Screech Owl

This morning the chickadees were complaining insistently in the spruce tree outside, and when I pointed my Therm-App at the ruckus, I saw this:

The spot on the left is a complaining chickadee fitting about, the blob on the right was something bigger. Here's another angle.

with the yellow box showing the frame of the Therm-App view. And the identity of our mystery guest?

an Eastern Screech Owl, very well hidden and difficult to find even when I knew where to look.

My first infrared-discovered owl! and a new yard bird, to boot.

Over the weekend I had occasion to test out another owl-spotting scenario with the infrared camera, a snowy owl in a field of snow:

The bird had been reported by others, and was somewhat far away, perhaps a hundred yards from the road:

Unfortunately, it does not show up that well: it is the little spot towards the upper-left.

One big thing working against me that day was that it had been sunny all day, and as I've come to learn, infrared from a sunny day can overwhelm even in a cold winter day (temps were under 20F), so I expect the surrounding trees to be cooler when overcast. Anyhow, I think it could still be an effective detector if the bird were closer, though one might then expect it to be noticed by the keen eye.

Interestingly, while I watched, this snowmobiler drove by:

passing probably within 30 yards of the bird, who did not flinch.

Before the Squall

Yesterday afternoon, with the temperatures still in the "balmy 20s" before an approaching snow squall was to bring in a sub-zero cold wave, I decided to take a stroll to Potter's Falls and the Upper Dam of Six-Mile Creek, snowshoeing through about a foot and a half of snow.

Fire and Ice

Along the way were some nice icicle formations, which was interesting when viewed with the Therm-App:

Here's another block of ice.

From afar, the ice walls look uniform, but the thermal image shows some parts warmer than others. Getting close to some of these ice walls I could see that the "warm" regions had water trickling through actively.

Before the Squall

When I reached the upper dam, my new "hotspot detection" tweak alerted me to this hotspot:

Yellow box shows Therm-App field of view shown on the right. Photo on the left reflects approximately how dark it was getting at the time.

This turned out to be an American Goldfinch, sitting motionless in a relatively low branch.

Continuing up through a pine grove, I saw this:

It was getting quite dark by now (darker than suggested by the photo), and a couple of Great Horned Owls started counter-hooting nearby. I was excited to find out what this hotspot was, but at this very moment the snow squall hit, and combined with the growing darkness I just could not get my binoculars onto the hotspot. So close!

The image above also shows a big challenge with IR birding: can you find the area of the photograph that corresponds with the IR image? It is just as difficult to do so in real time, combined with trying to use binoculars while holding on to the Therm-App and staring almost straight up warblering-style.

I'm definitely heading back to that stand of pines in the next "warm spell".

Christmas Lights

The birds were hanging out in good numbers at Sapsucker Woods this morning, and were decorating the trees like Christmas lights when seen via the Therm-App.

(Mostly) House Finches in the "staging trees" by the boardwalk near the Sapsucker Woods feeder garden.

(Mostly) House Finches in the "staging trees" by the boardwalk near the Sapsucker Woods feeder garden.

Mourning Doves in the left, Cedar Waxwings in the right. Not that I expect to be able to tell species apart with a thermal imager.

Meanwhile, I found this interesting warm spot in the woods:

The thermal image shows a slice of the tree back being warmer than the rest of the tree. Is this from rot? A cavity? Something else? I did not pry to find out. (Experience has also shown that tree cavities usually register colder rather than warmer than the surrounding wood, unless, of course, there is some warm inhabitant within.)

Keeping Track

Today was day 2 of a Keeping Track animal tracking field day with Sue Morse and Linda Spielman.

During lunch a few of us stumbled across some fresh mouse tracks in the snow, which all led to or from a tree with a nice cavity - a cavity that registered as a cold spot with my Therm-App. But I happened to swing the Therm-App across a smaller cavity and noticed something warm: it was a deer mouse peeking through a crack in the tree:

After I had my fun photographing the critter in the crevice, I'd walked away when another participant (Dylan from Binghamton University) noticed it coming out of the tree. I captured this thermal image of the mouse leaving the cavity as Dylan stood to the side photographing it with his point-and-shoot:

Slowly the mouse descended, crossed the snow, and climbed up the next tree:

Comparing IR Devices

In 2014, three infrared cameras for mobile devices entered the market:

1. Seek Thermal, $250-$300, 208x156 resolution
2. FLIR One, $350, 80x60 resolution
3. Therm-App, $1600, 384x288 resolution

Here I give a quick review of these products. To allay further suspense, let me just give the bad news: of the three only Therm-App is useful for birding.

Requirements

What I consider "useful for birding" is for the device to help detect a subject that would likely go undetected otherwise. This means roughly the ability to notice and highlight a hawk-sized subject at a distance of 100 feet (30m), at the minimum, since wildlife rarely approach closer than this, and when they do they can usually be spotted with the naked eye of an averagely observant person.

Note that I use the word "highlight", because it isn’t sufficient to just detect the subject. An infrared image of a forested scene looks rather messy, like this:

and it is not helpful to say that the few pixels in the upper left is a bird if it can’t easily be distinguished as such.

FLIR One

FLIR One, which fits only the iPhone 5 and 5s (support for Android and other iOS devices coming soon), is the first mobile-device attachment to come from the leading name in IR products. Its $350 price tag is much lower than any other FLIR product, but alas, it has a rather pathetic resolution of 80x60 pixels. On my first field test with the device on a cool November morning, I was lucky to encounter a pileated woodpecker and raccoon, both about 30 feet (10m) away from me. Both barely registered (if at all) in the FLIR One image. An interesting aspect of the FLIR One is that it combines the IR image with a higher-resolution visible-light photo to give the resulting image more detail. While I can imagine this being useful for close-range uses (to narrow down the hot-spot in an appliance, for example), and results in neat photos to share on Facebook, it does not help in bird/mammal detection. If anything, the overlaid detail from the photograph of the raccoon probably obscured any IR signature from the raccoon that might have been captured.

FLIR One image of Canada Geese grazing in Stewart Park, about 100 feet (30m) away. The black forms indicating the geese come from the visible camera and not the IR camera.

Seek Thermal

Seek Thermal is a small IR camera for both iOS and Android devices, seemed more promising. At $200 apiece (update: now $250) and a workable 208x156 resolution, I was really hoping it would be good enough for birding, and bring this technology to the masses. Unfortunately, field tests have found it to be lacking in range: it shows reasonable detail at close range, but beyond 20 feet the images are just a blur. This is likely due to the lens, which is fairly wide-angled and probably not set up to focus at distance. To its credit, it did highlight a squirrel on a tree about 100 feet (30m) away, though a crow quite a bit closer failed to register (more on crows later).

Seek Thermal image of Canada Geese grazing in Stewart Park, about 100 feet (30m) away. The geese are detected, but largely as unrecognizable shapes.

Seek Thermal XR

In January 2015, Seek Thermal released a new XR ("extended range") version of their camera, which brings it a step closer towards being useful for birding. My preliminary review is that it is close, but not quite satisfactory for birding purposes. When pointed at a known hot subject at moderate distance, you can certainly see it in the image, but it does not "stand out" at all; i.e., I would not have been able to "find" the subject. But I suspect that a different color palette could improve this some.
Here is a roosting goldfinch at about 50 feet away:

Left: "white-hot" palette. Right: "black-hot" palette. The subject is barely discernible here, and even less so with the other Seek palettes (not shown).

This bird was first detected at this distance with the Therm-App:

However, in walking closer to the subject (about 20 feet away), the subject does stand out in the Seek Thermal XR, here shown in six of Seek's assortment of palettes.

I hope to write a more detailed review soon.

Therm-App

With a 19mm lens and a resolution of 384x288 pixels, Therm-App is the only one of the three devices I've tested that has found things I wouldn't have otherwise noticed. Distant squirrels and raccoons show up as bright yellow spots, as do their roost sites. Birds seem to be better insulated than mammals, but still often register as bright spots. Following a Junco that had flown into a tangled bush, the IR image quickly pointed to a second bird nearby I hadn't noticed. Pointing towards a complaining crow showed the bright form of a red-tailed hawk - but curiously the crow itself did not register, even though I could easily see it with my naked eye. (Several encounters with crows were similarly undetected, leading me to hypothesize that either they're better insulated, or their more absorbent black color radiates less infra-red.)

Unfortunately, this capability comes with a $1600 price tag (I luckily bought mine during a $1000 promotional period), but it at least gives an indication of the minimum specs required for birding. In particular, the Seek would probably do quite well if installed with a lens comparable to the Therm-App’s 19mm (a fairly zoomy lens whose 19x14 degree FOV is comparable to a 60mm lens in a typical SLR). The Therm-App is also only available for newer Android devices (with USB On-The-Go support).

Therm-App image of Canada Geese grazing in Stewart Park, about 100 feet (30m) away. Note the higher zoom compared to the FLIR One and Seek, with a distant pedestrian showing up clearly. Note also that the hottest point in this image is an electrical unit on the utility pole; that same unit can be seen in the other two images, but being more distant does not register as being especially hot.

A hotspot in a tree stump coming from a cavity in which I could see, with binoculars, a roosting squirrel. Therm-App image on the left, Seek Thermal image on the right.