False triggers, resulting in photos and videos with no animals, are the bane of camera trappers everywhere. The PIR sensors in modern trail cameras do a pretty good job at triggering for animals. Unfortunately, they will also trigger for other things, like waving vegetation, a moving camera mount, and (in some circumstances) ripples on water. In this post I will cover the underlying causes of false triggers, as well as some practical tips for avoiding false triggers in camera trap sets. As in the cover photo, I will use real-life data, photos, and video taken from our too-large collection of captures without animals.

How PIR Sensors works

(Unlike the sensors in Star Trek) PIR sensors don’t (fundamentally) detect “lifeforms”.  They detect changes in temperature which appear between certain regions in the sensor’s field of view.  Mammals regulate their temperature, which is very likely to be different from the background temperature of other objects in a scene.  Thus, an animal passing in front of the camera will cause a change in temperature in different regions of the scene as it moves. This change in temperature is detected by the PIR sensor. When the change exceeds some threshold, it triggers the camera.

False triggers occur when something other than an animal causes a change of temperature within the PIR sensor’s field of view.

For a more detailed look at PIR sensors, see References a the end of this post.

Using File Metadata to Gain False Trigger Insight

Sometimes it’s hard to tell, just looking at a sequence of “empty” photos or videos, what exactly caused the false trigger. I’ve found it useful in these cases to look at the file meta-data. Specifically, the date and times each photo/video was taken (assuming the date and time on the camera are working correctly).

I created the graph below by copying the results of a “dir” (list directory) command on SD cards retrieved from 8 cameras from a recent deployment in Maine. Then, I pasted these into Excel. The graph shows the cumulative number of triggers as a function of date. These 8 cameras had been in the field since early April 2021. We retrieved them in late September of the same year. Critically, this period includes the peak of spring growth. You an see this definitely figured into the success (or failure) of some of these sets.

The graph clearly shows cameras which had an excess of false triggers. These are the 4 lines toward the left of the graph which shoot up, rapidly accumulating thousands of triggers. Careful examination of the slope of these lines shows that these cameras were triggering at near their maximum rate (yet not so much as to suggest, “run-on triggering”, described in How (some) Trail Cameras Fail). The other 4 lines have a nice, gentle slope upward. They have few vertical jumps, and contain a large fraction of “true” triggers of some type.

I have labeled the lines with high false trigger rates with their primary cause. I’ll discuss how to avoid these in the next section.

Avoiding False Triggers

Here area some tips for reducing false triggers.  The video below gives examples of not following some of these rules from the same set of 4 “false trigger” cameras in graph above. Sorry — no animals.

Sturdy Base

The PIR sensor can easily trigger if the camera moving. If the camera moves relative to the background, perhaps because it is mounted to a swaying tree, the PIR sensor will detect “moving heat” and will trigger.  Avoid this by securely anchoring the camera to a large, sturdy tree.  If a small tree is the best available, and if the site allows for it, cutting off the top of the tree just above the camera can substantially reduce camera movement due to wind.  In the case of “make do” sets (see The Art of the Treeless Set), make sure that the ad hoc camera support is sturdy – not just when you set it up, but over the change in seasons. 

A second problem can arise if the camera is loose inside a security box. Even a little movement caused by the wind can induce false triggers. We ran into this problem when we started using security boxes, screwed into a tree, and a padlock only. Without a cable lock securing the camera, there is a couple of mm “play” in the camera. It’s easy enough to also use a cable lock. Compressing a thin piece of foam rubber into the space between the camera and the security box also works. In the example in the video above, we made a make-do set with a security box and just a padlock. You can see that as the wind blows, the camera pivots up and down just enough to set off the trigger.   

Avoid Sun Facing Sets, Especially into Water

There several good reasons for not facing your camera southward, into the sun. Front lit photos and videos are often higher quality than those which look into the sun.  Further, having the PIR sensor face the sun can also increase the rate of false triggers.  We’ve found this to be especially true with cameras facing into the sun, and at a body of water.  In the video above I have an example of such a set in which IR radiation from the sun was reflected into the PIR sensor by wind-blown ripples on a beaver pond. The ripples created apparent changes in heat in the scene. These triggered the camera nearly constantly during the day, almost from the day we set it out.

Note “not pointing the camera into the sun” generally means pointing it towards the north in the northern hemisphere, and towards the south in the southern hemisphere. To be sure, use a compass (including the compass app on a smartphone). Thanks to Chile-based BernardoS for pointing this out in comments on original post.

Beware the Sunlit Rock Background

A rock background can provide a dramatic backdrop for many animal photos and videos.  It is especially useful for night-time color shots because it reflects some of the light from the flash to create a brighter photo or video.   Unfortunately, a varied rock face facing the sun can result in a background of different temperatures.  Even small waving vegetation between the camera and the rock can cause the PIR sensor to trigger.    Our most successful rock-backed sets have been in well-shaded areas, with practically no vegetation. Unlike the set in the photos below.

Avoid Growing Vegetation

Leafy green vegetation blowing in the wind can alternately shade and expose background objects with different temperatures. We carry a pair of clippers and a small saw to clear away vegetation from the set.  This is tedious, but quite effective.  Problems arise in sets pruned in advance of a major growth spurt in the surrounding vegetation.  In this case, it’s best to look for spots free of signs of earlier growth (e.g. tall, dead grass).  Another strategy is to visit and update sites after a major growth spurt.  The timing for this varies by region, of course, but the basic rule is that vegetation growth happens in early spring. 

Reduce Sensor Sensitivity

Newer cameras often have a setting for adjusting the sensitivity of the PIR sensor. That is, the threshold in temperature difference required to trigger the camera. For example, on Browning Recon Force Advantage and Spec Ops Advantage (BTC-7A and BTC-8A, resptively), this setting is listed under:

• Motion Detection: Options are for “60 ft” or “80 ft”

These options might as well be labeled, “low” and “high” sensitivity, respectively.

Unfortunately, although reducing sensitivity will decrease the number of false triggers described in this post, it is often not nearly enough. Moreover, reducing sensitivity also decreases the chances that the camera will trigger on animals.

Limit Operating Hours

In many case, the time at which animals are active is not the same time of day with the most potential for false triggers. Wind often reaches a maximum more or less in mid-day. By this time, the sun has heated up some objects, and not others, creating perfect conditions for false triggers.  If your target species are not active during this time period, one solution is to program the camera to “turn off” for certain times during the day.  Not all camera models support this feature.  

Below is a chart that shows the relatively frequency of true and false triggers by time of day for a set in Maine accrued between April and September. Setting operating hours for the camera to by outside of 9:00 am to 5:00 pm would have substantially eliminated false triggers on this set. Unfortunately, you might also miss exceptional appearances of your target species during this time.   Since these shots — like a lynx in the daytime — are highly prized, disabling the camera only during false trigger peak time is a better strategy. In the example below, we would have seriously reduced the number of false triggers by disabling the camera between 11:00 am and 1:00 pm.

The exact distribution of false trigger times during the day will vary with camera location, how the camera is facing relative to the sun, and other factors.

As far as when animals are active, the development of so-called “animal activity curves”, increasingly based on camera trapping data, is subject of current research. See References below. Knowing exactly when your target species, in your location, may require some of your own research.

Recovery Delay

Most trail cameras have a setting which allows you to adjust the minimum time between successive triggers.  One can reduce the number of false triggers a camera records by increasing this time.   Thus, in the worst case, a camera that triggers at most every 5 minutes during a windy day will take far fewer videos than one that can trigger every 20 seconds.   Unfortunately, increasing the recovery delay has a downside for true triggers.  With a delay of 5 minutes, for example, a camera set to take a 20 second video will only capture the first 20 seconds of an exceptional 2 minute animal encounter.    One could compensate for this by increasing the length of videos. Unfortunately, this also causes false triggers to consume more battery and SD card resources.

Maximally Provision and Check your Camera

When it’s not possible to follow the rules above, one can maximize the chances that the camera also catches at least some true triggers by:

• Using high capacity batteries at full charge: e.g. LiFeS2
• Using an SD card large enough to exhaust the batteries: 256 GByte for Browning Recon Force Advantage (BTC-7A) when using LiFeS2 batteries and taking “ULTRA” resolution video.
• Visiting frequently to “freshen” the set (especially during times of growing vegetation) 

Conclusion

Sifting though false triggers is time consuming and frustrating.  Missing an opportunity to capture an animal because false triggers have filled the SD card is even more frustrating.   In this post, I’ve attempted to summarize techniques to avoid false triggers in trail cameras.  If you have used these techniques successfully (or unsuccessfully!), or have others, please comment below.

References

Peter Apps, John Weldon McNutt, “How camera traps work and how to work them,” African Journal of Ecology, 2018.

Rowcliffe, J. M., Kays, R., Kranstauber, B., Carbone, C., & Jansen, P. A. (2014). Quantifying levels of animal activity using camera trap data. Methods in Ecology and Evolution, 5(11), 1170-1179.

Lashley, Marcus A., Michael V. Cove, M. Colter Chitwood, Gabriel Penido, Beth Gardner, Chris S. DePerno, and Chris E. Moorman. “Estimating wildlife activity curves: comparison of methods and sample size.” Scientific reports 8, no. 1 (2018): 1-11.