Hi Ben,

I haven't used the motion detection feature of the Mobius but it should work well if you can place the camera where a blast will cause enough motion in the scene. This video illustrates how it works: https://www.youtube.com/watch?v=A0cG5BuS2v0.

To demonstrate that the blast alters air quality, the light might have to be just right, so that will be an issue to work on. Ideally you would have photos or video right before the blast happened and then similar video after the blast. That will require taking a photo or short video every 10 minutes or so. A second Mobius could do that. The Mobius motion detect feature starts a video when motion is detected and then continues for a predetermined time after motion stops, but only up to one minute. It might be good to have a longer recording time than that. The camera placement has to exclude all the other things that might move so that the camera does not record a lot of irrelevant video.

I don't know how long the Mobius battery lasts when in motion detection mode. If you will need to power the Mobius via USB, then you could also power a Raspberry Pi and Pi camera. That will probably allow more complete control of interval recording and also motion detection recording because you can program it to do whatever you want. It would not be very difficult to program the Pi to start taking photos or video when a signal was received from a microphone. A blast probably produces a much more distinct sound signal than motion signal so it might be easier to eliminate false positives by using sound.

Do you think the dew problem could be solved if you put the camera under something? In other words protect the camera from the sky to block radiative cooling. Or in some way keep the camera warmer. The protective case might not make any difference because the dew will just form on that.

Chris

Hi @bkleist Thanks for posting on this! I know of a few other people who are interested in using cameras to watch for blasting events. So I'm really excited to watch this play out! I did see this research note about using a light fixture for housing, but I'm not sure how well this field tested.

Also, there is an interesting thread about visual emissions reporting and how there is a method for using cameras for this I know @gretchengehrke has been looking into.

If something with the mobius doesn't work out, game cameras are coming down in price and might work!

Hi @bkleist,

Interesting stuff! What is your electricity situation and need for being inconspicuous like where you have the camera? I wonder if a small space heater could keep the camera area above the dew point. That would require a lot of energy and would probably be fairly noticeable, so it's not likely to be an ideal solution. Maybe trying to house the camera in a hydrophilic housing material could attract the dew?

For light conditions and frequency of photos that could be useful, see EPA's Method 9 for visual emissions monitoring. There is a digital camera opacity technique (see the link to the discussion Stevie put in her comment) where you could use a certified camera and photo analysis system, and you yourself become certified, but it might be worthwhile to try to achieve the conditions specified in the method without being certified (which can become expensive). Generally, the sun should be behind the camera (which is tough with a stationary camera over the course of a day...), and photos taken every 15 seconds for six minutes when fugitive emissions occur. Do you know the highest frequency setting the camera has?

Please join the conversation on the air quality google group! (see: https://publiclab.org/lists).

Hi! A few of us tested out a <$100 trail cam which worked really well -- it has a lot of the features you might need:

https://publiclab.org/notes/cfastie/09-20-2016/zero-dark-thirsty-capturing-replenishing-nighttime-rain-events-with-a-trail-camera

We've tried to list out some of the considerations in choosing a camera here: https://publiclab.org/questions/stevie/09-06-2016/question-has-anyone-used-a-timelapse-camera-in-stream-monitoring

We're also trying to come up with a list of specs we want any camera to meet when doing this kind of work -- the fogging issue is a big one we didn't think of, but I guess the timelapse camera we found is designed for that, as it ran all night in a rain storm without trouble. Someone had originally posted these requirements:

• should be waterproof, able to survive inclimate weather.
• should be able to record photos at 10 minute intervals for minimum of 30 days without recharge.
• should be able to have data removed and battery replaced/recharged in the field by a non-technical user.
• should be able to be mounted to a variety of bases, in urban and natural environments.
• should be of a minimum of 8MP resolution.
• should have total BOM cost of less than $100.
• should provide accurate time stamp or ability to determine time/date for every photo

Can we send this to you for testing? Would you say that your setup has similar requirements, or different ones? Thanks for sharing!

Hi, @bkleist- One thought regarding the moisture obscured lens-if the camera is in an enclosure, a small fan to circulate air should help with condensation. There are little battery powered fans, though I don't know about battery life, especially when it's cold outside. A computer fan may be more efficient and reliable than something from a box store; one from a laptop that's not working could be free, plus easy to mount.