 I sent in a suggestion of two possible case studies without thinking too much about it. How to actually evaluate the value of information with regard to the cases. But this is one topic that I think is quite interesting in many respects. Most of my slides are basically on the background of this topic. And as the title says, it involves the effects of soil structure interaction on the excitation response of buildings, subjected to near and far field strong motion. And just for repeated information, I think I have showed you this some time before. But Iceland is pulled apart by the Eurasian and the American plate. And in the south of Iceland we have a transfer zone where we have fairly strong earthquakes. So we have the South Icelandic seismic zone as you can see there. And in the past 17 years we've had three big earthquakes in this area. In 2000 we had two earthquakes, approximately 6.5 in the eastern part and then the middle part here. And then in 2008 we had a 6.3 magnitude earthquake in just east of Queragalli. And of course when these earthquakes occur it's not a single event, it's a multitude of events. So you have one main event and then you have post events that may go on for a year or longer. So we have quite a bit of data available. But in 1999 this building was instrumented and it's instrumented in the basement with a triaxial accelerometer. And then on the top floor, the third floor just there we are measuring the north-south motion and the east-west motion. And this is basically the only building that is instrumented for response measurements in this seismic region except power plants are on the outskirts of the seismic zone and therefore not as sensitive. So what we discovered when we compared the response of the building in the June 21st earthquake which was about 15 kilometers away and then the response in the 6.3 earthquake on May 29th in 2008 which was 5 to 8 kilometers away from the building was that we had very strong dissimilarities in the structural response for these two events. And the response factor don't show that maybe two clearly but still we can see that the response for the latter event in 2008 has a very strong peak in the sort of about 0.5 second period which is not as strongly seen in the previous event. And the damage to the building was fairly limited. There was some cracking to be seen in 2008 but not that severe. There were some damage to interior because of high acceleration levels but this is actually in the basement of the building recently installed book rack. But when we looked at the response in 2000 the earthquake was further away and we had a horizontal peak ground acceleration of 13 and 11 percent G in each direction and then we had on the third floor we were measuring up to 36 percent of G. So we had about three times larger accelerations on the third floor than in the basement. However in 2008 the horizontal PGA was 54 and 33 percent G in each direction but on the third floor we had 72 percent of G and in general we had only about 1.4 times larger response on the third floor than in the basement which was surprising because you would expect this on a step foundation to be at least 2.5 times magnification from basement to the third floor. And when we looked at the spectral information these are just power spectral densities of the relative acceleration between basement and top floor. Then we see that basically the frequency content has shifted down in frequency from what we saw in June 2000. So what's the reason, what lies behind this? And it took us a while to sort of come up with a hypothesis but what is traditional in soil structure interaction studies is to use study or look at the horizontal vertical spectral ratios and when we evaluate those we see that in 2000 we had a very strong response at approximately 8 hertz but then in 2008 we had a very strong response at about 2 hertz and this is just a demonstration of analysis that I did in 2002 or 2003 where I was looking at the main events and post events from those previous earthquakes and there basically in all the data the 8 hertz frequency was very strong and the actual natural frequency of the structure was not as dominant but it's around 6 hertz the first natural frequency. So the reason or our hypothesis is that we are fairly confident that the fact is that during the Ice Age several interglacial periods occurred and then we had sea level rising up to about 100 meters above the present coastline. So a large part of South Iceland was under sea accumulating sea sediments on top of the rock foundation and then in after the Holocene after the glacial or the Ice Ages then we had a lot of volcanic activity in Iceland and in 8500 years ago there was a big lava flowing over a large portion of South Iceland and therefore flowing over the soft sediments. So and then of course these lavas which are even though it's 8500 years old it's not really a very old rock in the geological sense and that is also has different compositions. So we have a compound lava and we have scoria lava which have different characteristics. These are much weaker than the compound lava layers. And so this is a hypothetical rock soil profile based on information from a borehole near to the building and then we have a compound layer on top. So when people are building the structures they think they are building on solid rock. Underneath the compound lava there is scoria lava layer and underneath the scoria lava layer there is sediments gravel, sand and clay. And then below that there is older rock 1.5 billion years old. And this has been studied based on ground motion data from other areas such as Quiragirli where we have a seismic array and then it has been seen that we have to split up and identify two peaks in the data and about between two or three birds and then around eight birds. So this is fairly well established. So we are going to skip these. This is a ship tower that I did a very simple platform for considering the lava structure and I put the two peaks with the measured natural frequency from post events. And so the outstanding questions are how to distinguish between foundation frequencies and structural frequencies and how to properly evaluate the natural frequency of a structure and its position on such complex foundation layers. And the fact is that in the building in question the foundation actually acted as in my opinion as a seismic isolation. So I think the building would have been similarly damaged if it actually had a magnification of a factor three rather than 1.4. But of course if in a different type of building, a more high rise building with the natural frequency towards two birds and of course the magnification would have been even stronger. So there are several things to consider. And I'm not pretty proposing to study this case in detail. I think it's more sort of suitable to consider a simplified scenario that is reflecting reality as I said. So we could select quick action layers, levels, different applications of distance. We could look at different foundations, distance. For instance reflect these two cases where we have a stiff foundation as a reference or an 8 hertz stiff or foundation system with an 8 hertz natural frequency and 2 hertz natural frequency. And that would affect the hazards for a particular building. And then we could maybe study different building types. Low rise, medium rise, high rise and evaluate the response. And based on that we did risk. And then the value of information would basically be the difference in risk because if we don't have any monitoring nobody would know about this. So we can either say it or we can either underestimate the risk or for instance high rise buildings or we are overestimating the risk for low rise buildings. And then of course we could provide argumentation for the monitoring or further information coming up. Thank you very much. I think buildings are very important and it's the first building and the classical monitoring of a building in case of earthquake. Is it the purpose to have a class of buildings here? So medium rise, high rise buildings. To evaluate the value of information for a class of buildings for different. Is it another approach, not approach, another case study just to say only as a mark, not as a mark? Yeah, I think it's not practical to analyze or for this for the post-action I think it's maybe not practical to analyze the building and the local conditions in too much detail. I mean that might be from our scope of interest here but to do a sort of parametric type of study that would reflect the issue. Maybe I'm not, but you could select the limit state like the intra-story drift between two stories as a limit state and look what would be with and without the monitoring. I don't know. Yeah, well basically it stops a method without monitoring. If you don't have monitoring, you don't know about the effects. So it's not monitoring. Everyone assumes that the foundation is still. And then you build your buildings and you do your risk evaluation for instance insurance purposes. The value of monitoring is the highest for the foundation assessment. Yeah, it's basically that's the value. You know more. You understand better the real risks than you would without monitoring.