Darwin Harbour Habitat Mapping Program: Predicted seabed mud content

This resource contains a predicted seabed mud content grid for the greater Darwin Harbour region as part of a baseline seabed mapping program of Darwin Harbour and Bynoe Harbour. This project was funded through offset funds provided by an INPEX-led Ichthys LNG Project to the Northern Territory Government's Department of Environment and Natural Resources (NTG-DENR) with co-investment from Geoscience Australia (GA) and the Australian Institute of Marine Science (AIMS). The intent of this program is to improve knowledge of the marine environments in the Darwin and Bynoe Harbour regions by collating and collecting baseline data that enable the creation of thematic habitat maps and information to underpin marine resource management decisions. The predicted seabed mud content grid was derived from a compilation of multiple surveys undertaken by GA, AIMS and NTG-DENR between 2011 and 2017, including GA0333 (Siwabessy et al., 2015), GA0341 (Siwabessy et al., 2015), GA0351/SOL6187 (Siwabessy et al., 2016), GA4452/SOL6432 (Siwabessy et al., 2017), GA0356 (Radke et al., 2017), and GA0358 and GA0359 (Radke et al., 2018), adding to those from a previous survey GA0333 collected by GA, AIMS and NTG-DENR. This dataset provides spatially continuous predictions of seabed %mud (< 63 µm) content for the Darwin and Bynoe harbour region, northern Australian marine margin. Data are presented in 10 m resolution raster grids format and ascii text file. Predictions are based on 395 samples and nine environmental variables derived from high resolution multibeam sonar bathymetry and backscatter data. Accuracy of predictions is very high, with a VEcv = 62% for mud; and the predictive accuracy has been increased by 198.3% for mud in comparison with the commonly used method (i.e., IDW). Absences in predictions occur in this dataset as a result of non-availability associated with predictive variables. This dataset supersedes previous predictions of mud content for the Darwin and Bynoe harbour region with demonstrated improvements in predictive accuracy.

本数据集为达尔文港（Darwin Harbour）与拜诺港（Bynoe Harbour）基线海底测绘项目的组成部分，涵盖大达尔文港区域的预测海底泥沙含量栅格（seabed mud content grid）数据。本项目由INPEX牵头的伊契斯液化天然气（Ichthys LNG）项目提供补偿资金，资助北领地政府环境与自然资源部（Northern Territory Government's Department of Environment and Natural Resources, NTG-DENR）开展，并由澳大利亚地球科学局（Geoscience Australia, GA）与澳大利亚海洋科学研究所（Australian Institute of Marine Science, AIMS）共同出资。本项目旨在通过整理与采集基线数据，提升对达尔文港与拜诺港海域海洋环境的认知，进而生成专题生境地图与相关信息，为海洋资源管理决策提供支撑。本次预测的海底泥沙含量栅格数据，整合了GA、AIMS与NTG-DENR在2011至2017年间开展的多份勘测数据，包括GA0333（Siwabessy等，2015）、GA0341（Siwabessy等，2015）、GA0351/SOL6187（Siwabessy等，2016）、GA4452/SOL6432（Siwabessy等，2017）、GA0356（Radke等，2017）以及GA0358、GA0359（Radke等，2018），同时纳入了此前由GA、AIMS与NTG-DENR完成的GA0333勘测数据。本数据集为澳大利亚北部海域边缘的达尔文港与拜诺港区域提供了空间连续的海底泥沙（粒径<63μm）百分比含量预测结果。数据以10米分辨率栅格（raster grids）格式与ASCII文本文件形式提供。该预测基于395个采样点与9项环境变量，这些变量由高分辨率多波束声纳测深与后向散射数据提取得到。预测精度极高，泥沙含量预测的VEcv值达62%；相较于常用的反距离加权（Inverse Distance Weighting, IDW）方法，泥沙含量的预测精度提升了198.3%。本数据集中的预测缺失值由预测变量不可用导致。本数据集取代了此前针对达尔文港与拜诺港区域的泥沙含量预测数据，且经证实其预测精度有所提升。