150,000,000 Data Points

Thu, 09/24/2020 - 00:42

Last month we passed a pretty significant milestone, surpassing 150 Million data points in our public domain dataset. It was 2 years after launching when we hit 10 Million and 3 more years until we reached 50 million. To triple that in 4 years, half a year shy of our 10th anniversary is very exciting.

Over the last 9 years our radiation monitoring efforts have proven to be an excellent proof of concept of the value of communities sharing open data. We’ve inspired the creation of other like minded projects and encouraged existing initiatives to open up their data. We’ve provided materials for new research and better understanding of both our environment and how people can work together towards a shared goal. We’ve shown that there’s a need for open data like this, and that locking down devices or bricking them doesn’t need to be the norm. We’ve demonstrated that the work of a relatively small group of people can have huge implications, creating long lasting benefits for the entire world. If we’ve done all this correctly, we’ve tried to show how others can do the same.

However you can’t embark on something new and untested while traveling a traditional path, and because of our approach we’ve never fit into traditional funding models. Most funders and grantors don’t understand us, or what we’re trying to do. That’s fine because we were lucky early on to find some who did and we’ve been able to spend the vast majority of our time working on our mission, not worrying about how we’re going to fund it. Earlier this year I talked about how that created some unfair dependencies and how we’d like to move to a more community funded model. A number of people graciously contributed which was wonderful to see, though unfortunately within days of making that request COVID-19 brought much of the world to a standstill. Industry-wide we’ve seen foundations and grantors tighten their belts and postpone check writing until 2021 at the earliest. Those donations we received in February helped more than we could have imagined.

As we make our way towards the end of the year I have a much more humble ask. Safecast is an example of the few helping the many, but now we need the help of many as well. Small dollar recurring donations are a new thing for us, but the idea is that they provide a wonderful safety net allowing us to know that each month we’ll have money coming in to cover some basic costs. Our server and hosting bills are about $1500 a month, thanks to a number of people who make recurring donations we currently receive just over $300 each month. At this time we want to ask our community to step in as funders. If you can, will you consider making a recurring monthly donation of $25? If 50 people reading this can commit to this then we’ll have enough each month to cover our hosting bills to keep our sites and data online. If you’d be willing to join with us, please visit our donation page and choose the monthly option.

Safecast Diaries: from Rest Stops to Real-Time Images, from the First bGeigie Trips to the Aftermath of 3.11

Thu, 07/30/2020 - 04:01
David Kell in 2011 ahead of the first-ever Safecast drive, equipped with Geiger counter and smartphone.

David Kell was among a rag-tag group of volunteers taking supplies to the Tohoku region in the weeks after the 2011 3.11 Tohoku disaster. Onboard were the very first incarnations of what would, over time, become the bGeigie geiger counters.

On March 11, 2020, David Kell found himself, quite by accident, staring out over a very familiar Tokyo car park. It was the exact same space that he and colleagues had run to when the Tohoku Earthquake struck Japan in the early afternoon hours of March 11, 2011.

At the time, he was giving a presentation in an office on one of the top floors of an aging, nine-floor office building.

“I distinctly recall the deep rumble of the ground that we heard before the quake struck. I’d never heard anything like it – nor have I since. The building didn’t have quake dampers like there are in modern Tokyo buildings, so it shook all over the place. We didn’t really think, but just looked at each other, grabbed the laptops and ran for the stairs,” David Kell remembers.

Very few people outside Japan are likely aware of how massive aftershocks continued to rip through the country for weeks afterward. Even in the first hours after the main earthquake, they struck, leaving people with a feeling of moving through one gigantic, hours-long quake with ebbs and flows. Like being caught up in ocean swells.

The subsequent tsunami and  Fukushima Daiichi meltdowns left the region with two things in short supply: daily necessities and trustworthy information.

“In many instances, Japan has a strong, admirable track record of disclosing detailed, believable data, but that wasn’t the case here. Many people, both in Tokyo where I live and in Tohoku, which I visited extensively throughout the first months after the quake and Daiichi meltdown, were as affected by the lack of information as the daily goods shortages. That situation generated its own aftershocks, some of which we feel to this day, for example when it comes to people’s trust in information relating to the coronavirus outbreak,” David Kell says.

One of the most esteemed British institutions is the pub where many great ideas have arisen, (although often they have also been forgotten within the hour.) Thankfully, this wasn’t the case for a project started by David, Rob Keyworth, Andy Abbey, along with other patrons of the now-defunct Black Lion pub in Meguro. The group arranged fundraisers for the Tohoku area. Around 2.3 million yen poured in.

The group of rag-tag volunteers decided to secure supplies and fresh produce which were in short supply and drive it north to the places in need, focusing on the coastal towns of Iwate, the route to which passed nearby the Fukushima power plants. So began six weeks of getting off from work early on Friday afternoon, renting trucks, and filling them with supplies. Then, around 5 pm, begin the slog up Honshu towards Tohoku across the earthquake-damaged highways. Often, the group would arrive around 3 or 4 am, catnapping in the truck cabins before staring deliveries throughout Saturday. The night was spent at friends’ houses before driving back to Tokyo on Sunday to get ready for another work week.

“Some of those early trips are full of images that have seared themselves to my mind. A 300-foot fishing trawler that had been picked up by the tsunami, like it were a matchbox car, and put down inland. Soldiers moving through debris that was like puzzle pieces from the lives of families, neighbours, school friends, town groups, all mashed into one by the tsunami. I had a camera with me, and I took a lot of pictures during the early days. About six months later, I looked through the images and decided to delete them. I had never shown them to anyone. Perhaps they were somehow too private. Like someone else’s private life and grief laid bare,” David says.

Other memories are a testament to the incredible resilience of the people in the area.

“We would drive through areas where whole cities had literally been wiped off the map. Where there was nothing left. Then we would come around a bend and there would be a house, precariously toppled to the side, with washing hanging outside. People went through 3.11 and got back up saying, ‘Life goes on. We won’t let this defeat us.’ I found that, and the warmth we encountered, incredibly inspiring.”

Ahead of one of the early trips, Dave was contacted by Pieter Franken from what would, over time, become Safecast. The two met and talked about various relief efforts, the lack of trustworthy data, and the possibility of combining relief efforts and the collection of radiation readings. The two quickly reached an agreement, and the first Safecast radiation measurement equipment went to Fukushima.

The first trip involved Dave and other volunteers stopping at petrol stations and rest stops, noting the location, finding a good spot, and scooping up a soil sample for analysis. Just a few trips later, indicating the speed at which iterations of the devices were evolving, the first GPS-enabled Safecast device was put aboard, This version would log radiation data as well as location, providing initial data for the global Safecast radiation map.

“Those data formed the basis of what grew into Safecast. And I think that I, as well as my co-pilot at the time Robert Keyworth, and Andrew Coed, plus other members of that Black Lion Pub crew, are honoured that we took part in what has grown so impressively since. The early data showed readings that were higher than expected – and in some cases higher than those reported by the Japanese authorities – but not at Chernobyl levels. I think that for us, as well as many others in Japan, those facts and figures were a first sense of security, in that they provided an objective foundation for making decisions on what to do,” David says.

“Here, nine years later, you can look back and see that Safecast was born at an opportune moment when mobile connectivity and cost-effective sensor platforms were available thanks to technological advances. This ability to measure, analyse, and communicate about the data, about the environment a person or group of people live in, is incredibly powerful. Today, everyone has the power to collect data, look through it, and find patterns. Then the next step is to share that information. It is amazing to think that we all have the power to do this today.”

“I also think that one of the biggest things with Safecast is that it is a testament to how technology can help build communities on or around anything from a hyper-local to a truly global scale.



Honor among nuclear neighbors

Sun, 07/05/2020 - 07:49

We’ve received a lot of questions about radioactive isotopes that were detected in the atmosphere in late June by monitoring stations in northern Europe. We’ve been gathering information and touching base with knowledgeable experts to help clarify the situation. We will state right off that the levels of radiation that were detected are extremely small, in the microbecquerel per cubic meter (µBq/m3) range, and pose no risk to human health or to the environment. This was not a nuclear weapons test, or a major accident. Nevertheless, this is the third widely detected but undeclared radiation release in the region in almost as many years, and this should concern all of us.

In late 2017 a plume of Ruthenium 106 was detected over much of Europe. Despite Russian denials which continue to this day, researchers have been able to narrow down the possible origin to the trouble-plagued Mayak complex in the southern Urals. In fact, as we reported at the time, Mayak was a leading suspect from the beginning, but it has taken two and a half years for detailed analytical evidence to be assembled and published. Even then, in the absence of timely access to the site and physical evidence following the release in 2017, or the discovery of a definitive technical “signature” that identifies the site, researchers are ethically constrained from stating in absolute terms that Mayak was the source. In August, 2019, a fatal explosion involving a rocket engine test with an “isotope power source” occurred in Nyonoksa, near Severodvinsk in the Arkhangelsk region. This time, because of the number of witnesses and immediate local media coverage, the Russian government quickly admitted that an accident had occurred, but still refuses to provide accurate information about the radiation releases or risks.

This most recent release unfortunately seems to be following these alarming precedents. Our immediate take aways are: 1) We don’t know yet where it came from, and 2) unless the responsible government declares the release, it will probably take years for researchers to analyze the data and identify the source with any degree of certainty. This analysis began immediately and is still in its early stages.

To summarize what we know, during the week of June 8 –14, 2020, a radiation monitoring station in  Visby, Sweden detected higher than normal levels of Cs-134, Cs-137, Co-60, and Ru-103, all of them associated with nuclear fission (as opposed to natural radiation sources). The release was believed to have originated from somewhere on or near the Baltic Sea. The Swedish Radiation Safety Authority announced the detections in a tweet on June 23, and followed up with more detailed data on June 30 (published in English on July 3).  Finnish and Estonian radiation authorities also announced detections made during a similar time period. The former were in air samples collected in Helsinki between June 16 and 17; the latter between June 14 and 21. On June 27, Lassina Zerbo, Executive Secretary of the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO), which operates the global IMS network of extremely sensitive radiation and seismic monitors, made an official announcement via Twitter about additional detections made at the CTBTO station in Stockholm on June 22-23. Zerbo included an initial map with his tweet showing the possible source region of the release:

Source: CTBTO

(SEP63 is the Swedish CTBTO monitoring station in Stockholm. Please note that the orange-shaded area of this map shows the estimated area within which the source of the release was likely located. It does not show the extent of the release, a radiation plume, or fallout. We think some people are confused by this, and believe a simple dotted outline would have been better.)

The CTBTO uses extremely sensitive equipment, but while it announces issues of public concern like this one, the data itself is usually not made public, for nuclear security and diplomatic relations reasons. Although Sweden has published basic information about the isotopes detected and their levels, other detailed isotope measurement information would be welcome.

As far as publicly available information goes the data is limited. Based on the available confirmed evidence, however, our best guess at this point is that mishandling of used fuel caused the release. This is no more than an educated guess, however. Since the first flurry of notification and press activity a week or so ago, little new data has been made public. But this quiet does not mean that nothing is happening. Labs throughout the region and elsewhere are certainly now examining weather patterns and other factors that need to be considered while evaluating readings at various monitoring stations in order to ascribe them to the same event. Monitoring stations in the region are now undoubtedly closely scrutinizing air samples taken during the period in question to see if faint radiation traces that might be related were also detected. Confirming the location and timing of the detections, combined with understanding of the wind and other meteorological dynamics during those days, will help determine the likely dispersion of the release. This will help narrow down its possible origin and cause.

Cheryl Rofer, a retired chemist from the Los Alamos National Laboratory active in the open-source non-proliferation community, published a brief but informative initial breakdown several days ago. She observes:

“Iodine-131 was observed at more northerly stations and on different days than the other isotopes. It has a half-life of 8 days and is a fission product, as are the other isotopes except for cobalt. Cobalt is an activation product of the steel containment vessel for a reactor. It seems likely that these observations come from a leaking nuclear reactor, but where?”

On June 27, official Russian sources were reported as denying that there had been problems at either of two Russian nuclear power plants located in or near the CTBTO’s estimated release zone:

“Two nuclear power plants in northwestern Russia – the Leningrad NPP and the Kola NPP – operate normally, with radiation levels being within the norm, a spokesperson for Rosenergoatom Concern (part of the Rosatom state nuclear energy corporation) told TASS.”

Rosenergoatom is quoted further as stating, “There have been no complaints about the equipment’s work,” “Aggregated emissions of all specified isotopes in the above-mentioned period did not exceed the reference numbers,” and radiation levels around the two powers stations “have remained unchanged in June.”

This is, of course, almost identical to the repeated denials made by Rosatom and Russian media sources following the 2017 Ru106 release, which were eventually contradicted by the scientific evidence. In a sense, then, through these recent denials official Russian sources have publicly nominated suspects that warrant closer scientific examination.

Rosenergoatom specified that it checked the Kola NPP, near Murmansk which began operating in 1973, and is at the extreme northeastern edge of the CTBTO’s estimated release zone. It has a history of incidents. It also specified the Leningrad Nuclear Power Plant (Leningradskaya AES), which falls well within the estimated release zone, and has a history of minor accidents. It began commercial operation in 1974. One of the five reactor units onsite was shut down for decommissioning in 2018, and fuel unloading is expected to take until 2023. Once again, we must note that there is no publicly available evidence yet which links the release to either of these facilities, just that official Russian sources themselves have called attention to them.

On June 27, the IAEA issued a brief statement saying it is seeking information from member states:

“The International Atomic Energy Agency (IAEA) is aware of information from the Provisional Technical Secretariat of the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) that its International Monitoring System (IMS) detected elevated levels of three radioisotopes Ru-103, Cs-134 and Cs-137 in the air at an IMS monitoring station in Sweden.

“As per standard practice in such cases, the IAEA has contacted its counterparts requesting information on whether these radioisotopes have been detected in their countries, and if any event may have been associated with this atmospheric release.”

Since the IAEA essentially operates on the “honor system,” it can’t do much more than this. As of the time of writing, almost three weeks have passed since the first detections in Sweden, and no nation has come forward to give notification as stipulated by the IAEA Convention on Early Notification of a Nuclear Accident. The IAEA has followed up with brief statements acknowledging receipt of information from over 40 countries, and its preliminary assessment on July 2 that the releases are “likely related to a nuclear reactor that is either operating or undergoing maintenance.”

“Based on the IAEA’s technical analysis of the mix of artificial radionuclides that were reported to it, the release was likely related to a nuclear reactor, either in operation or in maintenance. The IAEA ruled out that the release was related to the improper handling of a radioactive source. It was also unlikely to be linked to a nuclear fuel processing plant, a spent fuel pool or to the use of radiation in industry or medicine.”

As in 2017, while not yet confirmed, most signs seem to point towards Russia. This is the third similar episode since late 2017 in which the country has flaunted its noncompliance with an important nuclear agreement. This, more than the radiation risk itself, is why this episode is of great concern. The honor system only works if it is upheld.