The Climate Response to Solar Forcing in the Last Millennium Ensemble Simulations

Interest in linking solar and climate variations has motivated many modeling and proxy based studies. Over the past millennium, most estimates of total spectral irradiance (TSI) suggest variations are generally small, on the order of 1 Wm^-2 between the Maunder Minimum and present day, and about half this for variations across the 11-year solar cycle.

I will present results from analyses of the Last Millennium Ensemble (LME; http://www.cesm.ucar.edu/projects/community-projects/LME/) that explore the climate response to solar forcing over the last millennium (C.E. 850-2005). The LME employs the CESM_CAM5 fully coupled climate model, which contains representations of the physical processes necessary to simulate the bottom-up mechanism linking solar variations with climate responses.

Analyses are performed that separate the response to high-frequency (HF; a few decades, containing the 11-year cycle) and low-frequency (LF; multi-decadal to multi-centennial) solar forcing. On both time scales, weak but statistically significant responses are detectable in surface and near surface air temperature (order of tenths of a ^oC), precipitation (order tenths of a mm day^-1) and sea-level pressure (order tenths of a hPa).

The responses to both LF and HF forcing are both global in scale but distinct from one another. The response to LF solar forcing has a simple spatial structure that may be characterized as global warming with polar amplification. Sensitivity experiments indicate that the polar signal amplitude varies non-linearly with forcing amplitude.

The spatial structure of the response to HF forcing is more complex. The most prominent feature is an Interdecadal Pacific Oscillation pattern. Equatorial SST, precipitation, SLP and near surface wind anomalies associated with this pattern and located in the Pacific and Indian Ocean basins indicate a Bjerknes feedback between the atmosphere and ocean. Other response features are located in the tropical Atlantic, mid-latitudes of both hemispheres, Siberia and the Arctic.

In spite of positive feedbacks, the simulated climate response to solar forcing is weak, with reliable detection possible because a large sample size was available for analysis. This suggests that it may be difficult to detect solar forced signals in proxy based and modern observed climate records.