Fossil fluorescence: UV laser drone fossil hunting

[synapsomorphy]

(totally new to this forum, let me know if there's a more appropriate place to post this!)
Hey all!

I found a video showcasing a fossil hunting method that I'm super fascinated by!

It's a drone that uses the fact that many fossils fluoresce under UV light to fly around and find fossils autonomously.

This seems like an absolutely fantastic way to hunt for fossils over a large area. I have experience with drones and electronics and I'm interested in making a similar drone, hopefully a bit simplified, maybe open-source and user friendly etc. If this method works well to find fossils, getting this tool into the hands of more researchers would be an awesome way to find many more fossils.

So I'm looking more into just how effective this method is; particularly how often it is that a fossil is fluorescent, especially vertebrate fossils. I did find this paper that gives some good info but it is largely focused on smaller vertebrates and only from a single formation.

Basically any information you have about fossil fluorescence under UV light I'd love to hear! I'm especially interested in figuring out how broad of a phenomenon this is. Do most vertebrate fossils from most formations fluoresce? Is it only smaller fossils or only some formations? The better this method is for finding fossils, the more effort I want to put into making a simple, reliable drone employing the method.

I'm also interested in automating the process with machine learning down the line. I read that fossils have a typical fluorescence wavelength that is different from most other things you might find out there, which gives you a clear signal to look for. The eventual result could be: you let a drone go fly around autonomously for a couple hours, recharging it as needed, and then it automatically gives you a list of GPS coordinates to go check out and the images associated with them.

(Also, if anyone wants to collaborate on this project, please let me know, whether your skills are more on the electronics/software or paleontology side!)
Disclaimer: I'm very aware that drone flight is not allowed in many typical fossil bed areas, definitely won't ever be flying anything without approval.

Edited February 10 by synapsomorphy
fix links

[jpc]

Tom in the videos is a good friend of mine.  It is very much in the state of development.  He has done a lot of really cool work with his lasers... they has to be green lasers... especially with fossils found on flat plates (think Green River Fm, Solnhofen and the Chinese Cretaceous deposits) where he and his colleagues are able to see sotf tissues that are not seen by the human eye.    

 

Your are referring to two different technologies here: green lasers and UV light.  UV is easier to do because you can buy UV lights fairly easily.  UV comes in a, b and c and I don't know much about them except that they are different wavelengths and that UV is actually quite dangerous and is NOT what you buy at Wally World to look for cat pee.  

 

As for how fluorescent fossils are... from my personal experience,, most fossils are not very fluorescent.   A major exception:  White River Fm teeth glow very well, a dull orange color.  Tom and I went out to the ranch he has access to one night 20 years go to look for UV sensitive fossils.  We collected everything that glowed... and there were a lot, but only a very few were fossils.  We had collected a lot of calcite which also glows and is common in the White River.

 

"Do most vertebrate fossils from most formations fluoresce? "

From my experience, my answer is quite the opposite.  As at least if you are looking for stuff that the human eye can detect. see below, though.    

 

"I'm also interested in automating the process with machine learning down the line. I read that fossils have a typical fluorescence wavelength that is different from most other things you might find out there, which gives you a clear signal to look for. "

Years ago Tom also invented a microfossil sorting machine.  I think he is perfecting it now with the folks at the Burke Museum.  It basically shone a green laser at a bunch of small fossils and shot a puff of air to separate out anything that registered different than the background levels.  The catch is you have to try this with a known quantity for each formation and maybe even for each quarry as chemistry is very variable.  I gave him some samples to check and it only worked on some.

 

"The eventual result could be: you let a drone go fly around autonomously for a couple hours, recharging it as needed, and then it automatically gives you a list of GPS coordinates to go check out and the images associated with them."

 

This is exactly what Tom is doing.

PM me if you want to me to see if Tom would like to discuss this with you.  I think he is happy to share ideas. He may also tell you that only half of what I said here is the actual tale.  

[patelinho7]

This is fascinating! I've often wondered if something like this is possible. I've been studying machine learning recently, so I've been thinking of applications to fossil hunting/sorting. I'm an electrical engineering undergrad so I would definitely be interested in seeing what you come up with. I reckon there is definitely something we could find to use as a "class" to separate fossils from everything else. And with a bit of machine learning and stored GPS coordinates, we could identify possible targets. I would think there would be a way to do this on a smaller scale for fun. Most people would bring up the inefficiency of a method like this, but not only would it be interesting to explore, it allows additional time hunting without expending energy and getting dehydrated. Keep me posted!

[deltav2]

2 hours ago, synapsomorphy said:

(totally new to this forum, let me know if there's a more appropriate place to post this!)
Hey all!

I found a video showcasing a fossil hunting method that I'm super fascinated by!

It's a drone that uses the fact that many fossils fluoresce under UV light to fly around and find fossils autonomously.

This seems like an absolutely fantastic way to hunt for fossils over a large area. I have experience with drones and electronics and I'm interested in making a similar drone, hopefully a bit simplified, maybe open-source and user friendly etc. If this method works well to find fossils, getting this tool into the hands of more researchers would be an awesome way to find many more fossils.

So I'm looking more into just how effective this method is; particularly how often it is that a fossil is fluorescent, especially vertebrate fossils. I did find this paper that gives some good info but it is largely focused on smaller vertebrates and only from a single formation.

Basically any information you have about fossil fluorescence under UV light I'd love to hear! I'm especially interested in figuring out how broad of a phenomenon this is. Do most vertebrate fossils from most formations fluoresce? Is it only smaller fossils or only some formations? The better this method is for finding fossils, the more effort I want to put into making a simple, reliable drone employing the method.

I'm also interested in automating the process with machine learning down the line. I read that fossils have a typical fluorescence wavelength that is different from most other things you might find out there, which gives you a clear signal to look for. The eventual result could be: you let a drone go fly around autonomously for a couple hours, recharging it as needed, and then it automatically gives you a list of GPS coordinates to go check out and the images associated with them.

(Also, if anyone wants to collaborate on this project, please let me know, whether your skills are more on the electronics/software or paleontology side!)
Disclaimer: I'm very aware that drone flight is not allowed in many typical fossil bed areas, definitely won't ever be flying anything without approval.

Here in Qatar, I have looked for vertebrate remains from 2 members in the Dammam formation (middle eocene), the midra shale member and dukhan alveolina limestone member, and all vertebrate fossils fluoresesce very well (even coprolites, if that can be called a vertebrate fossil). Lately in my past 2 or 3 trips I have been using a UV torch and hunting at night, and I have found much more fossils than hunting in daylight, and I also use it to look for micro fossils. The fossils here which have been bleached by the sun fluorescesce lesser, or it has a different color entirely than the usual yellow-green color from the ones protected by the formation. There are also some other inorganic stuff which fluorescesce like ferricete, which is dull and orangish, and also some other stones(usually around 5mm in length or less) which shine exactly like how the vertebrate remains do, but its easy to differentiate them because they're smooth and pebble shaped . I could see how automating it with a drone could be useful, especially for finding rare associated remains, or for surveying hard to reach areas for fossils. For finding the more common stuff like shark teeth for example, I'd say that using a handheld UV torch is much better, since they are easy to find.

My first time UV hunting: 

 

Edited February 10 by deltav2

[oilshale]

I used to climb around the waste heaps in Solnhofen with a UV lamp from time to time. You can find small fossils that way, larger and more distinct fossils have already been discovered by the quarry workers.

Solnhofen would be a suitable area, but the UV lamp would have to be very strong.

[synapsomorphy]

@jpc Wow, small world! Very good insights. I'd love to chat with Tom if he's up for it! I'll pm you.

@patelinho7 I'm in my last semester of a mechE degree and also getting into machine learning. Pm me if you wanna collaborate on this or similar project! I'm also starting a project right now to make a robotic walking (and maybe eventually flying) pterosaur using reinforcement learning.

@deltav2 Very cool! Nice that you got a bunch of super fluorescent stuff over there! It would be really cool if there was some survey of a bunch of fossils and their fluorescence to find which formations are the most fluorescent, if there's a size dependence, maybe some clade's bones fluoresce very well for some reason, etc. I think drones would be a huge advantage for finding rare stuff because they can cover a ton of ground. If you're looking for large fossils in a large area the drone could probably fly 40+ mph and still get decent images. You could cover a lot of ground like that, which seems to me to be the main challenge for formations that span hundreds of miles of barren land. If you're going full mad scientist, you could set up a recharging area that the drone can return to and just let it do its thing for days at a time (if your programming and autonomous flight is good enough of course). 30 mins of flight time flying at 40 mph (feasible for modern drones) gives you a 10 mile search radius centered around the charging pad, which would already take a long time to thoroughly search, and any speed/flight time improvements from there increase the search area with the square of the radius.

 

Edited February 10 by synapsomorphy

[Ammonite_hunter]

On 2/10/2024 at 6:00 PM, synapsomorphy said:

@patelinho7 I'm in my last semester of a mechE degree and also getting into machine learning. Pm me if you wanna collaborate on this or similar project! I'm also starting a project right now to make a robotic walking (and maybe eventually flying) pterosaur using reinforcement learning.

Hi there! I'm a software developer who is fascinated by this field. I have some experience using Google's Vertex ai and with Tensor flow image classification. I would love to be involved in something like this. Give me a shout if y'all are looking for collaborators

[Jaybot]

Where we are at, fossil teeth do not glow, whereas a 'cow that died just a month ago's' teeth will glow brightly in UV.

I'm sure it really depends where you are at.

[Bronzviking]

The drone sounds interesting, thanks for sharing. I am a rock and mineral collector and into fossils as well. I've just found out about fluorescent minerals a few years ago at a rock show and found it fascinating. So I bought a UV flashlight and was amazed how many of my minerals glowed. The fluorite glows purple, calcite glows pink, Ruby zoisite glows deep red, yooperlites glow orange to name a few. I find agatized coral here which is a fossil and it typically glows yellow. If it has banding like an agate it glows green and yellow. I just found one that glows a deep purple. I also just came a across a video about glowing fossilized seashells a guy stumbled upon when looking for calcite crystals. See Link below.

https://www.sciencetimes.com/articles/41291/20221209/240-million-year-old-seashells-glows-under-uv-light-revealing.htm

 

[patelinho7]

On 2/12/2024 at 2:17 PM, Ammonite_hunter said:

Hi there! I'm a software developer who is fascinated by this field. I have some experience using Google's Vertex ai and with Tensor flow image classification. I would love to be involved in something like this. Give me a shout if y'all are looking for collaborators

Revisiting this thread. Recently been considering small-scale ideas along this line of thinking that would achieve smaller issues I’ve come across recently in paleo. I could probably figure out a lot of the hardware stuff and messing around with the drone itself, but if we were to go down the classification route to identify fossils from the footage, we would need ML experience which is out of my pay grade. Maybe someone like you would be perfect for that!