Chronology errors and their effects on the recovery of characteristic time scales of the geodynamo from relative paleointensity

The intensity of Earth’s magnetic field has been modeled as a continuing sequence of growths and decays due to a rotating parametric instability (RPI) of the fluid core. Taking paleomagnetic intensity as a proxy for the turbulent fluid velocity field, timescales of motion in the fluid core and geodynamo can be inferred from estimates of growth and decay. While such inferences may be in error from several possible sources, this work examines the effect of age errors on relative paleointensity data from cores of marine sediments. True temporal variations in paleomagnetic intensity are distorted by stretching and compressing the observations in time and displacement in age of paleomagnetic features due to errors in the ages of paleomagnetic samples. For the interval 0–75 ka, growths and decays previously found in three single-site records are seen in one composite stack but not in a second composite stack over the same age interval. The coherence spectra for all pairs of individual records that make up the stack illustrate the apparent smoothing of relative paleointensity records that are stacked over a common age interval. We further demonstrate that the events seen in both the stack and the single core records are predictable from the same method of pairwise coherence. Thus single-site and composite stacks can both be used to estimate growth and decay rates of the paleomagnetic field, provided constraints from coherence of pairs making up the stack are respected. It is also concluded that temporal properties of a growth and decay process are best obtained from individual records and then stacked by event, even though these events may appear at different times as a result of discrepancies in age models. The main contribution of this work is therefore to confirm the previous analysis of growth and decay in sequences of relative paleointensity using a newly developed analytical model.