El Niño as a mediator of the solar influence on climate

Abstract

[1] Using a climate model of intermediate complexity, we simulate the response of the El Niño-Southern Oscillation (ENSO) system to solar and orbital forcing over the Holocene. Solar forcing is reconstructed from radiocarbon production rate data, using various scaling factors to account for the conflicting estimates of solar irradiance variability. As estimates of the difference since the Maunder Minimum range from 0.05% to 0.5% of the solar ''constant,'' we consider these two extreme scenarios, along with the intermediate case of 0.2%. We show that for large or moderate forcings, the low-pass-filtered east-west sea surface temperature gradient along the equator responds almost linearly to irradiance forcing, with a short phase lag (about a decade). Wavelet analysis shows a statistically significant enhancement of the century-to-millennial-scale ENSO variability for even a moderate irradiance forcing. In contrast, the 0.05% case displays no such enhancement. Orbitally driven insolation forcing is found to produce a long-term increase of ENSO variability from the early Holocene onward, in accordance with previous findings. When both forcings are combined, the superposition is approximately linear in the strong scaling case. Overall, the sea surface temperature response is of the magnitude required, and is persistent enough, to induce important climatic perturbations worldwide. The results suggest that ENSO may plausibly have acted as a mediator between the Sun and the Earth's climate. A comparison to key Holocene climate records, from the Northern Hemisphere subtropics and midlatitudes, shows support for this hypothesis.

Solar forced transient evolution of Pacific upper water thermal structure during the Holocene in an earth system model of intermediate complexity

“…With a warmer WPWP at surface, the stronger zonal SST gradient agrees well with the "ocean thermostat" hypothesis [3] and may indicate a solar forced ENSO-like response [14] at centennial scales. Further, the enhanced meridional TWT gradient is reversed with a meridional SST gradient (with warmer WPWP and relative cooler SST north of 30°N) in Western Pacific (Figure 2(a)).…”

“…In the ZC model [14], a zonal SST gradient linearly responds to TSI forcing and shows a short phase lag (about a decade) at periods of 200-500 and 800-1000 years. Similar periods appear in Mg/Ca-SST record of Northeast Pacific with 50-100 years' lag [13] and can be comparable with our SST results (Figure 4(b)).…”

“…The enhanced response of the Western Pacific Warm Pool (WPWP) is first recorded in surface temperature anomalies between the Medieval and the Little Ice Age [12]. The centennial zonal oscillations of Pacific SST forced by solar activities are also significant in the Mg/Ca records of early Holocene [13] and supported by a transient Holocene simulation of the El Niño-Southern Oscillation (ENSO) [14]. However, this simulation used a simple Zebiak-Cane (ZC) model constrained in tropical Pacific and it is still unknown whether their results are independent of model.…”

Palaeoclimate reconstructions reveal a strong link between El Niño-Southern Oscillation and Tropical Pacific mean state

“…2a). The variability of both Mg/Ca and d 18 O values obtained using repetitive analyses of small volume samples of N. dutertrei (Figs 2b,c and 3a-c) show similar changes across the deglaciation. Minimal variability is recorded in the interval representing the early Holocene (9.1 ky), whereas largest variability is found during intervals corresponding to the glacialinterglacial transition (12.5, 15.1 and 17.9 ky) (Fig.…”

“…Analysing ENSO data from observations and modelling experiments, he found 10-20-year cycles within the ENSO system, which can be described as alternating periods of enhanced ENSO variability and decreased zonal gradient with periods of subdued ENSO variability and increased zonal gradient. These decadal changes were attributed to the nonlinear dynamics of the ENSO system 12,18,19 . The nonlinear dynamic may be important for interpreting our results because the total variability reconstructed using proxy data represents a millennial average and consequently includes multiple components of ENSO variability.…”