Description

This dataset consist of two vector files which show the change in the building stock of the City of DaNang retrieved from satellite image analysis. Buildings were first identified from a Pléiades satellite image from 24.10.2015 and classified into 9 categories in a semi-automatic workflow desribed by Warth et al. (2019) and Vetter-Gindele et al. (2019).

In a second step, these buildings were inspected for changes based on a second Pléiades satellite image acquired on 13.08.2017 based on visual interpretation. Changes were also classified into 5 categories and aggregated by administrative wards (first dataset: adm) and a hexagon grid of 250 meter length (second dataset: hex).

The full workflow of the generation of this dataset, including a detailled description of its contents and a discussion on its potential use is published by Braun et al. 2020: Changes in the building stock of DaNang between 2015 and 2017

Contents

Both datasets (adm and hex) are stored as ESRI shapefiles which can be used in common Geographic Information Systems (GIS) and consist of the following parts:

• shp: polygon geometries (geometries of the administrative boundaries and hexagons)
• dbf: attribute table (containing the number of buildings per class for 2015 and 2017 and the underlying changes (e.g. number of new buildings, number of demolished buildings, ect.)
• shx: index file combining the geometries with the attributes
• cpg: encoding of the attributes (UTF-8)
• prj: spatial reference of the datasets (UTM zone 49 North, EPSG:32649) for ArcGIS
• qpj: spatial reference of the datasets (UTM zone 49 North, EPSG:32649) for QGIS
• lyr: symbology suggestion for the polygons(predefined is the number of local type shophouses in 2017) for ArcGIS
• qml: symbology suggestion for the polygons (predefined is the number of new buildings between 2015 and 2017) for QGIS

Citation and documentation

To cite this dataset, please refer to the publication

• Braun, A.; Warth, G.; Bachofer, F.; Quynh Bui, T.T.; Tran, H.; Hochschild, V. (2020): Changes in the Building Stock of Da Nang between 2015 and 2017. Data, 5, 42. doi:10.3390/data5020042

This article contains a detailed description of the dataset, the defined building type classes and the types of changes which were analyzed. Furthermore, the article makes recommendations on the use of the datasets and discusses potential error sources.

These datasets were obtained from ECMWF/GloFAS on November 13, 2017, to include the flood forecast (area grid) for Hurricanes Harvey and Irma in the USA from August 15 - September 15, 2017. These are contained in netCDF files, one per day.

Note that while folders and files may have the words "areagrid_for_Harvey" in the name, all the data here are for the southeast USA, encompassing both Harvey and Irma impact areas.

Dataset variables: - dis = forecasted discharge (for all forecast step 1+30 as initial value and 30 daily average values, with ensemble members as 1+50 where the first is the so-called control member and the 50 perturbed members) - ldd = local drainage direction within routing model - ups = upstream area of each point within routing model - rl2,rl5,rl20 = forecast exceedance thresholds for 2-, 5- and 20-year return period flows, based on gumbel distribution from ERA-interim land reanalysis driven through the lisflood routing.

Models used (see [2] for further details): - Hydrology: River discharge is simulated by the Lisflood hydrological model (van der Knijff et al., 2010) for the flow routing in the river network and the groundwater mass balance. The model is set up on global coverage with horizontal grid resolution of 0.1° (about 10 km in mid-latitude regions) and daily time step for input/output data. - Meteorology: To set up a forecasting and warning system that runs on a daily basis with global coverage, initial conditions and input forcing data must be provided seamlessly to every point within the domain. To this end, two products are used. The first consists of operational ensemble forecasts of near-surface meteorological parameters. The second is a long-term dataset consistent with daily forecasts, used to derive a reference climatology.

Suggestions for usage: - Selected software: ArcGIS or QGIS - Select dis for example, then any of the bands (51*31 in total), then set the range manually to 0-1000 or something like that.

Agency: GloFAS [1] From its public website: "The Global Flood Awareness System (GloFAS), jointly developed by the European Commission and the European Centre for Medium-Range Weather Forecasts (ECMWF), is independent of administrative and political boundaries. It couples state-of-the art weather forecasts with a hydrological model and with its continental scale set-up it provides downstream countries with information on upstream river conditions as well as continental and global overviews. GloFAS produces daily flood forecasts in a pre-operational manner since June 2011."

References [1] GloFAS home page [http://www.globalfloods.eu/] [2] Data and methods [http://www.globalfloods.eu/user-information/data-and-methods]

This dataset was generated by the TU Wien Department of Geodesy and Geoinformation.European Sentinel-1 forest type and tree cover density maps represent first continental-scale forest layers based on Sentinel-1 C-Band Synthetic Aperture Radar (SAR) backscatter data. For the year 2017 they cover the majority of European continent with 10 m and 100 m sampling for forest type and tree cover density, respectively. The maps were derived using the method described in https://www.tandfonline.com/doi/full/10.1080/01431161.2018.1479788.The forest type map shows the dominant forest type class (coniferous, broadleaf). Tree cover density map shows the percentage of forest canopy cover within the 100 m pixel.Please be referred to our peer-reviewed article at https://doi.org/10.3390/rs13030337 for details and accuracy assessment accross Europe.Dataset RecordThe forest type and tree cover density maps are sampled at 10 m and 100 m pixel spacing respectively, georeferenced to the Equi7Grid and divided into square tiles of 100km extent ("T1"-tiles). With this setup, the forest maps consist of 728 tiles over the European continent, with data volumes of 3.12 GB and 378.3 MB.The tiles' file-format is a LZW-compressed GeoTIFF holding 16-bit integer values, with tagged metadata on encoding and georeference. Compatibility with common geographic information systems as QGIS or ArcGIS, and geodata libraries as GDAL is given.In this repository, we provide each forest map as tiles, whereas two zipped dataset-collections are available for download below.Code AvailabilityFor the usage of the Equi7Grid we provide data and tools via the python package available on GitHub at https://github.com/TUW-GEO/Equi7Grid. More details on the grid reference can be found in https://www.sciencedirect.com/science/article/pii/S0098300414001629.AcknowledgementsThe computational results presented have been achieved using the Vienna Scientific Cluster (VSC).

Accurate coastal wave and hydrodynamic modelling relies on quality bathymetric input. Many national scale modelling studies, hindcast and forecast products, have, or are currently using a 2009 digital elevation model (DEM), which does not include recently available bathymetric surveys and is now out of date. There are immediate needs for an updated national product, preceding the delivery of the AusSeabed program’s Global Multi-Resolution Topography for Australian coastal and ocean models. There are also challenges in stitching coarse resolution DEMs, which are often too shallow where they meet high-resolution information (e.g. LiDAR surveys) and require supervised/manual modifications (e.g. NSW, Perth, and Portland VIC bathymetries). This report updates the 2009 topography and bathymetry with a selection of nearshore surveys and demonstrates where the 2009 dataset and nearshore bathymetries do not matchup. Lineage: All of the datasets listed in Table 1 (see supporting files) were used in previous CSIRO internal projects or download from online data portals and processed using QGIS and R’s ‘raster’ package. The Perth LiDAR surveys were provided as points and gridded in R using raster::rasterFromXYZ(). The Macquarie Harbour contour lines were regridded in QGIS using the TIN interpolator. Each dataset was mapped with an accompanying Type Identifier (TID) following the conventions of the GEBCO dataset. The mapping went through several iterations, at each iteration the blending was checked for inconstancy, i.e., where the GA250m DEM was too shallow when it met the high-resolution LiDAR surveys. QGIS v3.16.4 was used to draw masks over inconstant blending and GA250 values falling within the mask and between two depths were assigned NA (no-data). LiDAR datasets were projected to +proj=longlat +datum=WGS84 +no_defs using raster::projectRaster(), resampled to the GA250 grid using raster::resample() and then merged with raster::merge(). Nearest neighbour resampling was performed for all datasets except for GEBCO ~500m product, which used the bilinear method. The order of the mapping overlay is sequential from TID = 1 being the base, through to 107, where 0 is the gap filled values.

Permissions are required for all code and internal datasets (Contact Julian OGrady).