observations and developed a formula that is still used today to calculate the international sunspot number, also known as the Wolf number, which gives a measure of how solar activity is changing over time.

In 2011, Frédéric Clette became director of the Solar Influences Data Analysis Center, based at the Royal Observatory of Belgium in Uccle, which curates sunspot counts gleaned from photographs and hand drawings of the Sun's surface made by more than 500 observers since 1700.

The data are invaluable for predicting sunspot activity, says Leif Svalgaard, a solar physicist at Stanford University in California. The activity seems to wax and wane over the course of 11 years or so, and the streams of charged particles that the sunspots spray into space can affect satellites and electronics on Earth. The detailed records help researchers to understand why that cycle happens, and to refine predictions of particularly intense events. “The longer the time series is, the better we can check our theories,” Svalgaard says. Around 200 papers a year cite sunspot data, in fields extending beyond solar physics to geomagnetism, atmospheric science and climate science.

But the enterprise runs largely on goodwill. Each month, the Belgian centre collates sunspot numbers from about 90 observers, two-thirds of them amateurs, who use small optical telescopes no more powerful than those available 200 years ago. And although it is a World Data Centre recognized by the International Council for Science in Paris, it receives no funding from the organization. Clette works with one other part-time person to maintain the database, in addition to his 'night-job' as an astronomer at the Royal Observatory of Belgium.

Still, says Clette, it is fascinating to 'work' with colleagues from hundreds of years ago. For instance, he says that even though Galileo's coverage of the Sun was spotty because Galileo was “busy with planets and other things”, the drawings are detailed enough to reveal information about the magnetic structure of the sunspot groups and the size and tilt of the star's dipole. “You can extract from those drawings exactly the same information as from a drawing made today,” he says.

More than that, however, he is taken with his forebears' foresight. They faithfully recorded what they saw, thinking that it could be useful later on, he says. “It's a fundamental aspect of science,” he says, “not worrying what will be the final result.”

170 years: Monitoring an irritable giant

FRATELLI ALINARI MUS. COLLECTIONS, FLORENCE

Now a museum, the original building of the Vesuvius Observatory was used to monitor volcanic activity from the side of the mountain.

Although consistently active, every few thousand years, Mount Vesuvius erupts in spectacular style. The last time it did so, in AD 79, it consumed the city of Pompeii in the flames and before that, about 3,800 years ago, it covered all of present-day Naples in hot gas and rock (see Nature 473, 140141; 2011). The Vesuvius Observatory, the oldest volcano research station in the world, has been keeping an eye on its inhospitable subject since 1841, logging all the volcano's seismic rumbles to try to spot approaching danger. Originally perched 600 metres up the side of the volcano, far enough from the summit to be safe from ejected debris and high enough on a knoll to avoid the lava flows, the observatory has shaped the way that volcanology and geology is done, says Marcello Martini, its current director.

Macedonio Melloni, the observatory's first director, did pioneering work on the magnetic properties of lava that was crucial to later studies of palaeomagnetism — the history of Earth's magnetic field as recorded in rocks. In 1856, its second director, Luigi Palmieri, invented the electromagnetic seismograph, which was much more sensitive to ground tremors than previous machines and allowed him to predict eruptions. Under Palmieri and subsequent directors the observatory contributed to the development of much of the instrumentation used to monitor volcanoes worldwide. In the early twentieth century, for example, Giuseppe Mercalli developed the scale still used today to classify volcanic activity.

But the building itself no longer has the same role. “In the early stages it was important to be as close to the action as possible, but that's not necessarily the case any more,” says Haraldur Sigurdsson, a volcanologist at the University of Rhode Island in Kingston. Most of the monitoring is now done by remote, ground-based sensors, with the data sent back to the lab of the National Institute of Geophysics and Volcanology in Naples. The original buildings were turned into a museum in 1970.

In addition to informing scientific theory, the observations are used to predict trouble and protect the public — as they did successfully in 1944. The Naples lab, where scientists are on duty 24 hours a day, also keeps an eye on Mount Stromboli, on an island north of Sicily; the Campi Flegrei caldera west of Naples; and the island of Ischia. Sigurdsson, says, however, that the future of volcanology lies not with putting sensors on the volcanoes already known to be dangerous, but with satellite-based radar that can study ground deformation everywhere and pick out the risky regions irrespective of geologists' expectations. “We should be moving towards an internationally coordinated system of volcano monitoring that is not tied down to bricks and mortar on the side of a volcano, but looking at it in a really comprehensive way globally,” he says.

170 years: Harvesting data

ROTHAMSTED RESEARCH

Rothamsted has been home to experiments on the effects of fertilizer on the yield of wheat since 1843.

The stewards of long-term research projects are keen to maintain the integrity of the work, but also to keep it relevant. That is the case for Andy Macdonald who, in 2008, inherited a set of agricultural experiments that have been testing the effects of mineral fertilizers and organic manure on crop production since 1843.

Started by the fertilizer magnate John Lawes on the grounds of his country estate at Rothamsted, north of London, these studies have been used to test how nitrogen, phosphorus, potassium, sodium, magnesium and farmyard manure affect the yields of several staple crops, including wheat, barley, legumes and root crops.

“After 20 or 30 years, many of the basic questions about the relative importance of different fertilizers were pretty well answered,” says Macdonald, manager of the 'classical experiments' now run at Rothamsted Research. Nitrogen has the largest effect, followed by phosphorus. So the experiments are periodically updated to test new ideas and keep them relevant to current farming practice. In 1968, for example, the long-strawed cereal crops that had been grown since the experiments started were replaced by the higher-yielding short-strawed cereal crops being adopted by farmers. Macdonald says that these new crops turned out to need more fertilizer than the traditional cultivars because of the additional nutrients they were extracting from the soil, so farmers had to adapt.

“Rothamsted is the granddaddy of long-term agricultural research,” says Phil Robertson, director of the W. K. Kellogg Biological Station, a long-term agricultural research site at Michigan State University in Hickory Corners. The unbroken chain of data is invaluable, he says. Not only is Rothamsted able to study environmental and biological trends — such as carbon storage in soil or the effects of invasive species — that become apparent only over long timescales, but it also provides a platform for shorter studies on, say, nitrate loss in soil.

The Rothamsted archive holds about 300,000 preserved plants and soil samples that have been collected since the experiments began. In 2003, scientists extracted the DNA of two wheat pathogens from archived samples dating back to 1843 and showed that industrial sulphur dioxide emissions affected which one was dominant1.

It can be a struggle to keep funding agencies interested. Rothamsted gets by on a mixture of government funding, competitive grants and a trust fund set up by Lawes before he died. “As a funder you have to commit to maintaining observations even during periods where there may not be any exciting results,” says Robertson, who was involved in setting up the US Department of Agriculture's Long Term Agro-Ecosystem Research network last year. Macdonald and his team take pride in their history. “I think back sometimes to John Lawes,” Macdonald says. “I feel a great responsibility to ensure that the experiments are handed on in good condition for the next generations. They're not a museum piece, they're part of our living scientific community.”

90 years: Watching genius blossom

STANFORD UNIV. ARCHIVE

Lewis Terman started one of the lengthiest studies of human development.

In 1921, psychologist Lewis Terman of Stanford University in California started tracking more than 1,500 gifted children — as identified by the Stanford–Binet IQ test that he developed — born between 1900 and 1925. It was one of the world's first longitudinal studies and it has now accrued the longest in-depth record of human development, having tracked the participants for nine decades — looking at their home lives, education, interests, abilities and personality.

One of Terman's main goals in his 'Genetic Studies of Genius' was to disprove the then-common assumption that gifted children were sickly, socially inept and not well-rounded. But even by the standards of his day, the study design was plagued with problems. His selection method was haphazard, administering the tests based largely on recommendations from teachers, and the sample was far from representative (more than 90% were white and upper or middle class, and Terman even enrolled his own children). What is more, Terman skewed the very life outcomes he was trying to study, writing letters of recommendation for his 'Termites', as the participants became known, and helping several to get into Stanford.

By tracking the children into adulthood, Terman showed that they were just as healthy and well-adjusted as the general population and that they generally grew into successful, happy adults. And, as it progressed, researchers tweaked the study to try to overcome some of its flaws.

In the 1980s, for example, George Vaillant, a psychologist at Harvard Medical School in Boston, Massachusetts, began using Terman data to supplement his own long-running study of adult development and started collecting the death certificates of the Termites. Using these records, Howard Friedman, a psychologist at the University of California, Riverside, was able to come up with one of the Terman study's most significant findings. He showed that conscientiousness — namely prudence, persistence and planning — measured in both childhood and adulthood is a key psychological factor predicting longevity, and was associated with as many as six or seven extra years of life2. “It would not have been easy to discover this without a lifespan longitudinal prospective data set,” says Friedman.

Longitudinal studies evolve according to scientific fashions, says Laura Carstensen, director of the Stanford Center on Longevity. New investigators will add measures and modify or discard those that they think are no longer interesting or are obsolete. “Today we would measure emotional well-being, for example, in a very different way from in 1900,” she says. So in many ways “looking at a longitudinal data set is almost like writing a history of psychology”.

85 years: Waiting for the drop

JOHN S. MAINSTONE, UNIV. QUEENSLAND

The pitch-drop experiment has yielded just eight drops since 1927.

On his second day of work at the University of Queensland in Brisbane, Australia, in 1961, physicist John Mainstone came across a quirky little experiment that had been quietly running in a cupboard for 34 years. Fifty years later, he is still looking after it — and still waiting to witness its most dramatic activity.

The pitch-drop experiment started when Thomas Parnell, the university's first professor of physics, set up a demonstration for his students to show that a sample of pitch, a black tar distillate that is brittle enough to shatter with a hammer when cold, will act like a liquid and flow through a funnel, dripping out of the bottom like the world's slowest hour glass. It did, at a rate of about one drop every 6–12 years. Mainstone expects — cautiously — the ninth drop to fall sometime towards the end of this year.

The experiment is not exactly a hotbed of discovery. In 86 years, it has yielded exactly one scientific paper3, which calculated that the pitch was 230 billion times more viscous than water. And in 2005, it earned the dubious distinction of an Ig Nobel prize (see Nature 437, 938–939; 2005), a cheeky parody of the Nobels.

There is, however, still some science to be gleaned. No one has ever seen a drop fall — a webcam recording the experiment failed just when the most recent drop fell, in November 2000 — so exactly what happens when the drop detaches from the body of pitch above is unknown. It will also take another few decades to tease out how weather, the introduction of air conditioning and the vibrations from renovation work in the building influence the drip rate.

But Mainstone says that the experiment's value lies not in its science, but in its historical and cultural impact: it has inspired sculptors, poets and writers to muse about the passage of time and the pace of modern life. It also provides a link to scientific history, and a sense of constancy. “It's going about its business while the world is going through all sorts of turmoil,” Mainstone says. And with a great deal of pitch left in the funnel, it has the potential to serenely ignore that turmoil for another 150 years or so. Luckily, 78-year-old Mainstone has already convinced a younger colleague to watch over the experiment after he is gone.

Journal name:
Nature
Volume:
495,
Pages:
300–303
Date published:
()
DOI:
doi:10.1038/495300a

References

  1. Bearchell, S. J., Fraaije, B. A., Shaw, M. W. & Fitt, B. D. Proc. Natl Acad. Sci. USA 102, 54385442 (2005).

  2. Friedman, H. S. & Martin, L. R. The Longevity Project: Surprising Discoveries for Health and Long Life from the Landmark Eight-Decade Study (Hudson Street Press, 2011).

  3. Edgeworth, R., Dalton, B. J. & Parnell, T. Eur. J. Phys. 5, 198200 (1984).

For the best commenting experience, please login or register as a user and agree to our Community Guidelines. You will be re-directed back to this page where you will see comments updating in real-time and have the ability to recommend comments to other users.

Comments

Commenting is currently unavailable.

sign up to Nature briefing

What matters in science — and why — free in your inbox every weekday.

Sign up

Listen

new-pod-red

Nature Podcast

Our award-winning show features highlights from the week's edition of Nature, interviews with the people behind the science, and in-depth commentary and analysis from journalists around the world.