The evaluation of large amounts of digital image data is of growing importance for biology, including for the exploration and monitoring of marine habitats. However, only a tiny percentage of the image data collected is evaluated by marine biologists who manually interpret and annotate the image contents, which can be slow and laborious. In order to overcome the bottleneck in image annotation, two strategies are increasingly proposed: “citizen science” and “machine learning”. In this study, we investigated how the combination of citizen science, to detect objects, and machine learning, to classify megafauna, could be used to automate annotation of underwater images. For this purpose, multiple large data sets of citizen science annotations with different degrees of common errors and inaccuracies observed in citizen science data were simulated by modifying “gold standard” annotations done by an experienced marine biologist. The parameters of the simulation were determined on the basis of two citizen science experiments. It allowed us to analyze the relationship between the outcome of a citizen science study and the quality of the classifications of a deep learning megafauna classifier. The results show great potential for combining citizen science with machine learning, provided that the participants are informed precisely about the annotation protocol. Inaccuracies in the position of the annotation had the most substantial influence on the classification accuracy, whereas the size of the marking and false positive detections had a smaller influence.
Citizen Science and Crowdsourcing
Marine stakeholder groups have diverse relationships with the ocean and life within it, which can create conflict and distrust between them. Citizen science and social licence present promising means to develop dialogue between these diverse marine stakeholders and improve outcomes for marine management. Citizen science can be defined as public engagement in scientific research and activities and amongst other benefits, has been demonstrated to improve communication and relationships amongst resource management and stakeholder groups. Social licence is a concept that reflects unwritten permission from the public for others to use and manage natural resources, and has become an important theme for development in the marine realm. We explore a case-study of the marine citizen science programme Redmap Australia, utilising a mixed-methods approach to understand community perceptions of other marine user groups. We explore how marine users legitimise one another, and how this relates to building relationships and developing social licence. Our results show that participation in citizen science can allow users to display their marine citizenship and shared concern about the marine environment, and that this can allow them to earn trust from other user groups. We conclude that participation in citizen science improves perceptions of trustworthiness and can enhance social licence for marine user groups, with positive implications for marine and coastal management. These outcomes provide fruitful insights on marine resource user groups' perceptions that can help to advise future developments in the growing fields of citizen science practice and citizen science research.
Long-term measurements are imperative to detect, understand, and predict changes in coastal biological communities, but can be both costly and difficult to implement. Here, we compare measurement methods used to document community structure and assess changes in marine systems, and explore potential applications in citizen science. The use of photographs for species identifications and monitoring has become a popular and useful data collection tool, but its use requires evaluation of its effectiveness in comparison to data collected from live examinations. We used settlement panels in San Francisco Bay, a well-studied and vital coastal ecosystem, to compare standardized measures of the invertebrate fouling community through examination of live organisms in the field and via photographs. Overall, our study found that live measurements were more accurate and better represented these marine communities, having higher richness, and diversity measurements than photographic measurements. However, photographic analyses accurately captured the relative abundances of some species and functional groups. We suggest that highly recognizable target taxa or broad scale comparisons of functional group composition are easily tracked through photographs and offer the best potential for research conducted by citizen scientists.
Public efforts to support proper use and preservation of Florida’s historic shipwrecks began in earnest in the late 1980s. One of the most successful and popular programs developed by the state is the Underwater Archaeological Preserve system. As part of this process, state archaeologists begin by working with sport divers and local governments to establish the Preserve. From this point forward, archaeologists utilize the submerged sites to facilitate hands-on, non-disturbance survey and documentation trainings for these and other groups. Using the same framework, the Florida Public Archaeology Network continues to engage the public’s interest in Florida’s shipwrecks and other submerged cultural heritage sites. This chapter describes the Submerged Sites Education and Archaeological Stewardship (SSEAS) program and the Heritage Awareness Diving Seminar (HADS), and how these programs are intended to encourage divers to become active in monitoring wreck sites and making their own discoveries, in the process producing information instead of simply consuming information. While initial results have been encouraging, they also provide a lesson for orienting collaborative programs to the needs of the audience, rather than only to the needs of archaeologists.
There is considerable scientific and societal concern about plastic pollution, which has resulted in citizen science projects to study the scale of the issue. Citizen science is a cost-effective way to gather data over a large geographical range while simultaneously raising public awareness on the problem. Because the experiences of researchers involved in these projects are not yet adequately covered, this paper presents the findings from ten semi-structured qualitative interviews with researchers leading a citizen science project on micro- or macroplastics. Our results show it is important to specify the goal(s) of the project and that expertise on communication and data science is needed. Furthermore, simple protocols, quality control, and engagement with volunteers and the public are key elements for successful projects. From these results, a framework with recommendations was drafted, which can be used by anyone who wants to develop or improve citizen science projects.
The democratization of ocean observation has the potential to add millions of observations every day. Though not a solution for all ocean monitoring needs, citizen scientists offer compelling examples showcasing their ability to augment and enhance traditional research and monitoring. Information they are providing is increasing the spatial and temporal frequency and duration of sampling, reducing time and labor costs for academic and government monitoring programs, providing hands-on STEM learning related to real-world issues and increasing public awareness and support for the scientific process. Examples provided here demonstrate the wide range of people who are already dramatically reducing gaps in our global observing network while at the same time providing unique opportunities to meaningfully engage in ocean observing and the research and conservation it supports. While there are still challenges to overcome before widespread inclusion in projects requiring scientific rigor, the growing organization of international citizen science associations is helping to reduce barriers. The case studies described support the idea that citizen scientists should be part of an effective global strategy for a sustained, multidisciplinary and integrated observing system.
Divers have widely participated in citizen science (CS) projects and are one of the main groups of marine citizen scientists. However, there is little knowledge about profiles of, and incentives for potential divers to join CS projects. To date, most studies have focused on the SCUBA diving industry; nevertheless, there is a diversity of divers, not all using SCUBA, who engage in different activities during their dives. Differences in diver profiles could affect their willingness and ability to contribute to CS. In this study, we compare the diving profile, interests, preferences and motivations to participate in CS of five diver types (artisanal fishermen, recreational divers, instructors, scientific divers, and others). All divers have strong interests in participating in CS projects, with no major differences among diver types. In general, they are interested in a wide variety of themes related to CS but they prefer simple sampling protocols. Divers are motivated to participate in CS to learn about the sea and contribute to science. Some important differences among diver types were found, with artisanal fishermen having significantly more dive experience than other diver types, but less free time during their dives and limited access to some communication channels and technologies. These characteristics make them ideal partners to contribute their local ecological knowledge (LEK) to local CS projects. In contrast, recreational divers have the least experience but most free time during their dives and good access to cameras and communications channels, making them suitable partners for large-scale CS projects that do not require a high level of species knowledge. Instructors and scientific divers are well-placed to coordinate and supervise CS activities. The results confirm that divers are not all alike and specific considerations have to be taken into account to improve the contribution of each diver type to CS. The findings provide essential information for the design of different types of CS projects. By considering the relevant incentives and opportunities for diverse diver groups, marine CS projects will make efficient gains in volunteer recruitment, retention, and collaborative generation of knowledge about the marine environment.
Long-term environmental records are among the most valuable assets for understanding the trajectory and consequences of climate change. Here we report on a newly recovered time-series from Project Oceanology, a non-profit ocean science organization serving New England schools (USA) since 1972. As part of its educational mission, Project Oceanology has routinely and consistently recorded water temperature, pH, and oxygen as well as invertebrate and fish abundance in nearshore waters of the Thames River estuary in eastern Long Island Sound (LIS). We digitized these long-term records to test for decadal trends in abiotic and biotic variables including shifts in species abundance, richness, and diversity. Consistent with previous studies, the data revealed an above-average warming rate of eastern LIS waters over the past four decades (+0.45 °C decade−1), a non-linear acidification trend twice the global average (−0.04 pH units decade−1), and a notable decline in whole water-column dissolved oxygen concentrations (−0.29 mg L−1 decade−1). Trawl catches between 1997 and 2016 suggested a significant decrease in overall species diversity and richness, declines in cold-water adapted species such as American lobster (Homarus americanus), rock crab (Cancer irroratus), and winter flounder (Pseudopleuronectes americanus), but concurrent increases in the warm-water decapod Libinia emarginata (spider crab). Our study confirmed that Long Island Sound is a rapidly changing urban estuary, while demonstrating the value of long-term observations made by citizen-scientists, educators, and other stakeholders.
This study measured spatial distribution of marine debris stranded on beaches in South Eleuthera, The Bahamas. Citizen science, fetch modeling, relative exposure index and predictive mapping were used to determine marine debris source and abundance. Citizen scientists quantified debris type and abundance on 16 beaches within three coastal exposures (The Atlantic Ocean, Great Bahama Bank and The Exuma Sound) in South Eleuthera. Marine debris, (~2.5 cm or larger) on each beach was monitored twice between March–May and September–November 2013 at the same locations using GPS. Approximately, 93% of all debris items were plastic with plastic fragments (≤2.5 cm) being the most common. There were spatial differences (p ≤ 0.0001) in plastic debris abundance between coastal exposures. Atlantic Ocean beaches had larger quantities of plastic debris by weight and by meter (m) of shoreline. Stranded plastic may be associated with Atlantic Ocean currents associated with leakage from the North Atlantic sub-tropical gyre.
Central to appropriate wildlife management is an effective monitoring program. Monitoring wildlife in urban environments offers unique challenges in the form of barriers, prohibited access and crime. It also, however, provides a unique opportunity to enlist residential communities in collecting data on distribution of a number of species. Opportunistic sightings data has its flaws, including the lack of data on species absences, and unequal sampling effort. Yet these data may still provide reliable information on the distribution of species and complement localized, hypothesis driven research. Where possible opportunistic sightings data should be validated against traditional methods to determine their value for long term monitoring programmes. We use Maxent to model citizen-reported sightings to determine whether sightings of Cape clawless otters (Aonyx capensis) can complement standardized river occupancy surveys to monitor an elusive, widely distributed species living within a fragmented urban/natural matrix. The drivers of otter presence and the predicted distribution of otters modelled from citizen sightings mirrored that provided by previously published results based on occupancy models in the same system, and highlighted further areas of suitable otter habitat and routes for dispersal. Involving citizens in the monitoring of the urban otter population complemented standardized occupancy surveys and provided additional benefits. In addition to alleviating the pressure on local authorities to allocate resources to routine monitoring, citizen involvement provides an opportunity to gather supplementary data on behaviour and/or threats to the species; shed light on the potential dispersal routes, and promote awareness and encourage coexistence with urban adapted wildlife.