Fig 1.
General reference map showing the location of the ten marine ecoregions used in the analysis indicating the distribution of total reef area within each of the 423 reef units that were used in the larvae transport and settlement simulation.
Coral reef data used are from the Millennium Coral Reef Mapping Project [35] which represents the most accurate and consistently mapped global distribution of shallow coral reef systems.
Table 1.
Descriptions and values of coral larval biological parameters used in the dispersal simulations.
Fig 2.
Strength of reef connections based on modeled transported coral larvae.
These values represent an average of eight coral larvae dispersal simulations between 2008–2011. The width and color of the lines represent the strength of connection. The darker red and orange areas indicate high amounts of settled coral larvae transported along that connection, while the shades of blue represent smaller amounts of settled larvae.
Fig 3.
Visualization of a 30-day simulated coral spawning event based on NOAA’s Real-Time Ocean Forecast System (RTOFS) ocean current data starting on August 21, 2011.
The amount of coral larvae released was based on reef area. These maps represent time steps during the 30-day pelagic larvae duration model, representing coral larvae distribution and concentration after a) 10 days; b) 20 days; and c) 30 days. These data were used to create the hourly animations for each of the eight spawning events.
Fig 4.
a) Modeled larvae settlement rates by EEZ averaged across eight spawning events showing total received larvae (red) and contributed larvae (blue). The red bar indicates the total modeled larvae received and settled within each corresponding EEZ. The highest amounts of larvae received and contributed are largely influenced by an EEZ’s reef area, ocean current patterns, and geographic location. Refer to the individual country maps to see these results in map format. The blue bar represents the total modeled larvae that originated within each EEZ and settled anywhere. b) The same data, but ignoring larvae that originates and settles within the same EEZ. For example, according to the model, Belize receives very little larvae that originate outside its EEZ. However, Belize contributes more larvae to other EEZs than any other EEZ. Honduras on the other hand, receives the most incoming larvae from other EEZs, while contributing the second highest level of larvae to other EEZs. c) Larvae contribution ratio by EEZ showing the proportion of all settled larvae that originates within each respective EEZ and is contributed to other EEZs.
Fig 5.
a) Modeled larvae settlement that originates and settles within the same reef unit (i.e. local -retention); b) Modeled larvae settlement of that originates and settles within the same EEZ.
Fig 6.
Connectivity matrix by EEZ showing the relative strength of each country connection based on the amount received (x-axis) and contributed (y-axis) settled larvae.
Fig 7.
a) Betweenness centrality measures by reef unit indicating the importance of each reef unit’s role in maintaining network connectivity. The corresponding graph shows betweenness centrality measures summed by marine ecoregion; b) Closeness centrality measures by reef unit indicating how long it will take to spread something from a particular node to the other nodes in the network.
The corresponding graph shows closeness centrality measures summed by marine ecoregion.
Fig 8.
Typical Marxan best solution (not considering coral connectivity) that met a 30% target for reef area only as summarized by reef unit: a) regional assessment (no strata) and b) stratified by marine ecoregion.
These results were based on 100 repetitions using a million iterations per run with a calibrated BLM value of 0.17. A penalty factor of 10 was used and a boundary file based on the Euclidean distance between reef units. The calculated cost value by reef unit was derived from the Global Map of Human Impacts to Marine Ecosystems [54].
Fig 9.
Marxan coral connectivity best solution that met a 30% target set for local retention and betweenness centrality values by reef unit: a) regional assessment (no strata); and b) stratified by marine ecoregion.
These results were based on 100 repetitions using a million iterations per run with a calibrated BLM value of 0.17. A penalty factor of 10 was used and an asymmetric boundary file based on the amount of settled larvae that traveled between reef units. The calculated cost value by reef unit was derived from the Global Map of Human Impacts to Marine Ecosystems [54].
Fig 10.
Marxan coral connectivity selection frequency (i.e. summed solution) that met a 30% target set for local retention and betweenness centrality values by reef unit: a) regional assessment (no strata); and b) stratified by marine ecoregion.
These results were based on 100 repetitions using a million iterations per run with a calibrated BLM value of 0.17. A penalty factor of 10 was used and an asymmetric boundary file based on the amount of settled larvae that traveled between reef units. The calculated cost value by reef unit was derived from the Global Map of Human Impacts to Marine Ecosystems [54]. Reef units shaded in red and orange represent those areas that are likely to contribute more to coral reef connectivity.
Fig 11.
Results of the marine ecoregional coral connectivity best solution (30% target set for local retention and betweenness centrality values by reef unit), overlaid onto the World Database on Protected Areas [56] and The Nature Conservancy’s Marine Protected Area Database of the Insular Caribbean [57].
Table 2.
Breakdown of coral reef area numbers by marine ecoregion and the percentage of selected high value reefs within the existing MPA network.