23EC4E21E4412A1DE0632144CD9BD709EnglishData Requests OfficerDepartment of Lands, Planning and EnvironmentGeospatial Services Unit (on behalf of department)Datarequests.DLPE@nt.gov.au2025-08-26AS/NZS ISO 19115 Geographic Information - Metadata, ANZLIC Metadata Profile1.0 Mangrove extent mapping across the Darwin Harbour Integrated Marine Monitoring and Research Program (DHIMMRP)2024-10-08 Mangrove extent mapping is a dataset that consists of mangrove extent classification derived from annual and seasonal composites, starting from Year 2016 to 2023. The dataset is intended for applications like mapping mangrove monitoring, modelling mangrove prioritization, vegetation change detection analysis during the specified periods, and other scientific uses. The mangrove extent is mapped using machine learning random forest classification. Auxiliary data such as vegetation indices, Digital elevation models and Woody Vegetation Canopy Height Models also provide support to the decision trees classifying to the mangrove areas. The tiles captured to perform the mangrove extent classification were 52lfm 52lgm 52lfl 52lgl then mosaic in a single raster. Download the spatial data package (contains spatial data, maps and reports) via data.nt.gov.au/ You are licensed to use the DLPE geospatial products on the terms and conditions set out in: Creative Commons Attribution 4.0 International Public License (CC BY 4.0) at: creativecommons.org/licenses/by/4.0/legalcode If you do not agree to the terms and conditions you must not use the geospatial products. You are free to copy, communicate and adapt the licensed material, provided that you abide by the licence terms (including Attribution) and attribute the licensed material using the statement: Supplied by the Department of Lands, Planning and Environment. Copyright Northern Territory Government.English-12.778010017168-12.1268303629232130.333778778396131.3800752540962016-01-012023-12-31datasetnot known100% Completed 0not knownThe logical consistency is maintained from the Copernicus parent data 0not knownMangrove extent classification maps has a geolocation accuracy one pixel dislocation or 10-meter (Queenland Government, 2024). No geo dislocation was observed on the overlay imagery product.0not knownAccuracy of the mangrove classification was assessed by ground truthing at 3992 sample points against imagery. 95.02% of sample points returned coherent (Sun and Grant, 2019).0The mangrove extent mapping involved mangrove classification since baseline in 2016 by the methodology in Sun and Staben, 2019. Reviewed in 2023 (Salum and Roach, 2024). Data Acquisition - Copernicus Annual and seasonal Sentinel 2 satellite imagery for June to August, January to December for 2016 to 2023. Tiles 52lfm, 52lgm, 52lfl, 52lgl. AUSLIG Northern Territory coastline. Digital Elevation Model DEM S v1.0. Extent of the Darwin Harbour by Darwin Harbour Integrated Marine Monitoring and Research Program Preprocessing Annual and seasonal datasets were processed to a spatial resolution of 10m, including TM like bands (Blue, Green, Red, NIR, SWIR1, and SWIR2). Two bands originally captured at 20m resolution were resampled using cubic convolution. (Flood, 2017; Sun and Staben, 2019) Distance to coastline was extracted with a 10m spatial resolution from the AUSLIG and the DEM S v1.0. DEM S v1.0 was resampled to 10-meter spatial resolution. Vegetation indices were extracted from Red and NIR bands and stored in temporary memory to support mangrove extent classification (Sun and Grant, 2019). Woody vegetation structural parameters predicted from Sentinel imagery, were stored in temporary memory (Staben, 2018). All the above data were stacked. Training data was created and saved in a pickle file format. Classification Stacked layers were classified using a Random Forest machine learning algorithm of mangroves in Darwin Harbour. Merged 4 tiles of mangrove extent classification. Clipped mangrove extent to DHMMRP study site boundary. Manual editing was conducted to reclassify false classification for the extent in 2023. Pixel clusters larger than 3 by 3 were considered as mangrove during the manual editing process (Salum and Roach, 2024). Validation The mangrove extent were mapped, audited, and validated through fieldwork and Aerial photography from the Department of Infrastructure Planning and Logistics.