SEA SURFACE-WATER TEMPERATURE AND ISOTOPIC RECONSTRUCTIONS FROM NANNOPLANKTON DATA USING ARTIFICIAL NEURAL NETWORKS

Artificial neural networks (ANN) are computer systems that differ from other pattern recognition methods in their ability to ‘learn’ one or more target variables from a set of input variables. These systems learn by self-adjusting a set of parameters to minimize the error between the desired output and network output. To explore the potential of artificial neural networks for predictions of paleo-oceanographic parameters from relative abundances of calcareous nannoplankton species we analysed observations taken from the literature for (1) the prediction of sea surfacewater temperature (SST) in samples from offshore southern California, and (2) the prediction of oxygen isotopic values in a Quaternary core from the eastern Mediterranean. We employed a back propagation (BP) neural network to assess the ability of the network to predict SST and oxygen isotopic values. Each of the data sets was divided into five random training and test sets to assess the stability of the error rate estimates. For the California Bight samples we obtained an average Root-Mean-Square-Error of Prediction (RMSEP) in the test sets of 0.68, implying that an unknown SST can be predicted with a precision of ± 0.68°C. In the Mediterranean samples the average RMSEP in the test sets was 0.64; hence an unknown oxygen isotope value can be predicted with a precision of ± 0.64 δ18O‰ vs. PDB. ANN techniques can be seen as a complementary tool to more conventional approaches to paleoceanographic data analysis. Such techniques hold great potential for making predictions of various types of variables from paleontological data.

[1]  H. Okada,et al.  Distribution of coccolithophores in marginal seas along the western Pacific Ocean and in the Red Sea , 1975 .

[2]  D. Kroon,et al.  PHYTOPLANKTON AND FORAMINIFERAL FREQUENCIES IN NORTHERN INDIAN-OCEAN AND RED-SEA SURFACE WATERS , 1989 .

[3]  B. Malmgren,et al.  Application of artificial neural networks to paleoceanographic data , 1997 .

[4]  Charles L. Karr,et al.  Determination of lithology from well logs using a neural network , 1992 .

[5]  R. Thunell,et al.  Export production of coccolithophores in an upwelling region: Results from San Pedro Basin, Southern California Borderlands , 1995 .

[6]  A. Winter,et al.  Coccolithophores: a major phytoplankton component in mature upwelled waters off the Cape Peninsula, South Africa in March, 1983 , 1987 .

[7]  R. Thunell,et al.  Seasonal changes in coccolithophore densities in the Southern California Bight during 1991–1992 , 1995 .

[8]  F. Muller‐Karger,et al.  Plankton response to physical forcing in the Gulf of California , 1996 .

[9]  K. Gaarder.  Three new taxa of Coccolithineae , 1970 .

[10]  P. H. Roth,et al.  Floral and solution patterns of coccoliths in surface sediments of the North Pacific , 1982 .

[11]  P. Monteiro,et al.  Distribution and malformation of living coccolithophores in the northern Benguela upwelling system off Namibia , 1993 .

[12]  A. Mcintyre,et al.  Seasonal distribution of modern coccolithophores in the western North Atlantic Ocean , 1979 .

[13]  B. Malmgren,et al.  Application of artificial neural networks to chemostratigraphy , 1996 .

[14]  A. Pujos.  Calcareous nannofossils of Plio-Pleistocene sediments from the northwestern margin of tropical Africa , 1992, Geological Society, London, Special Publications.

[15]  A. Raiche.  A pattern recognition approach to geophysical inversion using neural nets , 1991 .

[16]  Björn A. Malmgren,et al.  Climate Zonation in Puerto Rico Based on Principal Components Analysis and an Artificial Neural Network , 1999 .

[17]  Brian S. Penn,et al.  Using neural networks to locate edges and linear features in satellite images , 1993 .

[18]  T. Probyn.  The inorganic nitrogen nutrition of phytoplankton in the southern Benguela: new production, phytoplankton size and implications for pelagic foodwebs , 1992 .

[19]  A. Mcintyre,et al.  Modern Pacific Coccolithophorida: a paleontological thermometer. , 1970, Transactions of the New York Academy of Sciences.

[20]  W. H. Hutson.  The Agulhas Current During the Late Pleistocene: Analysis of Modern Faunal Analogs , 1980, Science.

[21]  E. Parisi.  Carbon and oxygen isotope composition of Globigerinoides ruber in two deep-sea cores from the Levantine Basin (Eastern Mediterranean)☆ , 1987 .