GIS-modelling of long-term consequences after a nuclear accident.
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 183 | |
| Author | ||
| License | CC Attribution - NonCommercial - ShareAlike 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
| Identifiers | 10.5446/32124 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language | ||
| Producer | ||
| Production Year | 2015 | |
| Production Place | Seoul, South Korea |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
00:00
Nuclear spaceMathematical modelTerm (mathematics)Open sourceView (database)Mathematical modelDomain nameOpen sourceCartesian coordinate systemReading (process)Level (video gaming)Nuclear spaceComputer animationXMLUML
01:01
Mathematical modelRun-time systemComputer fontCASE <Informatik>Mathematical modelCodeSpecial unitary groupNuclear spaceAreaDifferent (Kate Ryan album)Presentation of a groupRun-time systemMathematical modelSquare numberReduction of orderCASE <Informatik>LiquidBitMetreField (computer science)Function (mathematics)Mereology2 (number)Computer fontGrass (card game)Internet forumMatching (graph theory)CausalityGame theoryQuadratic equationTheory of relativityDerivation (linguistics)Computer programmingHypothesisTerm (mathematics)Multiplication signGraph (mathematics)Focus (optics)Computer animation
04:38
Thermodynamic equilibriumMathematical modelLevel (video gaming)Population densityoutputSpecial unitary groupGame theoryHeat transferGleichverteilungLevel (video gaming)Expected valueArithmetic meanAreaMaxima and minimaHeat transferDivisorType theoryProduct (business)Mathematical modelDistribution (mathematics)Reduction of orderPopulation densityDifferent (Kate Ryan album)Group actionMereologyObservational studyMeasurementView (database)ConcentricSquare numberCASE <Informatik>Parameter (computer programming)MetreVarianceScaling (geometry)Position operatorMetropolitan area networkNormal distributionQuicksortTheoryIncidence algebraIRIS-TMathematical modelNormal (geometry)Term (mathematics)Internet forumThermodynamic equilibriumFigurate numberData structureRange (statistics)outputGame theoryUsabilityGraph (mathematics)Set (mathematics)Computer animation
12:51
Special unitary groupGradientExecution unitGrass (card game)InfinitySource codeScripting languageMathematicsAreaRule of inferenceProduct (business)Shape (magazine)Electronic visual displayClique-widthMaxima and minimaColor managementThermal radiationMathematical modelScripting languageFunctional (mathematics)Raster graphicsPopulation densityComputer fileLevel (video gaming)ResultantInformationComputer programmingNumberGoodness of fitDisk read-and-write headGrass (card game)CodeData storage devicePlug-in (computing)Online helpLiquidDifferent (Kate Ryan album)AreaMultiplication signObservational studyCASE <Informatik>Universe (mathematics)DatabasePoint (geometry)Compilation albumTheory of relativityPresentation of a groupMereologyMetropolitan area networkProduct (business)Position operatorHill differential equationSocial classPressureWordReading (process)Graph (mathematics)Parameter (computer programming)Noise (electronics)Limit (category theory)Open source
21:04
Computer animation
Transcript: English(auto-generated)
00:05
So, I think it's time, we should start. It's four to half past one. My name is Martin, I'm from Norway. I'll be the first talker today, and I'm also the chairperson. So first I will just introduce the people
00:22
talking in this session. So this is the session for various applications of open source GIS, GIS. I'll be having a talk about GIS modeling of long-term consequences after a nuclear accident. After that, we will have a spatial view in the culture heritage domain by Jakob
00:42
from Saxon State and University Library. And last, we will have mapping in GeoServer with SLD and CCS from Andrea. Okay, so are you ready?
01:02
As I mentioned, my name is Martin. I work at the Norwegian Radiation Protection Authority. I also do some work in the center of environmental reactivity. Yeah, so this is me.
01:21
When I'm not doing any programming, Python programming, I'm out in the field cutting some grass. That's actually from this year. I usually do that like one week every second year, so it's quite fun getting out, outdoor sometime.
01:40
Just try it. Okay. So the outline of this talk, first I will talk about the scenario, the case scenario. Also talk a bit about Chernobyl, because that's the reason why we have the institute and we have people working there, actually.
02:01
And the largest part will be how the model works, or how I focus on this part. And also, if you have time, I'll just show some easy code, how this model is made. So the Chernobyl accident is the largest
02:22
nuclear accident we have had. And Norway was one of the countries which were most affected, one of the countries in the world. And you can see the affected areas here, which is some areas in the middle of Norway,
02:40
or south there, almost reached one mega-bacrel of mega-bacrel per square meters. Who here does know anything about radioactivity? Raise your hand. Who knows what a bacrel is? Okay, a bacrel is one counting per second.
03:02
So when one mega-bacrel, it's per square meters, it's a lot, so that's much. And now, after 30 years, after the Chernobyl accident, we still have trouble with the Chernobyl deposition in these areas, and we have to do countermeasures
03:20
on different kinds of food products. Especially the reindeers, cheap, other game. Well, we'll come back to that. So that's the Chernobyl. This is the hypothetical scenario. I had one presentation where somebody asked,
03:40
when did this happen? This has not happened. This is a hypothetical scenario. And in Norway, we are worried about this cellophile nuclear plant in England, in the east coast of England. It's storing a lot of liquid reductive waste.
04:05
And the scenario is thinking about if we have a big explosion there, what will be the consequences for Norway? This is an output from a model made by the Norwegian Meteorological Institute.
04:22
It's called a SNAP model. It's a dispersion model for nuclear accidents, and also used for volcanoes and stuff.
04:42
So, the consequences. The first one to two years, we have what we call short-term consequences. They are really complex and seasonal-dependent. I won't go into that, because that's not the topic of this talk, and I don't know much about it either, so that's okay.
05:00
But after one to two years, we have a more or less equilibrium. We know from the experience from the Chernobyl accident that the main consequences was for animal grazing and an improved pasture, and the main isotopes are cesium and some strontium.
05:29
So, what do we do? Well, introducing the Stratos model, which will figure out the consequences. So, we have a deposition,
05:42
and we have reactivity in the animals grazing, the woodlands, the unimproved pasture. We know that from the experience from the Chernobyl accident. So, what this model tries to do is saying, is this food okay to eat?
06:04
Well, so, in the model, we have included, we have reindeer, sheep, other kinds of game, mushroom, and we have some berries. Because wild berries and fungi are also important
06:20
for the human consumption in Norway. Yes. So, the Stratos model is a very easy model. It has four inputs, sets of deposition, tags, intervention levels, and density maps. And I will go through all but four of them.
06:43
So, the first of all, it's the deposition map. You have already seen the Chernobyl. This is from the case scenario. We just filter out the part we're interested in, Norway, of course. It's the only thing we're interested in.
07:00
We see that the largest deposition in this scenario is at the west coast of Norway. It's marked one, Filke, in Norway, which we have a case study on, but I'm not going into that here. So, this is the deposition we're dealing with.
07:20
We see the levels are going up to almost one megabecorale per square meters. So, it's a Chernobyl. It's similar to Chernobyl. Also, the scale is logarithmical. I forgot to tell you, but if you have any questions,
07:42
I'm feeling happy about you interrupting. It's no problem. So, if you have a question, just pop up. So, first we'll talk about the tags. It's the tag is just the ratio between the concentration
08:03
in the meat or the product divided by the deposition. So, it's very easy. And we have a lot of those tags from the Chernobyl accident, lots of measurements were done. So, we have a lot of these tags,
08:21
and based on those tags from the experience of the Chernobyl accident, they are very, have a very high variance because it's so many parameters contributing to the tags.
08:42
So, it could like be areas where the tags are high and you have areas where the tags are low, and big variance, and it's very hard to model where you find the high and low tags. So, that's why we are using three different tags in our model. We're using an expected tag.
09:01
That's what like, okay, we're expected. It's the mean of the normal distribution, and we use the minimum and the maximum tag. And basically, you can say that we don't think we'll find a tag that is lower than the minimum, and we don't think we'll find a tag that's higher than the maximum.
09:22
So, this is an important part. And then introducing the intervention levels in Norway. We have different types of intervention levels, and they are listed here. Intervention levels indicate when the dose reduction actions must be taken.
09:44
So, food stuff about these levels can't be sold in Norway. So, now we can apply the deposition together with the tags, and we can filter out areas
10:01
where the products go above or under the intervention level. So, here we have the areas for moose in Norway with different kinds of tags. So, using a low transfer tag, we get this red area. And what we can say about this red area is that no matter what the tag is,
10:21
the transport from the soil to the moose, this area, all the moose will be above intervention levels. For the orange area, we expect to find moose above the intervention levels. But there can be some areas where they are not.
10:42
In the high transfer area, we expect not to find moose above intervention levels. But certain places where the sensitive areas you might find moose. But the green areas, you know, you won't find any moose above intervention levels.
11:03
So, if you hunt the moose in those area, you'll save. Okay, why did I have a question mark there? Yes, because the big question now is, are there any moose in these red areas?
11:20
We don't know, but that's why we introduced the density maps. So, as you see, this is a density map based on registered hunted moose in Norway. And I've made these density maps from all the different kinds of products that goes into the model. And here we see that the most moose
11:43
is on the east side of Norway. And there's almost no moose on the west coast where we have the highest deposition from this hypothetical scenario. Okay, yeah, so I won't go into
12:01
how I made those density maps, but. So, now we'll have a look at our view of the model. You have a deposition. You have this aggregated transfer factor that tags. You multiply them. You see, you get the red areas where from the three different tags, you get three areas like this
12:20
where you see where the product goes above the intervention levels. You have a regional distribution of the animals or the products. And this could be a municipality or a grazing area of some sort. And we have assumed a uniform distribution.
12:41
And combining those two layers, we can find out how many animals are above the intervention levels for the different products. And we also know where these animals are. So that's the overview of it.
13:01
So, you have a radiative deposition, tags, intervention levels, density maps, and here are the products, the results. You have a map for each product, and you have a text file which gives a number or the percentage, whatever.
13:23
So you see, for this case, the moose from the Sellafield scenario, we only have 1% of the moose, hunted moose in Norway. So that's no problem. And in the expected area, only 3%. And in the maximum area where we don't expect there to be many mooses
13:46
above the intervention levels. So that's 33%. We can also say something about time from this map. We know that in this area, you'll have a problem for a very long time.
14:01
While in the cocky colored area, you know that after like 10, 20 years, there won't be any problem anymore. So that's also information you can get out of this map. Yes. So, we'll have actually some time for a code.
14:23
So for doing this, I have mainly used Grass and Python. Python wrapping in some Grass code. I also used Python to make a GUI, but I'm not showing that here.
14:41
I also used a Postgres database for storing the tags, but I don't think that's really necessary, not in this case, or at this stage. So, just looking at some code. So it's, the easiest way is to start Grass, and then in your script, import Grass like this,
15:03
Grass script, that's Grass. And then you can use the different kinds of Grass functions. And here I'm showing a function, because Python likes, is very fond of defining functions, and making functions makes your code clearer,
15:21
better structured, and it's easy to reuse. Yeah, so here I've used the MapCalc function, which is a very good function to use if you're working with rasters. I have another. So this is another function.
15:42
I'll just show that I'm using this Grass run command, which is basically you can start most of the functions in Grass. So this is just for wrapping the Grass command, and this is the function in Grass.
16:01
And then it's the parameters, so it's very straightforward, and it's easy, easy scripting. So it's easy and fun. Yes? Thank you for attention. Any questions?
16:22
Yes? Yes? Grass GIS. Oh, someone else should be here and tell about it. It's a fantastical GIS tool for, you could say it's like QGIS, only it's more like,
16:43
I'm going to get knocked in the head of some Grass folk by saying this. But I think it's a really good GIS program. So you could load in the maps, and you could make maps, but it's mostly for doing map calculations, map analysis, and well, you can do almost anything.
17:05
But it's not, which is a dumb part, it's not made for like doing cartography, or it wasn't, so now it's getting better. But you have a plugin in QGIS for importing things you've done in Grass. So if you want to make a nice map,
17:20
which you have seen my maps, it's not that good, but if you want to import the maps you make in Grass, there is a plugin in QGIS, so you can easily get the layers and all the stuff you made in Grass, basically. So it's a GIS program, which is open source,
17:42
very good, and very good community. I'll have lots of help from the Grass community. I've emailed them, and I answer quite quickly, and it's, well, I wish I'd used it more. So yeah, hope that answered your question.
18:15
We have no commercial reactor in Norway, but we have two scientific reactors in Norway,
18:22
but they are very tiny, small. And Sellafield has very much, and they have a very big storage of red activity, liquid red activity. We also looked at other case study where we looked at landing ground nuclear power plants,
18:42
which is a power plant. And we basically looked at different kinds of scenarios, one which was a Chernobyl-type scenario, and some other which was a minor accident. And we looked at the consequences after those as well. But Sellafield has a lot of red activity waste
19:05
because they are doing reprocessing of old spent fuel. Also, the wind is going eastward towards Norway, so if something happens, it's very likely that you get something to Norway, as if. But in Leningrad, it's not so often
19:21
having wind going to Norway. But it has to be a big explosion. You have to get lifted the waste really high. So it's very, well, this could be discussed all day, but it's very hypothetical. And I don't think it will happen. Once you make a scenario, the scenario won't happen.
19:42
The next nuclear accident, you wouldn't know what it is. If you knew it, it wouldn't happen. That's my opinion. Yeah, thank you. It's like Fukushima. Nobody was prepared for it. The next one will be a similar, or it will be a shock like, oh, why didn't we think of it?
20:02
Once you think of it, it won't happen because then you're prepared for it, yeah. Silly question. What about fish? Yes, we have a different model for fish. So it's not included in this model, but it's a good point. And that's not, my institute is not working with it,
20:20
so I don't know much about it, but I met the guy who's working at it, and he's working at the university in Norway. Modera or something, it's a model. So it's, yeah, you have different kinds of model for different things, that's my problem, yes.
20:45
Okay, so I've started working on UV radiation, so I hope to see you next year, and then I'll be presenting something of UV radiation, and this is the ozone layer over Norway. So, okay, thank you.
21:01
And the next speaker is Jakob. Okay.