It’s time for another part of my review of palaeoecological evidence for solar forcing.
Rigozo et al (2007) have a tree-ring width record from the conifer Pilgerodendron cupressoides in Southern Chile in which they look for evidence of solar forcing. The raw data are presented in Rigozo et al (2006), a paper that differs from Rigozo et al (2007) essentially only in that it a) shows the raw data, b) uses periodiograms rather than wavelets.
There are ring width records from six trees, which seems rather thin. Ontogentic effects are removed with a wavelet smooth that removes trends and periods longer than 200 years. The success or otherwise of this operation will greatly affect long periodicities in the data, with less impact on shorter periodicities I want to focus on.
Rigozo et al (2007) apply wavelet analysis to their combined tree-ring record and find an ~11 year cycle during part of their record, which they relate to the 11 year solar cycle. Solar activity could affect the trees’ growth either directly through the increased dose of uv-radiation at solar maxima, or indirectly through climate. Rigozo (2006, 2007) only discuss the latter possibility, without suggesting a mechanism. They do not even report whether high or low solar activity is associated with wide tree rings. There might be a reason for that – the wavelet analysis detects an eleven year cycle only during the Maunder minimum, a period when there were few sunspots and no discernible sunspot cycle.
The authors do not discuss this conundrum, that the trees only display the characteristic eleven year cycle when the sun doesn’t. At least not in sunspot numbers: there are papers reporting that the eleven year cycle continues in aurora activity (Schröder 2008), and in radiocarbon production. Estimates of solar forcing are flat within the Maunder Minimum (IPCC, 2013).
So what to make of this paper? Noisy proxy data, overlain by the effects of El Nino, are going to struggle to detect the eleven year solar cycle at the best of times. And the Maunder Minimum is not the best of times – the consensus opinion seems to be that solar variability was limited within this period. Pilgerodendron-affecting solar variability could represent the manifestation of some unknown solar physics linked either directly or indirectly to the trees. Alternatively, the eleven year periodicity within the Maunder Minimum could represent random fluctuations.
At the moment, I’m going to go with the latter explanation as there is a huge problem with publication bias. Tree-ring data with 11-year cycles get published as evidence of solar forcing. Tree-ring data without 11-years cycles are filed in a drawer or published with some other story.
Rigozo et al (2006) Spectral Analysis of Tree Ring Width Time Series from Chile (1587-1994 A.D.). Trends in Applied Sciences Research 1, 73-78.
Rigozo et al (2007) Solar and climate signal records in tree ring width from Chile (AD 1587–1994). Planetary and Space Science 55, 158–164