Monitoring through many eyes: Integrating disparate datasets to improve monitoring of the Great Barrier Reef

Abstract

Numerous organisations collect data in the Great Barrier Reef (GBR), but they are rarely analysed together due to different program objectives, methods, and data quality. We developed a weighted spatio-temporal Bayesian model and used it to integrate image-based hard-coral data collected by professional and citizen scientists, who captured and/or classified underwater images. We used the model to predict coral cover across the GBR with estimates of uncertainty; thus filling gaps in space and time where no data exist. Additional data increased the model's predictive ability by 43%, but did not affect model inferences about pressures (e.g. bleaching and cyclone damage). Thus, effective integration of professional and high-volume citizen data could enhance the capacity and costefficiency of monitoring programs. This general approach is equally viable for other variables collected in the marine environment or other ecosystems; opening up new opportunities to integrate data and provide pathways for community engagement/stewardship.

Monitoring of Coral Reefs Using Artificial Intelligence: A Feasible and Cost-Effective Approach

“…Looking forward, this work comprises a specific network model for each country/region (Table 1) and a more advanced system, accounting for the geographic origin of the data, will facilitate its applications in global monitoring. Furthermore, propagating the classification uncertainty from the machine into statistical analyses will ensure a more robust integration and interpretation of monitoring data (e.g., [41]).…”

Coral Reef Community Changes in Karimunjawa National Park, Indonesia: Assessing the Efficacy of Management in the Face of Local and Global Stressors

“…Firstly, ecological monitoring data are not always sufficient in coverage or detail to confidently draw conclusions [38]. Reefs host complex benthic communities of diverse algae and invertebrate taxa, and in-water assessments relying on SCUBA can be time-consuming and require expertise for identification.…”

Correcting Misclassification Errors in Crowdsourced Ecological Data: A Bayesian Perspective

“…In addition, monitoring the GBR is especially challenging because it extends over 346, 000 km 2 and traditional marine surveys are expensive (Nichols & Williams, 2006;Nygård et al, 2016;Roelfsema & Phinn, 2010;Vercelloni et al, 2020). To address these issues, Peterson et al (2020) demonstrated how image-based coral cover data elicited from citizens can be combined with other professional data sources to improve the spatio-temporal data coverage in the GBR and increase the information gained to inform management. This approach was operationalised in the Virtual Reef Diver project using the weighted approach (https://www.virtu alreef.org.au/), without accounting for misclassification bias.…”

“…Approaches that pool or integrate CS elicited data with those obtained from professional monitoring programmes are gathering momentum (Peterson et al., 2020). This idea revolves around meta‐analysis principles and has been extensively studied in many areas including medicine, social sciences, and the environment (see Claggett et al., 2014; Higgins et al., 2009; Rice et al., 2018), as well as ecological settings (Koricheva et al., 2013).…”

“…The second approach is to use weighted linear regression (i.e. weighted approach) with observation weights proportional to the user’s accuracy or performance measures (Bird et al., 2014; Peterson et al., 2020). This ensures that data from users with poorer accuracy receive lower weights in the regression model.…”