Wetland Vegetation of the Macquarie Marshes 2022
Published by NSW Department of Climate Change, Energy, the Environment and Water
About this dataset
This wetland vegetation map of the Macquarie Marshes was produced using aerial photographic interpretation (API) of high-resolution aerial imagery. Key wetland and floodplain vegetation communities were delineated through expert interpretation, supported by existing vegetation survey data, environmental spatial layers and object-based image segmentation. The product provides a landscape-scale baseline of wetland vegetation extent and distribution, with mapped communities assigned to relevant NSW Plant Community Types (PCTs), Vegetation Classes, Vegetation Formations and vegetation functional groups aligned with the Macquarie Marshes Long Term Watering Plan (LTWP). It is intended to support environmental water planning, monitoring and reporting, and long-term assessment of wetland vegetation extent and change. Imagery and supporting datasets The primary mapping input was high-resolution, orthorectified aerial RGB (red-green-blue) imagery acquired in June 2022, with a ground sampling distance of 40 cm across the study area. This imagery provides the primary temporal reference for the vegetation map. Additional information and spatial datasets used to support vegetation interpretation included. - Existing vegetation survey data, including BioNet species records, and other relevant reports and grey literature. - Environmental spatial layers, including flood frequency, elevation and geomorphology. - Oblique aerial photographs captured from a helicopter between January 2022 and April 2023, selected to provide additional vegetation context close to the primary imagery acquisition period. - An object-based segmentation layer generated using Definiens eCognition from raster inputs at 5 m spatial resolution. Mapping and aerial photographic interpretation Vegetation mapping was undertaken through expert API, with trained interpreters identifying and delineating vegetation communities based on colour, texture, structure, spatial patterns and landscape context. An object-based image segmentation layer was generated in Definiens eCognition using spectral and textural characteristics of the input raster datasets (Roff et al., 2022). The resulting segments were overlaid on the 40 cm aerial imagery and manually selected, grouped and attributed by interpreters to delineate vegetation patterns. This approach supported consistent polygon delineation and reduced the need for manual digitisation of vegetation boundaries. Vegetation was interpreted at a scale of approximately 1:10,000 for wetland communities and 1:25,000 for non-flood-dependent vegetation, with a minimum mapping unit (MMU) of 2 ha. Vegetation patterns were initially assigned to Vegetation Photo Pattern (VPP) classes based on visually distinguishable image characteristics and landscape context, consistent with the VPP classification approach described in the NSW State Vegetation Type Map: Technical Notes (Roff et al., 2022). These were subsequently aligned with NSW PCTs through expert interpretation informed by existing vegetation survey data, environmental information and the NSW BioNet Vegetation Classification. Class hierarchy Vegetation communities were assigned at PCT level using the NSW BioNet Vegetation Classification and aggregated to Vegetation Class and Vegetation Formation levels within the NSW vegetation classification hierarchy. To support Monitoring, Evaluation and Reporting (MER) and environmental water planning, wetland PCTs were also aligned with indicative vegetation functional groups corresponding to the vegetation objectives of the Macquarie Marshes LTWP. Key fields dictionary *‘PCTID’* (NSW PCT Code); *‘PCTName’* (NSW PCT Name); *‘vegClass’* (NSW Vegetation Class); *‘vegForm’* (NSW Vegetation Formation); *‘FG_MER’* (MER Functional Group aligned with LTWP reporting); *‘Shape_Area’* (polygon area); *‘Shape_Leng’* (polygon perimeter length) Context classes (*‘Cleared'*) are included in the final map for completeness but were excluded from the reporte