A spatial analysis of seagrass habitat and community diversity in the Great Barrier Reef World Heritage Area

Alex B Carter,Barbara J Robson,Michael A Rasheed,Rob Coles,Emma Lawrence,Catherine Collier

doi:10.1038/s41598-021-01471-4

Abstract

The Great Barrier Reef World Heritage Area (GBRWHA) in north eastern Australia spans 2500 km of coastline and covers an area of ~ 350,000 km2. It includes one of the world’s largest seagrass resources. To provide a foundation to monitor, establish trends and manage the protection of seagrass meadows in the GBRWHA we quantified potential seagrass community extent using six random forest models that include environmental data and seagrass sampling history. We identified 88,331 km2 of potential seagrass habitat in intertidal and subtidal areas: 1111 km2 in estuaries, 16,276 km2 in coastal areas, and 70,934 km2 in reef areas. Thirty-six seagrass community types were defined by species assemblages within these habitat types using multivariate regression tree models. We show that the structure, location and distribution of the seagrass communities is the result of complex environmental interactions. These environmental conditions include depth, tidal exposure, latitude, current speed, benthic light, proportion of mud in the sediment, water type, water temperature, salinity, and wind speed. Our analysis will underpin spatial planning, can be used in the design of monitoring programs to represent the diversity of seagrass communities and will facilitate our understanding of environmental risk to these habitats.

Highlights

Coastal marine habitats are some of the most at-risk ecosystems in the world[1]
Using random forest (RF) statistical models we identified approximately 88,321 km[2] of potential seagrass habitat in the Great Barrier Reef World Heritage Area (GBRWHA) (Fig. 1) as a function of 12 environmental variables
Within regions of potential seagrass habitat, we identified 36 seagrass community types defined by their species assemblages (Figs. 3 and 4, Table 3), based on the results of Multivariate Regression Trees (MRTs)

Summary

Introduction

Coastal marine habitats are some of the most at-risk ecosystems in the world[1]. Proximity to land-based anthropogenic activities exposes these habitats to threats from multiple s tressors[2]. Challenges include describing diversity, distribution and connectivity within ecosystems, defining desired state and assessing ecosystem condition to understand long-term trends and in evaluating short-term impact-recovery cycles[7] This is exacerbated in Australia’s Great Barrier Reef World Heritage Area (GBRWHA) by the vaguely defined objective and high bar set by the reef management authority in 2015 to “maintain diversity of species and ecological habitats in at least a good condition and with a stable to improving trend”[8] and updated in 2018 to “[facilitate] adaptive management for the Reef that is effective, efficient and evolving”[9]. They extend north and south of GBRWHA boundaries into the Torres S trait[31] and south-east Queensland[32,33]

Methods

Results

Conclusion