Project Hayano is a project to understand the risk of the children in Fukushima and surrounding areas to get thyroid cancer due to internal contamination of Iodine 131.
The team mashed up simulation of Iodine plume emission from the nuclear power plant with traffic congestion map which shows where people actually were.
The purpose of this project, at the end of the day, is to work on a legislation so that when someone gets thyroid cancer in the future, the government will need to compensate the medical fee without the citizens to prove the causal relationship between the cancer and the nuclear power plant accident. Calculations done during Project311 will be used for the financial estimations related to that bill.
Slides used at the presentation is as following:
According to data from 1964, Japanese person's body had about 10 becquerel of cesium per kilogram of body weight.
If we compare this to the data from Fukushima, we can see that most people in the prefecture are far below the level of data back in 1964.
However, that doesn't mean that everything is OK. Iodine 131 has a half-life of only 8 days, and there could have been many who inhaled it shortly after the accident. Data from that period of time barely exists, so many people are anxious about this. It is known that after Chernobyl accident, many children
developed thyroid cancer from the iodine intake. So we need to know what the risk of developing
thyroid cancer is for the children in Fukushima and the surrounding prefectures.
Professor Hayano started "Project Hayano" aiming to evaluate openly the risk of internal contamination by radioactive iodine. In order to do so, he had to know how many people inhaled how much iodine at what time at each location. We have to rely on simulations such as SPEEDI etc to guess the amount of iodine (since real data does not exist). The problem until now was to estimate when and how many people were present at each location. Thanks to the congestion data provided by ZENRIN DataCom at this workshop, they were able to solve that problem.
Risk evaluations are made by multiplying radiation dose times the number of people. For example, if 1 person had a thyroid radiation dose of 100 millisieverts, the risk would equal that for 10 people with 10 millisieverts each. This is the hypothesis they used.
Of course, the federal and local governments have also done population behavior studies, asking where the people were on paper surveys, and their survey results were used for evaluation of radiation doses. However, because this is private information, the data will not be open to public and third-party evaluation is impossible. Therefore, they used the data provided by ZENRIN DataCom and SPEEDI to conduct a risk evaluation that is verifiable by third-parties.
There are many kinds of simulations, he used 4 of them.
1) Data from SPEEDI, which is open to public, movie with timeline
2) Data compiled by JAMSTEC and mashup movie with ZENRIN data
3) Data calculated by the National Institute for Environmental Studies and mashup movie with ZENRIN data, and another mashup movie with ZENRIN data (iodine * people)
4) The JAEA calculations.
Professor Watanabe made a mashup map on Google Earth and overlaid the data with the congestion data from ZENRIN DataCom. For example, in the screenshot below, the green bars and circles underneath is data from ZENRIN, showing where people were at about 3 p.m. on March 15 -and you can move the sliders to change the time frame. The red bars represent the concentrations of iodine in the atmosphere close to the surface.
You can see the simulation here on the web: http://speedi.mapping.jp/
-Time slider at top left will enable you to move the timeline
-Navigation tool on the top right will enable you to zoom in/out on specific place
-You can toggle on/off the following using the checkbox on the menu bar
1) Iodine simulation by National Institute for Environmental Studies (NIES, default on)- red bars
2) Iodine simulation by Japan Agency for Marine-Earth Science and Technology (JAMSTEC, default on)- color coded on the surface
3) Iodine simulation by Japan Atomic Energy Agency (JAEA, default off) - color coded on the surface
4) Iodine simulation from System for Prediction of Environmental Emergency Dose Information (SPEEDI, default off)- orange bars
5) Traffic congestion data by ZENRIN DataCom (default on) - green dots, for places with more than 500 people 3D bar chart is displayed
6) Automobile drive map by HONDA (default off)
7) Phone service map (default off)- red represents areas where mobile phones were down, grey represents areas where fixed phone lines were down
Before we move on, we need to understand how reliable the basic data we are using is.
1) Understanding ZENRIN's data
Prof. Hayano plotted how many people were within radius of 5 kilometer intervals from the reactor along with the time. From the chart, on 3/10-12, we can see people actively entering and leaving the 5-kilometer zone, going to and from work, before and during the accident .
On 3/12-14 we can see fewer and fewer people within 5, 10, and 20 kilometers, due to the evacuation order with people fleeing from the reactor, and the population is decreasing. On 3/13-14 it looks like the population is decreasing, but this was due to the cell phone base stations going down, so we don't have accurate data during this time.
However between March 14th and 15the the base stations started working again and it looks like the number of people is increasing. The most problematic time of the reactors were from the early morning of March 15th to March 16th, when the concentration of iodine was at its highest. Fortunately the ZENRIN data for this time is from after the base stations restarted, so it appears to be useable.
Further investigation on ZENRIN's data: this population frequency distribution chart shows how many people were within this 250-meter grid, and we see peaks at 200, 400, 600, and 800.
What we can guess about this data is that here were certain number of people with automatic GPS features on their phones within the grid (like 1, 2, 3 people) , and we can see that ZENRIN is multiplying them by the proportion of contracted cell phone users, and guessing that there are 200, 400, or 600 people there. Then, the intervals are filled in.
Taking a closer look at the data for March 10, 3 p.m., as an example- people aren't moving much, and the peaks are sharp. But at 8 p.m., when people start changing locations the peaks become flatter. So we can imagine that the data has been derived using the fractions of people leaving and entering the grid. Due to this reason, prof. Hayano didn't use the 250-meter grid, instead made the space-time grid coarser.
2) Understanding SPEEDI's data
Here, Professor Hayano evaluated 2 kinds of SPEEDI calculations (in red and blue), and overlaid them with the calculations from the National Institute for Environmental Studies (in yellow) to check the difference. Unfortunately, the data is extremely different. The source terms must have been pretty different. Therefore, they made the grid coarser here as well.
He published the population figures for a 10-kilometer grid openly, and asked many people to help check the calculations. As a result he was able to approximate the magnitude.
The purpose of this project, at the end of the day, is to work on a legislation so that when someone gets thyroid cancer in the future, the government will need to compensate the medical fee without the citizens to prove the causal relationship between the cancer and the nuclear power plant accident. Calculations done during Project3111 will be used for the financial estimations related to that bill.
Q (Professor Murai): What do you think the relationship between these data presented today should be - data published and analyzed by the authorities, and data that is publicly available and can be analyzed by anyone?
A: That is a difficult issue. Ever since March 11th, there were many questions surrounding the way authorities that are supposedly owning data- how they communicate or publish data, how those data are utilized or not utilized. It is important that our project published the data openly in a way anyone can see what was happening and verify.
Project Hayano's documentation can be seen here [ja]:
You can see how cautious Professor Hayano is about the data itself. Getting reliable data is extremely important in analyzing data, because if the data is rubbish all of the analysis will be rubbish as well. Therefore, he takes a lot of efforts verifying the data, making the data available and asking many people to check and verify his calculations.
The other important aspect about this project is that it is goal oriented- Prof. Hayano has the aim to make a legislature to save the victims get compensated if they get cancer in the future, and he is very focused on making that happen. Data, research and analysis is tools for making bigger decisions, not the goal.
He was also very successful in involving other people help his project- many experts provided data outside Project311 itself, due to his credibility. He was also able to get many people to help verify his calculations.
We learned so much from professor Hayano through watching his projects.
He continues working on his projects, and the latest presentation slides can be seen as below, it is a comparison of thyroid radiation intake between Chernobyl and Fukushima.
It looks like Fukushima (red) is extremely lower than Chernobyl (blue) and the distribution differs greatly as well.
BTW, Project Hayano was actually the very first project that launched at Project311. Upon launching Project311, we weren't sure how many people would participate. We didn't know how many valuable projects will start from our data. On 9/19, we did an event called "office hour". We gathered participants and all of the data providers in our office, and made this opportunity for the participants to ask questions to the data providers. At the after party of this office hour, Professor Hayano started talking about what he wants to accomplish from this project- which was so concrete and so compelling. The moment I thought there definitely will be great findings coming out of the workshop was when I talked with Professor Hayano. Thanks again!
Disclaimer: The opinions expressed here are my own, and do not reflect those of my employer. -Fumi Yamazaki
The purpose of this project, at the end of the day, is to work on a legislation so that when someone gets thyroid cancer in the future, the government will need to compensate the medical fee without the citizens to prove the causal relationship between the cancer and the nuclear power plant accident. Calculations done during Project311 will be used for the financial estimations related to that bill.
Slides used at the presentation is as following:
You can see Professor Hayano's presentation video here, audio is in Japanese, you can turn on English subtitle, and machine translation on other languages:
He started this project because many people feared the internal contamination of radiation due to the
Fukushima nuclear reactor accident. He proposed to do tests of school lunches and other meals, and
combined those with measurements from whole body counters (WBC). From those data, he learned that the internal contamination due to radioactive cesium in contaminated foods was extremely low. For example, the following diagram shows the results of whole body counter scans since April 1st within Fukushima Prefecture. 99.1% of scans were below the detection threshold. And for children, 100% of the scans were below the detection threshold.
If we compare this to the data from Fukushima, we can see that most people in the prefecture are far below the level of data back in 1964.
However, that doesn't mean that everything is OK. Iodine 131 has a half-life of only 8 days, and there could have been many who inhaled it shortly after the accident. Data from that period of time barely exists, so many people are anxious about this. It is known that after Chernobyl accident, many children
developed thyroid cancer from the iodine intake. So we need to know what the risk of developing
thyroid cancer is for the children in Fukushima and the surrounding prefectures.
Professor Hayano started "Project Hayano" aiming to evaluate openly the risk of internal contamination by radioactive iodine. In order to do so, he had to know how many people inhaled how much iodine at what time at each location. We have to rely on simulations such as SPEEDI etc to guess the amount of iodine (since real data does not exist). The problem until now was to estimate when and how many people were present at each location. Thanks to the congestion data provided by ZENRIN DataCom at this workshop, they were able to solve that problem.
Risk evaluations are made by multiplying radiation dose times the number of people. For example, if 1 person had a thyroid radiation dose of 100 millisieverts, the risk would equal that for 10 people with 10 millisieverts each. This is the hypothesis they used.
Of course, the federal and local governments have also done population behavior studies, asking where the people were on paper surveys, and their survey results were used for evaluation of radiation doses. However, because this is private information, the data will not be open to public and third-party evaluation is impossible. Therefore, they used the data provided by ZENRIN DataCom and SPEEDI to conduct a risk evaluation that is verifiable by third-parties.
There are many kinds of simulations, he used 4 of them.
1) Data from SPEEDI, which is open to public, movie with timeline
2) Data compiled by JAMSTEC and mashup movie with ZENRIN data
3) Data calculated by the National Institute for Environmental Studies and mashup movie with ZENRIN data, and another mashup movie with ZENRIN data (iodine * people)
4) The JAEA calculations.
Professor Watanabe made a mashup map on Google Earth and overlaid the data with the congestion data from ZENRIN DataCom. For example, in the screenshot below, the green bars and circles underneath is data from ZENRIN, showing where people were at about 3 p.m. on March 15 -and you can move the sliders to change the time frame. The red bars represent the concentrations of iodine in the atmosphere close to the surface.
You can see the simulation here on the web: http://speedi.mapping.jp/
-Time slider at top left will enable you to move the timeline
-Navigation tool on the top right will enable you to zoom in/out on specific place
-You can toggle on/off the following using the checkbox on the menu bar
1) Iodine simulation by National Institute for Environmental Studies (NIES, default on)- red bars
2) Iodine simulation by Japan Agency for Marine-Earth Science and Technology (JAMSTEC, default on)- color coded on the surface
3) Iodine simulation by Japan Atomic Energy Agency (JAEA, default off) - color coded on the surface
4) Iodine simulation from System for Prediction of Environmental Emergency Dose Information (SPEEDI, default off)- orange bars
5) Traffic congestion data by ZENRIN DataCom (default on) - green dots, for places with more than 500 people 3D bar chart is displayed
6) Automobile drive map by HONDA (default off)
7) Phone service map (default off)- red represents areas where mobile phones were down, grey represents areas where fixed phone lines were down
Before we move on, we need to understand how reliable the basic data we are using is.
1) Understanding ZENRIN's data
Prof. Hayano plotted how many people were within radius of 5 kilometer intervals from the reactor along with the time. From the chart, on 3/10-12, we can see people actively entering and leaving the 5-kilometer zone, going to and from work, before and during the accident .
On 3/12-14 we can see fewer and fewer people within 5, 10, and 20 kilometers, due to the evacuation order with people fleeing from the reactor, and the population is decreasing. On 3/13-14 it looks like the population is decreasing, but this was due to the cell phone base stations going down, so we don't have accurate data during this time.
However between March 14th and 15the the base stations started working again and it looks like the number of people is increasing. The most problematic time of the reactors were from the early morning of March 15th to March 16th, when the concentration of iodine was at its highest. Fortunately the ZENRIN data for this time is from after the base stations restarted, so it appears to be useable.
Further investigation on ZENRIN's data: this population frequency distribution chart shows how many people were within this 250-meter grid, and we see peaks at 200, 400, 600, and 800.
What we can guess about this data is that here were certain number of people with automatic GPS features on their phones within the grid (like 1, 2, 3 people) , and we can see that ZENRIN is multiplying them by the proportion of contracted cell phone users, and guessing that there are 200, 400, or 600 people there. Then, the intervals are filled in.
Taking a closer look at the data for March 10, 3 p.m., as an example- people aren't moving much, and the peaks are sharp. But at 8 p.m., when people start changing locations the peaks become flatter. So we can imagine that the data has been derived using the fractions of people leaving and entering the grid. Due to this reason, prof. Hayano didn't use the 250-meter grid, instead made the space-time grid coarser.
2) Understanding SPEEDI's data
Here, Professor Hayano evaluated 2 kinds of SPEEDI calculations (in red and blue), and overlaid them with the calculations from the National Institute for Environmental Studies (in yellow) to check the difference. Unfortunately, the data is extremely different. The source terms must have been pretty different. Therefore, they made the grid coarser here as well.
He published the population figures for a 10-kilometer grid openly, and asked many people to help check the calculations. As a result he was able to approximate the magnitude.
The purpose of this project, at the end of the day, is to work on a legislation so that when someone gets thyroid cancer in the future, the government will need to compensate the medical fee without the citizens to prove the causal relationship between the cancer and the nuclear power plant accident. Calculations done during Project3111 will be used for the financial estimations related to that bill.
Q (Professor Murai): What do you think the relationship between these data presented today should be - data published and analyzed by the authorities, and data that is publicly available and can be analyzed by anyone?
A: That is a difficult issue. Ever since March 11th, there were many questions surrounding the way authorities that are supposedly owning data- how they communicate or publish data, how those data are utilized or not utilized. It is important that our project published the data openly in a way anyone can see what was happening and verify.
Project Hayano's documentation can be seen here [ja]:
You can see how cautious Professor Hayano is about the data itself. Getting reliable data is extremely important in analyzing data, because if the data is rubbish all of the analysis will be rubbish as well. Therefore, he takes a lot of efforts verifying the data, making the data available and asking many people to check and verify his calculations.
The other important aspect about this project is that it is goal oriented- Prof. Hayano has the aim to make a legislature to save the victims get compensated if they get cancer in the future, and he is very focused on making that happen. Data, research and analysis is tools for making bigger decisions, not the goal.
He was also very successful in involving other people help his project- many experts provided data outside Project311 itself, due to his credibility. He was also able to get many people to help verify his calculations.
We learned so much from professor Hayano through watching his projects.
He continues working on his projects, and the latest presentation slides can be seen as below, it is a comparison of thyroid radiation intake between Chernobyl and Fukushima.
It looks like Fukushima (red) is extremely lower than Chernobyl (blue) and the distribution differs greatly as well.
BTW, Project Hayano was actually the very first project that launched at Project311. Upon launching Project311, we weren't sure how many people would participate. We didn't know how many valuable projects will start from our data. On 9/19, we did an event called "office hour". We gathered participants and all of the data providers in our office, and made this opportunity for the participants to ask questions to the data providers. At the after party of this office hour, Professor Hayano started talking about what he wants to accomplish from this project- which was so concrete and so compelling. The moment I thought there definitely will be great findings coming out of the workshop was when I talked with Professor Hayano. Thanks again!
Disclaimer: The opinions expressed here are my own, and do not reflect those of my employer. -Fumi Yamazaki
0 件のコメント:
コメントを投稿