Editorial, Geoscientist 17.3, March 2007
Nick Petford pays homage to one of the greats and calls for geochemists to embrace the new information age.
The feature in this issue focuses on Victor Goldschmidt (1899-1947), the man generally acknowledged as the father of modern geochemistry. Today geochemistry encompasses a wide range of themes that mirror to some extent its grown-up parent chemistry, with its traditional sub-disciplines of inorganic, organic and physical. Its more exotic branches reach out into cosmology (cosmochemistry) and astrophysics (astrochemistry). Away from the purely material, organic geochemistry informs petroleum chemistry, environmental science and ultimately biochemistry and the composition of living things.
A mighty subject indeed, then. Geochemistry has changed in two ways since Goldschmidt. One is the sheer number of compositional data available to researchers, and the other is their ability to access it. There must by now be millions (at least) of major, trace and isotopic rock and mineral analyses held on personal and institutional spreadsheets and databases around the world. Traditionally, these data have been stored in journal articles and books. However, the information age has paved the way for global sharing of geochemical data in ways about which Goldschmidt could only have dreamt.
Recent years have seen an explosion in websites set up to host large compositional datasets. Some examples include www.geokem.com where users can download spreadsheets of data on volcanic arcs and lunar basalts, along with summary notes - a great resource, especially for students. Another useful site,
www.geochemistry.com, is more than just a database. It has links to upcoming conferences, laboratories, jobs and even a section labelled “fun” (check out the
Crude Oil Story starring an Arab in full desert garb and a diving dinosaur!). The
site hosted by Cornell University is another very useful resource with links to GERM, the Geochemical Earth Reference Model Home Page, and access to large and mostly useful data.
But despite this, geochemists have yet to make the most of available data. For example, in igneous geochemistry, the standard way of plotting and visualsing geochemical data is still the Harker plot, or ratio/ratio variants thereof. The strength of this approach is that such plots are (for the most part) easily understandable. Their weakness however is that they can only analyse trends based on a maximum of four dimensions (a/b + y/z), meaning that covariance and global patterns in the data are not fully captured. Many geochemists remain conservative in adopting the full range of statistical techniques available to them to help glean information and ultimately new knowledge from their data, despite the wide availability of commercial packages such as SAS and SPPS.
Time perhaps for more creative thinking?
Prof. Nick Petford is former Chair of the Volcanic and Magmatic Studies Group of the Geological Society and Vice President of the Mineralogical Society.