The ocean plastic library of Caribbean Cleanup — an initiative of Hands for Honduras, Inc. · Back to the Flagship
Every claim on this site stands on published research. This library holds it — 80 studies and reports, each with its publisher, its year, and what it found. Push a shelf button to open that shelf. Push any study to read the original article.
Kikaki et al. 2020 and Healthy Reefs 2024 (entries 13, 20)
Cross-reference — see the entries named
the regional timescale: Caribbean-bound river plastic pulses every rainy season into a reef system already graded "poor" overall, so each decade of delay adds legacy damage to a reef with little resilience left.
7. THE SOLUTION, AND THE PROOF IT WORKS (19 entries)
System change works on paper — and its components work in measured practice:
existing
technologies and a system change (reduce, reuse, collect, recycle, safely dispose), the annual flow of plastic into the ocean can be cut by more than 80% by 2040.
Lau et al. 2020 (entry 47)
Cross-reference — see the entries named
the peer-reviewed version: all feasible interventions together cut pollution 40% versus 2016 rates and 78% versus 2040 business-as-usual; the modeled gains come overwhelmingly from upstream reduction, collection and disposal — not from ocean cleanup.
Removing the pressure pays — recovery and avoided-loss evidence:
NOAA Economic Loss page (entry 43)
Cross-reference — see the entries named
the local business case in reverse: near-zero debris scenarios are worth hundreds of millions in tourism, and derelict-gear removal is worth tens of millions in crab harvest, in single US regions alone.
but the grade remains "poor", and the improvement cannot be attributed to plastic removal specifically. No study was found that removed plastic alone and measured reef or fish recovery — see Claims Checked and Not Carried.
reducing plastic entering the ocean directly reduces disease-associated coral mortality
prevention is the intervention the disease science points to.
Flagged / Contested Claims — not settled facts
"More plastic than fish by 2050" (Ellen MacArthur Foundation/WEF 2016) — AAP FactCheck verdict: Unproven. It is a scenario comparing projected plastic weight (850–950 Mt) with an estimated fish biomass (812–899 Mt) that AAP says significantly underestimates global fish stocks. Present only as a business-as-usual warning scenario.
"Humans eat a credit card (5 g) of plastic a week" — Refuted on re-analysis: Pletz (2022) found "severe errors" and estimated ~one credit card per ~23,000 years; Mohamed Nor et al. (2021) modeled median intake in micrograms per week. The figure does not hold (entry 40).
Motagua = "2% of global ocean plastic" / "world's most polluting river" — A single-source operator/advocate claim built on The Ocean Cleanup's ~20,000 t/yr estimate. Tonnages conflict across sources (~8,000 t/yr Guatemala environment ministry via 4ocean; ≥8,500 t/yr waste via ICSF; 20,000 t/yr via Ocean Cleanup/Kia), and against Meijer's global riverine total (0.8–2.7 Mt/yr) the share is ~0.7–2.5%. In Meijer's peer-reviewed ranking the top river was the Pasig (Philippines), not the Motagua. A range is published with attribution, never the ranking.
Ghost gear "640,000 tonnes/yr" and "10% of ocean waste" — Legacy FAO/UNEP figures repeated in FAO's 2021 feature, but the underlying 2009 FAO/UNEP report states its data do not allow global estimates. Richardson et al.'s loss rates (6% of nets, 9% of traps, 29% of lines per year) are the defensible numbers.
"90% of seabirds have eaten plastic" — Wilcox et al. 2015 is a model-standardized estimate; Kühn & van Franeker (2020) explicitly warn that species-level records are routinely misworded as individual prevalence. It is quoted as a model estimate.
Population-level harm to marine megafauna — Senko et al. 2020: individual harm proven, direct population-level evidence absent over 50 years of study. Population declines caused by plastic alone are not proven.
Microplastics and human health — Established: exposure via food, water and air, and detection in stool, placenta (small pilot) and blood. Not established: any causal disease effect in humans at current environmental levels. WHO (2019): no apparent risk on limited data — not proof of safety.
River-concentration models disagree — Schmidt 2017 (top 10 rivers = 88–95%) vs Lebreton 2017 (top 20 = 67%) vs Meijer 2021 (>1,000 rivers = 80%). Use Meijer as the current peer-reviewed synthesis and disclose that these are models with wide ranges (global riverine totals span 0.41–4 Mt/yr across studies).
All headline money figures are models — $13bn/yr (UNEP 2014, authors call it an underestimate), $3,300–$33,000/metric ton (Beaumont, "conjectured"), $3.7tn lifetime (WWF/Dalberg, ±$1tn), $1.27tn/$3.25tn savings (UNEP 2023 scenario). Label method and range every time; none is a measured expenditure or a guaranteed saving.
Interception effectiveness is thinner than interception tonnage — Certified/operator tonnages (Mr. Trash Wheel, Interceptors) prove devices capture plastic; peer-reviewed, independently monitored capture-efficiency studies for river interceptors are scarce, and the Chao Phraya efficiency results are preliminary university-partner findings, not yet a peer-reviewed paper. Ocean Cleanup removal totals (~tens of thousands of metric tons to date, per secondary summaries) equal only a few percent of one year of river inflow — cleanup complements prevention; it cannot substitute for it.
Scheld et al. derelict-pot dollars — The peer-reviewed paper reports +$21.3m; NOAA reports +$33.5m for the wider program; a stock-assessment committee review disputed the methodology. The direction is the defensible claim; the dollar figures are contested (entry 78).
Everaert threshold nuance — The same risk assessment often cited for microplastic danger projects that average floating concentrations stay below its derived safe level to 2100. The irreversibility case rests on accumulation, fragmentation, sediments and hotspots — not on a claim that the whole ocean crosses a toxicity threshold by a set date. No source supports a single dated "point of no return" for ocean plastic; the defensible statement is MacLeod's: damage in accumulation zones becomes practically irreversible, and every year of delay adds stock that cannot be retrieved.
Claims Checked and Not Carried
No peer-reviewed head-to-head cost-per-metric ton comparison of river interception vs open-ocean cleanup was found. The comparative case (entries 66–68) is built from placement modeling (Sherman & van Sebille), a surface-cleanup impact model (Hohn et al.) and a system-dynamics cost model (Cordier & Uehara) — strong convergent evidence, but not a field trial with measured $/metric ton for both approaches.
No study found that removed plastic alone and measured ecosystem recovery (coral, fish, reef) attributable to that removal. Recovery evidence is indirect: derelict-pot removal and fishery gains (Scheld), reserve/disease correlations (Lamb), and a regional report-card improvement (Healthy Reefs 2024) that cannot be attributed to plastic.
Motagua tonnage and global share remain unresolved across operator, ministry and NGO sources (8,000–20,000 t/yr); no peer-reviewed Motagua-specific plastic flux measurement was found — Kikaki et al. 2020 is the closest peer-reviewed source and does not give a Motagua flux total.
Roatán (Honduras) mangrove litter study (Zuniga-Lopez et al., via JoVE Visualize: 3,417 items, 98.7% plastic, litter found in a deceased American crocodile) — the publication year was not shown in the source captured, so it is listed here rather than cited with a guessed year.
Thompson et al. 2004 ("Lost at Sea: Where Is All the Plastic?", Science — the paper that coined "microplastics") — no clean, verifiable URL was captured in the sources read, so it is not linked with a constructed URL.
Exact publication years for a few undated institutional web pages (UNEP riverine interactive, EEAS Honduras post, ICSF news item, 4ocean and EcoHubMap pages, TOMRA comparison, Mr. Trash Wheel operator page) are given as "n.d." — the pages themselves carry no visible date in the sources read.
Sources
All sources read in October 2026. [read in full] = the page was opened and read completely; all others are cited from the publishers' own listings. Full working notes are kept with the organization's research file.
