For over 150 years traditional bathymetric and topographic measurements have been collected independently to serve different purposes. Today depth and height data tend to be referred to different vertical datums, creating inconsistency across the land-sea interface. Depth datums in particular are usually referred to local tidal datums (Chart Datums) connected to the local height datum at a limited number of discrete points. The UKHO/UCL Vertical Offshore Reference Frame project will provide the capability to seamlessly join land and sea surfaces across the shore, extend the knowledge of vertical datums and reference surfaces offshore and enable the transformation of height and depth between a number of vertical datums and reference surfaces, without the need for tide gauges.

The Commission
The requirement for VORF arose from earlier research and projects involving the creation of an integrated coastal dataset from existing data, such as ICZMap. There was increasing difficulty in assimilating land and sea data in the vertical dimension due to poorly understood relationships between the datums and surfaces that cross the land-sea boundary. In July 2005 a contract was awarded to University College London to develop a technical demonstrator.

Subcontracts were issued to Proudman Oceanographic Laboratory and the Danish National Space Centre. This is enabled by a mathematical model that underpins the VORF concept, focusing on home waters. Led by Dr Jonathan Iliffe and Dr Marek Ziebart of the Department of Geomatic Engineering, with full-time research fellows Dr James Turner and Mr João Oliveira, the first software demonstrator version will be delivered to the UKHO by the end of this year.

The Challenge
The aim is to refer vertical height/depth data to a consistent reference frame (such as ETRF89). On land this is (relatively) straightforward to resolve. At sea, however, data is referred conventially to Chart Datum, which approximates to the tidal level of approximately Lowest Astronomical Tide. Chart Datum is not a seamless reference surface. It varies from location to location and is established based on local water-level measurements at discrete locations, but its surface offshore is often less well known. Key elements of the challenge lie in developing a methodology for relating Chart Datum to the ellipsoid underlying the global datums ised in GPS data acquisition. This will result in not only an accurate representation of Chart Datum, and its relationship with all other relevant surfaces, but also in a system that runs efficently whilst handling the vast amout of data acquired by modern survey instruments.

Seven Hundred Datums
On the horizontal datum front, the UKHO has made great progress relating charts to WGS84 Datum (ETRF89 in UK Waters). However, for a full understanding of the connections between datasets the relationships between vertical datums must be known. Heights on land in the UK are established to around a dozen different datums, whilst depths at sea are given with respect to over seven hundred Chart Datums. Connecting land, sea and satellite datums needs to be established to achieve maximum interoperability between datasets and to exploit the latest data-acquisition technology. Integrating GPS technology into the acquisition of hydrographic data, using Liddar in coastal zones and matching Ordnance Survey data on land with bathymetric data, has important consequences. With increasing interest in coastal zones and projects ranging from leisure developments to flood-prevention schemes and offshore wind farms, this gap urgently needs plugging.

VORF Approach
To model the relationship between Chart Datum and other vertical reference surfaces UCL collected and validated a variety of sources including:

• mean sea-surface model in the open oceans derived from satellite altimetry


• Geoid model OSGM05, derived from OSGM02 model combined with long-wavelength gravity-field data from the GRACE


• tide-gauge data from the UK Permanent Service for Mean Sea Level (PSMSL) for all UK primary tide-gauges, circa sixty datasets; data comprising monthly mean sea-level, typically spanning ten years or longer


• tide-gauge data from Admiralty Tide Table (ATT) stations, comprising some seven hundred datasets; observations typically span short periods of time (one to twelve months) and go back as far as 1855


• GPS-derived ellipsoidal heights at specific tide-gauge locations


• bathymetric models for tidal modelling.