Abstract

A set of numerical simulations has been carried out to evaluate the influence of coupled land–atmosphere and ocean–atmosphere interactions on natural climate variability. The baseline experiment was a long integration of a state-of-the-art-coupled atmosphere–ocean–land general circulation model (GCM). A sensitivity experiment was conducted in which the ocean and atmosphere were fully interactive but the soil moisture was specified. This paper describes a connection found between land–atmosphere coupling and midlatitude sea surface temperature (SST) variability in the North American–North Atlantic sector. Specifying soil moisture results in a reduction in surface and atmospheric temperature variability and also an increase in net heat flux variability. Surface temperature variance is reduced because it is constrained by the fixed soil moistures. Since the surface temperature cannot equilibrate with a given atmospheric anomaly, the resulting heat flux will be quite large and will act to damp the atmospheric anomaly. This is consistent with larger heat flux variance and reduced temperature variance in the simulation with suppressed land processes. SST anomalies in the midlatitude Atlantic are sensitive to air temperature and moisture anomalies modified over the North American continent, so it is not unexpected that SST variance is significantly reduced when land temperature variability decreases. Oceanic ‘re-emergence’ operates in both simulations but is weaker in the fixed soil moisture integration, particularly in a region of the western North Atlantic contiguous with North America. Reemergence is the mechanism by which late winter ocean temperature anomalies are sequestered below the stable summer ocean mixed layer and reentrained into the deepening autumn mixed layer. The larger oceanic anomalies in the fully coupled simulation decay more slowly and are a partial explanation for stronger reemergence. However, during the second winter, the atmospheric forcing favors the same sign of SST anomalies as those reemerging and, therefore, acts to reinforce the anomalies in the fully coupled simulation. An area averaged SST index was constructed for the region of the western North Atlantic where reemergence was most notably reduced. This index was used to construct composites which suggest that, in the fully coupled model, land surface temperature and SST anomalies both reemerge the second winter, whereas in the suppressed land processes simulation, there is no winter-to-winter reappearance of land surface temperature anomalies. The late winter land temperatures are able to reemerge in fall because of the persistence of soil moisture anomalies.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call