About Me

My photo
Native Californian, biologist, wildlife conservation consultant, retired Smithsonian scientist, father of two daughters, grandfather of four. INTJ. Believes nature is infinitely more interesting than shopping malls. Born 100 years too late.
Showing posts with label equipment. Show all posts
Showing posts with label equipment. Show all posts

Sunday, August 10, 2008

Solving a parallax problem


What's wrong with this picture?

It was supposed to be a full frame picture of 4 playful rodents centered in the frame.

Why did I screw up?

The cam was too close, and I misaimed. Taking a little more time might have helped, too.

How can you miss a close-up like this?

Parallax.

The PIR's narrow zone of detection was aimed at the burrow, but the camera was aimed above it. This is one of the problems you encounter when you try to use a camera trap for close-ups. The center of the camera's picture frame and the narrow detection zone of the PIR are not in alignment. It doesn't matter at a greater distance.

Some commercial trail cameras use a built-in laser pointer for aiming the camera; so I just made my own laser pointing device.



This is a snap-on mount made from a hacksaw blade, and a laser pointer with a magnetized base. (They are 'posed' on my table saw). You can adjust the hacksaw blade mount and then move the magnetized laser from the PIR window to the lens window for fine adjustment.

It helps.

Saturday, February 9, 2008

On making bear guards for camera traps


Our first bear guard design. When Bruin paws the hinged treadle the canary (a $1 security alarm) screams. The terrified and paw-pricked bear never touches another camera trap. Well, that's the theory, anyway.


I finally repaired the camera trashed by the bear. I dried it above the wood stove for a few days, and last weekend I resoldered the torn wires. When I turned it on for the walk test the controller and camera communicated like a charm.

I'm a little smarter now and not taking chances. Many camera trappers have no problems with bears at all, but I am not one of them. Now I only put cameras in the woods that are equipped with homemade bear guards. I have more camera traps than bear guards, however, so it was time to make some more.


I don't use junction boxes or spiked collars for camera traps any longer. Though functional, this set up was too much work to get to the camera.

Neighbor Richard and I have come up with several designs for bear guards, and all of them work. Any spiked metal frame attached to the camera case deters curious bears, but it took several designs to arrive at the quick-set bear guards seen below. If bruin used only his claws he might be able to disarm a camera case of its bear guard, but bears are ham-handed and lose interest as soon as they get pricked.

Richard improved the bear guard design with the addition of angled clamps which make it very easy to attach and remove the guard from Pelican cases. The guard is made of 3/4 x 1/8 inch angle iron. The clamps are excellent fasteners, but you have to cut off the strap slot on the top of the Pelican case. You can find pdf instructions for building it here (registration is free).


The fixed clamp of the bear guard grips the ridge on top of the Pelican case. The adjustable clamp uses a wing nut on a 1/4" threaded rod to grip the ridge on the bottom of the case.

Richard suggested making a more light weight version of the bear guard using 1/8 x 3/4 inch angle aluminum, instead of angle iron. It differs in a few details from the angle iron version described in the pdf. The joints are mitered rather than lapped, and the spikes are made of galvanized nails glued in place with J-B Weld. The iron version used sheet metal screws as spikes (which do not have to be glued--much easier to make).

To cut the aluminum accurately I made a cutting jig for the table saw (top photo, below). After drilling holes for the nails, the 4 pieces of the frame were clamped for welding (middle photo, below). Then the nails were glued in place and allowed to cure for 24 hours (bottom photo, below).



Excess J-B Weld was ground off with a Dremel tool, and the inner surface of the frame was coated with soft polyurethane to minimize scratching the Pelican case. Three pop rivets hold the top clamp in place, and the 1/4" x 20 nut for the bottom clamp is threaded and glued in place.

The cost of materials is about $12 for one, and I bought enough materials to make three. I'm lucky to have a neighbor like Richard who can build practically anything and always offers a helping hand.

I spray painted them this afternoon, and they'll be ready for use tomorrow.

Thursday, January 31, 2008

Canary in the garbage




The garbage collector gave us a new garbage can. The previous one was too big, so we opted for the smaller and cheaper one.

"That's a dink can", scoffed the redhead on seeing the replacement.

"But small is beautiful", I replied wondering where she came up 'dink can'. "It's a cute little can, just look at it!"

Neighbor Richard suggested we arm our 'dink cans' with "screaming canaries" to keep the bears away. He is convinced the one dollar security alarms deterred the garbage-scrounging bear last year.

Maybe so. Our neighbors had garbage-bear problems, but we didn't. Our cans were armed with "screaming canaries". Presumably the shock effect of the siren is enough to scare the bear. It sure worked on me everytime I forgot it was on. I didn't do a backward roll and chop my jaws in retreat, but I cursed it somewhat affectionately.

The screaming canary is actually a cheapo security alarm with a piezoelectric oscillating siren. It's a low-power-consuming sound alarm like those used in smoke detectors. It sounds like a German roller canary on steroids and methamphetamines. The alarm goes off when you remove the magnet.

Here's how Richard armed the garbage cans.



He mounted the alarm on the lip of the can, and fixed the magnet on the lid. The alarm is fastened to the metal bracket with double-sided tape, and the bracket and magnet are bolted to the can.



Spacing is critical, so he used scrap plexiglass as a spacer to align alarm and magnet. The red color is fingernail polish. (Richard has an artistic streak.) It also locks the nut and bolt together.



The plastic material is a rain shield.

One of these days I'll get around to experimenting on the effects of the alarm on pestiferous wildlife. Meanwhile, we'll keep the faith that they actually work.

Sunday, January 20, 2008

Camera Trap Protection

"Be advised, Captain Kirk that nowhere in the galaxy are your camera traps safe from tampering." Spock

A new publication on camera trap protection is always of interest because camera traps are never safe from wildlife and people. This paper in the Journal of Wildlife Management touts the use of a plate steel security box.

The authors, who study wildlife traffic in flood underpasses along a busy stretch of state route 58 in the Mohave desert, had the wisdom to contract a local welder to "armor" their 7 Cuddeback Digital Scouting cameras. The welder was not only paid for his services, but was made a coauther, which is nice to see.

The security box was made of 2 mm steel plates with arc welded joints. A sliding front door with windows for the lens, flash and sensor apertures gives access to the camera, and can be padlocked. I am certain that our state highway department, Caltrans would not approve the use of anchor bolts to fasten the cameras to the culvert walls. So epoxy was used to glue rotating camera mounts on the underpass walls. The cost of materials and labor was $90.

Each camera was deployed with a brief note in Spanish and English explaining the purpose and the ownership of the camera.

The cameras took 107 photographs over a period of 170 days, which by my calculations is 0.14 photos per day or 1 photo every 7 days. The culverts were not a hot area for animal traffic. The wildlife "catch" was black-tailed jackrabbits, desert cottontails, desert woodrats, domestic dogs and a domestic cat. The cameras also photographed 4 people in the culverts, and no traps were damaged, moved or apparently touched over the 6 month period.

With the obvious exception of people, none of these species is a threat to cameras, but I suspect this security box would be an excellent deterrent to bears. A black bear could probably break the wall mount, but it would have to use bolt cutters to cut the padlock.

Most people are not interested in exploring highway culverts, but this changes when nature is calling. A culvert is about the only place in the desert that affords potty privacy. Hitchhikers and hobos also appreciate culverts for shade. The researchers found ample evidence of these human activities. In this situation, the written notice probably had the desired effect.

The online hunting and camera trapping forums, such a Pixcontroller and Real Deal Hunting Chat are good places to learn about camera trap theft, reactions to it, and the ingenuity used to counteract it.

It's my impression that poachers and trespassers pose the greatest threat, because they realize the camera has all the evidence needed for conviction. A chainsaw, however, can even defeat the Python lock. (Just cut the tree down.)

It would be interesting to conduct an experiment to find the most effective deterrent to camera trap thieves. But who's going to do it? It's a lot more fun getting pictures of wildlife.


Reference

Fiehler, C.M., B.L. Cypher, S. Bremner-Harrison, and D. Pounds. 2007. A theft-resistent adjustable security box for digital cameras. Journal of Wildlife management, 71(6):2077-2080

Thursday, September 13, 2007

A codger's kit



A faithful reader requested a post about my kit -- he was curious about the baggage I haul around in order to set the cameras, and the musical background to the hammer and pipe. (Go ahead and click it!)

The redhead took this picture of me a couple days ago when I got home. I had just gathered two cameras that had been out for over a month -- both had bear pictures, including (soto voce) some very intimate movies of a bathing she-bear. For those of you who are old enough to have read Frank Norris's "Octopus", they remind me of Annixter's Peeping Tom venture when his wife-to-be took her ablutions in the creek. (I think you'll like the paparazzi intrusiveness of the film, provided I can figure out how to edit the footage and post it.)



This should also give you an idea of the kit needed to set two camera traps.

To the left:
two cameras in Pelican cases (one camou, one black with cable for attachment)
two metal posts for staking cams (I can't always find a tree in the right place)
two "bear protectors" disarmed with wine corks (another good reason to drink wine)


Middle column of tiles
:
Pruning shears (long and short) and folding saw for "landscaping" trails and sets,
pipe and hammer (for alerting the bears that I am coming--this is carried in my back pocket, and can used to accompany oneself singing the Happy Wanderer),
Swiss army knife (for cutting rope and as a size reference for photos of scat etc.)
Cree crooked knife (for stripping branches when you need to make a post, stake, or walking stick)
Tool roll with screwdriver, nut drivers, and ratchet wrenches (for attaching hardware)

To the right:
water bottle (filled with iced Gatorade in summer)
surveyors tape (for trail marking)
Miranda's Jim River beaver scent (a scent lure)
Plastic container of hardware (for tacking bait and jury rigging)
string and nylon cord
rope ratchet (for attaching cams to trees when using the cord)
Off Clean Feel (contains Picaridin, less mutagenic than than DEET)
Furfindr with chain
two-way radio (to call the boss in case of emergency)
Rechargeable batteries and memory sticks (for cameras already out there)
notebook and pencils (if I am feeling poetic)
compass
GPS for marking locations and recording tracks

I always carry an "Alpenstock" of California bay, and sometimes take a camera along. I gave up on carrying a multimeter -- when in doubt, replace batteries.

Thanks for the suggestion, Nigel.

Saturday, May 12, 2007

The submariner's solution



Ever tried to aim a camera trap on the end of a pole at a hole in a tree 12 feet above your head?

What is that nut talking about, you ask?

I'm talking about the trials and tribulations of photographing cavity-nesting birds. You are looking up into the foliage, the sun usually in your eyes, and from one angle the camera looks dead on target. Then you move for a different take, and it looks off target. Photographing the comings and goings of a small bird up close with a camera trap takes the patience of Job. It's a bear adjusting the camera so that the passive infra-red sensor (or PIR) detects the critical area of activity.

If you're not dozing yet, you may be wondering "what's the problem?"

Well, the fresnel lens helps the PIR detect moving heat in a cone-shaped volume that expands with distance from the sensor. At short distances the sensor's area of reception is quite small. I measured it in one of my cameras and found that when my hand was moving only 3 feet away the sensor detected it only within a 5 inch circle.



This means that it takes a lot of trial and error eye-balling to get the PIR 'spot-on' the nest hole.

The other day I bought a cheap laser pointer with the idea of solving the problem. I was going to make a mount for the laser on top of the camera case, so I could aim the PIR quickly and with more certainty.

Then I thought I'd better run the idea by my neighbor Richard.

Richard is a submarine veteran (remember the mechanic Johann in "Das Boot?"), and he's a wizard at all things mechanical.

Well, Richard suggested going with a periscope. At first I was skeptical, but he rummaged around his shop, and in no time assembled a miniature periscope from a piece of square metal tubing, a piece of aluminum bent at 45 degrees, and a small mirror. We velcroed the device to the back of the camera, and aimed the laser up the tube. It worked like a charm.

Next I had to measure the distance between the laser spot and the PIR sensor. To correct for parallax the reflected laser spot should be several inches above the target (depending on the distance between the reflected beam and the center of the PIR). Then you adjust the camera's position on the pole till you get it right. This is pretty easy when using a telescoping fiberglass pole. It's a more tedious process when using square tubing with connecting sleeves.

The next day I hauled the gear down the hill to a cavity located in the end of a broken limb. The periscope indicated the camera was too low. I raised the camera on the pole, but I couldn't tell if I had it right. You see, the trunk wasn't vertical, and after I raised the camera I couldn't see the laser beam from the ground.

The joke was on me. I could only conclude that while technology can solve a lot of problems, nature doesn't always cooperate.

Wednesday, April 18, 2007

Recluse in the snag



Three days ago at 6:13 PM a squirrel climbed down the snag and aroused the resident screech owl (who looked pretty scary to me). During the next two nights the owl took 143 pictures of its own comings and goings. Or should I say owls? If there is a pair, I can’t tell them apart.

The cavity is in a black oak and looked unused. So I was prepared for disappointment this afternoon when I went to check the camera. For nearly two weeks I have been using a new toy -- a telescoping fiberglass pole -- to raise the camera up to the level of tree cavities. This tree was second choice.

I expected a bonanza of pictures from my first choice -- a much-pecked flicker-sized cavity about 25 feet up. It had all the signs of occupancy, except the presence of a bird. The edge of the cavity had been chipped away and enlarged recently.

It was a bear of a job getting the camera into position, because when it is 25 feet overhead you can’t really see exactly where the camera's sensor is in relation to the cavity. From one angle they look like they are face to face. Then you move uphill to confirm it, and the camera looks too low. So you make adjustments and check again. And again. The other problem is pole-wobble. Even when anchored with guy lines, the slightest breeze wobbles the pole, which fools the sensor, and the camera goes wild taking pictures of moving vegetation.

After failing with a cavity that had "active nest" written all over it, I settled for one that was 12 feet up.



It looked abandoned, but I got a nice surprise. There were many shots of the little guy checking out the camera and the scenery.



And then there were shots of it clinging to the entrance.



And (whoohoo!) I got three pictures of the owl in flight.



Now to improve the set-up so I can get that picture of mini-raptor with limp deer mouse in talons.

Wednesday, February 28, 2007

Camera trap pioneers: O.P.Pearson


Game trails have always interested naturalists, because they tell us where wild animals move and how they use their habitat. Trails come in all sizes, are usually "engineered" for energetic economy, and are most pronounced where activity is concentrated--near feeding areas, water, or mineral licks. Climb a tree or hide near a game trail, and sooner or later you will see who uses it.

In the 1950s, Dr. Oliver Pearson of the University of California embarked on a traffic survey of mouse runways on his property in Orinda, California. Professor Pearson was a distinguished and particularly inventive scientist who studied mammals and birds. His interests ranged from taxonomy to physiological ecology and predation. He was also a pioneer in the use of camera traps.

Pearson wanted to know what goes on in the microcosm of mouse runways. To a mouse, a patch of weeds or a grassy field is like a jungle, and species like meadow mice bushwhack and maintain their own runways with their teeth. In a clear runway they can zip along at a brisk clip and tend to their daily business. Harvest mice enjoy the labors of their larger neighbors.

Pearson's property harbored populations of meadow voles and harvest mice only 50 feet from the house, and the epicenter of his study was a 20 x 20' weed patch with a brush pile in the middle. Could a professor of mammalogy ask for anything more?

He used two 16mm movie cameras in tandem with weather stations to survey the traffic. The camera shutter was electrically activated whenever a mouse or even a large insect passed the camera lens. He used two trigger mechanisms with equal success, a treadle and a photoelectric cell.

The cameras were synchronized with electronic flash units. Nowadays we can snoop on sensitive species using infra-red cameras. To avoid frightening the mice, Pearson masked the flash with 18 layers of red cellophane, and noted that "a muffled clunk made by the mechanism seemed not to alarm the mice unduly."

Powered with alternating current, the cameras ran day and night, but there was a problem with differential exposure. Daytime pictures were overexposed, because the shutter was slower than 1/30th of a second and the f-stop was adjusted for night. Again the solution was red cellophane, this time over the camera's lens. The camera units were sheltered in glass-fronted housings. Overhanging eaves and a blackened light bulb counteracted the effects of rain and dew.

Pearson positioned a camera unit on one side of a mouse runway. On the other side, and framed within the camera's view, were the instruments of a miniature weather station: a dial thermometer, a hygrometer, a ruler, and an electric clock with a sweep second hand. Thus, whenever a creature took its own portrait it also recorded its body length, the time, and a weather report. Since he live-trapped the mice and gave them distinctive "haircuts" he could identify many of them individually.

Not long after he deployed the equipment, the neighbor's Siamese cat discovered that the camera housings were superb perches for mousing. He solved the problem by fencing the weed patch. Then a slender salamander electrocuted itself while short-circuiting the treadle. Delusional song sparrows were another problem. They wasted a lot of film shadow boxing their own reflections on the weather station window. Curiously, one bird side-stepped an oncoming meadow mouse, a gesture of road etiquette rarely seen on California highways.

In nineteen months of operation (111 recorder weeks) the camera traps generated 8,495 photographs. In other words, weed-patch wildlife made an average of 11 passages per day. The professor noted that, "A patient, non-selective predator waiting for a single catch at runways such as these could expect, theoretically, a reward each 2.2 hours." The cameras mined enough information for three scholarly articles.

Pearson's first paper on the traffic survey contains the information of most interest to camera trappers. I'll go out on a limb here, and state my belief that the results also apply to larger mammals, especially other herbivores like deer.

His "pleasant surprise" was that 26 species of animals, from weasels to snakes and mole crickets--made use of the runways. He never saw three of these species within at least a mile of the study area, proving that camera traps often disclose nature's little secrets. Ironically, he never photographed three other species he commonly saw within 100 feet of the recorders! The take home message here is that placement of camera traps is critical.

Though two species of mice used the runways, Pearson's camera traps revealed that meadow mice were the "public works custodians". When meadow mice stopped using a runway, weeds and seedlings quickly filled them in. Another curious finding was that mice apparently used runways selectively even though nearby runways seemed equally suitable.

Pearson's camera trap studies of rodent ecology are still among the most elegant and detailed studies of their kind. The work also seemed to stimulate Pearson's interest in predator-prey relations, for he went on to study predation of mice in Tilden Park not far from the UC campus. During the next twenty years a smattering of biologists, including your's truly used homemade camera traps for wildlife studies and surveys. Then in the 1980s camera traps became commercially available, and wildlife biologists developed camera-trap-fever.

Often we don't discover common interests till it's too late. This camera trap codger wishes he had mustered the courage earlier in his career to ask Pearson about his camera trapping days. Many times I saw the professor at meetings of the American Society of Mammalogists, but stood speechless and in awe. I guess we are a lot like meadow mice, politely passing each other through the runway of life. Mice and men are alike that way.

Acknowledgement: My appreciation to the Allen Press for permission to reproduce the plate from Pearson's 1959 paper, and to Kathleen Berge of the California Academy of Sciences for scanning it.

References
Osterberg, D.M. 1962. Activity of small mammals as recorded by a photographic device. Journal of Mammalogy, 43:219-229.

Pearson, O.P. 1959. A traffic survey of Microtus-Reithrodontomys runways. Journal of Mammalogy., 40:169-180.

Pearson, O.P. 1960. Habits of Microtus californicus revealed by automatic photograph records. Ecological Monographs, 30:231-249.

Pearson, O.P. 1960. Habits of harvest mice revealed by automatic photographic recorders. Journal of Mammalogy, 41(1):58-74.

Thursday, December 1, 2005

More background

When I was a kid thumbing an ancient issue of Popular Mechanics Magazine I encountered a morbid high-speed photo of a mouse execution. The luckless mouse was airborn as the bar of the mousetrap closed in for the kill. I think that's when the obsession took root. In college I made my first camera trap under the guidance of the late Professor Joe Hall of San Francisco State University. That was the "doorbell model". The trigger--a whisker switch--was wired to a battery-operated doorbell. Activating the switch made a little wooden bar press the shutter button. At the time, it was a wonderful gadget, but you had to hand crank the film after each exposure. If you were lucky you got one picture a night.



In 1980 a couple of my colleagues from the National Zoo and I went to Indonesia where we used an improved homemade camera trap to photograph the little known Sulawesi civet. Camera technology was becoming electronic. We wired a photoelectric trigger to plug directly into the electrical shutter outlet on the front of the camera--a Nikon FE. With battery powered film advance, you could take many pictures in a night. Two months later we were back in the states impatiently waiting for our developed film. When it finally arrived, we found that box after box had slides of night-flying moths and falling leaves. But there were also a few photos of the civet, and we considered it a great success.

Now in the age of the "home-brew scouting camera" you can get hundreds of digital pictures a day. I have slowly advanced from easy-to-build "plug and play" systems, to those that require taking the camera apart and soldering internal contacts. Luckily there are some great support groups on the web, whose mentors patiently assist greenhorns.

Affordable digital cams have one irritating deficiency--the camera has to be turned on to take the first picture when an animal activates the passive infra-red sensor. The delay of a few seconds can give you a lot of cut-off images.

The latest control boards however keep the camera awake when there is infra-red activity, which means you no longer have to wait for the camera to wake up for each picture. This is an acceptable challenge. You just have to use your "jungle lore" to compensate for the camera's shortcoming.