Australia: The Land Where Time Began

A biography of the Australian continent 

Mantle Plumes - Are They Periodic?

According to the authors1 evidence has recently been published of a number of biological and geological events that appear to be cyclical, showing cycles of similar length. Some examples include the diversity record of marine organisms over the past 540 My which shows evidence of 2 cycles in the data, cycles of about 62 My and of about 140 My (Rohde & Muller, 2005). Another study of long-term sequences in sedimentary basins showed a 56 My cycle (Myers & Peters, 2011), and a study of the marine strontium isotope record that demonstrated a 59 My cycle (Melott et al., 2012), and there may possibly be a similar period of the atmospheric carbon dioxide concentration over the past 542 My (Franks et al., 2012). There also appears to be a cycle of about 140 My in the long-term fluctuations of the global climate (Veizer et al., 2000; Mayhew et al., 2008).

The authors1 suggest that if the regular cycles around 60 and 140 My are real they would indicate large-scale Earth processes may be responsible for the cycles. As these cycles are much longer than the astronomical fluctuations that affect the Earth, such as changes in the Earth's orbit and axial tilt, the authors1 say scientists have increasingly looked to phenomena such as mantle convection and plume activity as possible explanations for these cycles.

Large igneous provinces (LIPs), large-volume, short-lived, eruptions that are primarily rich in iron that are not associated with the typical processes at plate boundaries, that are produced by eruptions associated with upwelling plumes. LIPs have been suggested to form when hot spots are initiated by upwelling of mantle plume heads impinging on lithosphere, either continental or oceanic (Richards et al., 1989). Various effects of LIPs on the environment have linked them to mass extinction events (Wignall, 2001; Arens & West, 2008), and it has been suggested they mal also be connected to changes in other records, their occurrence possibly following some as yet unrecognised pattern instead of each occurrence being an isolated event. Other authors have studied correlating events and underlying causes that could possibly link LIPs to cycles that have been recognised on the surface of the Earth.

Correlating mantle plume activity with cycles that occur in geological records

A study has been carried out searching for regular periods in LIP eruptions to test the proposal that the activity of mantle plumes follows cycles (Prokoph et al., 2013), in which they applied continuous wavelet transform analysis to data on fossil diversity of Rhode & Muller (2005), as well as on data on the distribution probability of the ages of LIP events that are known (Courtillot & Renne, 2003). The transformation of a time series into wavelet coefficients in "time" and "scale" (or frequency) domains is allowed by wavelet analysis. This method screens out values that are chaotic and simultaneously transforms these domains to search for cycles in the data, by the use of short filtering functions of various shapes and functions that have been called wavelets (Grossman & Morlet, 1984). Before 250 Ma the numbers of LIPs used in this analysis trails off, as a result of older Palaeozoic LIPs being more difficult to recognise.

The results of the wavelet analysis of data on fossils indicated 2 strong periods, at 62 My and at 140 My, both of which agree with the earlier spectral analysis (Rohde & Muller, 2005). Another cycle was also found in the fossil data at about 35 My, especially in the last 135 My since the Early Cretaceous. Wavelet analysis of LIP ages (Prokoph et al., 2013) also detected a cycle of about 62 My and and another of 140 My that was somewhat weaker. In the ages of LIPs a shorter cycle of 30-35 My was also detected, that was more pronounced during the last 135 My. These results of wavelet analysis agree with the results of Fourier analyses reported (Prokoph et al., 2013). It is suggested these results are real, not simply edge effects of time series truncation, by the strength and simultaneous onset of the 30-35 My cycle at 135 Ma in both data sets. It had been proposed that the 62 My cycle was the result of sampling (Smith & McGowan, 2005), but the authors1 say the discovery of the 62 My period of the marine strontium- and sulphur-isotope records (Melott et al., 2012; Prokoph et al., 2013) supports the conclusion that the period is not the result of sampling. Based on the 3 cycles in the LIP eruptions and the diversity of fossils, are negatively correlated, with times of high lava production being generally times when the diversity of fossils is low.

Correlation to cause

Rampino & Prokoph suggest cyclicity that may be found in mantle plume activity that directly produces LIPs is a possible driver for the biodiversity and other geologic cycles (Richards et al., 1989), though the question of why mantle plumes would exhibit regular periodic activity must be answered before such a conclusion is reached.

Regular periodic or quasiperiodic plumes may be generated in the mantle. Repeated generation of instabilities in the thermal-boundary layer, the D-layer, just above the boundary between the core and the mantle, has been suggested as a mechanism for the formation of deep mantle plumes (Olson et al., 1987). A flux of heat which marks the D-layer from the core into the lower mantle, with the more buoyant material accumulating, which eventually becomes unstable and rises as narrow plumes (Loper & Stacey, 1983). It has been estimated that such instabilities in the boundary layer can reach a critical threshold, then grow into plumes in about 50-100 My (Olson et al., 1987).

Subduction, in which cold plate material sinks into the mantle, and accumulates at the mantle discontinuity at 670 km (Solheim & Peltier, 1994). The authors1 suggest periodic avalanches of material from the upper mantle down to the lower mantle where a return flow from the deeper mantle may result from the buildup of this unstable material (Machetel & Humler, 2003).

Plate geometry is another proposed driver of periodicity in plumes and mantle convection. According to this proposal a supercontinent acts like a blanket over the mantle resulting in a buildup of heat in the mantle beneath the continental lithosphere, which eventually leads to the generation of hot spots and LIP eruptions (O'Neill et al., 2009). Therefore supercontinents may carry the seeds of their own destruction when rifting results from hot spot activity (Storey, 1995). In some numerical models periodic behaviour of convection systems has been observed (Zhang & Libchaber, 2000). It has been suggested, based on past findings, that non-linear convection, that is time-dependent, in the mantle may exhibit periodic or quasiperiodic behaviour before more chaotic flow occurs (Hansen & Ebel, 1988).

The authors1 suggest that before a consensus can be reached by the scientific community on the existence of regular cycles in the activity of mantle plumes (Hannisdal & Peters, 2011; Smith & McGowan, 2005), more statistical analyses of possible periodic components of other events in the geologic record are required.

Sources & Further reading

  1. Rampino, Michael R., and Andreas Prokoph. "Are Mantle Plumes Periodic?". Eos, Transactions American Geophysical Union 94, no. 12 (2013): 113-14.


Author: M. H. Monroe
Last updated  14/06/2013
Journey Back Through Time
Experience Australia
Aboriginal Australia
National Parks
Photo Galleries
Site Map
                                                                                           Author: M.H.Monroe  Email:     Sources & Further reading