Premium
Prioritizing localized management actions for seagrass conservation and restoration using a species distribution model
Author(s) -
Adams Matthew P.,
Saunders Megan I.,
Maxwell Paul S.,
Tuazon Daniel,
Roelfsema Chris M.,
Callaghan David P.,
Leon Javier,
Grinham Alistair R.,
O'Brien Katherine R.
Publication year - 2016
Publication title -
aquatic conservation: marine and freshwater ecosystems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.95
H-Index - 77
eISSN - 1099-0755
pISSN - 1052-7613
DOI - 10.1002/aqc.2573
Subject(s) - seagrass , environmental science , bay , thalassia testudinum , benthic zone , threatened species , habitat , sediment , range (aeronautics) , metapopulation , ecology , fishery , oceanography , biology , geology , population , biological dispersal , paleontology , materials science , demography , sociology , composite material
Abstract Seagrass habitat is globally threatened from the cumulative impact of human activities on coastal ecosystems. Successful conservation and restoration of seagrass requires knowledge of the environmental factors that limit seagrass presence; however, quantifying the relative importance of these environmental factors is a significant challenge. To resolve this issue, a species distribution model (SDM) for seagrass was constructed from a range of data collected at different spatial and temporal resolutions in Moreton Bay, Australia. Mean annual benthic light availability, significant wave height, and sediment settling time (calculated from sediment size distribution) were the environmental factors predicting seagrass presence in the model, which was calibrated and validated using maps of seagrass cover from 2004 and 2011, respectively. The SDM correctly classified seagrass presence/absence in 85% of cases for calibration to 2004 data and 88% of cases for validation to 2011 data. Application of the SDM to 12 regions of 10–100 km 2 size within the study area demonstrated that the environmental factors limiting seagrass presence (i.e. either threatening seagrass that is present or hindering its colonization if seagrass is absent) varied substantially between these regions. For example, seagrass presence in the regions Waterloo Bay and Central West was predicted to be limited by their high mud content, while in the region Deception Bay South, light availability was the major limiting factor. The results demonstrate that SDMs need to be developed on a sufficiently broad scale to capture significant variability in environmental conditions. These models can then be applied at finer scales to assess management options based on the local environmental conditions. Copyright © 2015 John Wiley & Sons, Ltd.