Trend AnalysisEnvironment & Earth Sciences

Citizen Science for Invasive Species: Can Crowdsourced Monitoring Scale to Global Threats?

Platforms like iNaturalist and Pl@ntNet have mobilized millions of non-expert observers for biodiversity monitoring. Scientometric analysis shows citizen science publications in aquatic ecology doubled between 2019 and 2024โ€”but data quality, taxonomic bias, and geographic gaps limit scientific utility.

By Sean K.S. Shin
This blog summarizes research trends based on published paper abstracts. Specific numbers or findings may contain inaccuracies. For scholarly rigor, always consult the original papers cited in each post.

Lameira, Guerrero-Moreno & Cruz da Silva (2025), with 8 citations, provide a scientometric analysis of citizen science in aquatic ecology, mapping publication trends, geographic distribution, and thematic clusters across 185 articles. Key findings:

  • Publication volume doubled between 2019 and 2024, reflecting both genuine growth in citizen science programs and increased academic interest in the approach. - Geographic concentration: The United States, Australia, and the United Kingdom are the most productive countries in citizen science aquatic ecology research. Africa, South America, and much of Asia are severely underrepresentedโ€”regions where monitoring capacity is most limited and where citizen science could add the greatest value. - Taxonomic bias: Fish (30.64%) and aquatic mammals (13.87%) dominate citizen science records. Invertebrates, algae, and microorganismsโ€”often more ecologically informative indicators of ecosystem healthโ€”are poorly represented because they require taxonomic expertise that most volunteers lack. - Thematic clusters: The largest research cluster concerns invasive species detection; the second concerns water quality monitoring; the third concerns threatened species surveys. ### Integration with High-Tech Methods
Boyse, Clark & Carr (2025), with 3 citations, deploy eDNA metabarcoding to expand monitoring capacity for potential invasive species along the Northwest Passage in Arctic Canada. Their study uses environmental DNA samplingโ€”collecting and analyzing water samples for traces of species DNAโ€”rather than visual observation or citizen science reporting. Key findings:

  • eDNA metabarcoding detected a broad range of taxa across sampling sites, including species not previously documented in the region through conventional surveys. - The approach is particularly valuable in remote Arctic waters where professional survey teams can visit only infrequently and where visual detection of aquatic species is difficult. - The authors note that integrating eDNA methods with community-based monitoring (such as citizen science) could further expand coverage in the future, but this integration was not tested in the current study. The study highlights the potential of molecular methods to fill critical monitoring gaps in understudied regions. By deploying eDNA sampling along the Northwest Passage, the researchers established baseline biodiversity data that can support early detection of invasive species as Arctic waters warm and shipping traffic increases. ### Data Quality: The Central Debate
Hsu, Kang & Chang (2025), with 2 citations, address the data quality concern directly through a community-based citizen science program on the Taoyuan Algal Reef in Taiwan. By comparing citizen-generated species records against expert surveys conducted simultaneously, they demonstrate that community-trained observers achieve species identification accuracy of 85โ€“the vast majority for common taxaโ€”sufficient for monitoring trends in species richness and community composition, though insufficient for taxonomic research requiring definitive identification. The key finding is that data quality depends not on whether observers are professionals, but on training quality, protocol standardization, and verification systems. Programs that provide species identification guides, use structured data collection protocols, and implement expert review of submitted records achieve substantially higher data quality than open-submission platforms. Bonnet, Affouard & Chouet (2025) raise an emerging challenge: as AI-powered identification tools (Pl@ntNet, iNaturalist's computer vision) become more accurate, they generate enormous volumes of species records. But using these AI-assisted records as evidenceโ€”for regulatory enforcement, environmental impact assessment, or scientific publicationโ€”raises questions about provenance, reproducibility, and accountability that the citizen science community has not yet resolved. ## Critical Analysis: Claims and Evidence

<
ClaimEvidenceVerdict
Citizen science publications in aquatic ecology doubled 2019โ€“2024Lameira et al. scientometric analysisโœ… Supported
eDNA metabarcoding expands invasive species monitoring in remote Arctic watersBoyse et al. Northwest Passage eDNA studyโœ… Supported โ€” single study, specific context
Trained community observers achieve high species ID accuracy (comparable to expert accuracy in controlled comparisons)Hsu et al. Taiwan reef comparisonโœ… Supported โ€” for common taxa
Citizen science eliminates need for professional monitoringNone of the reviewed studies make this claimโŒ Refuted โ€” complement, not replacement
Geographic coverage is globally equitableLameira et al.: US, Australia, UK dominateโŒ Refuted โ€” severe Global South underrepresentation

The Motivation Problem

A challenge underaddressed in the reviewed literature is volunteer retention. Citizen science programs typically experience high initial enrollment followed by rapid dropout. Most iNaturalist users contribute fewer than 10 observations; most eBird users stop reporting within a year. For monitoring programs that require consistent data across years and locations, this dropout rate creates temporal gaps that undermine trend detection. Programs that maintain volunteer engagement tend to share common features: visible impact (volunteers see their data used in management decisions), social community (regular meetups, online discussion groups), and personal benefit (learning about local nature, contributing to place-based conservation). Programs that treat volunteers as unpaid data collectorsโ€”without feedback, recognition, or communityโ€”struggle with retention regardless of the scientific value of their data. ## Open Questions and Future Directions

  • AI + citizen science integration: Can AI identification tools serve as real-time quality filters for citizen-submitted records, flagging uncertain identifications for expert review while automatically validating confident ones? 2. Incentive design: What motivational structures sustain long-term volunteer participation? Financial incentives, gamification, social recognition, and educational outcomes each have different effectiveness profiles. 3. Global South expansion: How can citizen science programs be designed for contexts with limited internet connectivity, different cultural attitudes toward volunteerism, and competing livelihood demands on potential participants' time? 4. Regulatory acceptance: Under what conditions should environmental regulators accept citizen-generated data as legally sufficient evidence for species presence, habitat condition, or environmental compliance? 5. Integration with formal monitoring: Rather than treating citizen science as a standalone program, how should national monitoring agencies integrate citizen data into their existing frameworks? ## Implications for Researchers and Conservation Managers
  • Citizen science for invasive species monitoring has moved past the question of "does it work?" to "under what conditions and for what purposes?" For conservation managers, the practical value is clear: citizen networks extend spatial and temporal monitoring coverage at costs that professional surveys cannot match. For program designers, the evidence argues for structured programs with training, protocols, and verificationโ€”rather than open-submission platforms where data quality varies widely. For researchers, citizen science data is valuable for detecting distribution patterns and temporal trends but should be treated with appropriate caution for abundance estimation and fine-grained ecological analysis. The integration of citizen observation with high-tech methods (eDNA, remote sensing, AI identification) represents the productive frontierโ€”combining the spatial reach of human observers with the taxonomic precision of molecular and computational tools. ## References

    [1] Lameira, H.L.N., Guerrero-Moreno, M.A. & Cruz da Silva, E. (2025). Citizen Science as a Monitoring Tool in Aquatic Ecology: Trends, Gaps, and Future Perspectives. Sustainability, 17(11), 4972. https://doi.org/10.3390/su17114972

    [2] Boyse, E., Clark, M. & Carr, I.M. (2025). Expanding Monitoring Capacity for Potential Invasive Species in Arctic Canada With Environmental DNA Metabarcoding. Global Change Biology, 31, e70452. https://doi.org/10.1111/gcb.70452

    [3] Hsu, C., Kang, J. & Chang, Y.-M. (2025). Reliable Data From Community-Based Citizen Science for Coastal Biodiversity Research in the Taoyuan Algal Reef, Taiwan. Aquatic Conservation, 35, e70138. https://doi.org/10.1002/aqc.70138

    [4] Bonnet, P., Affouard, A. & Chouet, M. (2025). Challenges in Using AI-Based Citizen-Generated Plant Observations as Forensic Evidence in Biodiversity Investigations. Biodiversity Information Science and Standards, 9, 181619. https://doi.org/10.3897/biss.9.181619

    References (4)

    [1] Lameira, H.L.N., Guerrero-Moreno, M.A. & Cruz da Silva, E. (2025). Citizen Science as a Monitoring Tool in Aquatic Ecology: Trends, Gaps, and Future Perspectives. Sustainability, 17(11), 4972.
    [2] Boyse, E., Clark, M. & Carr, I.M. (2025). Expanding Monitoring Capacity for Potential Invasive Species in Arctic Canada With Environmental DNA Metabarcoding. Global Change Biology, 31, e70452.
    [3] Hsu, C., Kang, J. & Chang, Y.-M. (2025). Reliable Data From Community-Based Citizen Science for Coastal Biodiversity Research in the Taoyuan Algal Reef, Taiwan. Aquatic Conservation, 35, e70138.
    [4] Bonnet, P., Affouard, A. & Chouet, M. (2025). Challenges in Using AI-Based Citizen-Generated Plant Observations as Forensic Evidence in Biodiversity Investigations. Biodiversity Information Science and Standards, 9, 181619.

    Explore this topic deeper

    Search 290M+ papers, detect research gaps, and find what hasn't been studied yet.

    Click to remove unwanted keywords

    Search 8 keywords โ†’