Spatial models and risk assessments to inform marine planning at ecosystem-scales: seagrasses and dugongs as a case study

摘要: Informing marine planning and the management of species at ecosystem-scales is difficult because data are generally lacking that scale. Collecting empirical information on distribution and/or abundance across broad spatial scales expensive logistically difficult. Accurate efficient monitoring programmes assess response to actions often cannot be conducted due time, expertise cost constraints. The Great Barrier Reef World Heritage Area (GBRWHA) Queensland, Australia, world’s largest (approximately 348,000 km2) second protected area (MPA). The region supports a variety habitats including coastal seagrasses globally significant populations dugong (Dugong dugon), threatened mammal. Seagrasses, dugongs their exposed multiple anthropogenic threats along much 2,300 km coastline GBRWHA. Assessing effectiveness current arrangements for informing design new regimes challenging difficulties associated with collection scale GBRWHA. My thesis goal was overcome in GBRWHA by using models risk assessments geographical systems (GIS). My objectives were to: (1) develop GBRWHA; and, (2) use these estimate from threats. This approach allowed me compare rank identify most severe risks, locate specific sites require conservation actions. I used predictor variables ecological theory expert knowledge inform Bayesian belief network, predictive seagrass habitat model. network quantified relationship (dependencies) between eight environmental drivers: relative wave exposure, bathymetry, extent flood plumes, season, substrate, region, tidal range sea surface temperature. outputs modelling exercise probabilistic GIS-surfaces suitability entire coast both wet dry seasons unit 2 * km. Quantitative impact incomplete or unavailable, cumulative measure predict. In light this uncertainty, I evaluate hazards. Vulnerability scores derived opinion, delineate areas low, medium high habitats. found whilst units remote Cape York classified as low risk, almost two thirds urban activities. Reducing 13 identified action would require: improving quality terrestrial water enters mitigating impacts port infrastructure development dredging; (3) addressing hazards shipping accidents recreational boat damage. I spatially explicit population model collected 20 year time-series aerial surveys. Data surveys corrected differences sampling intensity sampled prior interpolated geostatistical estimation method universal kriging. estimated density (the same model). each medium, value basis frequency analysis. I compared commercial gill-netting activities bycatch introduced 2004 appreciably reduced reducing total where netting permitted. Restructuring industry further conducted. Netting currently prohibited 67% value, 56% improvement over former arrangements. four still value. Conservation closures modified fishing practices should considered regions. In addition gill-netting, Indigenous hunting, trawling, vessel traffic, poor runoff. developed rapid GBRWHA, evaluated options ameliorate risk. Expert opinion Delphi technique potential adversely all (96%) (93%) human Decreasing hunting banned runoff areas. The able GIS quantify species; action. achieve outcome data-inadequate environment combining qualitative threat distributions. Implementing will provide greatest positive result Future research directed understanding constraints opportunities ensure effective implementation can achieved.