Testing the impact of stratigraphic uncertainty on spectral analyses of sedimentary series

Abstract

Abstract. Spectral analysis is a key tool for identifying periodic patterns in sedimentary sequences, including astronomically related orbital signals. While most spectral analysis methods require equallyspaced samples, this condition is rarely achieved either in the field or when sampling sediment cores. Here, we propose a method to assess the impact on the resulting power spectra of the uncertainty or error made in the measurement of sample stratigraphic position. We apply a Monte-Carlo procedure to randomise the sample steps of depth series using a gamma distribution. Such a distribution preserves the stratigraphic order of samples, and allows control of the mean and variance of the sample step distribution after randomisation. We test the Monte-Carlo procedure on two geological datasets and find that at 5 % uncertainty in the sample positions, the power of the spectral peaks is signifi cantly affected in all frequencies above ~1/3 of the Nyquist frequency. The randomisation process progressively affects the lower frequencies when increasing the level of uncertainty in the sample position. With 10 % sample position uncertainty, the change in relative power exceeds 10 % in all frequencies above ~1/5 of the Nyquist frequency. For robust applications of the power spectrum, we suggest strongly controlling the measurement of the sample position, keeping the variance of the sample distribution to a maximum of 10 % of the average sample step. In addition, the simulations indicate that taking at least 6-10 samples per precession cycle should allow calculation of robust power spectra estimates in the Milankovitch band.

Detection of Strong Precession Cycles from the Late Pliocene Sedimentary Records of Northeastern Tibetan Plateau

“…Although the fluvio‐lacustrine strata reliably record orbital cycles at 100‐kyr frequency, these sedimentary records may miss reliable higher‐frequency cycles due to autogenic processes and sedimentary hiatuses associated with the erosional nature of rivers, and potentially large differences of sedimentary accumulation rates (SARs) between fluvial and lacustrine environments. These factors, together with nonlinear climate response and a not perfectly uniform sampling rate, tend to decrease spectral power of higher frequency orbital cycles (i.e., precession and obliquity cycles) and enhance the spectral power of lower frequency ones (i.e., short and long eccentricity cycles), resulting in potential artifacts of stronger eccentricity power and weaker precession and obliquity power (Martinez et al, ; Nie et al, ), preventing a confident interpretation of climate changes at orbital timescales in the northeastern Tibetan Plateau.…”

Depositional changes during the Danian–Selandian transition in Loubieng (France), Zumaia (Spain) and Sidi Nasseur (Tunisia): insights from and limits of rock magnetism

“…Batch B2 contains 266 samples, making Loubieng the most completely sampled section of the three. However, the sampling resolution is not sufficient to allow the robust detection of short-term cycles (Martinez et al 2016), which is beyond the scope of this study.…”

Multi-proxy record of orbital-scale changes in climate and sedimentation during the Weissert Event in the Valanginian Bersek Marl Formation (Gerecse Mts., Hungary)

“…The power spectra and the evolutive spectra are then interpreted together. Variations in the sedimentation rate can result in one Milankovitch-cycle being expressed over several frequencies on the 2p-MTM spectra of the sedimentary series (e.g., Weedon, 2003;Martinez et al, 2016). Changes in the sedimentation rate can be recognised in the evolutive spectra by the deviation of the spectral bands, assigned to the cyclicity in question (Martinez et al, 2015).…”

“…Low-density sampling in the studied time series can cause distortion of the spectrum in the precession band, making the record of the orbital cycles unclear (Weedon, 2003). Furthermore, a highly fluctuating sedimentation rate can smooth the power spectrum at high frequencies and decrease the power and significance levels of the spectral peaks in an important proportion of the spectrum (see Martinez et al, 2016). The effect of the low-density sample distance, and the variations in the sedimentation rate on the spectral analyses were tested on four ETP (Eccentricity e Tilt e Precession)series calculated for the Late Valanginian using the La 2004 (Laskar Fig.…”