Transcript Aug 2007

Gareth Mitchell: From Europe's leading science university this is the official Podcast of Imperial College London.

Hello, I'm Gareth Mitchell, presenter of the BBC's Digital Planet and a lecturer here in Imperial's Science Communication Group. In this the August edition -the catastrophic flood that made Britain an island. And, as Imperial celebrates its 100th we look back in time to the College's earliest days.

Hannah Gay: The Imperial Institute was being demolished when I was a student and people of my generation thought it was a hideous thing and were pleased to see it come down. But I think, after we saw some of the buildings that went up, the Sherfield Building in particular, we had second thoughts about this.

GM: Plus in more modern times - technology gets everywhere, literally everywhere. So enter Imperial's new Pervasive Sensing Group dedicated to putting listening, watching, measuring and monitoring devices in everything from corridors to concrete.

Eric Yeatman: You could relate pervasive sensing in one application area to do with the term smart building or intelligent building but that term has been around long enough and unsuccessfully enough to have gotten rather a bad name. You know, you sit down quietly at your desk and all the lights switch off because it thinks you're not there anymore.

GM: That's all to come right here on the official Podcast of Imperial College London.

How a mega-flood split the UK from Europe

All right, well let's start with that catastrophic flood that sliced Britain away from the greater European continental land mass. Dr Sanjeev Gupta has published his work on this in the journal Nature. And we're in his office here in the Earth Science and Engineering Department at Imperial. So just take us back. We don't know exactly how far back but there was a time basically when Britain was part of Europe effectively?

Sanjeev Gupta: Yes, absolutely. Actually for most of Britain's geological past it's actually been attached. It's been a promontory of Europe so Britain's island status is actually very recent in geological time.

GM: Just give us a sense as to how Britain was attached to this greater land mass. I mean, was it a fairly narrow land bridge stretching from what we think of now as the southeast of England to the north of France or was it something a bit different to that?

SG: Well, I think one has to consider what sea levels were doing. So we know that sea levels in the past million years have actually fluctuated quite markedly. It's something quite remarkable. Certainly, for example, in the last 500,000 years our sea level has fallen to 120 metres below the present five times. And so actually Britain's island status has a very narrow window of time in fact. So in general Britain has been connected. But particularly what we recognise is that there's an extensive chalk ridge extending from southern England, from Kent, into northern France and this ridge actually forms the famous white cliffs of Dover.

GM: And at one point that was a ridge that was intact linking Britain to France. That's now gone, do we have any insights as to how it went?

SJ: Yes, it's actually quite exciting. I mean, this is an old idea. We haven't developed this hypothesis ourselves. We've actually tested an hypothesis that was actually proposed as a sort of outrageous idea in the mid 1980s. An English geologist called Alex Smith suggested that a huge lake that existed where the southern North Sea is at the moment formed and this was bounded to the north by the ice sheets. So the British ice sheet and the Scandinavian ice sheet had actually coalesced and formed a wall of ice across the North Sea. At the southern boundary there was this rock ridge, the ridge I've talked about, made out of chalk. And so you had basically a bath tub, a basin, which got filled up with water from the Rhine and the Thames and all the glacial melt water. And at some point there was so much water in this lake, this huge lake that was probably something like the size of Wales, maybe even bigger, it actually over spilled this chalk ridge and basically there was a catastrophic breach in this ridge.

GM: You've been, and your colleagues here, doing some fascinating work with very accurate sonar imaging of what is now the English Channel to give a few ideas as to how that ridge was breached then?

SG: Yeah, that's it exactly. When I first read this paper when I came across it in the library one day I realised that we actually had the technology now to test this problem. Initially actually what we wanted to do was to go and survey in the Straits of Dover but my geophysicists colleagues said that's the world's busiest shipping lane. There's no way I'm taking our equipment out there. So we actually started working further to the west. We actually developed a collaboration with the UK Hydrographical Office who presented us with this remarkable data set collected over 24 years, which is high resolution sonar data which actually produces the first detailed map of the sea floor. So it's exactly like taking the sea away and actually being able to see the landscape below the English Channel.

GM: You now then have this incredibly detailed idea of what's going on, on the floor of the English Channel, and what it looks like so what have you seen from these images?

SG: Well, it's quite remarkable. We can see this extraordinarily large valley that's carved into the centre of the English Channel. This is actually carved into rock, into chalk and limestone and sandstone. And this valley is 50 metres deep, up to 10 to 15 kilometres wide and it has very distinctive features that are characteristic of landscapes that have been carved by huge floods. Now, in particular what we've done is we've compared the landscapes we see under the Channel with a landscape in western North America, an area which was carved by a huge catastrophic flood about 15,000 years ago. But the landscapes we see there are analogous to the ones we see in the Channel so we can be very confident about our interpretation. It's likely to be one of the most significant or amongst the handful of most significant floods. It really is the dimensions. The dimensions of the valley are so big that we know that there's no major rivers feeding into this valley so it couldn't have been formed by some of the smaller rivers that we see in Southern England or in France. It's really quite impossible to generate something like this through normal river processes.

GM: So you're now saying with a very high degree of confidence that Britain effectively got sliced off mainland Europe by this incredible flood? Like the North Sea breached this land bridge effectively and with such violence that the two land masses separated?

SG: Well, it's not quite as simple as that. So what this did was this event actually set up the conditions for Britain to become separated as an island from Europe. Prior to this event, as we've said, there would have been a rock ridge joining Britain to the continent. Now, even at times of h igh sea level this rock ridge was actually higher than the highest se a levels and the refore even at high sea levels there would have been a sea embayment in the North Sea and a sea embayment in the Channel and they wouldn't have been able to meet together. By cutting this gap in the Straits of Dover it allowed, when sea levels rose when the climate warmed, the sea from the English Channel to actually connect with the North Sea and therefore at these times Britain could actually become an island.

GM: And so it kind of set up the conditions that I'm assuming over geological timescales eventually then separated Britain for good?

SG: Yeah, absolutely. So every time there's high sea levels, particularly for example at our present time, it will be the case that Britain will be an island. Of course if sea levels fall then Britain actually becomes connected. So if we were to go into a major glaciation now and sea levels started falling eventually the sea level would drop below the lowest point that we see in the Straits of Dover and you could walk across.

GM: I know it's hard to gauge exactly what it was like when this catastrophic flood happened but can you give an idea of what people might have seen if they were standing on the southeast coast of Britain? Because I assume this wasn't just a huge gush of water and then that was that, it was all over in ten minutes? It went on for a bit longer than that?

SG: Yeah. It's difficult to estimate how long exactly it went on for but I would say something like several months. It would have been a huge torrent of water cutting through. And we have several examples in the geological past of similar events. This isn't a one-off event. This has happened before. So there would have been huge eddies in this flow. Giant boulders flying around and all of these sorts of things. It would have been a terrifying experience and extremely noisy I would imagine. It's unlikely that people were hanging out at that time because this probably happened during a cold phase and it would have been likely too cold to be anywhere near Britain. At those times people actually retreated to southern France or southern Spain.

GM: But just give us a sense then of how profound this event was in subsequent British history?

SG: Well, if you think about it actually it makes all the difference. This event changed the course of British history. Imagine that we were not an island. Our whole history has been generated or controlled by our island status. Our whole psyche is controlled. If this hadn't happened the M20 would have actually continued across the rock ridge into France all the way to Paris. There wouldn't have been the invasions. Britain would have been an integrated part of Europe. We would have had the Euro and everything. And we think that the truncation of this land bridge basically formed a profound barrier to human migration. So firstly, once this event happened and the rivers diverted, it may have been difficult for early humans, for Neanderthals, to actually cross these swollen rivers that would have been fed by melting ice sheets. And secondly, once sea levels rose and Britain actually became an island it would have been very difficult. They would have got to Calais and found the route barred to them essentially.

GM: Sanjeev Gupta going back a few hundred thousand years.

Well, in a moment more from the past. Not quite so dim and distant as that great flood but a long time in Imperial College terms as we trace the history of the college with one of its former students who's just written a book on the subject.

Headlines from around the College

But before all that let's have a quick look at some of the other stories making the news from around the College.

What do human beings and the great pond snail have in common? Yes, insert your own tasteless joke here about a politician of your choice. But beyond that it turns out that we share a similar pain mechanism with our pond dwelling cousins. In our brains we both have a protein called the potassium channel. It regulates the way neurons communicate with each other blocking messages, including pain, unless the protein is deactivated. So the flipside to that is that if the potassium channel is switched on pain signals don't get through, which is pretty handy if you're trying to develop an anaesthetic. In the lab the team of biophysicists in Imperial's Division of Cell and Molecular Biology have been firing anaesthetics at experimental potassium channels hybridised from humans and great pond snails. The trials seem to support the long held suspicions that the channels are key targets when it comes to blocking pain. And if the team is right it could lead to highly accurate anaesthetics targeted at this one protein with far fewer side effects than current compounds.

And briefly, Imperial researchers have linked up with colleagues at King's College London to study the link between asthma and obesity. The Imperial and King's scientists say that they've found that cells that cause inflammation in the lungs of asthma sufferers may also release a protein that tricks the brain into thinking it's hungry. Pinning down this mechanism may well explain why asthma is more common in obese people.

And you can stay up-to-date with news from the College on our Press Office website at imperial.ac.uk/news.

Writing the history of Imperial College

Sound of merry-go-round music

GM: Well, that's not a sound you here everyday at Imperial but that merry-go-round was one of a number of attractions on campus last month as the College celebrated the exact day that it became 100 years old. The sun shone, the bands played and the merry-go-round went round as Imperial staff marked the College's Centenary. It was one of a number of events being held this year for the anniversary. And another has been the commissioning of a book telling the history of Imperial. The author, Hannah Gay, is a former student here and she's been speaking to current Science Communication MSE student Charlotte Bathe about her book and the daunting task of compiling it.

HG: The way in which an institution emerges can have profound long term consequences for its future development. The three colleges that joined to form Imperial in 1907 were pioneering institutions that offered advanced level education in science technology with a pedagogical emphasis on the practical. Even before it became widely understood that skilled scientists and technologists were important to the nation's economic well being the Royal School of Mines, the Royal College of Science and the City and Guilds Central Technical College were producing graduates who were to play their part in government, academic and industrial science.

CB: Dr Hannah Gay reading from her book, The History of Imperial College London 1907 to 2007. So here we are in the archives and there are lots of bookcases here with old books and I can see records of the Governing Body and Council just behind me. In one of the other rooms there were artefacts and various scientific instruments and gowns. So lots of different things here. Did you spend much time in the archives while you were writing your book?

HG: Yes, I must have spent at least two to three years working in the archives.

CB: Now, I've got a copy of your book here and it's over 800 pages and pretty heavy. Was it difficult to assimilate all the information that we have around us here in the archives into a single book?

HG: Well, yes. First of all I had to make choices about which things to read. So I didn't read everything in the archives. Possibly another historian would have chosen other thi n gs to read and written a different kind of book. Also after I h ad collected material and gathered all my notes together I had to reduce those as well, and that's quite a difficult process. It's hard to know what to put in a book and what to leave out.

CB: But Dr Gay must have made many good decisions because Professor Andrew Warwick from the Centre for the History of Science, Technology and Medicine is full of praise.

Andrew Warwick: One of the really great achievements of Hannah's book is that she's managed to write a book which is accessible to a very wide audience and is credible with professional historians of science. And one of the ways she's achieved that is by engaging in a dialogue with myself and other members of the department. We've helped to advise on how a book can be made highly readable and yet make some really important incisive points about the history of 20th century science.

CB: Dr Gay, you were a student at Imperial College in the 1960s. What did you study?

HG: I studied chemistry.

CB:And what was it like being a woman at the College in the 60s?

HG: Well, there were few women. Women were treated reasonably well when it came to doing science. In other words then you were expected to do as well as men and people believed that you could. The prejudice was more against women in a worldly sense. They weren't consulted on anything important. They weren't included in meetings or any kind of decision making.

CB: Do you remember the Imperial Institute?

HG: The Imperial Institute was being demolished when I was a student and people of my generation thought it was a hideous thing and were pleased to see it come down. But I think after we saw some of the buildings that went up, the Sherfield Building in particular, we had second thoughts about this.

CB: I've left the archives for a moment and come outside the Sherfield Building to the Queen's Tower which is now all that remains of the Imperial Institute. I have with me a copy of Dr Gay's book and I'm looking at an aerial photograph taken in 1951. In the photograph the Queen's Tower forms the centrepiece of a long and magnificent looking building. Now it stands alone and has become an icon of the College. Back in the archives what I really want to know is whether there was much controversy surrounding the demolition of the Institute.

HG: Somebody whose name is associated with opposition to the demolition was the poet John Betjeman and he kept up his opposition for many years. And for a while the government teetered on the edge of keeping the Institute but the forces from Imperial College won out in the end. Imperial College had some connections to the Imperial Institute. Both were imperial institutions and they were founded with imperial interests in mind so that Imperial College was expected to help the industries, not just of the United Kingdom but also the industries of Empire. And in the early days that meant agricultural industries such as cotton, rubber, tea, coco, coffee. And this is why the biological departments were strong in the early days.

CB: The strength of the biological sciences in the early years was one of the surprises thrown up by Dr Gay's research in the archives. During and after the Second World War the physical sciences became more dominant. But recently the biological sciences have once again become important as London medical schools have joined the College. Deputy Rector, Les Borysiewicz, was Principal of the Faculty of Medicine from 2001 to 2004. He explains the impact of bringing medicine to the College.

Leszek Borysiewicz: Previously in small medical schools most of the research is very directly focused to a single clinical problem. The big opportunity that has happened to the Faculty of Medicine has been the opportunity to engage in far broader problems and to utilise all the expertise that exists in the other faculties. For example, working with Paul French in optics we've now been able to develop systems whereby we can start looking endoscopically for tumours using what the Department of Physics is developing as new imaging systems. So these are the sorts of opportunities that suddenly open up, and they open up across the College to the benefit not just to those in the Faculty of Medicine but also in the other faculties of the College as well.

CB: According to Dr Gay the bringing of medicine to Imperial is the most significant event in the recent history of the College. Another new development is the opening of the Tanaka Business School. Traditionally Imperial has kept close ties with industry and commerce and the Business School represents a renewed move in this direction. Professor Warwick again.

AW: I think one of the interesting things in College at the moment and one that's very controversial is that under the present Rector it has moved again very closely to industry and commerce, now more than ever setting up of the Business School. The focus on bringing in research money. The focus through Imperial Innovations on getting product out into the industrial and commercial world. And one of the things I think is important about that is that it will ensure the financial security and intellectual security of the College over the next century.

GM: Andrew Warwick ending that report from Charlotte Bathe. And Hannah Gay's book is called The History of Imperial College London 1907 to 2007.

Imperial's new centre for talking sensors

So as one century ends it could be said that another era is possibly beginning with the Centre for Pervasive Sensing, which is partly a venture of the Electrical Engineering Department, which is where I'm now sitting with Eric Yeatman. So, Eric, tell us about this new Centre? It's actually across a number of centres within Imperial isn't it?

EY: That's right. It's one of these virtual centres, if you like. There are a number of researchers in different departments, mainly in engineering departments, that have got an interest in this theme. And we decided to bring a few of those activities together to form this new Centre as sponsored by the Faculty of Engineering and to use it as a vehicle to spread the good word I guess. To get more people involved and to encourage new collaborations both within the Faculty and with other faculties and outside.

GM: So pervasive sensing. I assume it's something to do with being able to put up all kinds of temperature sensors, pressure and visual sensors in a pervasive way? Tell me more.

EY: Pervasive sensing goes a step beyond simply having lots of types of sensors and lots of sensors. You could think of it as a related discipline or sub-discipline of pervasive computing, which is a concept that intelligent electronic devices will become so cheap and so ubiquitous in our environment they'll start to be able to interact with each other and with us in kind of indirect ways that we don't need to be so conscious of. So at the moment when you have electronic devices, including sensors, you are usually aware that they're there and you usually interact with each sensor individually and you'd look at it and you'd see what the temperature is or something like that. But you can imagine a home security system or a building comfort system which used lots of different kinds of measurements at lots of different physical locations and brought them all together without the users having to interact with that process but still getting the benefit of it. So the system has its own intelligence.

GM: And so if we stick with, say, just as an example, the temperature sensor theme. The idea is then at the moment a temperature sensor is literally just something that gauges what the temperature is and maybe feeds that information into your central heating system or into your air conditioning. How would a pervasive temperature sensor work? How would that differ from the kind of thing that we have on our walls already?

EY: Probably the individual sensor doesn't need to be so different. From the point of view of what they do the important thing is probably that instead of having a single temperature sensor in a building or half a dozen you might have 50 t hat interact with each other or communicate with each other and also communicate with other measurement types. Like measuring humidity. Measuring whether there are people present in the room or not. Measuring lighting levels and so on.

GM: And I suppose an obvious application of that might be, you know, for instance where you have motion sensors that can gauge whether we've walked into a room. And we all know that situation where you walk into a room, the lights come on automatically and then you sit at your computer and you stay still for two or three minutes and all t he lights go off. In a pervasive sensing world hopefully that kind of thing wouldn't happen so much if at all?

EY: Yeah. So you could relate pervasive sensing in one application to do with the term smart building or intelligent building but that term has been around long enough and unsuccessfully enough to have gotten rather a bad name for exactly the kind of reasons you say, you know, you sit down quietly at your desk and all the lights switch off because it thinks you're not there anymore. That's because you've got an intelligent system that's not intelligent enough really. If you want to automatically determine whether the lights should be on or off it's not good enough to have one or two motion sensors in a room because there are a number of different things. You have to remember whether the person is there. You might have some video or some heat detector. If you've got a number of things working together and you've got a system that can combine that intelligence in the way that if you employed somebody to stand by the light switch they would probably have a pretty good idea when they needed to turn the lights on and when they don't, you know, with their simple sort of human sensors in their relatively rudimentary brainpower for that particular task. I guess if we can get a system that has that level of function but is cheaper than employing a person to stand by every light switch then we've succeeded.

GM: Well, also in here with us is Paul Mitcheson who's a lecturer here in the Electrical Engineering Department. So what pervasive expertise are you bringing to this field then?

Paul Mitcheson: Well, I did my PhD on one of the technological aspects of pervasive sensing. One of the great barriers to implementing these sorts of sensors is how do you power them? If you think about networks where we're now talking about hundreds of sensors and we're deploying them throughout buildings or one of the scenarios we haven't discussed yet is implantable sensors in the human body. Powering these devices becomes quite difficult. If we're powering them with batteries they have a finite life. In your mobile phone, if you only have one phone, you can charge it everyday. It's not a problem. If you had 100 sensors to look after you can't go around charging them or replacing the batteries. So we need some sort of perpetual powering mechanism. And I've concentrated actually with Eric on looking at how we power these devices using a technique called energy harvesting. This is essentially taking energy from the environment. What otherwise would be waste energy in the environment such as heat radiated from desktop computers, or something like that, or vibrations of walls and doors and air conditioning ducts or motion of the human body and turning this into useful electrical energy to power the sensors.

GM: And that might be a surprise to some people because we do tend to assume that most small devices are powered by batteries. So you might be talking about something that for instance could be powered just from the movements in the building then, something like that?

PM: Exactly. One of the other things that we've looked at in the Electrical Engineering Department is powering things through air flows in air conditioning ducts. So one of our colleagues, actually Andrew Homes, has designed micro turbines to fit in our conditioning ducts to generate energy.

GM: And I suppose this has all come about now, I mean the reason why we have a Pervasive Sensing Centre now and for instance not ten years ago, is that devices have become plenty enough and small enough and crucially low enough power for us to be able to use them in this pervasive way?

EY: Yeah, that's exactly right. Probably the biggest leap has been on the power consumption of the electronics. And if you want to integrate some intelligence with your electronics then you need some low power electronics. But also the wireless communication has advanced enormously and that's also made a lot of applications practical that wouldn't have been practical a few years ago.

GM: Another issue as well as trying to power the device is it's no use having a sensor unless it can communicate with some central unit or with other sensors. And so the primary mode of communication you're thinking then is wireless is it?

EY: Yeah. In most cases in these circumstances we're thinking of wireless. You'd like the whole sensor to be wireless, including both communication and power. Even in an environment like a building where there is mains electricity available in the wall, gaining access to that and converting it to a form suitable for electronics is expensive. You don't want to have to run wires to power plugs. You don't want to have to drill into the walls to gain some power access every time you install something. So we'd like to be able to just take the sensor as a standalone module and just stick it to the wall or shove it underneath the skirting board and it's happy forever. The nice thing from the communication point of view is the amount of data is very small and maybe doesn't have to transmit very far, maybe only a few meters, to talk to its neighbour and maybe doesn't have to do that all the time. It can wake up occasionally and make a little transmission and go back to sleep. So that's the kind of operation we're looking at.

GM: And these sensors can obviously be found in buildings and in rooms but, Paul, also in the human body then?

PM: You'd probably be surprised actually as to the size of implants that you can put in the body to do measuring of, say, considering a diabetic, levels of sugar in the blood. You might want to monitor heart rate in elderly people or something like that. And actually you can put quite large devices into the body. In fact large enough that we can put energy harvesting devices in them.

GM: And that means that your body can almost become its own self contained monitoring centre and then also communication out of the body into whatever monitoring equipment might be necessary?

PM: That's exactly right. So, again, one of the things that we see with a lot of the pervasive sensors is that you have a hierarchy in the number of sensors you have. So you have some very primitive sensors which are the low power sensors, which in this case would be the ones implanted in the body. And then maybe this transmits over a low power short range wireless link to some device maybe a little bit like a mobile phone. Or maybe a mobile phone with some add-on which can do the medical monitoring for you and it can phone your GP or it can warn you if there's some vital sign that you need warning about.

GM: So, Eric, as we embark on the Centre for Pervasive Sensing, have you set any grand objecti ves as to what you hope to achieve within the nex t five or ten years? Are we at that stage?

EY: The main objectives of the Centre are probably not kind of technical and research objectives. They're more to encourage new applications, new collaborations and new ideas. And that's already starting to happen. We had our kick-off event a couple of weeks ago and we've opened a call for proposals for brief projects. We invited a lot of people along to this initial half day workshop who hadn't been involved before and we're already starting to see new collaborations come about. So we've had meetings, in fact today, with people we've not collaborated with before talking about unexpected applications of sensors in their applications. Some civil engineering applications for example and some other kinds of bio applications. So I think that would be my objective over the next couple of years, that we're working on things that we hadn't even thought of.

GM: Eric Yeatman there, which pretty much wraps it up for this edi tion. But we'll have more for you in September. The official Podcast of Imperial College London is updated on the first working day of each month and is a co-production of the Science Communication Group and the Imperial College Press Office.

Ozgur Buldum is the composer who's very kindly let us use this rather nice tune called Lila as our podcast theme music. Hear more of his work at ozgurbuldum.com.

So see you next time but for now from me Gareth Mitchell thanks for listening and goodbye.