Sea Turtle Trends: Understanding Patterns with Predictive Models

Dutch below

In their 2019 paper, Sea Turtle Conservation Bonaire (STCB) used sea turtle data collected between 2003 and 2018 to estimate abundance and predict future population trends for green and hawksbill turtles on Bonaire’s west coast. The non-profit organization has now expanded that research in their latest paper to determine the population trend between 2019 and 2022.

Hawksbill turtle. Photo credit: Brenda Kirkby

Sea turtles are an iconic species that face a wide variety of threats.  Since these species can travel vast distances, they are susceptible to a number of transboundary threats, including climate change and poaching as well as local concerns such as pollution and habitat loss.  Monitoring nesting and foraging grounds is a critical part in understanding long-term sea turtle trends throughout the Caribbean.

Unfortunately, sea turtles can be difficult to accurately monitor due to issues with detection and hence incomplete counts. To combat this, research statisticians have developed methods to estimate detection and population size, allowing for more realistic counts. STCB has been working many years to monitor local turtle populations, and in collaboration with Frank Rivera-Milan, to determine trends and test the accuracy of prediction models. For this particular research, the goal was to determine whether the population increased, decreased on remained the same between 2019 and 2022, compared to 2003-2018 and predictions for 2019-2030.


Between 2019 and 2022, surveys were conducted along western Bonaire and Klein Bonaire. In total 703 green turtles and 56 hawksbill turtles were recorded.  Researchers also looked at how easy it was to spot the turtles, and they found that some factors, like the substrate (e.g., sand, rubble) or the level of disturbance (e.g., boats, divers), influenced the number of turtles that were detected.

Green turtle. Photo credit: Brenda Kirkby

When comparing historical data to the prediction data, this study yielded interesting results.  For green turtles, the number of turtles counted between 2019 and 2022 did not change much but was comparatively lower than data collected from 2003-2018 and lower than estimates predicted in the 2019-2030 model.  The number of hawksbill turtles, on the other hand, fluctuated between 2019 and 2022 but was similar to previous surveys (conducted between 2003-2018) and was closer to the predictive models.


The study’s density estimates aligned with previous research on sea turtles in other locations, and the researchers emphasized the importance of accounting for detection when estimating population numbers. This research concluded that this methodology provided reliable detection and population estimates for monitoring sea turtles within foraging grounds in the Caribbean, therefore this approach could be valuable for similar studies in coastal areas.


The Dutch Caribbean Nature Alliance (DCNA) supports science communication and outreach in the Dutch Caribbean region by making nature related scientific information more widely available through amongst others the Dutch Caribbean Biodiversity Database, DCNA’s news platform BioNews and through the press. This article contains the results of one of those scientific studies, but this study is not a DCNA study. No rights can be derived from the content. DCNA is not liable for the content and the in(direct) impacts resulting from publishing this article.



In hun wetenschappelijk artikel uit 2019 gebruikte Sea Turtle Conservation Bonaire (STCB) gegevens over zeeschildpadden die tussen 2003 en 2018 waren verzameld om de populatie te schatten en toekomstige populatietrends voor groene en karetschildpadden aan de westkust van Bonaire te voorspellen. STCB heeft dat onderzoek nu uitgebreid in hun nieuwste wetenschappelijk artikel om de populatietrend tussen 2019 en 2022 te bepalen.

Karetschildpadden. Photo credit: Brenda Kirkby

Zeeschildpadden zijn een iconische soort die met een breed scala aan bedreigingen wordt geconfronteerd. Omdat deze soorten grote afstanden kunnen afleggen, zijn ze vatbaar voor een aantal grensoverschrijdende bedreigingen, waaronder klimaatverandering en stroperij, maar ook voor lokale problemen zoals vervuiling en verlies van leefgebied. Het monitoren van broed- en foerageergebieden is een cruciaal onderdeel om inzicht te krijgen in de langetermijntrends van zeeschildpadden in het Caribisch gebied.

Helaas kunnen zeeschildpadden moeilijk nauwkeurig te volgen zijn vanwege problemen met detectie en dus onvolledige tellingen. Om dit tegen te gaan, hebben onderzoeksstatistici methoden ontwikkeld om de detectie en de populatiegrootte te schatten, waardoor meer realistische tellingen mogelijk zijn. STCB werkt al vele jaren aan het monitoren van lokale schildpaddenpopulaties, en in samenwerking met Frank Rivera-Milan, aan het bepalen van trends en het testen van de nauwkeurigheid van voorspellingsmodellen. Voor dit specifieke onderzoek was het doel om te bepalen of de bevolking toenam, daalde of gelijk bleef tussen 2019 en 2022, vergeleken met 2003-2018 en voorspellingen voor 2019-2030.


Tussen 2019 en 2022 zijn tellingen uitgevoerd langs westelijk Bonaire en Klein Bonaire. In totaal werden 703 groene schildpadden en 56 karetschildpadden geregistreerd. Onderzoekers keken ook naar hoe gemakkelijk het was om de schildpadden te spotten, en ze ontdekten dat sommige factoren, zoals de ondergrond (bijvoorbeeld zand, puin) of de mate van verstoring (bijvoorbeeld boten, duikers), van invloed waren op het aantal schildpadden dat werd gedetecteerd.

Groene schildpad. Photo credit: Brenda Kirkby

Bij het vergelijken van historische gegevens met de voorspellingsgegevens leverde dit onderzoek interessante resultaten op. Voor groene schildpadden veranderde het aantal getelde schildpadden tussen 2019 en 2022 niet veel, maar was relatief lager dan de gegevens verzameld van 2003-2018 en lager dan de schattingen die in het 2019-2030-model waren voorspeld. Het aantal karetschildpadden fluctueerde daarentegen tussen 2019 en 2022, maar was vergelijkbaar met eerdere onderzoeken (uitgevoerd tussen 2003-2018) en lag dichter bij de voorspellende modellen.


De schattingen van de dichtheid tijdens de studie kwamen overeen met eerder onderzoek naar zeeschildpadden op andere locaties, en de onderzoekers benadrukten het belang van het verantwoorden van detectie bij het schatten van populatieaantallen. Dit onderzoek concludeerde dat deze methodologie betrouwbare detectie- en populatieschattingen opleverde voor het monitoren van zeeschildpadden in foerageergebieden in het Caribisch gebied, daarom zou deze aanpak waardevol kunnen zijn voor soortgelijke studies in kustgebieden.


De Dutch Caribbean Nature Alliance (DCNA) ondersteunt wetenschapscommunicatie en outreach in de Nederlandse Caribische regio door natuurgerelateerde wetenschappelijke informatie breder beschikbaar te maken via onder meer de Dutch Caribbean Biodiversity Database, DCNA’s nieuwsplatform BioNews en via de pers. Dit artikel bevat de resultaten van een van die wetenschappelijke onderzoeken, maar dit onderzoek is geen DCNA-onderzoek. Aan de inhoud kunnen geen rechten worden ontleend. DCNA is niet aansprakelijk voor de inhoud en de indirecte gevolgen die voortvloeien uit het publiceren van dit artikel.




Published in BioNews 67

Data type
Other resources
Education and outreach
Geographic location

Chapter 12. Status and trends of coral reefs of the Caribbean region

Regional Context:

The Caribbean Region represents only 1% of Earth’s marine surface but hosts 10% of the world’s coral reefs, including fringing reefs, which are most common, barrier reefs such as the Mesoamerican Reef, which is the largest barrier reef in the Western Hemisphere, bank reefs, patch reefs, and a few atolls.

Caribbean shallow and mesophotic reefs are characterized by relatively low coral species diversity (70 hard coral species including two Acroporid species: Acropora palmata and A. cervicornis) and high levels of endemism, making them unique among the world’s reefs.

The physical geography of the Caribbean region is also complex with continental coasts (north, central, and south America), large continental islands (Greater Antilles), numerous small sandy islands (The Bahamas), volcanic islands (most of the Lesser Antilles), and coral islands (some Lesser Antilles islands).

The Caribbean is politically and culturally diverse with 30 sovereign states (continental and insular) and 16 European overseas territories or outermost regions (British, Dutch, and French), and considerable economic disparities between nations (e.g. per capita Gross Domestic Product in the USA was USD63,544 compared with less than USD1,200 in Haïti)1 .

About 70% of people in the Caribbean live near the coast. Indeed, Caribbean economies depend heavily on coral reefs and associated ecosystems (seagrasses and mangroves) for recreation and tourism (e.g., sandy beaches, snorkeling, and SCUBA diving), livelihoods, food (e.g., fishes, queen conch, lobsters), and other social, cultural, and economic benefits. Socio-economic monitoring (SocMon) in the Caribbean region, carried out largely according to the GCRMN SocMon protocol, is in use as an approach for coral reef managers and provides valuable insights on how coastal communities value and depend on coral reefs. Thus, SocoMon assessments have been conducted for almost 20 years in the region, including a series of workshops conducted recently beginning in 2016 (Jamaica) to the most recent in 2019 (MesoAmerica) by SPAW-RAC and supported by a NFWF-funded project to develop and refine a set of integrated coral reef monitoring guidelines that explicitly include human dimensions characteristics. For a detailed analysis of the SocMon Caribbean socio-economic assessments, please see the Global SocMon report that is forthcoming in 2022.

Socio-economic monitoring is important in order to understand the human interactions with coral ecosystems so that we can mitigate negative effects to coral reefs while promoting positive benefits that reefs provide []. SocMon has been part of the wider GCRMN effort since 1997 and was developed with the intent for socio-economic monitoring to complement biophysical monitoring. While SocMon data are not included in the present analysis, future work should and will seek to integrate Caribbean node socio-economic data with biophysical data.

The Caribbean is divided into 10 Marine Ecoregions of the World (MEOW) Ecoregions2 that were grouped into five subregions for the analyses underpinning this report (Tab. 1). There are coral reef marine protected areas (MPAs) in many countries in the Caribbean, as well as MPA networks such as MPAConnect and CaMPAM. The MPAs are usually small and generally located in nearshore areas. Efforts to support coral monitoring and capacity-building are underway with support from partner organisations such as the UN Environment Programme/ Cartagena Convention Secretariat, the National Oceanic and Atmospheric Administration (United States of America), the Gulf and Caribbean Fisheries Institute, the Specially Protected Areas and Wildlife protocol and its regional activity center (SPAWRAC), through regional projects and via multi-national programmes. MPA financing, enforcement, fisheries management, monitoring and communications are among the top management capacity building needs identified by coral reef managers to implement effective marine protection.

Data type
Research report
Research and monitoring
Geographic location
Saba bank
St. Eustatius
St. Maarten