I am so intrigued by the whole goal of Safecast. It is so empowering to put precision tools in the hands of “citizen scientists” (good word). As I peel the layers back and learn more about Safecast, the more I want to get involved.
I work in the nuclear field, and deal with fission products on a daily basis. Comparing our work practices and tribal knowledge to the common perception that ‘radiation is bad’ ultimately frustrates me, because I have experienced first-hand just how safe reactors are, and that radiation is present in every location on earth. Sievert units are a measure of effective biological dose, and are useful in measuring a radiation dose to human tissue. I immediately honed in on your setup because, to me, it implies that at that specific location, a human being would be absorbing ~0.1uSv/hr of ionizing radiation. However, alphas can be blocked by human skin, and betas can usually be blocked by clothing, so they are not externally damaging to humans. Yet, because they are being measured by a non-discriminatory GM tube, they are treated as gamma radiation and bump the dose rate way up.
I agree with you, though, that simply reporting CPM is hard to interpret. I am just fearful that someone who is uninformed may misinterpret the data and assume that being in Seattle is harmful, or they are exposed to “dangerous” radiation levels.
Hi,
Could you please explain what you are measuring in your picture? I am a little confused as to why this counter is so high off the ground. Wouldn’t background radiation levels be taken at ground level? Thanks
Great question. When safecast has the opportunity to do a “survey” style of measurement, indeed we try to uniformly measure at a height of 1m, facing downward. If given the opportunity we like to do it without the shield so that we can pick up alpha/beta and not just gamma.
But sometimes safecast also has the opportunity to do a “relative change” style of measurement. In those cases we focus on secure weatherproof mounting, and simply ensure that the device isn’t pointing into space. In these cases we’re looking for differences in reading as a function of time, as opposed to absolute values.
In that case, I guess I am even more confused. Inherently, a GM tube cannot discriminate between radiation types. When you take the shielding off, your raw output from your counter is a CPM of alpha, beta, and gamma ionizing events. From what I have researched, Safecast Geiger counters are calibrated for the gamma decay of Barium-137m, and effective doses are calculated and reported as Sieverts based on this decay energy. Because we live in a radioactive world, a Geiger counter facing out a window by a wall of concrete shielding in a room many stories up would mainly detect radiation from naturally occurring Radon (beta), cosmic ray interaction, and other naturally occurring radioactive materials. To me, your observation room seems like a steady-state environment that would never really change. Reporting your data in microSieverts is really misleading because these counters are calibrated for the gamma decay of fission products.
The unit displayed in the photo is completely outdoors. This probe is attached to the rear of the unit and it extends outward/downward beyond the rail that you see in the photo. You can find technical characteristics of the probe on the spec page: https://medcom.com/product/hawk-radius/
I don’t mean to sound aggressive, and I apologize for that if I do. I am just trying to understand what this data means. The Hawk Radius probe uses a LND pancake GM tube very similar to Safecast counters. Are you just monitoring to see if CPM rises above background? And why are you reporting in Sieverts if you are counting alpha, beta, and gamma events?
ray ozzie
No need to apologize; you know what you’re talking about. The primary “business-level goal” is to empower citizens to reliably gather and openly disseminate sensor data, as inexpensively and yet as accurately as possible, so that experts can make their own interpretations.
The nature of citizen science is inherently imperfect; there are many potential imperfections as compared with professional scientific measurement. However, even with this imperfect data, and with imperfect souls gathering that data, surprisingly compelling results can be achieved. Certainly in most cases better than no data at all.
Realistically, for our fixed sensors the short term goal is that if one (or ideally multiple) of the sensors detects changes above what the normal background level would be, that citizens and pros alike would immediately go in and look at the situation in more detail.
Indeed for any given data point gathered, the most important things here are for us
a) to know precisely the technical specs of the monitoring probes, what shield was used, and other aspects of how the data is specifically being gathered by the device
b) to track, with reasonable efforts, the provenance of each piece of data submitted, enabling someone to query the data and filter down only to sources that they trust in their own judgment.
Indeed, as you point out, the best and most relevant value to record and report is counts-per-unit-of-time. Your points about Sieverts are well taken. We tend to convert to Sieverts when we think that consumers may be looking at the data, simply because it’s easier for them to interpret than CPS/CPM. The goal is not to confuse people, and to the extent that Sieverts confuses people it is indeed arguable that it should be ommitted.
I hope this adds a bit of clarity.
Kent Noonan
Hi Billy and Ray, I’m coming late to the party, but I can help a little. I work with Medcom and Safecast. A couple of reasons for using Sieverts: It normalizes readings for tubes with different response. There are many GM tubes in use that have about 1/3 of the response of the more popular pancake tube. So this conversion yields a comparable measurement regardless what the tube is. Many users don’t understand what all this means and are only looking for good/bad criteria. Another reason is that during the March ’11 crisis, there was a major shift in how radiation detectors were used, designed, and talked about in media, and that resulted in most information being available in microseiverts. So to avoid confusing things more, it was an opportunity to present users with a more uniform data set, rather than sometimes using mR/hr, or uSv. Since some users do understand, and the CPM data really is different, we present both whenever possible. Many of the Hawk Radius have both a shielded energy compensated gamma-only detector, and a beta/gamma type (with optional alpha) so discrimination is possible if you compare the two readings taken simultaneously. In that case, the uSv reading is truly gamma uSv, not just converted CPM. Also as Ray indicated, relative measurements are usually the best use, since siting issues, maintenance, and other factors can make the absolute measurement skew from normal. Noticing a sudden increase from a long term flat line is pretty easy to interpret.
I strongly agree with Ray, that not confusing people in a very confusing subject is critical to the mission. They need simple answers.
I hope this helps.
20 Responses to “USA, WA, Seattle”
Billy Burnquist
I am so intrigued by the whole goal of Safecast. It is so empowering to put precision tools in the hands of “citizen scientists” (good word). As I peel the layers back and learn more about Safecast, the more I want to get involved.
I work in the nuclear field, and deal with fission products on a daily basis. Comparing our work practices and tribal knowledge to the common perception that ‘radiation is bad’ ultimately frustrates me, because I have experienced first-hand just how safe reactors are, and that radiation is present in every location on earth. Sievert units are a measure of effective biological dose, and are useful in measuring a radiation dose to human tissue. I immediately honed in on your setup because, to me, it implies that at that specific location, a human being would be absorbing ~0.1uSv/hr of ionizing radiation. However, alphas can be blocked by human skin, and betas can usually be blocked by clothing, so they are not externally damaging to humans. Yet, because they are being measured by a non-discriminatory GM tube, they are treated as gamma radiation and bump the dose rate way up.
I agree with you, though, that simply reporting CPM is hard to interpret. I am just fearful that someone who is uninformed may misinterpret the data and assume that being in Seattle is harmful, or they are exposed to “dangerous” radiation levels.
Billy Burnquist
Hi,
Could you please explain what you are measuring in your picture? I am a little confused as to why this counter is so high off the ground. Wouldn’t background radiation levels be taken at ground level? Thanks
Ray Ozzie
Great question. When safecast has the opportunity to do a “survey” style of measurement, indeed we try to uniformly measure at a height of 1m, facing downward. If given the opportunity we like to do it without the shield so that we can pick up alpha/beta and not just gamma.
But sometimes safecast also has the opportunity to do a “relative change” style of measurement. In those cases we focus on secure weatherproof mounting, and simply ensure that the device isn’t pointing into space. In these cases we’re looking for differences in reading as a function of time, as opposed to absolute values.
Billy Burnquist
In that case, I guess I am even more confused. Inherently, a GM tube cannot discriminate between radiation types. When you take the shielding off, your raw output from your counter is a CPM of alpha, beta, and gamma ionizing events. From what I have researched, Safecast Geiger counters are calibrated for the gamma decay of Barium-137m, and effective doses are calculated and reported as Sieverts based on this decay energy. Because we live in a radioactive world, a Geiger counter facing out a window by a wall of concrete shielding in a room many stories up would mainly detect radiation from naturally occurring Radon (beta), cosmic ray interaction, and other naturally occurring radioactive materials. To me, your observation room seems like a steady-state environment that would never really change. Reporting your data in microSieverts is really misleading because these counters are calibrated for the gamma decay of fission products.
ray ozzie
The unit displayed in the photo is completely outdoors. This probe is attached to the rear of the unit and it extends outward/downward beyond the rail that you see in the photo. You can find technical characteristics of the probe on the spec page: https://medcom.com/product/hawk-radius/
Billy Burnquist
I don’t mean to sound aggressive, and I apologize for that if I do. I am just trying to understand what this data means. The Hawk Radius probe uses a LND pancake GM tube very similar to Safecast counters. Are you just monitoring to see if CPM rises above background? And why are you reporting in Sieverts if you are counting alpha, beta, and gamma events?
ray ozzie
No need to apologize; you know what you’re talking about. The primary “business-level goal” is to empower citizens to reliably gather and openly disseminate sensor data, as inexpensively and yet as accurately as possible, so that experts can make their own interpretations.
The nature of citizen science is inherently imperfect; there are many potential imperfections as compared with professional scientific measurement. However, even with this imperfect data, and with imperfect souls gathering that data, surprisingly compelling results can be achieved. Certainly in most cases better than no data at all.
Realistically, for our fixed sensors the short term goal is that if one (or ideally multiple) of the sensors detects changes above what the normal background level would be, that citizens and pros alike would immediately go in and look at the situation in more detail.
Indeed for any given data point gathered, the most important things here are for us
a) to know precisely the technical specs of the monitoring probes, what shield was used, and other aspects of how the data is specifically being gathered by the device
b) to track, with reasonable efforts, the provenance of each piece of data submitted, enabling someone to query the data and filter down only to sources that they trust in their own judgment.
Indeed, as you point out, the best and most relevant value to record and report is counts-per-unit-of-time. Your points about Sieverts are well taken. We tend to convert to Sieverts when we think that consumers may be looking at the data, simply because it’s easier for them to interpret than CPS/CPM. The goal is not to confuse people, and to the extent that Sieverts confuses people it is indeed arguable that it should be ommitted.
I hope this adds a bit of clarity.
Kent Noonan
Hi Billy and Ray, I’m coming late to the party, but I can help a little. I work with Medcom and Safecast. A couple of reasons for using Sieverts: It normalizes readings for tubes with different response. There are many GM tubes in use that have about 1/3 of the response of the more popular pancake tube. So this conversion yields a comparable measurement regardless what the tube is. Many users don’t understand what all this means and are only looking for good/bad criteria. Another reason is that during the March ’11 crisis, there was a major shift in how radiation detectors were used, designed, and talked about in media, and that resulted in most information being available in microseiverts. So to avoid confusing things more, it was an opportunity to present users with a more uniform data set, rather than sometimes using mR/hr, or uSv. Since some users do understand, and the CPM data really is different, we present both whenever possible. Many of the Hawk Radius have both a shielded energy compensated gamma-only detector, and a beta/gamma type (with optional alpha) so discrimination is possible if you compare the two readings taken simultaneously. In that case, the uSv reading is truly gamma uSv, not just converted CPM. Also as Ray indicated, relative measurements are usually the best use, since siting issues, maintenance, and other factors can make the absolute measurement skew from normal. Noticing a sudden increase from a long term flat line is pretty easy to interpret.
I strongly agree with Ray, that not confusing people in a very confusing subject is critical to the mission. They need simple answers.
I hope this helps.
Ray Ozzie
Watching over the city.
robouden
Nice..
robouden
From the other side ot the earth and from the other side of the building.
Ray Ozzie
Deployed and operational.