RECONSTRUCTING PALEOECOLOGY AND PALEOENVIRONMENTAL VARIABLES USING FACTOR-ANALYSIS AND REGRESSION - SOME LIMITATIONS

We test the success of Principal Components, Factor and Regression Ana lysis at recovering environmental signals using numerical experiments in which we control species environmental responses, the environmental conditions and the sampling scheme used for calibration. We use two g eneral conditions, one in which sampling of a continental margin for b enthic foraminiferal assemblages is done in a standard grid and the dr iving environmental variables are correlated to one another, and the o ther where sampling is done so that the environmental variables are un correlated. The first condition mimics many studies in the literature. We find that where the controlling environmental variables are correl ated, Principal Components/Factor Analysis yield factors that reflect the common variance (correlation) of those variables. Since this commo n variance is largely a product of the sampling scheme, the factors ex tracted do not reliably present true species ecologic behavior. This b ehavior cannot be accurately diagnosed and faulty interpretations may lead to substantial error when using factor coefficients to reconstruc t conditions in the past. When the sampling scheme is constructed so t hat the controlling environmental variables for the calibration data s et are uncorrelated the factor patterns will reflect these variables m ore accurately Species responses can be more successfully interpreted from the Principal Components/Factor Analysis structure matrices. Addi tionally, regression analysis can successfully extract the independent environmental signals from the biotic data set. However, matrix closu re is a confounding effect in all our numerical results as it distorts species' abundances and spatial distribution in the calibration data set. Our results show clearly that a knowledge of the controlling envi ronmental variables, and the correlations among these variables over a study area, is essential for the successful application of multivaria te techniques for paleoenvironmental reconstruction.