PEM-West A: Meteorological overview

. Phase A of the NASA Pacific Exploratory Mission in the western North Pacific (PEM-West A) region was conducted during September-October 1991. The background meteorology of eastern Asia and the western North Pacific region during the PEM-West A study is described. Mean large-scale flow patterns are discussed along with transient synoptic scale features (e.g., midlatitude cyclones, anticyclones, and frontal systems) responsible for long-range transport of trace species over the study region. Synoptic summaries are given for each of the 18 data flights, together with selected examples of meteorological processes that gave rise to some of the changes observed in the measured trace gases. Examples of large-scale ozone features observed above and below the DC-8 flight altitude by an onboard lidar system are also related to meteorological processes such as stratospheric-tropospheric exchange and upward transport of air from the boundary layer.


Biomass burning annually occurs over parts of western and central Indonesia prior to and during the September-October period. A factor that may have contributed to enhanced biomass burning during PEM-West A was the El Nino/Southern Oscillation (ENSO)
event which was under way, which induced an eastward shift of major convection leaving portions of the Indonesia region with below-normal rainfall [Janowiak, 1993]. Media reports indicated that over 70,000 hectares had burned, including 30,000 hectares in southern Borneo alone. This burning produced a massive smoke pall across that region which was often evident on visible GMS satellite imagery (

Transient Synoptic Scale Features
In the PEM-West study region, the major transient synoptic scale features affecting long-range transport into and throughout the study area are migratory cyclones and their associated frontal systems and tropical cyclones. During PEM-West A, migratory midlatitude cyclones would frequently form (an average of one per week) and intensify along the east coast of Asia or over the interior of the continent. This coastal area is a region of preferential cyclone formation due to the strong lower-tropospheric temperature gradient in the vicinity of the warm Kuroshio ocean current [Carlson, 1991]. As these cyclones moved across the Pacific, their trailing cold fronts formed the boundary between continental air of midlatitude (Asian) origin and marine air of either midlatitude or tropical (Pacific) origin. The mean positions of the two primary cold frontal zones (subjectively averaged from an ensemble of 12 hourly NMC frontal analyses) and the southernmost frontal locations are illustrated in Figure 6, together with the mean tracks taken by migratory cyclone and anticyclone centers. Cyclones tended to track eastward and, eventually, poleward as they crossed the Pacific. Anticyclones tended to form over northwestern China, then move eastward or southeastward in the wake of the transient cyclones.

The analysis reported by Gregory et al. [this issue
] suggests two basic transport pathways of aged marine air into the PEM-West A study area: (1) "north marine" air masses moving eastward, southward, and then westward around the periphery of the Pacific subtropical anticyclone; and (2) "south marine" air masses from the southeasterly trade wind flow or from the southern hemisphere. Pacific marine air into the study area. A weaker Pacific anticyclone in tandem with a stronger Asian anticyclone would tend to move the cold front to its southernmost position, bringing continental flow to a larger portion of the sampling region. Tropical cyclones recurving close to the east coast of Asia also induced a stronger continental flow along the western portion of the storm circulation, owing to the increased pressure gradient between the cyclone and the Asian anticyclone which often followed in its wake. Talbot et al. [this issue] also stress that at higher altitudes continental outflow can originate much farther upstream than eastern Asia. The reason for this is illustrated in Figure 7a which shows the latitudinal distribution of the polar jet (PJ) and the subtropical jet (STJ) and figure 7b shows a meridional cross section of the mean zonal wind illustrating the higher wind speeds encountered with increasing altitude.
During this flight and on flights 9 and 14, dimethly sulfide (DMS) mixing ratios were observed in the free troposphere (115 pounds per trillion volume (pptv)) comparable to those observed in the boundary layer (,--160 pptv). The free troposphere measurement was made at 2100 UT at 54øN, 171øW, and an altitude of 8 km, between the Kamchatka and Bering Sea low-pressure areas ("FT" in Fig-ure 11). The boundary layer measurement was obtained at 2230 UT, near 53øN, 174øE at an altitude of 365 m ("BL" in Figure 11). The aircraft down-looking video showed convective clouds reaching ahnost up to the aircraft flight level, suggesting a mechanism for the vertical transport of boundary layer The subtropical ridge had amplified slightly and built northward just to the east of Japan (Figure 13d). In the midtroposphere the deep cyclonic circulation associated with Typhoon Mireille was just southwest of Japan. The eye and extensive spiral bands and cirrus outflow of Typhoon Mireille were very prominent and covered much of Japan, the Sea of Japan, and the Yellow Sea. From satellite animation the best outflow aloft appeared to be from the northwestern quadrant, although outflow from the southeastern quadrant was also well defined. A satellite image and a discussion of the trace constituents for this case and their interpretation is given by Newell  FLIGHT 17-Oct. 14-15, 1991 60N :::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::  (26ø-31øN, 141ø-137øE). It was associated with potential vorticity values of 4-6 x 10 -7 K m 2 kg 4 s q within the 320-330 K potential temperature layer (Figure 16). The winds shown in Figure 16 indicate that the flow within this layer was from the southwest. Typhoon Orchid (10.4øN, 130o4øE at 0000 UT, October 8) and Typhoon Pat (17.3 ø, 151.3øE at 0000 UT, October 8) were beginning to move northwestward and accelerate. The boundary between the continental air mass moving eastward and southward off the Asian continent and the cyclonic flow of marine air around Typhoon Orchid was becoming diffuse and difficult to define as a frontal feature (Figure 12i).
An anticyclone aloft along the southern China coast was responsible for weak off-continent flow over that region (Figure 13i). The deep cyclonic circulations associated with Typhoons Orchid and Pat were eroding the well-defined ridge of high pressure which had persisted over the central North Pacific during the pl•viou• IIIUIltIl.
Patches of low-and middle-level clouds (cumulus, stratocumulus, altocumulus) were over the South China Sea and to the south of Taiwan. Extensive cloudiness associated with Typhoons Orchid and Pat was prominent over much of the western North Pacific, with clusters of organized convection located between the two storms. Based on the motion of cirrus clouds on satellite imagery loops, the outflow aloft appeared to be along the southern quadrants of both typhoons. This flight encountered several major synoptic systems, includ-ing passing within 300 km of the center of Typhoon Orchid. As normal ascent rate of a rawinsonde balloon. At this rate the whole was the case for Typhoon Mireille [Newell et al.,

Conclusions
A brief description of the synoptic situation for each of the PEM-West A missions has been presented. In addition, meteorological data have been used to give background information for some of the more marked atmospheric trace constituent features observed. In particular, in several cases where stratosphere-to-troposphere exchange was suspected, comparisons were made between the DIAL ozone and PV cross sections, which were found to support the explanation. Three cases where boundary layer air was found within the middle to upper troposphere have been identified by virtue of DIAL low ozone together with low PV. Two more cases were found from the flights through Typhoons Mireille and Orchid, where the identification of boundary layer air was made from in situ observations of DMS at nearly the same mixing ratios in the boundary layer and in the upper troposphere. From a total flying time above 6 km of 60 hours, 5.6 hours (about 9%) had values of DMS greater than one quarter of the boundary layer values on the same flight. This is a rough measure of the fraction of time the boundary layer was in communication with the upper troposphere; some distortion exists because flights were often planned to avoid active storm areas. Several of these constituent features are treated in detail in papers that follow in this volume, and the topic of layering is also explored further.