On using the seasonal cycle to interpret extratropical temperature changes since 1950

Abstract

Extratropical near‐surface air temperature variability is explored on three different time scales: the seasonal cycle, observed changes in temperature since 1950, and the equilibrium response to increasing CO2 in an atmospheric general circulation simulation with fixed sea surface temperatures. Exploration is undertaken using an energy balance model (EBM) that parameterizes advective land‐ocean heat fluxes. The EBM is tuned only to the climatological seasonal cycle yet captures 47% of the variability in observed multidecadal temperature changes in the extratropics and 78% of the variability in the equilibrated model simulation. The subseasonal time scale of atmosphere‐surface heat fluxes explains, at least in the context of this EBM, the ability to infer patterns of multidecadal change using information primarily drawn from the seasonal cycle.

Millennial-scale climate variability over land overprinted by ocean temperature fluctuations

“…If the macroweather-type behaviour over land held to longer timescales, internal variability would only play a minor role on the uncertainty of regional climate projections at multi-decadal and longer timescales 22,23 . However, a different scaling behaviour of ocean and land at longer timescales would imply an increasingly large variability discrepancy between terrestrial and marine regions, which seems physically implausible given their coupling by the atmosphere, and in contradiction with both diffusive energy balance models (EBMs) 24,25 and GCMs 17 . This leaves us with the conundrum that we must either reject altogether the marine proxies or see a fundamental change of variability in the terrestrial domain on longer timescales.…”

“…EBMs suggest that this parallel behaviour of land and oceans on long timescales is due to heat exchange between the land and ocean compartments. In such models, land air temperature can be described as a linear combination of the SST and a time-dependent forcing over land 25,38 ; the resulting variability spectrum over land is then a linear combination of the spectra of each term (Fig. 1b) when the two are uncorrelated (Supplementary Note 2).…”

“…Also, the discrepancy between RLI and the sub-decadal variability over East Asia suggests that RLI might not be an appropriate metric of continentality there. Notably, East Asia is one of the few regions where an EBM using RLI as a mixing parameter does not reproduce the seasonal gain (associated to continentality) skilfully 25 . The P value was obtained in the same way as for the sub-decadal variability binning, accounting for the spatial autocorrelation and the distribution of the fields.…”

Temperature annual cycle variations and responses to surface solar radiation in China between 1960 and 2016

“…One way of achieving this is to use the Fourier transform method, which can transform temperature change in time to different frequencies and then use the 1 year frequency to characterize the annual temperature cycle (Prescott and Collins, 1951; Thomson, 1995; Stine et al ., 2009). Many researchers have studied the variations in the annual temperature cycle using observational data or model simulations that utilize the Fourier transform method and have found that the warming rate for winter is greater than for summer (Dwyer et al ., 2012; McKinnon and Huybers, 2014). A small phase delay and an amplitude increase in the annual surface air temperature cycle have been found at the global scale (Wallace and Osborn, 2002; Stine et al ., 2009).…”