By Chen Zhu & Wande Li in the Thousand Island Lake

Biodiversity monitoring is the process of determining status and tracking changes in living organisms. Effective biodiversity monitoring needs a reliable tool in the field. When we go into the forest, many ecological processes (e.g., animal-mediated seed dispersal) are occurring above us because there are many animals living in the arboreal habitats. However, it’s not easy to find them and know what they’re doing. Arboreal camera trapping provides us an avenue to record arboreal animals and track their activities.

Fleshy fruits are very attractive to frugivorous animals as an important food resource. The frugivory mediated by animals can help plants to spread seeds, and has a positive effect on the maintenance of plant and animal communities. Traditionally, arboreal animals have been inventoried using ground-based direct observation. However, given the success of camera traps in detecting secretive terrestrial animals, we thus expected camera trapping could also be a valuable tool for inventorying arboreal frugivore species at monitoring sites with high food densities in the trees. Compared to the direct observation, we think arboreal camera trapping is a non-invasive, cost-effective and relatively labor-saving technique for collecting interaction data between plant and frugivores. Because we can’t directly observe a target tree 24 hours a day and night, and some cryptic species may avoid visiting the target tree due to the existence of observers. Furthermore, our study is conducted in a fragmented landscape, it’s very inconvenient to travel among islands in this artificial reservoir. Arboreal camera trapping makes us record plant-frugivore interactions remotely simultaneously.

Next, we wanted to test whether this method is a reliable tool to record plant-frugivore interactions. We also didn’t know whether we could take a relatively complete frugivore inventory. Knowledge starts with practice. We set above 300 camera traps in fruits trees on 22 islands to record frugivore species. And we selected one of the islands to test the sampling completeness.

After retrieving all cameras, we checked all photos and videos. We found arboreal camera trapping had recorded many frugivorous birds and their foraging behaviors, including a ground-foraging bird species, Sliver pheasant(Lophura nycthemera). Many arboreal mammals like red-bellied squirrel (Callosciurus erythraeus), masked civet (Paguma larvata) and some rodents (Muridae) were also recorded. In this way, we think arboreal camera trapping has much potential to be a reliable tool for monitoring species interactions in the arboreal habitats.

From the sampling completeness, we suggest that after monitoring approximately a third of the fruit maturation periods, a complete frugivorous species inventory can be achieved. We have recommended an effective sampling method according to fruit production and fruiting periods. Our aim is that to sampling plant-frugivore interactions adequately and effectively in the patch level. Arboreal camera trapping can realize it.

However, we found some problems in the field, like any other study about camera trapping. But most problems could be identified and solved at the first routine check. Hence, we kept our routine check about once a fortnight to kept cameras running until the end of the fruit period. Furthermore, we proposed a framework to classify interaction types for the first time, which will help us better understand potential plant-animal interactions occurring on a focal island.

The aim of the current study was to build unbiased frugivory and seed-dispersal networks on different islands. We think that this technology will make our ongoing work easier and more accessible. More importantly, we are very glad to introduce arboreal camera trapping to many researchers who are interested in frugivory or other ecological processes in the arboreal habitats.

Read the paper here