- A team of camera trapping experts surveyed researchers and conservation professionals to identify limitations to their successful use of remote cameras, assess their wish list of technological developments, and predict what next-generation camera trapping will look like.
- Their recently published study revealed that cost, theft, vulnerability of the cameras to environmental conditions, and several ongoing technical issues may be limiting the effectiveness of this popular technology in providing the utility the users seek.
- The survey respondents offered numerous predictions for next-generation camera trapping, including solar and lithium-ion power sources, a wider range of sensors, and software-driven automation.
Camera trapping has become an important conservation and research tool worldwide. Photos from remote cameras have afforded us insights into the lives of rare, shy, cryptic, nocturnal, or otherwise seldom-seen animals.
|Remote cameras can capture images of a variety of rare, cryptic, and shy animals that would otherwise be impossible to view in their natural habitat.These three cameras in a forest in Borneo found an orangutan, a pangolin, and a mousedeer. Images by Oliver Wearn.
The idea is simple: Buy some cameras equipped with motion or heat sensors that automatically take an image or video when triggered by a passing animal, set the cameras out where you think the animal(s) will go, and let the animals take the images for you. Retrieve the images, and record all the interesting animals in them.
Unfortunately, for many camera trap users, the reality has not been so easy.
Cost, theft, and leakage
A team of camera trapping experts surveyed researchers and conservation professionals to identify limitations to their successful use of remote cameras, assess their wish list of technological developments, and predict what next-generation camera trapping will look like.
The 258 survey respondents from 52 countries use remote cameras for a variety of tasks: to inventory wildlife communities, answer questions about species’ distributions, abundance and behavior, obtain photos for artistic and outreach work, and monitor people, including potential wildlife poachers.
It turns out that certain camera trap weaknesses identified by participants in an informal 2016 Mongabay survey continue to challenge users three years later.
“We were surprised just how widespread, and varied, the problems with camera trapping are,” co-author Oliver Wearn told Mongabay. “Our survey shows that they’re an everyday part of using camera traps, which is really frustrating given that they’re now a mature technology with a large user base.”
The recently published study revealed that cost, theft, vulnerability of the cameras to environmental conditions, and several ongoing technical issues may be limiting the effectiveness of this popular technology in providing the utility the users seek.
“We knew from colleagues that their projects were taking a really big hit from thieves and vandals,” said Wearn, an AXA research fellow at the Zoological Society of London, “but we were really surprised to see that it was a truly global problem. Overall, it was outranked only by cost, which we already knew was a big constraint given tight research budgets.”
Researchers use grids of tens or hundreds of camera traps to generate enough observational data to statistically analyze data on species’ occupancy, abundance, or behavior, exacerbating cost and theft problems.
Beyond their frustration with cost and theft, the respondents collectively identified several aspects of camera traps where technological improvements would make a big difference. They most commonly identified: sensor performance, especially the speed and sensitivity of the image trigger; resistance to temperature, humidity, and other environmental conditions; and automated filtering to exclude blank images.
More than half the respondents considered improvement in each of these areas highly important for the next generation of camera traps. Wearn said they hoped their survey results would encourage camera developers to innovate in the priority areas identified by the participants.
“Sometimes it isn’t the most novel or exciting developments which will make the biggest difference,” he said. “As researchers and conservationists, we need the basics to work better, and this really came out in the survey results — the priorities were, for example, better resistance to humidity, faster trigger speeds, and fewer blank images. Solving these age-old problems will transform our effectiveness overnight.”
The survey revealed that these “age-old problems” still plague camera trap users. General technological progress and a broader offering of makes and models mean that camera traps cost less now than they did several years ago, although the price is still a barrier to widespread installation of these cameras in many areas.
Respondents using newer models didn’t give higher ratings on average than those using older equipment when it came to reliability, image quality, sensor quality, usability or value for money.
“The lack of any trend in camera trap ratings over time, despite improvements in the technology, was highly surprising to us,” said co-author Carolina Soto-Navarro, a postdoctoral scientist with the Swiss-based Luc Hoffmann Institute and the U.N. Environment Programme’s World Conservation Monitoring Centre (UNEP-WCMC). “We think that’s due to the fact that we haven’t witnessed any major breakthroughs since the modern digital camera trap (i.e. a camera trap with a passive-infrared sensor, infrared flash etc.) emerged in the mid-2000s.”
Despite their frustrations, the survey respondents recognized the value of camera traps as a tool, describing a wide range of solutions they’ve devised in the field to address the various technical challenges they identified.
To discourage theft, for example, researchers have disguised cameras with camouflage or locked them to supports or within security cases; and modified the flash or antenna to be less conspicuous. In a few cases, they publicly explained the cameras’ purpose through informative signs attached to the devices or in conversations with local residents.
They’ve also addressed environmental hazards by replacing lenses damaged by the sun or sand; using glue, zinc or duct tape to keep out moisture; placing cameras high up to avoid flooding; inserting silica or tampons inside cameras to absorb humidity; and drying camera traps with dehumidifiers and lights between deployments.
To maximize battery life, they’ve set longer time delays between photographs or modified power supplies by integrating large 12-volt batteries or solar panels.
Respondents described several techniques they’ve used to prevent empty images (false positives) from using up batteries and filling the cameras’ on-board memory cards, from positioning the devices to avoid direct sunlight and shadows, to moving vegetation from the field of view, and even putting out branches to keep cows away from the cameras, among other modifications.
Despite researchers’ best efforts, remote cameras still generate large numbers of false positives. To filter out these unwanted images and otherwise manage the huge image or video data sets that can result from even a single camera trapping effort, some respondents have enlisted dedicated platforms, such as eMammal, Camera Base, or Agouti. No particular software package stood out as a favorite. Others have involved citizen scientists, through programs such as Penguin Watch, Snapshot Serengeti, Instant Wild, Zooniverse, or MammalWeb, to help process the data sets. A few developed their own custom software.
Camera trappers also want to avoid false negatives, where the cameras miss animals that walk past them, due either to slow trigger speed, high temperatures, or poor camera placement. Survey respondents have learned to determine their cameras’ field of view and trigger sensitivity, calculate effective detection distances for target species, and position cameras at the correct distance from a species’ path, sometimes by mounting cameras on poles or branches.
The study authors shared a large number of these custom solutions in their paper, in part, they said, to help novices avoid some common frustrations.
“We’ve seen many a budding camera trapper come to the technology with wild hopes of beautiful images and flawlessly functioning equipment,” Wearn said. “By collating and reporting on the vast number of ways camera trapping can go wrong, we hope that those new to camera traps have more realistic expectations and can plan to minimize potential problems from the outset.”
Soto-Navarro added that as the camera trap community grows, standardizing data management, metadata collection, and image cataloging become increasingly important to enable coordination among individual camera trapping projects.
Specifically, the authors highlighted in their paper the benefits of building a crowdsourced platform for sharing information and experiences among camera trappers, including reviewing camera trap models.
“Harnessing people power through the creation of a platform where scientists, conservationists and amateur naturalists around the globe can collaborate and share experiences,” Soto-Navarro said, will help build a richer knowledge base that could be instrumental to advance conservation efforts.
Camera trapping 3.0
The survey respondents offered numerous predictions for next-generation camera trapping. For example, respondents said they believed integrated solar power and lithium-ion batteries (like those used in mobile phones and laptops) would soon power remote cameras with more functions for longer periods. They also predicted that sensors for acoustics, temperature, and additional environmental variables would be integrated into camera traps, either on board the camera or as part of a networked system.
Software that automates the detection and identification of target objects in photographs received the most votes, however, followed by automated communication of camera trap images, via cellular or satellite networks. Ongoing efforts to test and refine machine-learning algorithms that identify species automatically are advancing, the authors said, and open-access camera trap data sets, such as those from SnapShot Serengeti, or Wildlife Insights, will be key to providing data to train AI algorithms.
In fact, Soto-Navarro said, “If we repeat this study in 20 years’ time, we would expect major improvements in the technology, such as the incorporation of solutions from AI and increasingly moving camera trapping into the realm of big data, which will revolutionize the way we study and monitor wildlife.”
Glover‐Kapfer, P., Soto‐Navarro, C. A., & Wearn, O. R. (2019). Camera‐trapping version 3.0: current constraints and future priorities for development. Remote Sensing in Ecology and Conservation.
Burton, A. C., Neilson, E., Moreira, D., Ladle, A., Steenweg, R., Fisher, J. T., … & Boutin, S. (2015). Wildlife camera trapping: a review and recommendations for linking surveys to ecological processes. Journal of Applied Ecology, 52(3), 675-685.
FEEDBACK: Use this form to send a message to the editor of this post. If you want to post a public comment, you can do that at the bottom of the page.
This story first appeared on Mongabay
South Africa Today – Environment
This article is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.
You may republish this article, so long as you credit the authors and Mongabay, and do not change the text. Please include a link back to the original article.