The PESGB thank Lynx Information Systems and Ophir Energy Company Ltd for sponsoring this London Evening Wine Reception

'Strata' Smith - his two hundred year legacy within
a modern geographic information system

Peter Wigley, Lynx Information Systems

Tuesday 13th May 2008, doors open 5.45 pm for 6 pm
 Burlington House, Geological Society, London

For explorationists in the hydrocarbons industry, stratigraphy underpins much of our endeavour. Two centuries have passed since William Smith worked on his classic geological map of England and Wales.  For those working in the North Sea today there is a direct link back to Smith because “it was on the Jurassic rocks that William Smith, Father of Historical Geology, founded the science of stratigraphy, enunciated the law of superposition, identified fossils with particular strata, and named the classic formations” (W.J Arkell, 1956).

It has also been two hundred years since the founding of the Geological Society.  To mark the occasion,  my co-authors and I decided to make a contribution to the celebrations which would span the age of the Society.  We decided to take William Smith’s famous 1815 map, A delineation of the strata of England and Wales, with part of Scotland, together with other contemporaneous maps and geological sections and to enhance these data with modern remote-sensing information and present the results in a Geographic Information System (GIS).  The project was completed in 2007 and is now available on a DVD published by the Geological Society.

Information from the DVD together with modern digital geological mapping from the British Geological Survey and seismic data from the UK Onshore Geophysical Library (UKOGL) has been used in this retrospective view of Smith’s achievements.

Historical Data
The geological maps included in this project were from the collections of the National Museum of Wales as was the copy of Smith’s original 1815 prospectus and Geological Table of British Organized Fossils.  The geological sections came from the Geological Society.  Johns Cary’s Reduction  of his ‘Turnpike Roads’ map made for the Post Master General in 1796 is from Peter Wigley’s collection.

Two Smith maps are included in the project, Map ‘100’ and Map’Y’.  Map ‘100’, the last Series II early issue map, was signed and numbered by William Smith on 17^th  December 1815.   Map ‘Y’ is one of the late unnumbered Series V maps, originally thought to have been issued around 1817. However it is now known from its watermarks to have been produced in or after 1834. It is almost identical to the map now hanging in the Geological Society’s apartments in Burlington House which carries a watermark date of 1836.

Two geological  maps by George Greenough are also included, a 1^st Edition dated 1819 and a 2^nd Edition dated 1839, although published in 1820 and 1840 respectively.

From 1817-1819, Smith and Cary issued a series of cross sections to complement Smith’s map. In addition to showing the strata in cross-section, these illustrations display a three-dimensional view of the topography along the line of section. The sections are:

Geological section from London to Snowdon, showing the varieties of the strata, and the correct altitude of the hills by William   Smith Civil Engineer 1817
Vertical Section of Strata in Surry [sic] dipping northward
Section of Strata in Sussex dipping southward
Geological View and Section in Essex and Hertfordshire
Geological View and Section of the country between London and Cambridge
Geological View and Section of Norfolk Geological View and Section through Suffolk to Ely
Section of the Strata through Hampshire and Wiltshire to Bath, on the road from Bath to Salisbury
Geological View and Section through Dorsetshire and Somersetshire

 The Richard Thomas  (1819) section reproduced is a Geological View of the Mining Districts of Cornwall corresponding with the Map from Chasewater to Camborne.

Modern Remote-sensing Data
Two modern remote-sensing datasets were used to enhance Smith’s ‘Y’map. These were Shuttle Radar Topography Mission (SRTM) elevation data (90m)  from the U.S. National Geospatial-Intelligence Agency (NGA) and U.S. National Aeronautics and Space Administration (NASA) and Landsat TM (14.25 metres) imagery from NASA.

Image Processing
Because the original maps are both valuable and fragile they were first photographed and scans made from the analogue film transparencies. The images were then composited into single continuous images using Adobe® Photoshop® 7.0.

GIS Processing
The map images were incorporated into the GIS using ESRI ArcGIS software. In order that maps could be overlain with each other and with modern data an attempt was made to determine the original projections used.  There was no indication of projection shown on any of the maps.  The only coordinate data shown were longitude and latitude graticules around the margins of the maps.  Various late 18^th and early 19^th century projections were tested on the map images.  Based on 1^st order polynomial transformation (affine) residual errors, best fit was obtained using Bonne and Cassini projections.  Greenough’s basemap proved to be more accurate than the base provided to Smith by John Cary.  It is suggested that this is because Greenough’s base may have included early results from the 1^st Principal Triangulation of Great  Britain.

Because of positional inaccuracies in the Smith “Y” map it was necessary to carry out some more complicated georectification before modern remote sensing data could be incorporated. Initially, some 2,000 points from capes and bays around the country were referenced from the “Y” map to their actual locations. However, after transformation, the resulting map was severely distorted and did not fit well with place names and topographic features. Subsequently, the “Y” map was split into a number of regions (north, north-central, south and southeast.) and each of these regions was georectified using a 2nd order polynomial transformation to Cary’s place names. Whilst not particularly accurate on coastlines or indeed to actual positions of places, Cary’s basemap is a tour de force regarding relative positions of places. Using this technique, each region was georectified to the British National Grid. There was some overlap between regions but this was considered acceptable.

Modern datasets in the GIS
SRTM data were converted to an ESRI Grid format and imported into the GIS.  From this grid a  hillshade layer was generated and used to enhance Smith’s ‘Y’ map.  Landsat TM imagery was also processed and used as a backdrop for the map images.  Smith’s cross-sections have been located on the ‘Y’ map and have been hyperlinked to images of the sections. Modern geology from BGS (DiGMapGB, 1:650,000), seismic shotpoint data  from UKOGL and cultural data are shown as overlays.

Stratigraphy in 3D
William Smith conceived stratigraphy in the course of his perambulations around England and Wales.  He had a keen eye for topography and his work as a canal surveyor allowed him access to hitherto unseen stratigraphic cross-sections exposed by the canal navigators.  All this added to his three-dimensional stratigraphic concept.  He endeavoured to delineate stratigraphy in three dimensions in the planar view on his legendary map.  Unlike Greenough’s, his basemap had very little indication of elevation on it apart from some spot heights.  The authors believe that his unique technique of colouring his maps using a dense tint at the outcrop edge with progressively fainter tinting updip was an attempt to give a three-dimensional effect.  Using GIS technology we have been able to enhance Smith’s map with computer generated hill shades.  We have also been able to drape the map image on a elevation model derived from the SRTM data.  This allows us to fly through the map in 3D and hopefully see the geology as Smith envisioned it (see figure 1).

top