- cross-posted to:
- winnipeg
- cross-posted to:
- winnipeg
With apologies for the colour scheme (I’m so sorry if you’re colour blind!). This is the lower layer of a two layer 1D inversion model, interpolated. Was playing with maximizing the contrast to try to find the foundation of the Richardson Mansion that existed on this property prior to it being bulldozed in the 70s and donated to the City of Winnipeg to become Munson Park. I’m pretty sure I captured it in the red square in the middle.
Short form: it is a soil conductivity meter - it measures variations in electrical conductivity.
Long form, and assuming you have a little physics background. There are two coils in the device, one transmitter and one receiver. You pass an oscillating electrical signal through the first coil and it creates a magnetic field that is oscillating all around the device. This oscillating magnetic field causes electrons in the ground to oscillate back and forth, but how much they oscillate will depend on how electrically conductive the material is. These oscillating electrons in the ground will themselves create a (much smaller) secondary magnetic field, which can be detected on the receiver coil on the instrument. So you basically walk with this thing and make a map of electrical conductivity variations in the subsurface.
System costs about $40k. Very useful for finding archeological things, buried debris, mapping contaminated water (salty water is more electrically conductive, for example), for gravel exploration, fertilizer distribution mapping in agriculture, soil moisture content, and sometimes for utility detection (if the utility is conductive).
Thanks, I hopped down a rabbit hole while you were answering and found this: https://www.ars.usda.gov/ARSUserFiles/50701000/cswq-0167-148589.pdf
The reason I ask is that my wife and I are scientists (I’m remote sensing, she’s a biogeochemist), and we’ve collaborated on a number of projects where we have to do very involved field sampling, digging or drilling of cores, etc… and the time it takes to get that data just massively limits what we can do or say.
So I’m interested in all things remote sensing when it comes to sub-surface. We’re interested in anything that can speed up/ scale/ increase the density of the data we work with.
Yeah, that’s where geophysics comes into play. In the end you still need to sample, but you have to sample a lot less. With the geophysical maps, you can interpolate strongly between samples with a high degree of confidence. But the toys aren’t cheap.
In the medical context, it’s like medical imaging. An image with a blob on it can be anything, until you take a biopsy. But the medical imagine sure cuts down the total number of samples you need to take :)
With the EM38, a lot of people mount them in a toboggan behind an ATV or something, so you can record at low driving speeds across terrain that supports it. Or you can walk it with the device in hand for rougher terrain. It’s a poor method to mount on a drone, since the data is best when it’s collected within about 30 cm of the surface. At 1.5m from the surface, you get null results. (In fact, you hold it up in the air to calibrate your zero prior to use.)
Oh I know on the price, I’ve spent plenty of time I. the field with a half million dollars in gps and hyperspectral equipment on my back. My background was originally in sonar but underwater, didn’t get into aerial and spaceborne remote sensing till my second career.
Ideally I’d like to be able to measure soil moisture, soils density, CO2 and vocs at the same time. I’ve been looking into some of the recent advances in millimeter wave stuff, sonar, infrared, and also infernometrics using WiFi radio antenna. We’ve got a couple grant proposals in to fund some proof of concept stuff but nothing funded so far. Even if it was a fixed position imaging device, anything that could capture spatial variance over time would be a major improvement.
Unfortunately science equipment is really not a big money maker, so very little private investment.
It seems like the EM38 could be used to characterize a site over time with multiple passes. Looks since it’s based on electrical conductivity I would expect moisture content to drive variation.
Honestly if we could just get soil depth that would be huge.
If the soil is sandy, GPR is probably your golden ticket. But it only works in electrically resistive regimes.
We’ve got a new toy coming in a few months – the EM31-8. In theory it’ll give us a real time four layer soil model to a depth of six metres as we walk. I haven’t had it in my hands, so can’t speak with confidence about the model it’ll produce (yet). Commercial unit #1. Living on the edge, expecting to bleed a little. :)
Whats your relationship with the company?
I co-own a small business. Two people, no employees. We do geophysical equipment, in the general sense. Sales, rentals, service, training, customization, long term repair work for discontinued products, and some R&D when we have time. It’s super niche stuff, and we probably have less than a dozen competitors globally, all of whom are very small (like us).
With have no official relationship with Geonics (the above equipment manufacturer) and do not formally represent them in any way. We are a customer.
However, we have about a dozen of their instruments in our pool. This instrument is new, both to us, and as a product from the manufacturer.
I’m donating time on the equipment later this month to a Planetary Society research project – but I’ve never used it before. When we’re supporting the equipment in our pool, I need to be able to provide tech support and such, and to do that, I need experience surveying with them. This was one such survey. No client, just me testing a new toy. I borrowed my SO and had her film my setup and teardown, and I’ll make a tutorial video out of that when I have time to splice footage. We spent our evening together in the park on New Year’s Day. Fortunately she is also a geophysicist and can tolerate that sort of date night ;)
There isn’t much going on in c/geophysics so I’ve been sort of posting random work stuff. Like the above test data, which was a quick and dirty QGIS plot.
Well we’re not quite there yet, but I’d like to keep in contact with you regarding renting. My wife and I both have university association as well as private and non-proft relationships where we sit as PI or co-PI.
If nothing else I’d love to get into a call with you can chat about the state of the industry/ technology. Like I said, our work is primarily biogeochemical in nature, but we’ve got an ongoing conversation in our house about how we’re fundamentally limited by the tool-kits and culture of the domain. Its roots are largely in agriculture/ forestry and we are primarily interested in massively scaling the number of samples we can take, reducing the time it takes to get samples, increasing the various resolutions of sampling, and non-destructive sampling. PM me directly and I’ll shoot you my contact information.