The authors recently proposed (Nigam et al. 2017) that this seasonality originates from land surface–hydroclimate interactions and, to an extent, also from the secular change in low-level atmospheric circulation and related thermal advection. It was argued that the winter dormancy and summer vigor of the hydrologic cycle over the middle-to-high-latitude continents elicits different responses to the additional incident radiative energy received at the surface from increasing greenhouse gas concentrations. The considerable overlap of the observed summer-minus-winter evapotranspiration trends with the winter-minus-summer SAT trends over northwestern North America was cited in support. Also encouraging for the advocated mechanism was the analysis that showed that summer evapotranspiration trends consumed approximately two-thirds of the additional incident radiative energy that needed latent disposition.
The state-of-the-art climate system models that informed the IPCC-AR5 assessment, unfortunately, show both weak seasonality and unrealistic structure in the historical simulation of SAT trends – effectively precluding their use in investigation of the seasonality in observed SAT trends.
The presentation will describe new analyses of the observed secular temperature trends, this time focusing away from the surface, into the troposphere; as the authors continue to seek additional corroboration of their hypothesis for SAT seasonality from the analysis of related observations.
Seasonally-stratified secular trends in upper-air temperature in the very region showing notable seasonality in surface temperature trends – northern Canada, where winter trends are much larger than summer ones (e.g., Nigam et al. 2017) – are examined in the post-IGY period (1959-2015) from radiosonde data and atmospheric reanalysis (and historical climate model simulations) to shed mechanistic insights on the origin of surface and upper-air trends.
The upper-air secular trends in NCEP Reanalysis are shown to be remarkably similar to those computed from radiosonde data, providing a basis for the use of NCEP Reanalysis in the characterization of regional vertical structure. The focus is not on the canonical structure of trends – tropospheric warming and stratospheric cooling – but on its seasonal variation.
The winter-minus-summer difference in secular trends over northern Canada is positive and large in the lower troposphere (500-hPa) and stratosphere (50150-hPa); it is largest at the surface and smallest at the tropopause. The surface-trapped vertical structure of the lower tropospheric trend difference supports the attribution of the notable seasonality of surface temperature trends to land-surface–hydroclimate interactions (Nigam et al. 2017).
In the lower stratosphere, a cooling trend is evident in all seasons – not unexpectedly – but a pronounced seasonality is again apparent, with the strongest cooling in summer.
Examination of the period trends in historical climate simulations reveals that models are more challenged in the lower troposphere than in the lower stratosphere, likely, from deficiencies in the representation of land-surface–hydroclimate interactions. The near-zero trend tropopause region is a rare point of confluence for the simulated and observed trends.
The seasonal cycle of climate, despite its monotony, provides an expanded phase space for the exposition of the dynamical and thermodynamical processes generating secular warming, and an exceptional cost-effective opportunity for benchmarking climate projection models.
Nigam, S., N. Thomas, A. Ruiz-Barradas, and S. Weaver, 2017: Striking Seasonality in the Secular Warming of the Northern Continents: Structure and Mechanisms. J. Climate, 30, 6521-6541 (15 August).
Thomas, N., V. Ravi, and S. Nigam, 2017: Seasonality in the Vertical Structure of Secular Temperature Trends over North America. To be submitted to the J. Climate in August 2017.