Towards an unbiased estimate of fluctuations in reef abundance and volume during the Phanerozoic

Abstract. The globally preserved number and volume of ancient biogenic reefs is strongly biased by two factors: geological history and research intensity. These biases are sufficiently strong to cast doubts on the biological meaning of the recorded raw pattern. Without adjustment, it is hard to reliably identify factors potentially controlling the waxing and waning of this important ecosystem through time. Although it is currently impossible to completely compensate for the biases, I demonstrate here, based on a comprehensive database of ancient reefs, that spatiotemporal heterogeneities of the biases can largely be evened out by: (1) omitting oceanic reef sites and reef sites only known from subsurface exploration; (2) standardizing for economic factors known to affect research intensity; and (3) adjusting for sedimentary cycling processes. The resulting curves of fossil reef abundance and volume appear quite different from the original ones but the patterns of waxing and waning of the time series are not significantly altered and the overall volatility is not reduced. This suggests that both the raw curves and the adjusted curves correctly depict the basic timing of major reef blooms and declines. Nevertheless, the general pattern of the new curves (maximum proliferation in the middle Paleozoic, decline thereafter) is more in line with patterns of global cratonic carbonate sedimentation than the original curves. The adjusted curves should thus be preferred over the original ones for future tests of potential extrinsic controls of Phanerozoic reef development.