This site uses cookies. By continuing to use this site you agree to our use of cookies. To find out more, see our Privacy and Cookies policy.
Skip to the content

Free weekly newswire

Sign up to receive all our latest news direct to your inbox.

Physics on film

physicsworld.com's multimedia channel features exclusive video interviews with leading figures in the physics community.

Visit our multimedia channel to see the latest video.

Michael Banks: April 2009 Archives

By Michael Banks

The Twitter bandwagon keeps on rolling.

Earlier this month Physics World joined the likes of Barack Obama, 10 Downing Street and the US rapper Snoop Dogg on Twitter — a website where people can post an answer to the question “what are you doing?” in under 140 words.

While some people actually do write — or “tweet” — what they are doing in every detail, most use it to post interesting links to stories that appear on the web. For example, Physics World tweets links to stories and blog entries that appear on our website.

Now, Robert Simpson, a PhD student from Cardiff University in the UK, has created a website that ranks papers appearing on the arXiv preprint server according to their popularity on Twitter.

His website searches Twitter for tweets that mention an arXiv url or posts that are tagged “#arxiv” and include the paper’s unique identifier.

The website retrieves and lists all the tweets and produces a table of the most popular papers, authors and arXiv categories ranked by how many tweets they have received.

The website has only been active since 16 April, but already there have been 75 tweets quoting arXiv papers.

This week’s top three papers include an introduction to machine learning, a 3D study of the photosphere of HD99563 and power-law distributions in empirical data.

The paper ranked fourth in the table, however, as far as I could tell was an April fool’s joke, which proclaimed that pi has changed since 1900 BC. So maybe think twice before taking such a ranking seriously.

graph1997.jpg
Map of physics in 1997…

By Michael Banks

Producing maps of science seem to be a popular pastime for researchers these days.

Only last month we reported a map made by using over a billion so-called “click-throughs” - produced when going from a web portal like Elseiver’s Science Direct to the full text of a paper or to the abstract on the journal’s website.

Now physicists have made a map using the Physics and Astronomy Classification Scheme (PACS) codes produced by the American Institute of Physics (AIP).

PACS codes contain four numbers and two letters, which are used to classify papers according to the research they contain. Each paper usually includes two or three PACS numbers listed just after the abstract.

The first two numbers of a PACS code denote the subject area, which are grouped in tens. For example, 20-29 is nuclear physics and 30-39 is atomic and molecular physics while 60-69 is condensed matter: structure, mechanical and thermal properties.

Then within nuclear physics, say, there are up to 10 subfields such as nuclear structure (21) and nuclear astrophysics (26). These subfields are also split in tens, so nuclear structure runs from 21.10 (properties of nuclei; nuclear energy levels) to 21.90 (other topics in nuclear structure).

Bear with me a little longer, nearly there. Next is where the letters come in. Within, say, 21.10 there are then a list of topics, such as 21.10.Tg (lifetimes, widths) or 21.10.Dr (binding energies and masses).

Taking papers from the last two decades in the AIP’s database, Mark Herrera from the University of Maryland, David Roberts from Los Alamos National Laboratory and Natali Gulbachce from the Northeastern University in Boston, have created a map using the links between different PACS numbers.

For example, when a paper quotes two PACS codes, these topics are then linked. Running all this data through an algorithm then sorts and finds clusters of nodes and groups them.

The above images show the full maps from 1997 and 2006 with subfields of a subject area shown (that is the first two numbers in a PACS code). The nodes represent the subject area, and its size is proportional to the amount of single PACS codes it contains. The thickness of the links indicates how many papers have PACS codes corresponding to both nodes.

According to the researchers, the map lets you see how areas shrink and grow and how they merge with one another.

For example, in 1997 crystallography (61) was one central node, but in 2006 it had split into five (as deduced by the algorithm). Even though each one is strongly linked with the others, the algorithm didn’t deem it one single node, which may indicate five separate fields emerging within this area.

lquila.jpg
Earthquake shockwaves (credit: IREA-CNR)

By Michael Banks

Researchers have released the first satellite images showing the effect of the L’Aquila earthquake that struck central Italy earlier this month.

Measuring 6.3 on the Richter scale, the earthquake killed over 290 people with an epicentre only a few kilometres away from the Gran Sasso National Laboratory located between L’Aquila and Teramo, which is best known for studying the properties of neutrinos and searching for dark matter.

Scientists at the Instituto per il Rilevamento Elettromagnetico dell’ Ambiente in Napoli have now started analysing data taken from the Environmental Satellite (Enivsat) operated by the European Space Agency (ESA).

Launched in 2001, Enivsat carries ten instruments for Earth observation, which can measure, for example, sea surface temperatures and the amount of sunlight transmitted, reflected and scattered by the Earth’s atmosphere.

Enivsat’s on board radar can detect changes in the Earth’s surface with millimetre accuracy. Data taken just after the earthquake on 6 April and compared to an image made a few months before show a set of nine fringes originating a few kilometres from L’Aquila.

Each fringe on this ‘interferogram’ represents a ground movement of 2.8 cm, meaning the ground moved by 25 cm at the centre of the earthquake a few kilometres from L’Aquila.

The results from Envisat have also been confirmed by the movement of five GPS receivers located around the affected area that were moved as a result of the earthquake.

ESA is also making the satellite’s data taken in the L’Aquila region freely available for scientists to analyse, which is available to download here.

lab-esterno.jpg
The Gran Sasso National Laboratory (credit: LNGS)

By Michael Banks

Researchers at the Gran Sasso National Laboratory in central Italy are best known for their experiments that are designed to study the properties of neutrinos and search for dark matter.

The underground lab, however, also lies around 20 kilometres away from the town of L’Aquila, which was hit by an earthquake in the early hours of Monday morning. Measuring 6.3 on the Richter scale, the earthquake has so far killed over 200 people. But there are reports saying that Giampaolo Giuliani, a physicist based at Gran Sasso, predicted the earthquake would happen more than a month ago.

Predicting earthquakes is a tricky business as a feature we ran in January points out. But developing a system that could predict when and where they happen, although being a pipe dream at the moment, could save thousands of lives per year.

Reports yesterday say Giuliani predicted the earthquake would happen after nearby sensors picked up excess radon gas escaping from the ground last month.

There is, however, no reliable proof that radon emitted by smaller tremors could be used to predict an earthquake. Guiliani was also apparently told to remove videos and information from the internet warning that an earthquake could hit the region.

Physics World is currently looking into the full details of Guiliani’s story, so stay posted for updates.