Assessing the Limits: Citizen Science Data Accuracy in Underwater Monitoring

doi:10.21203/rs.3.rs-7177400/v1

Download PDF

Research Article

Assessing the Limits: Citizen Science Data Accuracy in Underwater Monitoring

https://doi.org/10.21203/rs.3.rs-7177400/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Citizen science has become a powerful tool for generating extensive ecological data and fostering public engagement in environmental monitoring and conservation initiatives. In marine environments, where monitoring is challenging due to logistical demands and financial constraints, engagement of recreational divers to citizen science projects offers a promising avenue for expanding data collection efforts over large spatial and time scales. Here, we evaluate the accuracy of ecological data collected by volunteer divers in the Mediterranean region by comparing their observations with those of expert scientists. Using standardized monitoring protocols across multiple sites, we assessed volunteers’ abilities to identify marine species, threats and impacts, estimate abundances, and correctly report species’ absence. Our results reveal considerable variability in the performance of volunteer divers, who frequently misidentified sessile taxa and fish species and failed to detect critical environmental stressors, such as tissue necrosis associated with mass mortality events. Diving experience and site familiarity were found to significantly influence the quality of information reported by non-experts, with experienced and site-acquainted divers demonstrating higher data accuracy. While citizen science has the potential to enrich ecological research and inform management and conservation initiatives, our results demonstrate that, without a prior training and proper data validation procedures, generated records often fail to reliably capture the complexity of variables required.

Citizen science, defined as the broad practice of including members of the public in scientific research (Dickinson et al. 2010; Roche et al. 2020), has become a popular and rapidly growing discipline in recent decades (Follett and Strezov 2015; Fraisl et al. 2022), offering numerous benefits to both participants and the scientific community (Bonney et al. 2009; Connors et al. 2012; Fraisl et al. 2020). The rising interest in participatory citizen science approaches stems partly from the growing need of large-scale, long-term monitoring datasets for policy and management (Conrad and Hilchey 2011; Hyder et al. 2015; McKinley et al. 2017). This is particularly the case when dealing with the vast marine areas where data collection over extensive marine areas is challenging and costly (Bauer-Civiello et al. 2018). Citizen scientists help to overcome the challenges posed by limited funding and available research effort (Delaney et al. 2008; Fritz et al. 2019) by generating large volumes of data across broad spatial and temporal scales (Poisson et al. 2020), providing scientists and resource managers a cost-effective way to expand their data collection capacity. Beyond its function as a tool for data collection, citizen science plays a multifaceted role in enhancing public awareness, facilitating experiential learning for participants, and strengthening societal engagement with scientific research and outreach activities (Bonney et al. 2014; Forrester et al. 2017; Turrini et al. 2018; Hecker et al. 2019).

Data collection and volunteer participation in monitoring of the marine environment present unique challenges. While both terrestrial and marine projects require volunteer training, marine projects usually have the additional requirements of swimming, snorkelling or SCUBA diving skills, along with the use of specialized equipment (e.g., underwater photography) (Goffredo et al. 2010; Gillett et al. 2012; Forrester et al. 2015). Recent decades have witnessed a surge in recreational divers (Cabral et al. 2025), prompting research programmes to recruit these divers as volunteers, leveraging their inherent interest in marine life. Consequently, citizen science initiatives encompass a broad spectrum of data collection efforts, ranging from the measurement of basic abiotic variables (e.g., seawater temperature) to the acquisition of complex biodiversity data involving multiple taxa, habitats and associated threats (Gerovasileiou et al. 2016; Kelly et al. 2020). Although marine citizen science projects constitute specialized applications with diverse scopes, focal subjects, and methodological approaches, they share the fundamental principle common to all citizen science initiatives: a commitment to generating reliable data to support scientific enquiry and inform policy-making processes (Forrester et al. 2015; van der Velde et al. 2017).

Despite the obvious merits of citizen science projects (Dickinson et al. 2010; Fraisl et al. 2022) scientists and decision-makers often question whether volunteers can produce the high-quality data required for rigorous scientific research (Kosmala et al. 2016; Aceves-Bueno et al. 2017; Brown and Williams 2019). Indeed, while several studies suggest that volunteers can perform comparably to professionals (Cox et al. 2012; Forrester et al. 2015; van der Velde et al. 2017), others report that volunteers produce highly variable and inaccurate data (Newman et al. 2010; Moyer-Horner et al. 2012; Bird et al. 2014; Ratnieks et al. 2016). These concerns about data quality underscore the need for volunteer training, expert validation, and systematic evaluation of citizen science data accuracy (Vermeiren et al. 2016; Aceves-Bueno et al. 2017; Figuerola-Ferrando et al. 2024).

To contribute to this discussion, we conducted a field exercise evaluating the performance of citizen scientists in compiling both quantitative and qualitative data, and then we compared their records to those of professional scientists. This assessment was carried out across three sites located at three Mediterranean countries, which offer the background to explore variations in performance under differing levels of complexity in data reporting both for the marine biota components (sessile vs. motile species) and also threat categories (e.g., easily recognised treats such as marine litter vs. complex impacts like necrosis/mortality events). Additionally, we investigated the influence of diver experience and site familiarity on data accuracy. While supportive of the involvement of citizens in marine data collection, our study raises serious concerns that hinder the broader adoption of citizen science datasets in the monitoring of marine species and the complex threats they face, without prior training and validation of their actual skills and reporting accuracy.

2.1. Study area, participant selection and diving procedures

Implementation of the monitoring protocols took place in three countries: France (Marseille), Greece (Zakynthos Island) and Türkiye (Ildır Bay/Çeşme) (Fig. 1). The study involves two monitoring exercises to evaluate effectiveness of citizen scientists in the identification of species and threats, and in implementing a standardized protocol, compared to professionals. The study also aimed to determine which factors were associated with performance. Monitoring sites and their associated coralligenous assemblages are described in detail by Çinar et al. (2020).

To assess the accuracy of data collected by volunteers, records from 27 participants were directly compared with those collected by expert scientists during several dives conducted at the three sites in July and September 2015, and from May to November 2016. Divers followed a standardized protocol designed to guide data collection and implementation (CIGESMED for divers – Gerovasileiou et al., 2016). This protocol captured an array of ecological and environmental information (e.g., coralligenous habitats and associated species, invasive species, human-induced pressures, marine litter) in addition to general site information (e.g., coordinates, observation depth, visibility, habitat continuity, slope, etc.). Prior to dives, participants received an electronic educational module (https://cs.cigesmed.eu/) containing guidelines, photographic material and detailed descriptions of the species and threats to be documented, along with general instructions on how to implement the protocol (Gerovasileiou et al. 2016). Local dive centers and expert scientists provided pre-dive briefings, during which participants could ask questions for clarification. During the dives, expert scientists accompanied one to five volunteer divers, without interfering with their observations. Divers and experts recorded observations at the same marked 2x2m segments on the seabed. For each dive, expert scientists generated an inventory of taxa and threats (with an abundance rating), which was then compared with the inventory created by each volunteer diver to assess data accuracy.

An additional performance assessment study involving 91 divers was conducted at the National Marine Park of Zakynthos, Greece. Participants implemented a standard underwater visual census protocol for fish species (Harmelin -Vivien et al. 1985) at fixed transects within popular diving sites during 2012 (Koester 2013). The purpose of this exercise was two-fold: (a) to assess the extent of false-positive species reporting by asking divers to confirm the presence of two fish species known to be absent from the area at the time (Sparus aurata and Lagocephalus spp.), and (b) to evaluate false-negative reporting by examining whether divers failed to report the presence of a very common and abundant invasive species (Siganus luridus), whose occurrence in the area was verified by experts through continuous monitoring. In the case of Sparus aurata, it was assumed that it could be easily misidentified as other morphologically similar species within the Sparidae family, such as Diplodus sargus or D. vulgaris. Conversely, it was assumed that the conspicuous morphological characteristics of Lagocephalus spp. would reduce the likelihood of misidentification. The protocol included a section for collecting background information of participants, such as diving experience and number of dives at the specific location, to determine which factors most strongly influenced performance.

2.2. Data analysis

Data collected on the abundance of both species and threats were transformed into ordinal categories: 0 – absent, 1 – limited, 2 - moderate, 3 – extended. These data were then converted to binary data (0 – absent, 1 – present). The presence/absence data allowed to determine whether divers could reliably identify species or threat occurrence, while species abundance estimation and threat intensity estimation demonstrated their ability to provide quantitative assessments. Species were further categorized by mobility status, creating two groups for both presence/absence and abundance data: mobile and sessile species.

Expert divers’ observations served as the baseline for comparison with volunteer divers’ data. With respect to presence/absence data analysis, we evaluated whether volunteer observations at each location were correct (true) or incorrect (false) compared to expert data. We calculated the frequency of true and false answers for each species or threat, and then we convert them into percentage success scores across all dives. Based on the data categorizing volunteer observations as true or false, we performed one-sided binomial tests to evaluate the ability of volunteers to accurately identify species and threats in the marine environment. The hypothesis tested was that true and false identifications occur with equal probability (p = 0.5). A p-value less than 0.05 indicates that volunteer divers performed significantly better than 50% accuracy, while a p-value greater than 0.05 suggests equal probability of correct and false identifications, indicating poor performance by volunteers. For abundance estimation analysis, we only included species/threat that was correctly identified. The filtered scores were grouped by category (motile species, sessile species, and threats) and a non-parametric paired Wilcoxon signed-rank test was conducted to compare volunteer and expert data.

For the performance assessment evaluating the false presence and absence of fish species, divers were tasked with labelling the three species (Sparus aurata, Lagocephalus spp., and Siganus luridus) as either present or absent. Based on expert-verified data regarding the true presence or absence of these species, each participant’s response was classified as false (0) or true (1). Success scores were calculated for each species based on the proportion of correct responses. Additionally, we recorded diver characteristics, including diving experience level (low = 1, high = 2) and number of dives performed at the specific location (low = 1 to high = 3). Using chi-squared tests, we examined relationships between diver success and their diving experience or site familiarity.

The accuracy of volunteer divers in threat identification ranged from 48–100%, with most threats catalogued in the protocol receiving scores above 70% (Fig. 2a). Lower scores were observed for sedimentation (63%) and necrosis/mortality events (48%), with p-values greater than 0.05 in their binomial tests, indicating that volunteers were unable to reliably identify these threats (Fig. 2a). For sessile species identification, volunteer divers achieved an average success score of 79%, with individual scores ranging from 38–100%. The lowest success scores for sessile species (Scleractinians: 48%, Axinella spp.: 56%, Cliona spp.: 38%, other erect bryozoans: 67%) corresponded with higher p-values on the binomial test, indicating poor volunteer performance (Fig. 2a). For the remaining 11 out of 15 sessile taxa, volunteers’ success scores exceeded 73%, whereas in only five of them detection success exceeded 90% (Fig. 2a). Motile species identification maintained high success rates overall (> 70%, average 80%), although the “other sea urchins” group (sea urchins other than the conspicuous hatpin urchin Centrostephanus longispinus, such as Arbacia lixula, Paracentrotus lividus, and Sphaerechinus granularis) recorded a lower success score of 59%, volunteer records for the latter group indicated insufficient identification accuracy (p-value > 0.05) (Fig. 2a).

For correctly identified species, abundance estimates of volunteer divers aligned closely with the assessment of expert scientists for both threats and motile species. Wilcoxon signed-rank tests (V = 4 for threats, V = 9 for motile species) revealed no statistically significant differences (p-value > 0.05) between the estimates of volunteer divers and expert scientists (Fig. 2b). However, for sessile species, volunteers showed a lower success rate in abundance estimation, with a statistically significant difference (p-value < 0.05) compared to expert assessments, typically underestimating abundance (Fig. 2b).

Divers frequently miss-reported the presence of Sparus aurata when it was absent at the diving sites (42% success score), similarly to the false absence reporting of Siganus luridus (49%) (Fig. 3). Volunteer divers performed better in the case of the morphologically conspicuous Lagochephalus spp., though the success score was still limited to 70%. Both diving experience and site familiarity (i.e. the number of dives conducted at the same site) significantly influenced accuracy (p-value < 0.05) (Fig. 2). Performance improved with diving experience, with scores increasing by 54% from beginners to experienced divers for S. aurata and by 58% for S.luridus. Site familiarity also improved accuracy: success score for S. aurata, rose from 5% with low dive frequency to 92% with high dive frequency (Fig. 3). Similarly, success score for S. luridus increased from 9% with low dive frequency to 82% with medium frequency, reaching 100% with high frequency. Accuracy for Lagocephalus spp. also, showed significant dependence (p-value < 0.05) on diving experience and site familiarity (Fig. 3).

Our findings revealed that citizen scientists, failed to achieve comparable levels of accuracy to those of experts when recording presence/absence data for various marine taxa and threats. Nevertheless, volunteer divers provided abundance estimates for threats and motile species that were aligned with expert evaluations. Significant discrepancies were observed in the volunteers’ estimations of sessile species abundance, and they were frequently susceptible to both false-positive and false-negative reporting, often misidentifying fish species that were either absent from or present at the dive sites. These outcomes underscore the importance of rigorously assessing data quality in citizen science projects or setting limits under which divers can be reliably involved in complementing professional scientific data gathering.

Divers demonstrated limited effectiveness in recognizing environmental stressors commonly included in several citizen science programs, such as instances of tissue necrosis linked to marine heatwaves and broader impacts of climate change—parameters that are critical for assessing ecosystem health and resilience (Starko et al. 2024; Trégarot et al. 2024). Conversely, volunteer divers demonstrated an improved performance in recording marine litter, anchoring, or discarded fishing gear, as these are relatively straightforward to detect and do not require specialized expertise. This likely highlights a limitation in the ability of citizen scientists to detect more complex environmental stressors or impacts that are not immediately apparent. These findings raise critical concerns regarding the potential boundaries and limitations of citizen science programmes in capturing complex ecological phenomena.

Disparities in accuracy emerged between sessile and motile species. For sessile taxa, volunteers reporting deviated significantly from those of experts in four out of the 15 taxa assessed. In contrast, discrepancies in motile taxa were primarily confined to “other sea urchins”. Abundance estimates for motile species were generally in agreement with expert scientists, whereas estimates for sessile taxa showed greater variability. These findings align with previous studies, which demonstrated that volunteer performance often varies by species type and may be influenced by individual interests (Branchini et al. 2015). For instance, macro photography enthusiasts might focus on benthic organisms, while those interested in larger species may overlook smaller or less conspicuous taxa (Meschini et al. 2021). Species detectability further complicates this dynamic, as less common or elusive species are inherently more challenging to identify, leading to more accurate data for common and easily recognizable taxa (Cox et al. 2012; Kosmala et al. 2016; Farr et al. 2023).

Our analyses revealed a significant dependence of volunteer data accuracy on participants’ diving experience and familiarity with dive sites. According to various studies (Hermoso et al. 2021; Lucrezi et al. 2018; Cerrano et al. 2017; Martin et al. 2016) experienced divers demonstrated greater precision, potentially due to enhanced confidence in their equipment and underwater skills, enabling them to concentrate more effectively on their surroundings. Additionally, familiarity with specific sites likely facilitated the recognition of local marine species and environmental features, thereby reducing errors and enhancing data reliability. We, therefore, encourage that specific diving skills and experience with the study sites should be used as the initial criteria for filtering citizen science collected data in demanding marine contexts.

Another important concern arises from the observation that divers were more prone to false-positive species identifications than to false negative ones. This tendency suggests a potential bias toward overreporting, which may reflect an eagerness to contribute or a lack of taxonomic certainty. Such patterns could compromise data reliability by inflating perceived biodiversity, thereby underscoring the need for improved training protocols and validation mechanisms within citizen science frameworks. Without rigorous validation methods, inconsistencies and observer bias can undermine the scientific credibility and usefulness of citizen science data, limiting its potential contribution to further applications such as ecological modelling, conservation planning, and policy development. Systematic evaluation not only helps identify potential sources of error but also guides the development of training protocols, data verification tools, and quality control mechanisms. This ultimately strengthens the role of citizen science in advancing scientific data production (Lukyanenko et al. 2016; Stevenson 2018; Anhalt-Depies et al. 2019).

For citizen science to make a meaningful contribution to ecological research, both the scientific community and the public must have confidence in the accuracy of the data. In fact, the proportion of published studies relying on citizen science data does not reflect the abundance and diversity of active citizen science programmes (Kullenberg and Kasperowski 2016; Davis et al. 2023), potentially due to concerns about data quality among peer reviewers (Theobald et al. 2015; Davis et al. 2023). Verification processes, whether applied to entire datasets or selected subsets, can enhance confidence in the reliability of citizen-collected data. However, implementing verification is not straightforward, often requiring means of comparison (e.g., images, videos or even the presence of experts along with citizens). Therefore, while verifying subsets of data may enable researchers to estimate error rates and identify additional sampling needs for hypothesis testing, the associated costs (time, effort, professional availability) raise questions about whether verification or comprehensive training for citizens prior to their involvement is the more effective approach. Ultimately, citizen science offers considerable prospects for advancing ecological and conservation research. Understanding, quantifying and eliminating biases in these data is an essential step towards their widespread application in addressing ecological questions and monitoring biodiversity.

Enhancing data quality in citizen science initiatives involves several strategies, including targeted training programmes, skill-based prequalification and on-going feedback for long-term participants (Kosmala et al. 2016; van der Wal et al. 2016). However, intensive training may inadvertently reduce participation, potentially limiting the educational and engagement benefits of such programs. Albeit, citizen scientists often produce data comparable to marine professionals (Forrester et al. 2015; van der Velde et al. 2017), they face greater challenges with specialized tasks, such as difficulty in species identification (Farr et al. 2023; Díaz-Calafat et al. 2024) or abundance estimation (Gillett et al. 2012; Done et al. 2017). Certain attributes are inherently more subjective or complex, further complicating data collection. Therefore, in light of the findings of our study, we caution that citizen science projects, and the users of the data collected under such projects, should carefully consider the trade-off between higher data reliability and broader public engagement.

The authors would like to thank the anonymous divers who participated in this study.

Author Contribution

C. D.: Methodology, Investigation, Formal analysis, Data curation, Conceptualization, Writing – review & editing, Writing – original draft, Supervision. K. K.: Visualization, Validation, Methodology, Writing – original draft, Writing – review & editing. M. E. Ç.: Investigation, Data curation, Writing – review & editing, Project administration. A. D.: Investigation, Data curation, Writing – review & editing. V. G.: Investigation, Data curation, Writing – review & editing. G. G.: Investigation, Data curation, Writing – review & editing. L. T. de V. d’A.: Investigation, Data curation, Writing – review & editing. M. S.: Investigation, Data curation, Writing – review & editing. K. S.: Investigation, Data curation, Writing – review & editing. A. D. M.: Validation, Methodology, Writing – original draft, Writing – review & editing. D. K.: Writing – review & editing, Project administration, Resources.

Acknowledgement

This study was carried out in the framework of CIGESMED project and was supported by: FRANCE – CNRS – ANR (Centre National de la Recherche Scientifique – Agence Nationale pour la Recherche) convention n° 12-SEAS-0001 – 01/LIGAMEN - ANR convention n° 12-SEAS-0001-02/IFREMER – ANR convention n° 12-SEAS-0001-03; GREECE – GSRT (General Secretariat for Research and Technology) 12SEAS-2-C2; TÜRKIYE – TUBITAK (Türkiye Bilimsel ve Teknolojik Araştırma Kurumu) project number: 112Y393. The authors have no conflicts of interest to declare.The authors would like to thank the anonymous divers who participated in this study.

Aceves-Bueno E, Adeleye AS, Feraud M., Huang Y, Tao M, Yang Y, Anderson SE (2017) The Accuracy of Citizen Science Data: A Quantitative Review. Bull Ecol Soc Am 98: 278–290. https://doi.org/10.1002/bes2.1336
Anhalt-Depies C, Stenglein JL, Zuckerberg B, Townsend PA, Rissman AR (2019) Tradeoffs and tools for data quality, privacy, transparency, and trust in citizen science. Biol Conserv 238: 108195. https://doi.org/10.1016/j.biocon.2019.108195
Bauer-Civiello A, Loder J, Hamann M (2018) Using citizen science data to assess the difference in marine debris loads on reefs in Queensland, Australia. Mar Pollut Bull 135: 458–465. https://doi.org/10.1016/j.marpolbul.2018.07.040
Bird TJ, Bates AE, Lefcheck JS, Hill NA, Thomson RJ, Edgar GJ, Stuart-Smith RD, Wotherspoon S, Krkosek M, Stuart-Smith JF, Pecl GT, Barrett N, Frusher S (2014) Statistical solutions for error and bias in global citizen science datasets. Biol Conserv 173: 144–154. https://doi.org/10.1016/j.biocon.2013.07.037
Bonney R, Cooper CB, Dickinson J, Kelling S, Phillips T, Rosenberg KV, Shirk J (2009) Citizen Science: A Developing Tool for Expanding Science Knowledge and Scientific Literacy. BioScience 59: 977–984. https://doi.org/10.1525/bio.2009.59.11.9
Bonney R, Shirk JL, Phillips TB, Wiggins A, Ballard HL, Miller-Rushing AJ, Parrish JK (2014) Next Steps for Citizen Science. Science 343: 1436–1437. https://doi.org/10.1126/science.1251554
Branchini S, Pensa F, Neri P, Tonucci BM, Mattielli L, Collavo A, Sillingardi ME, Piccinetti C, Zaccanti F, Goffredo S (2015) Using a citizen science program to monitor coral reef biodiversity through space and time. Biodivers Conserv 24: 319–336. https://doi.org/10.1007/s10531-014-0810-7
Brown ED, Williams BK (2019) The potential for citizen science to produce reliable and useful information in ecology. Conserv Biol 33: 561–569. https://doi.org/10.1111/cobi.13223
Cabral RB, Millage KD, Mayorga J, Kordell T, Kelkar M, Caughman AM, Favoretto F, Schuhbauer A, Aburto-Oropeza O, Sala E, Bradley D (2025) Marine protected areas for dive tourism. Sci Rep15:1923. https://doi.org/10.1038/s41598-024-83664-1
Cerrano C, Milanese M, Ponti M (2017) Diving for science - science for diving: volunteer scuba divers support science and conservation in the Mediterranean Sea. Aquat Conserv: Mar Freshw Ecosyst 27: 303–323. https://doi.org/10.1002/aqc.2663
Çinar ME, Feral J-P, Arvanitidis C, David R, Taşkin E, Sini M, Dailianis T, Doğan A, Gerovasileiou V, Evcen A, Chenuil A, Dağli E, Aysel V, Issaris Y, Bakir K, Nalmpanti M, Sartoretto S, Salomidi M, Sapouna A, Açik S, Dimitriadis C, Koutsoubas D, Katağan T, Öztürk B, Koçak F, Erdogan-Dereli D, Önen S, Özgen Ö, Türkçü N, Kirkim F, Önen M (2020) Coralligenous assemblages along their geographical distribution: testing of concepts and implications for management. Aquat Conserv: Mar Freshw Ecosyst 30: 1-17. https://doi.org/10.1002/aqc.3365
Connors JP, Lei S, Kelly M (2012) Citizen Science in the Age of Neogeography: Utilizing Volunteered Geographic Information for Environmental Monitoring. Ann Assoc Am Geogr 102: 1267–1289. https://doi.org/10.1080/00045608.2011.627058
Conrad CC, Hilchey KG (2011) A review of citizen science and community-based environmental monitoring: issues and opportunities. Environ Monit Assess 176: 273–291. https://doi.org/10.1007/s10661-010-1582-5
Cox TE, Philippoff J, Baumgartner E, Smith CM (2012) Expert variability provides perspective on the strengths and weaknesses of citizen-driven intertidal monitoring program. Ecol Appl 22: 1201–1212. https://doi.org/10.1890/11-1614.1
Davis LS, Zhu L, Finkler W (2023) Citizen Science: Is It Good Science? Sustainability 15: 4577. https://doi.org/10.3390/su15054577
Delaney DG, Sperling CD, Adams CS, Leung B (2008) Marine invasive species: validation of citizen science and implications for national monitoring networks. Biol Invasions 10: 117–128. https://doi.org/10.1007/s10530-007-9114-0
Díaz-Calafat J, Jaume-Ramis S, Soacha K, Álvarez A, Piera J (2024) Revealing biases in insect observations: A comparative analysis between academic and citizen science data. PLOS ONE 19: e0305757. https://doi.org/10.1371/journal.pone.0305757
Dickinson JL, Zuckerberg B, Bonter DN (2010) Citizen Science as an Ecological Research Tool: Challenges and Benefits. Annu Rev Ecol Evol Syst 41: 149–172. https://doi.org/10.1146/annurev-ecolsys-102209-144636
Done T, Roelfsema C, Harvey A, Schuller L, Hill J, Schläppy ML, Lea A, Bauer-Civiello A, Loder J (2017) Reliability and utility of citizen science reef monitoring data collected by Reef Check Australia, 2002–2015. Mar Pollut Bull 117: 148–155. https://doi.org/10.1016/j.marpolbul.2017.01.054
Farr CM, Ngo F, Olsen B (2023) Evaluating Data Quality and Changes in Species Identification in a Citizen Science Bird Monitoring Project. Citiz Sci: Theory Pract: 24. https://doi.org/10.5334/cstp.604
Figuerola-Ferrando L, Linares C, Zentner Y et al (2024) Marine Citizen Science and the Conservation of Mediterranean Corals: The Relevance of Training, Expert Validation, and Robust Sampling Protocols. Environ Manage 73: 646–656. https://doi.org/10.1007/s00267-023-01913-x
Follett R, Strezov V (2015) An Analysis of Citizen Science Based Research: Usage and Publication Patterns. PLOS ONE 10: e0143687. https://doi.org/10.1371/journal.pone.0143687
Forrester G, Baily P, Conetta D, Forrester L, Kintzing E, Jarecki L (2015) Comparing monitoring data collected by volunteers and professionals shows that citizen scientists can detect long-term change on coral reefs. J Nat Conserv 24: 1–9. https://doi.org/10.1016/j.jnc.2015.01.002
Forrester TD, Baker M, Costello R, Kays R, Parsons AW, McShea WJ (2017) Creating advocates for mammal conservation through citizen science. Biol Conserv 208: 98–105. https://doi.org/10.1016/j.biocon.2016.06.025
Fraisl D, Campbell J, See L, Wehn U, Wardlaw J, Gold M, Moorthy I, Arias R, Piera J, Oliver JL, Masó J, Penker M, Fritz S (2020) Mapping citizen science contributions to the UN sustainable development goals. Sustain Sci 15: 1735–1751. https://doi.org/10.1007/s11625-020-00833-7
Fraisl D, Hager G, Bedessem B, Gold M, Hsing PY, Danielsen F, Hitchcock CB, Hulbert JM, Piera J, Spiers H, Thiel M, Haklay M (2022) Citizen science in environmental and ecological sciences. Nat Rev Methods Primers 2: 64. https://doi.org/10.1038/s43586-022-00144-4
Fritz S, See L, Carlson T, Haklay M et al (2019) Citizen science and the United Nations Sustainable Development Goals. Nat Sustain 2: 922–930. https://doi.org/10.1038/s41893-019-0390-3
Gerovasileiou V, Dailianis T, Panteri E, Michalakis N, Gatti G, Sini M, Dimitriadis C, Issaris Y, Salomidi M, Filiopoulou I, Doğan A, Thierry de Ville d’Avray L, David R, Ҫinar ME, Koutsoubas D, Féral J-P, Arvanitidis C (2016) CIGESMED for divers: Establishing a citizen science initiative for the mapping and monitoring of coralligenous assemblages in the Mediterranean Sea. Biodivers Data J: 4. https://doi.org/10.3897/BDJ.4.e8692
Gillett DJ, Pondella DJ, Freiwald J, Schiff KC, Caselle JE, Shuman C, Weisberg SB (2012) Comparing volunteer and professionally collected monitoring data from the rocky subtidal reefs of Southern California, USA. Environ Monit Assess 184: 3239–3257. https://doi.org/10.1007/s10661-011-2185-5
Goffredo S, Pensa F, Neri P, Orlandi A, Gagliardi MS, Velardi A, Piccinetti C, Zaccanti F (2010) Unite research with what citizens do for fun: “recreational monitoring” of marine biodiversity. Ecol Appl 20: 2170–2187. https://doi.org/10.1890/09-1546.1
Harmelin-Vivien ML, Harmelin JG, Chauvet C, Mellon-Duval C, Galzin R, Lejeune P, Barnabé G, Blanc F, Chevalier R, Duclerc J, Lasserre G (1985) Evaluation visuelle des peuplements et populations de poissons : méthodes et problèmes. Revue d’Écologie (La Terre et La Vie) 40: 467–539. https://doi.org/10.3406/revec.1985.5297
Hecker S, Wicke N, Haklay M, Bonn A (2019) How Does Policy Conceptualise Citizen Science? A Qualitative Content Analysis of International Policy Documents. Citiz Sci: Theory Pract 4: 32. https://doi.org/10.5334/cstp.230
Hermoso MI, Martin VY, Gelcich S, Stotz W, Thiel M (2021) Exploring diversity and engagement of divers in citizen science: Insights for marine management and conservation. Mar Policy 124: 104316. https://doi.org/10.1016/j.marpol.2020.104316
Hyder K, Townhill B, Anderson LG, Delany J, Pinnegar JK (2015) Can citizen science contribute to the evidence-base that underpins marine policy? Mar Policy 59: 112–120. https://doi.org/10.1016/j.marpol.2015.04.022
Kelly R, Fleming A, Pecl GT, von Gönner J, Bonn A (2020) Citizen science and marine conservation: a global review. Philos Trans R Soc B Biol Sci 375: 20190461. https://doi.org/10.1098/rstb.2019.0461
Koester A (2013) Dive Tourism, Sustainable Management and Biodiversity Monitoring in Marine Protected Areas: The Case of the Nation al Marine Park of Zakynthos (SW Greece). Dissertation, Brandenburg University of Technology Cottbus
Kosmala M, Wiggins A, Swanson A, Simmons B (2016) Assessing data quality in citizen science. Front Ecol Environ 14: 551–560. https://doi.org/10.1002/fee.1436
Kullenberg C, Kasperowski D (2016) What Is Citizen Science? – A Scientometric Meta-Analysis. PLOS ONE 11: e0147152. https://doi.org/10.1371/journal.pone.0147152
Lucrezi S, Milanese M, Palma M, Cerrano C (2018) Stirring the strategic direction of scuba diving marine Citizen Science: A survey of active and potential participants. PLOS ONE 13: e0202484. https://doi.org/10.1371/journal.pone.0202484
Lukyanenko R, Parsons J, Wiersma YF (2016) Emerging problems of data quality in citizen science. Cons Biol 30: 447–449. https://doi.org/10.1111/cobi.12706
Martin V, Smith L, Bowling A, Christidis L, Lloyd D, Pecl G (2016) Citizens as Scientists. Sci Commun 38: 495–522. https://doi.org/10.1177/1075547016656191
McKinley DC, Miller-Rushing AJ, Ballard HL et al (2017) Citizen science can improve conservation science, natural resource management, and environmental protection. Biol Conserv 208: 15–28. https://doi.org/10.1016/j.biocon.2016.05.015
Meschini M, Machado Toffolo M et al (2021) Reliability of Data Collected by Volunteers: A Nine-Year Citizen Science Study in the Red Sea. Front Ecol Evol 9. https://doi.org/10.3389/fevo.2021.694258
Moyer-Horner L, Smith MM, Belt J (2012) Citizen science and observer variability during American pika surveys. J Wildl Manage 76: 1472–1479. https://doi.org/10.1002/jwmg.373
Newman G, Crall A, Laituri M, Graham J, Stohlgren T, Moore JC, Kodrich K, Holfelder KA (2010) Teaching Citizen Science Skills Online: Implications for Invasive Species Training Programs. Appl Environ Educ Commun 9: 276–286. https://doi.org/10.1080/1533015X.2010.530896
Poisson AC, McCullough IM, Cheruvelil KS, Elliott KC, Latimore JA, Soranno PA (2020). Quantifying the contribution of citizen science to broad-scale ecological databases. Front Ecol Environ 18: 19–26. https://doi.org/10.1002/fee.2128
Ratnieks FLW, Schrell F, Sheppard RC, Brown E, Bristow OE, Garbuzov M (2016) Data reliability in citizen science: learning curve and the effects of training method, volunteer background and experience on identification accuracy of insects visiting ivy flowers. Methods Ecol Evol 7: 1226–1235. https://doi.org/10.1111/2041-210X.12581
Roche J, Bell L, Galvão C, Golumbic YN, Kloetzer L, Knoben N, Laakso M, Lorke J, Mannion G, Massetti L, Mauchline A, Pata K, Ruck A, Taraba P, Winter S (2020) Citizen Science, Education, and Learning: Challenges and Opportunities. Front Sociol 5. https://doi.org/10.3389/fsoc.2020.613814
Starko S, van der Mheen M, Pessarrodona A, Wood GV, Filbee-Dexter K, Neufeld CJ, Montie S, Coleman MA, Wernberg T (2024) Impacts of marine heatwaves in coastal ecosystems depend on local environmental conditions. Glob Chang Biol 30. https://doi.org/10.1111/gcb.17469
Stevenson R (2018) A three-pronged strategy to improve trust in biodiversity data produced by citizen science programs. Biodivers Inf Sci Stand 2: e25838. https://doi.org/10.3897/biss.2.25838
Theobald EJ, Ettinger AK, Burgess HK, DeBey LB, Schmidt NR, Froehlich HE, Wagner C, HilleRisLambers J, Tewksbury J, Harsch MA, Parrish JK (2015) Global change and local solutions: Tapping the unrealized potential of citizen science for biodiversity research. Biol Conserv 181: 236–244. https://doi.org/10.1016/j.biocon.2014.10.021
Trégarot E, D’Olivo JP, Botelho AZ et al (2024) Effects of climate change on marine coastal ecosystems – A review to guide research and management. Biol Conserv 289: 110394. https://doi.org/10.1016/j.biocon.2023.110394
Turrini T, Dörler D, Richter A, Heigl F, Bonn A (2018) The threefold potential of environmental citizen science - Generating knowledge, creating learning opportunities and enabling civic participation. Biol Conserv 225: 176–186. https://doi.org/10.1016/j.biocon.2018.03.024
van der Velde T, Milton DA, Lawson TJ, Wilcox C, Lansdell M, Davis G, Perkins G, Hardesty BD (2017) Comparison of marine debris data collected by researchers and citizen scientists: Is citizen science data worth the effort? Biol Conserv 208: 127–138. https://doi.org/10.1016/j.biocon.2016.05.025
van der Wal R, Sharma N, Mellish C, Robinson A, Siddharthan A (2016) The role of automated feedback in training and retaining biological recorders for citizen science. Cons Biol 30: 550–561. https://doi.org/10.1111/cobi.12705
Vermeiren P, Munoz C, Zimmer M, Sheaves M (2016) Hierarchical toolbox: Ensuring scientific accuracy of citizen science for tropical coastal ecosystems. Ecol Indic 66: 242–250. https://doi.org/10.1016/j.ecolind.2016.01.031

No competing interests reported.

Download PDF

Editorial decision: Revision requested
20 Nov, 2025
Reviews received at journal
19 Nov, 2025
Reviews received at journal
04 Nov, 2025
Reviewers agreed at journal
04 Nov, 2025
Reviewers agreed at journal
27 Oct, 2025
Reviewers invited by journal
23 Oct, 2025
Editor assigned by journal
29 Jul, 2025
Submission checks completed at journal
23 Jul, 2025
First submitted to journal
21 Jul, 2025

You are reading this latest preprint version

Assessing the Limits: Citizen Science Data Accuracy in Underwater Monitoring

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods

2.1. Study area, participant selection and diving procedures

2.2. Data analysis

3. Results

4. Discussion

Declarations

Author Contribution

Acknowledgement

References

Additional Declarations

Status:

Version 1