Search

Ecological data are collected over vast geographic areas using digital sensors such as camera traps and bioacoustic recorders. Camera traps have become the standard method for surveying many terrestrial mammals and birds, but camera trap arrays often generate millions of images that are time-consuming to label. This causes significant latency between data collection and subsequent inference, which impedes conservation at a time of ecological crisis. Machine learning algorithms have been developed to improve the speed of labelling camera trap data, but it is uncertain how the outputs of these models can be used in ecological analyses without secondary validation by a human. Here, we present our approach to developing, testing and applying a machine learning model to camera trap data for the purpose of achieving fully automated ecological analyses. As a case-study, we built a model to classify 26 Central African forest mammal and bird species (or groups). The model generalizes to new spatially and temporally independent data (n = 227 camera stations, n = 23,868 images), and outperforms humans in several respects (e.g. detecting ‘invisible’ animals). We demonstrate how ecologists can evaluate a machine learning model's precision and accuracy in an ecological context by comparing species richness, activity patterns (n = 4 species tested) and occupancy (n = 4 species tested) derived from machine learning labels with the same estimates derived from expert labels. Results show that fully automated species labels can be equivalent to expert labels when calculating species richness, activity patterns (n = 4 species tested) and estimating occupancy (n = 3 of 4 species tested) in a large, completely out-of-sample test dataset. Simple thresholding using the Softmax values (i.e. excluding ‘uncertain’ labels) improved the model's performance when calculating activity patterns and estimating occupancy but did not improve estimates of species richness. We conclude that, with adequate testing and evaluation in an ecological context, a machine learning model can generate labels for direct use in ecological analyses without the need for manual validation. We provide the user-community with a multi-platform, multi-language graphical user interface that can be used to run our model offline.

Ecological data are increasingly collected over vast geographic areas using arrays of digital sensors. Camera trap arrays have become the ‘gold standard’ method for surveying many terrestrial mammals and birds, but these arrays often generate millions of images that are challenging to process. This causes significant latency between data collection and subsequent inference, which can impede conservation at a time of ecological crisis. To address this, machine learning algorithms have been developed to improve data processing speeds, but these models are not considered accurate enough for fully automated labeling. Here, we present a new approach to building and testing a high performance machine learning model for fully automated labeling of camera trap images. As a case-study, the model classifies 26 Central African forest mammal and bird species (or groups). The model was trained on a relatively small dataset ( c .300,000 images) but generalizes to fully independent data and outperforms humans in several respects (e.g. detecting ‘invisible’ animals). We show how the model’s precision and accuracy can be evaluated in an ecological modeling context by comparing species richness, activity patterns and occupancy derived from machine learning labels with the same estimates derived from expert labels. Results show that fully automated labels can be equivalent to expert labels when calculating these widely-used ecological metrics. We provide the user-community with a multi-platform user interface for running the model offline, and conclude that high performance machine learning models can fully automate labeling of camera trap data. Significance statement Large-scale ecological monitoring can be used to detect ecosystem change. Ecological sensors such as camera traps are deployed across large spatial and temporal scales to monitor species and communities. Camera trap data are often vast (millions of images) and manual processing times cause significant latency between data collection and ecological inference. Existing machine learning models can reduce processing times but are rarely used in fully automated workflows for ecological analyses, mainly because users lack confidence in the model’s precision and accuracy. Here, we show a new, high performance machine learning model can be used to make ecological inference that is equivalent to using manually generated, expert labels. These results pave the way for large-scale, fully automated biodiversity monitoring and forecasting using camera trap arrays. ### Competing Interest Statement The authors have declared no competing interest.

Despite being heavily exploited, pangolins (Pholidota: Manidae) have been subject to limited research, resulting in a lack of reliable population estimates and standardised survey methods for the eight extant species. Camera trapping represents a unique opportunity for broad-scale collaborative species monitoring due to its largely non-discriminatory nature, which creates considerable volumes of data on a relatively wide range of species. This has the potential to shed light on the ecology of rare, cryptic and understudied taxa, with implications for conservation decision-making. We undertook a global analysis of available pangolin data from camera trapping studies across their range in Africa and Asia. Our aims were (1) to assess the utility of existing camera trapping efforts as a method for monitoring pangolin populations, and (2) to gain insights into the distribution and ecology of pangolins. We analysed data collated from 103 camera trap surveys undertaken across 22 countries that fell within the range of seven of the eight pangolin species, which yielded more than half a million trap nights and 888 pangolin encounters. We ran occupancy analyses on three species (Sunda pangolin Manis javanica, white-bellied pangolin Phataginus tricuspis and giant pangolin Smutsia gigantea). Detection probabilities varied with forest cover and levels of human influence for P. tricuspis, but were low ( < 0.05) for all species. Occupancy was associated with distance from rivers for M. javanica and S. gigantea, elevation for P. tricuspis and S. gigantea, forest cover for P. tricuspis and protected area status for M. javanica and P. tricuspis. We conclude that camera traps are suitable for the detection of pangolins and large-scale assessment of their distributions. However, the trapping effort required to monitor populations at any given study site using existing methods appears prohibitively high. This may change in the future should anticipated technological and methodological advances in camera trapping facilitate greater sampling efforts and/or higher probabilities of detection. In particular, targeted camera placement for pangolins is likely to make pangolin monitoring more feasible with moderate sampling efforts.

First paragraph: Camera trapping is now firmly established as one of the most useful tools with which to study biodiversity, and terrestrial mammals in particular. This themed issue brings together reviews and empirical studies that illustrate the breadth and depth of camera trap applications, with a focus on the African continent.