Torres Strait has some of the world’s most extensive and diverse seagrass meadows. They protect coasts from erosion, store carbon, and provide habitat and food for fishes and invertebrates. They also support two significant mega-herbivore species: the world’s largest dugong (Dugong dugon) population; and, large numbers of green turtles (Chelonia mydas). Seagrass meadows are also culturally important to regional communities.

Although well adapted to grazing pressures, herbivory can change seagrass meadow structure and species composition, so that there is less above-ground biomass, lower canopy height, and faster growing species better adapted to disturbance. Elsewhere, sustained overgrazing by turtles and dugongs has led to loss of seagrass meadows and their associated ecosystem functions.

In 2019 and 2020 the Torres Strait Seagrass Monitoring Program’s researchers were concerned to detect dramatic declines in seagrass meadow condition, at a number of locations. There was no evidence of disease or significant environmental change, but there were unusually large numbers of grazing green turtles and dugongs in the areas where seagrass meadow declines were most dramatic. It was suspected they were a likely cause.

To gain better information, this project quantified the effects of dugongs and turtles on seagrass meadows over 7 months.

Approach and findings


Study sites were selected at two locations where seagrass declines had been pronounced: the intertidal meadows at Mabuyag Island, and atop Koey Maza (Kai Reef) in the Orman Reef complex.

Two-metre square exclusion cages were used to prevent green turtles and dugongs from grazing within. Measurements of a range of seagrass parameters were recorded inside and outside the cages for comparison; initially, and then at two and seven months.


Measurements indicated that large-animal herbivory was suppressing seagrass at both study sites. Exposed plots had much less, and shorter seagrass, than within the mega-herbivores exclusion cages. At the Orman Reef site, after seven months, canopy was almost twice as high and biomass over five times greater in the exclusion cages. At the Mabuyag Island site, both canopy height and biomass were around double.

The exposed seagrasses at both locations were cropped, without extensive disruption of the substrate, suggestive of green turtle grazing. Dugongs generally consume both above and below-ground seagrass material and leave distinctive feeding trails through meadows. No trails were observed at either site in this study.

At Orman Reef, the impacts of grazing could be seen after just two months, and were significantly greater, indicating that grazing pressure at this reef-top seagrass meadow was much higher than at the inter-tidal meadows of Mabuyag Island.


This project confirmed that mega-herbivore grazing is a key element shaping seagrass dynamics in Torres Strait. Grazing pressure continues to reduce seagrass abundance at both study sites. However, this does not necessarily mean that these meadows are in a state of decline.

Seagrasses evolved under intense herbivory and are well adapted to grazing. A mosaic of differentially grazed meadows, is likely to have been the natural state prior to the rise of human impacts. Despite appearing to be in a poor state to human observers, the meadow itself might still be productive and the plant-herbivore system in balance. The point at which grazing exceeds the threshold recovery balance, requires better understanding.

Research data is also needed to better understand: regional mega-herbivore movements and any associated changes in meadow distribution; other potential drivers of change such as wind or sediment movements; the role of mega-herbivore predation in population regulation; and long-term monitoring to understand if the very low levels of seagrass observed at the time of the study are part of a normal cycle at these sites. Monitoring at the time of the study had occurred for just four years at Orman, and ten at Mabuyag.


  • An expanded understanding of regional seagrass resources and associated herbivory.
  • Development of a simple and repeatable method for seagrass herbivory assessment.
  • A foundation for future Indigenous-led monitoring.

Project location