GFDL - Geophysical Fluid Dynamics Laboratory

LARGE-SCALE ATMOSPHERIC & OCEANIC VARIABILITY

Introduction

This is an area which has served as a focal point for many climate researchers over the past several decades and will continue as such for the indefinite future. While much of my earliest work was geared towards climate prediction (long-range forecasting), a necessary element was diagnostic analysis. In this area I discuss only the purely diagnostic studies, although the predictive ones certainly did explore the nature of atmospheric and oceanic variability along with their interactions. I have divided these diagnostic studies into two categories: (1) Extratropical and (2) Tropical & Tropical/Extratropical Interactions.

Extratropical

The work reported in Lanzante (1983) was aimed at examining, within a simple framework, the communication between the extratropical oceans and atmosphere. It examined the extent to which horizontal gradients in sea surface temperature and the (geostrophic) wind in the atmosphere vary in a consistent fashion on monthly and seasonal time scales in the North Pacific and North Atlantic Oceans. The expected relationship was found and was stronger for longer (seasonal) time scales. Lag analyses indicated that the relationship holds contemporaneously and with the atmosphere leading the ocean, but not when the ocean leads the atmosphere.

A more sophisticated approach with broadly similar objectives (Lanzante 1984; Lanzante and Harnack, 1984) utilized an eigenvector analysis technique which was a forerunner of singular value decomposition (SVD). This study identified large-scale SST patterns (covering both the North Pacific and North Atlantic Oceans) and their coincident atmospheric circulation patterns. In essence these are statistically coupled atmosphere-ocean patterns. Several well-known atmospheric teleconnection patterns (such as the North Atlantic oscillation and Pacific/North American pattern) appeared. In addition, two distinct warm-season “drought” modes were found with a striking similarity to patterns identified by Namias. The lag relationships showed that for the most part the dominant sequence is the atmosphere leading the ocean; this was found to be stronger in the Pacific than in the Atlantic. Again, there is not much support to the notion that the ocean leads the atmosphere in the extratropics.

Another study which falls into this category is Lanzante and Harnack (1983), which reports the results of my M.S thesis project. Since this also relates to the seasonal cycle it is discussed above in the appropriate section. This study examined some of the mechanisms responsible for the development of SST anomalies in the extratropical oceans during summer. It considered the effects of changes in the depth of the oceanic mixed layer as well as oceanic advective processes induced by wind-driven currents. In considering both anomalous and climatological mean SST and wind driven-currents it was found that the dominant term is the advection of the mean SST field by the anomalous wind-driven upper ocean current.

Tropical & Tropical/Extratropical Interactions

One of the earliest research projects in which I participated (Harnack et al. 1982), involved an examination of the relationship between some tropical parameters (anomalies of trade winds and tropical SSTs) and the extratropical atmospheric circulation. The now well known fact that the ENSO related signal in the tropics is transmitted to the extratropics at a lag up to several seasons later was found. This study also examined the seasonality of these relationships and was perhaps the first to identify the so called “spring barrier” whereby lag relationships involving ENSO signals break down at lag during boreal spring.

Subsequent work of mine  (Lanzante 1991; Lanzante 1996) has concentrated on ENSO related SST variability in the tropics. The latter study was motivated by the inconsistencies in the literature concerning lag relationships between tropical Pacific and tropical Atlantic SST’s. There were a variety of investigators who claimed either a positive, negative or no relationship between these two areas. I investigated the lag relations between SST anomalies throughout the tropics using a long historical record and identified four key areas. I found that ENSO related SST anomalies appear first in the eastern Pacific (EPAC), then appear several months later in the central Pacific (CPAC), and finally several months after this they appear in the Indian (IND) and North Atlantic (NATL) Oceans. The weakest link is the appearance in the Atlantic. I suggested that the SST anomalies in the more remote locations of the Indian and Atlantic Oceans are the result of the “atmospheric bridge” mechanism proposed by my Climate Diagnostics Group group colleagues Gabriel Lau and Mary Jo Nath.

Around the turn of the century I completed a suite of integrations of a unique coupled model involving a complex GFDL R30 atmospheric model and a simple model of the oceanic mixed layer. This coupled model was created by Mike Alexander of ESRL, in Boulder, CO, along with his colleague Jamie Scott. Mike created the oceanic component of the model and has used it quite a bit, in particular to study the “reemergence mechanism”, whereby extratropical SST anomalies which exist in the late winter and early spring become cut off from the surface and isolated below the mixed layer during summer and early fall, only to reemerge in late fall. The coupled model integrations which I ran incorporated prescribed forcing in the tropical Pacific (mimicing observations from the past ~50 years) but allowed the ocean model to predict the SSTs elsewhere. Mike is the lead on a paper which serves as a review of the “atmospheric bridge” and incorportes some results from these coupled model runs (Alexander et al. 2002).

Plans

Although my personal involvement with the recent coupled model integrations is on the “back-burner” I hope to return to this in the future.


Return to John Lanzante’s Home Page.