A 2.70-Year / 2.70-Month Cycle Pair

Oct 11, 2026 · @Ray Tomes

Introduction

I have found a cycle of 2.701 years in C14 data and a cycle of 2.697 months in Milan rainfall. Their ratio is 12.018 ± 0.022, consistent with an exact 12:1 harmonic relationship. Because Dewey's common cycles have ratios of 2 and 3 there are many instances of ratio 12 between them (2x2x3).

Neither period is one of Dewey's well-known cycles. A search of the literature and the FSC cycles catalogue shows they are not rare either. This report lists every report found within ±2% of either period (2.646–2.754 years and 2.646–2.754 months, i.e. 80.5–83.8 days), plus near misses.

My findings

The ratio is 32.412 / 2.697 = 12.018 ± 0.022, under 1σ from exactly 12. A factor of 12 (2² × 3) is one of the most commonly expected harmonic ratios.

The monthly period is not simply three solar rotations. Three Carrington synodic rotations (3 × 27.275 days) give 81.83 days, about 0.2 days or 2σ short of 82.03 days. Nor is it related to the solar Rieger cycle, which is close to 154.7 days; half of that is 77.35 days.

Reports near 2.70 years (32.4 months)

The closest independent match is an FSC lake-levels cycle at exactly 2.701 years, identical to my C14 value.

Kononovich & Shefov note the solar QBO was first found in the early 20th century, and describe it as a wave train whose period and amplitude decrease, which explains the spread of reported periods. The 33-month FSC entries sit at the edge of the window and are likely Dewey-era round-number estimates.

Near misses: 2.8 yr in the Sun's equatorial rotation rate around 1990 (wavelet study, +3.7%); FSC lake levels at 2.59 yr (−4%) and 2.905 yr (+7.5%).

Reports near 2.70 months (82.2 days)

All independent reports near 82 days come from solar and cosmic-ray data; none appears in the FSC catalogue.

Near misses:

• 84 d (+2.2%): strongest peak in Bai & Sturrock's 1966–76 spectrum (NASA NTRS); also sunspot activity in cycle 23, with 85 d in the southern hemisphere (cycle 23 study).

• 86 d (+4.6%): sunspot areas in cycle 22 (Springer).

• 78 d is excluded: it is half the ~154.7-day solar cycle, not this period.

Discussion and suggested tests

The 2.70-year period appears across solar, atmospheric, hydrological and economic data; the 2.70-month period so far only in solar and cosmic-ray data, plus Milan rainfall. That pattern fits a real but minor cycle, which may be why Dewey never singled it out.

Suggested follow-ups:

• Trace the FSC 2.701-year lake-levels entry to its original source: which lake, what span, who reported it.

• Look for other subdivisions of 2.70 years in the same data: 1/2 (16.2 mo), 1/3 (10.8 mo), 1/4 (8.1 mo), 1/6 (5.4 mo). Matches at similar precision would strengthen the harmonic family.

• Check whether the solar 79–83 day reports cluster near 82.0 days when re-measured with longer data.

The FSC entries come from the parts of the catalogue I had to hand; a full export may contain more. Readers who know of other reports near either period are invited to add them here.

Sources

• Kononovich & Shefov – Quasi-biennial variations of solar activity (2002)

• Zharkova et al. – Periodicities of solar activity and solar radiation (2023)

• Time variations of the solar neutrino flux (wavelet analysis)

• Oláh et al. – Multiple and changing cycles of active stars II

• Short-term periodicities in the Sun's surface rotation

• Özgüç & Ataç – North and south major flare periodicities, cycle 20

• López Comazzi & Blanco – Periodicities in neutron monitor counting rates (2022)

• Search for a clock related to the solar 154-day complex (Bai & Sturrock)

• Periodicities in sunspot activity during solar cycle 23

• Short-term periodicities in sunspot areas during solar cycle 22

• Tomes – Solar 154-day cycle (CRI)

• Foundation for the Study of Cycles period tables, via the Cyclos catalogue