Gutzler, D., G. Kiladis, G. Meehl, K. Weickmann, and M. Wheeler, 1994: The global climate of December 1992-February 1993. Part II. Large-scale variability across the tropical western Pacific during TOGA COARE. J. Climate, 7, 1606-1622.


INTRODUCTION

Recently, scientists from more than a dozen countries carried out the field phases of a project called the Coupled Ocean-Atmosphere Response Experiment (COARE), devoted to describing the ocean-atmosphere system of the western Pacific near-equatorial warm pool (Webster and Lukas 1992). The project was conceived, organized, and funded under the auspices of the International Tropical Ocean Global Atmosphere (TOGA) Program. Although COARE consisted of several field phases, including a year-long atmospheric enhanced monitoring period (1 July 1992-30 June1993), the heart of COARE was its four-month Intensive Observation Period (IOP) extending from 1 November 1992 through 28 February 1993.

This paper, Part II of the Seasonal Climate Summary for the boreal winter of 1992-93, places the weather and climate observed in the IOP into context with regard to large-scale, low-frequency fluctuations of the ocean-atmosphere system. Consequently, we consider aspects of tropical variability beginning in August 1992 and extending through March 1993, with some soundings data for April 1993. The post-IOP period also includes most of the field phase of CEPEX, the Central Equatorial Pacific Experiment (CEPEX Editorial Advisory Board 1993). We shall refer to Part I of the Seasonal Climate Summary (Bell and Basist 1994, hereafter referred to simply as Part I) for placing the western Pacific variability into global context. Several complementary overviews of initial COARE results are now being prepared by other authors, including one emphasizing the upper oceans (Lukas et al. 1994), and one featuring results from the Australian BMRC Tropical Analysis and Prognosis System (McBride et al. 1994).

Several western Pacific domains were defined for study during COARE (Webster and Lukas 1992). In this paper, we shall emphasize variability over the large-scale surroundings array (LSA) and the intensive flux array (IFA). The LSA is bounded by latitudes 10°S and 10°N and longitudes 140°E and 180°. A network of sounding sites was established for COARE within the LSA, including both operational radiosonde stations and temporary research sounding sites, with the aim of maintaining a soundings array with 10° latitude by 10° longitude resolution for studies of the large-scale circulation.

Embedded within the LSA just south of the equator is the IFA (Fig. 1). The IFA is bounded by a quadrangle formed by two islands, Kavieng (3°S, 151°E) and Kapingamarangi (1°N, 154°E), and two Chinese research vessels, the Kexue 1 (which maintained a slightly varying position about 4°S, 156°E) and the Shiyan 3 (whose position was maintained about 2°S, 158°E). The IFA was the target of a succession of aircraft missions and ship-based oceanographic and radar operations during the IOP, aimed at the better understanding of ocean-atmosphere interactions and studies of the development and organization of tropical convective systems over the warm pool.

In this overview of large-scale atmospheric variability during COARE, we will rely for the most part on the same operational data products, produced at the U.S. National Meteorological Center (NMC), that are used in Part I. The circulation data are derived from final initialized NMC analyses. It should be noted that as a result of COARE, many more soundings were taken across the western Pacific during the IOP than are usually available to NMC. Most of the special COARE soundings were transmitted via the Global Telecommunications System in time to be fed into the assimilation scheme at NMC, but operational constraints preclude determination of how many of these extra soundings were recognized and accepted by the assimilation algorithm. Thus, at present we cannot assess the extent to which the special COARE data affected the NMC analysis. Reanalysis experiments could be performed, first including and then excluding the special COARE soundings to quantitatively address this issue, and such experiments are currently being contemplated.

Our description anticipates more detailed analyses, forthcoming from COARE researchers, based on the soundings and surface data from the LSA as they are postprocessed. In Section 2 of the paper, we describe the low-frequency evolution of SST and atmospheric circulation across the western tropical Pacific during the months leading up to the IOP, culminating in the observed seasonal mean and anomaly fields during the IOP itself. Consideration of this low-frequency variability of the coupled ocean-atmosphere system is essential for describing the higher-frequency transient variability within the COARE LSA and IFA.

Subseasonal fluctuations within the COARE domains are discussed in Section 3, focusing on variability of convection and winds across the IFA. The wind data used in this section come from the network of Integrated Sounding Systems (ISS) deployed for COARE. An ISS is a complete portable weather station that includes surface observations, radiosonde balloon capability, and a lower-tropospheric wind profiler (Parsons et al. 1994). The ISS units provide extensive observations of the vertical structure of the atmosphere. An ISS was deployed at each of the four stations forming the boundary of the IFA (Fig. 1). Two other island ISS sites, Manus (2°S, 147°E) and Nauru (1°S, 166°E), are located to the west and east of the IFA, thus forming a line of near-equatorial stations extending across the almost 20° of longitude within the LSA.