Feature photo: A camera trap photo © MammalWeb (CC BY-SA 2.0)
In their latest article, Pen-Yuan Hsing et al. describe the processes involved in launching and running MammalWeb – a successful camera trapping project that has produced over 440,00 classified image sequences and videos over the past decade.
It is unlikely to come as a shock to readers of the Applied Ecologist that the world is undergoing an unprecedented decline in biodiversity. We need new methods and ways of working together to understand and tackle this challenge. One of these ways is partnering with local communities to monitor biodiversity which is what motivated the MammalWeb (@MammalWeb on Twitter) citizen science network as described in our recent article in Ecological Solutions and Evidence.
As the name implies, we founded MammalWeb in 2014 with a focus on wild mammal monitoring. In the UK, knowledge of mammal diversity and distribution is often spotty. In one case, the dearth of reliable data led to an estimate of the rabbit population that ranged from 2 to 255 million. In this post, I will describe how we started MammalWeb, its growth to encompass most of Great Britain, key lessons learned, plus some of my own reflections.
Growing a citizen science biodiversity monitoring network
We started MammalWeb in partnership with the Durham Wildlife Trust where we enlisted the help of community members in North East England to set up motion-sensing camera traps to observe local mammals. These citizen scientists would upload their camera trap photos to the MammalWeb website where anyone with an Internet connection can help classify the observed wildlife. Citizen science is a nuanced topic and includes many forms of collaboration, of which MammalWeb is the “contributory” type where a project crowdsources data collection and processing.
Over the past 7 years, MammalWeb has grown from the Durham area into a biodiversity recording network covering most of Britain and beyond. Citizen scientists have deployed camera traps at 2,300+ sites for more than 340 years’ worth of cumulative observation time and have uploaded ~2 million photos to our web platform. More recently, we have added the ability to upload and classify video clips.
Our article describes how data collected this way could inform a better understanding of mammal distribution, abundance, and temporal patterns.
Much has been said about what makes a successful and inclusive contributory citizen science project, such as in this recent review of “Citizen Science in Environmental and Ecological Sciences”. In my view, a critical thing for professional/institutional scientists is to treat and recognise all participants as collaborators, each bringing their own set of expertise to the project. For our article in Ecological Solutions and Evidence, we highlight a few innovations that set MammalWeb apart:
The first starts with bringing MammalWeb into the classroom. After a fruitful pilot with Belmont Community School where students co-authored a peer reviewed paper about their experiences and were featured in a short documentary, we worked with the British Ecological Society to provide engagement programmes and lent camera traps to a large network of primary schools. Experiencing nature proved especially valuable for students during the COVID-19 pandemic. Our education and outreach experience also informed the creation of “Mobile MammalWeb units” with touchscreens that have been taken to public spaces such as museums or even the New Scientist Live event at London’s O2 arena.
Another key mechanism is that the MammalWeb platform hosts camera trapping projects conducted by other organisations like conservation NGOs or even those initiated by citizen scientists. This greatly expands the awareness and reach of MammalWeb by decentralising the administrative burden of running one big national-scale project. It has also made it easier for MammalWeb to initiate international partnerships, such as with MammalNet to monitor mammals across several other European countries.
Lessons for biodiversity monitoring
In my opinion, what’s equally useful about our article is that we highlight ongoing challenges faced by citizen science biodiversity monitoring projects. This includes exploring ways to maintain engagement when crowdsourcing data collection; enhancing feedback for contributors through the aid of artificial intelligence technologies; or optimising camera trap deployment sites to help estimate wildlife abundance.
What’s particularly worrying to me is the lack of long-term funding to support established, ongoing projects. Many projects, including MammalWeb, rely on small, ad-hoc funding to build up their online platforms in a piecemeal fashion. This prevents higher-level planning and sustained impact. Instead of only chasing “novelty” and flashy outputs, more resources should be invested in the less glamorous, but equally critical maintenance of existing infrastructure and supporting people who do the work.
And to maximise the impact of our efforts, we also call for making online biodiversity monitoring platforms interoperable, easing the flow of data between them. When this is done with fully open source code, data, and hardware, it will be easier to pool our observations to achieve large-scale monitoring. Only with the open sharing and reuse of research outputs can we collaboratively address the incredible challenges facing our world today!
Read the full article: “Large-scale mammal monitoring: The potential of a citizen science camera-trapping project in the United Kingdom” in Issue 3:4 of Ecological Solutions and Evidence.
Blog post by Pen Yuan-Hsing, shared under the Creative Commons Attribution-ShareAlike 4.0 International license (CC BY-SA 4.0).
