This article is from
Journal of Creation 37(1):6–7, April 2023

Browse our latest digital issue Subscribe

How accurate are marine microfossil transfer functions?

by Michael Oard

Secular scientists apply uniformitarianism to almost everything that happened in the past. For instance, they are interested in past oceanic variables, especially sea surface temperature (SST), during their many ice ages or their warm periods of the Cretaceous and Cenozoic. It is widely believed the species composition of the marine plankton predicts SSTs. So, they use present-day distributions as an analogue to calculate SSTs from fossil assemblages from deep-sea cores, applying straightforward uniformitarianism.

The development of transfer functions

Uniformitarian scientists have drilled thousands of deep-sea cores, which usually have microfossils. In order to determine what these microfossils mean in regard to environmental variables, they employ transfer functions.

Image: Hannes Grobe/AWI, Wikimedia / CC BY 3.0

Transfer functions commonly use foraminifera (figure 1), and the most common environmental variable of choice is SST. This is assumed to be either the temperature at the 10 m level or the average temperature over 75 m.¹ They measure present-day SSTs and take samples of the water and analyze the foraminifera. In this way, they obtain a relationship, i.e. an equation, between SST and species distribution. This equation is then applied to the marine microfossils in the deep-sea cores to calculate past SSTs, which can be related to other paleoclimatic variables, such as atmospheric temperature.

Transfer functions can be inaccurate

A problem that has not been considered until recently is that the species of foraminifera used to develop the transfer function may bias the equation.² It is common practice to use as many species of foraminifera as possible, collected in the surface layer, to determine the transfer function. However, it is now known that “not all species are sensitive to temperature, and their distribution may be governed by other parameters.”³ The researchers discovered that a majority of the foraminifera do not carry useful temperature information, although those scientists who developed the transfer function have assumed all, or nearly all, do contain useful information. They also found that transfer functions, developed with a smaller sample of species, resulted in different SSTs. And the difference is not trivial: “The differences can be up to several degrees Celsius and variable in time and space, with important implications for paleoceanographic interpretations.”⁴ It also “adds a previously unrecognized source of uncertainty to quantitative assemblage-based reconstructions.”⁵

Thus, only a small number of species appear to indicate SSTs. Some of the taxa have a neutral effect, while other species included can have a major effect on the transfer function. It may be difficult to determine which foraminifera should be used. Various combinations of species could be used, but it would be difficult to decide which transfer function is correct without ‘independent’ information: “Taken at face value and in the absence of independent evidence, such inherent ambiguity renders it impossible to decide which of the reconstructions is more realistic.”⁵ At this point in time (2022), the uncertainty cannot be determined for any transfer function. Therefore, all transfer functions must be viewed with caution. These results not only apply to foraminifera and SSTs, but also other planktonic microorganisms and other paleoecological reconstructions.

Need more understanding for microorganisms and their fossils

The researchers believe that they also “need a more mechanistic understanding of the factors that determine species assemblage composition in the sediments.”⁶ Species inhabit certain vertical and seasonably distinct habitats, “adding even more complexity to deriving a single environmental variable from an assemblage of different species.”⁶ In fact, it is now known that planktonic foraminifera live in a broad depth range.¹ And since oceanic temperatures decrease downward, especially in the tropics, this vertical zonation can have a drastic effect on species abundance. So, species abundance of planktonic foraminifera is a poor measure of SSTs.

Then there is the problem that ocean temperatures were different from today during glaciation or during supposed warm periods. This temporal change can cause different species abundances that would not be picked up with a transfer function. For instance, different species abundances in the South China Sea during the last glacial maximum are unique, with no modern analogues, throwing off transfer functions developed from present-day distributions.⁷ The transfer functions gave glacial SSTs about the same as today or a little warmer, while geochemical information and terrestrial proxies gave temperatures 2–5°C cooler than today. One would think such results would disqualify the use of transfer function to determine past conditions.

There are additional ecological and geological processes on the bottom of the ocean that can affect the fossil foraminifera composition,³ throwing off transfer function deductions. Some of these variables are variable shell dissolution deep in the ocean and even within the sediment.⁸ Also, bottom currents can erode micro-organisms and deposit them elsewhere, and bioturbation can mix the sediments.⁹

Creation scientists should be aware that the uniformitarian quantitative estimates of past paleoenvironmental variables using transfer functions can be off considerably. This should make us wary of uniformitarian paleoceanographic deductions.