Population estimates of photo-identified individuals using a modified POPAN model reveal that Raja Ampat’s reef manta rays are thriving

Abstract

The 6.7-million-hectare Raja Ampat archipelago is home to Indonesia’s largest reef manta ray (Mobula alfredi) population and a representative network of nine marine protected areas (MPAs). However, the population dynamics of M. alfredi in the region are still largely unknown. Using our photo-identification database, we fitted modified POPAN mark-recapture models with transience and per capita recruitment parameters to estimate key demographic characteristics of M. alfredi from two of Raja Ampat’s largest MPAs: Dampier Strait and South East (SE) Misool. A total of 1,041 unique individuals were photo-identified over an 11-year period (2009–2019) from Dampier Strait (n = 515) and SE Misool (n = 536). In our models, apparent survival probabilities and per capita recruitment rates were strongly linked with El Niño–Southern Oscillation (ENSO) events. Our models also estimated high apparent survival probabilities and significant increases in (sub)population sizes in both MPAs over a decade. In Dampier Strait, the estimated population size increased significantly (p = 0.018) from 226 (95% CI: 161, 283) to 317 (280, 355) individuals. Likewise, the estimated population size in SE Misool increased significantly (p = 0.008) from 210 (137, 308) to 511 (393, 618) individuals. Regardless of variation in the percentage change in population size between years throughout the study, the estimated overall population change shows a compound growth of 3.9% (0.7, 8.6) per annum in Dampier Strait and 10.7% (4.3, 16.1) per annum in SE Misool. Despite the global decline in oceanic sharks and rays due to fishing pressure in the last five decades, our study demonstrates the positive impact of a suite of long-term conservation efforts, coupled with the influence of ENSO events, on increasing M. alfredi abundance in Raja Ampat MPAs. Our study also underscores the importance of long-term monitoring to evaluate the effectiveness of conservation management measures on manta ray populations. Our modification of the standard POPAN model by incorporating per capita recruitment and transience parameters represents an important advance in mark-recapture modelling that should prove useful when examining other manta ray populations and other highly migratory species that are likely to have a substantial percentage of transient individuals.

Genomic evidence indicates small island-resident populations and sex-biased behaviors of Hawaiian reef Manta Rays

“…We reject the alternative hypothesis that the lack of exchange between island groups results from isolation by distance. The shortest distance between Maui Nui and Hawaiʻi Island is only 49 km, and linear movements up to 91 km have been documented elsewhere in Hawaiʻi [ 14 ] and more than 200 km between nearby islands in other archipelagos (Table S3 ) [ 11 , 12 , 16 , 60 , 75 , 76 ]. Along continental shelves with continuous coastlines, linear movements over 500 km are common [ 18 , 60 , 77 , 78 ].…”

“…Long-term photo-identification studies have produced minimum estimates (i.e., catalog sizes) as low as 54 animals in Yap, Micronesia [ 102 ] and up to 4,411 individuals in the Maldives [ 86 ]. Larger population sizes (over 600) tend to be associated with continental shelves such as the East Coast of Africa [ 8 ], Australia [ 10 ], as well as archipelagos consisting of many islands connected by shallow water like the Maldives [ 59 , 86 ], and Indonesia [ 16 , 76 ]. Smaller populations tend to be associated with remote archipelagos with islands separated by deep-water including Hawai’i [ 9 , 14 ], French Polynesia [ 41 ], New Caledonia [ 35 ], and Seychelles [ 11 , 103 ] and generally have small home ranges.…”

“…In addition, an assessment of the impact of commercial tourism activities should be conducted, including evaluating mitigations such as establishing codes of conduct [ 116 ], setting carrying capacity for boats/divers and implementing rest-periods in heavy use areas. Third, robust mark-recapture population models should be built by expanding photo-ID and acoustic tagging efforts to improve survival and abundance estimates (e.g., [ 117 ]) or adopting a modification of the standard POPAN model that incorporates per capita recruitment and transience parameters to estimate annual population sizes [ 76 ]. Fourth, genetic sampling should be expanded across the archipelago and years to evaluate gene flow, monitor changes in contemporary N e [ 154 ], and assess population trends.…”

Spatial connectivity of reef manta rays across the Raja Ampat archipelago, Indonesia

“…Each acoustic receiver, approximately 2 m above the substrate, was securely cable-tied to buoyed moorings that were attached to the substrate [ 26 ]. To optimize acoustic detections, we strategically deployed the acoustic receivers within 150 m of M. alfredi feeding and cleaning sites or other known aggregation sites based on the results of range tests conducted in two previous studies in Raja Ampat using the same acoustic receiver and transmitter specification which suggested 150–200 m as a maximum distance for reliable acoustic detections in this environment [ 24 , 26 ]. These acoustic receivers were maintained and downloaded every 6 months.…”

“…The Raja Ampat archipelago in eastern Indonesia is home to a large population of M. alfredi , with numbers increasing over the past decade [ 24 ]. The species has been fully protected in the region since 2012 and is essentially managed by the Raja Ampat Marine Protected Area Management Authority as a single and homogeneous population [ 25 ].…”