access_HighResolutionAustralianOrography - ACCESS-NRI/accessdev-Trac-archive GitHub Wiki
High Resolution Australian Orography
Martin Dix
The Met Office ancillary file generation package includes the GLOBE30 orography dataset. This has 30 second or approximately 1 km resolution. The orography ancillaries for APS0 were generated from the GLOBE1 1 minute resolution dataset which is all we had at the time. The Met Office documentation now calls this an interim resolution that's discouraged. A quick comparison suggests it's close to a simple average of the 30 minute data.
Data and documentation for GLOBE1 are available from http://www.ngdc.noaa.gov/mgg/topo/globe.html. The documentation says
"DTED seeks to use World Geodetic System 84 (WGS84) for horizontal reference, and Mean Sea Level as vertical reference. However, this may not always be the case as cartographic sources may not be completely or accurately described, and materials from such sources may not be accurately converted to WGS84 and Mean Sea Level.
Histograms of heights for various regions commonly show periodic spikes, suggesting derivation from contoured maps.
Hilton et al 1983; Comparison of digital elevation models over Australia and external validation using ERS-1 satellite radar altimetry, Aust. J. Earth Sciences, 50, 157-168, http://www.cage.curtin.edu.au/~will/j.1440-0952.2003.00982.x.pdf, compared 5 global datasets and the Australian 9 second resolution dataset from GA (https://www.ga.gov.au/products/servlet/controller?event=GEOCAT_DETAILS&catno=66006)
The AUSLIG spot height database could not be used in the open access version of the GLOBE data for licencing reasons so the Australian data is largely from 1:10^6^ scale maps converted to raster grids. This degraded version is what we have from the Met Office.
There are a large number of odd features in the GLOBE30 data. E.g. the highest value over Australia is 2431 m, c.f. 2228 in the 9 sec data (matching the actual spot height). There are also some bizarrely deep valleys in this region.
Over the Melbourne region, most of the Mornington Peninsula has near zero height. The problem here is over a large enough region to affect even a 5 km model. There is also a point glitch in the Dandenongs.
