ABSTRACT
 
 

Surface meteorological measurements were made in the Japan/East Sea during January 16-February 4 abroad the R/V Revelle and February 28- March 17 aboard the R/V Khromov. The measurement suite included wind velocity, air temperature, pressure and relative humidity, sea surface temperature, and incident short- and long-wave radiation. The measurements were made on the Revelle with the ship's IMET system, additional SDSU sensors, self-contained WHOI ASIMET radiation recorders, and a WHOI sonic anemometer mounted on the bow mast, and on the Khromov with WHOI ASIMET sensors and the ship's anemometers. Time series estimates of the surface momentum and heat fluxes made using both bulk aerodynamic and direct covariance methods show a consistent spatial pattern associated with air flow across the subpolar surface front and large increases in the surface cooling caused by "cold air outbreaks" of relative dry continental air from Asia.
 
 
 
 

INTRODUCTION
 
 

The wintertime circulation and water mass formation in the Japan/East Sea (JES) are thought to be strongly driven by surface wind momentum and heat fluxes. Strong winds off Siberia during "cold air outbreaks" and orographic effects due to the coastal mountain range and gap at Vladivostok may combine to produce large spatially concentrated surface wind stresses and associated heat and moisture fluxes that have direct and possible large effects on the ocean. Despite the potential importance of surface forcing on the JES, very little is actually known about the marine boundary layer and surface forcing over the JES during winter conditions. To help remedy this, high-quality meteorological data were collected during two research cruises in the JES during winter 2000 as part of the ONR Japan/East Sea Directed Research Initiative field effort. Preliminary results from these measurements are presented here.
 
 

INSTRUMENTATION and METHODS
 
 

Meteorological data were collected on the R/V Revelle winter SeaSoar cruise (January 16-February 4) and the R/V Khromov winter survey cruise (February 28- March 17) (Figure 1). The ships' standard meteorological instruments were augmented by self-contained SDSU and WHOI sensor systems. This sensor redundancy proved critical to collect a complete set of high-quality wind, air temperature, pressure and relative humidity, incident short- and long-wave radiation, and sea surface temperature data for both cruises (Table 1).
 
 

Table 1. Instrumentation used to obtain best surface data during winter cruises. The R/V Revelle is equipped with an IMET sensor system as its standard instrumentation. The WHOI ASIMET units were developed recently as stand-alone systems for use on volunteer observing ships and buoys. The ship's gyro and P-code GPS navigation data were used to compute true winds from measured relative winds for the Revelle data. True winds were estimated on the R/V Khromov every 6 hrs for real-time submission to WMO.
 
 

 

Variable	R/V Revelle	R/V Khromov
Wind	SDSU (23m, main mast)	Ship (13m, bow mast)
AT	IMET (15m, bow mast)	ASIMET (14.5m, bridge roof)
RH	IMET (15m, bow mast)	ASIMET (14.5m, bridge roof)
BP	IMET (15m, bow mast)	ASIMET (12m, bridge roof)
SW	ASIMET (12m, O1-deck)	ASIMET (13m, bridge roof)
LW	ASIMET (12m, O1-deck)	ASIMET (13m, bridge roof)
SST	IMET (-2m, bow intake)	ASIMET (-1m, bow)
Sonic	WHOI (14.6m, bow mast)

Figure 1. Ship tracks from R/V Revelle (red) and R/V Khromov (green) cruises.
 
 
 
 

This data set is used here to describe the surface meteorological conditions during these two cruises and estimate the surface momentum and heat fluxes using the MATLAB Air-Sea Toolbox (Pawlowicz et al, 2001). The net heat flux into the ocean is given as
 
 

Q_NET = Q_SW + Q_LW + Q_SEN + Q_LAT,
 
 

where the sensible and latent fluxes are estimated using the TOGA/COARE bulk aerodynamic formulation (Fairall et al., 1996; Beardsley et al, 1998). Uncertainty in the heat flux components is roughly 10-20 W/m^2.
 
 

In addition, turbulent velocity and temperature measurements were made with a WHOI sonic anemometer /thermometer system mounted on the Revelle bow mast during the winter SeaSoar cruise. This system featured a 3-axis sonic anemometer sampling at 20 Hz plus motion sensors and GPS (Edson et al., 1998). The sonic system returned good data, allowing direct covariance estimation of surface forcing during high wind cooling events.
 
 
 
 

RESULTS
 
 

The basic meteorological data collected on the winter Revelle and Khromov cruises are shown in Figures 2 and 3, and the resultant surface heat flux components computed using the bulk aerodynamic method shown in Figures 4 and 5. Three cold air outbreaks occurred during the Revelle cruise; OB1 (yd 18-22), OB2 (yd 23-29), and OB3 (yd 29-34). Each outbreak is characterized by low air temperature and relative humidity and increased southeastward winds as a low-pressure center passes eastward, intensifying the flow of cold, dry continental air over the central JES. Skies became cloudy during each outbreak, causing a reduction in incident insolation.
 
 

Figure 2. Meteorological conditions during the Revelle winter cruise. The panels show from the top wind speed (m/s), wind direction in degrees relative to E, air temperature and SST (red) (^OC), relative humidity (%), barometric pressure (mb), and incident shortwave and longwave (red) radiation (W/m²). Three cold air outbreaks cause the minimum air temperatures observed on yd 20 (OB1), yd 25 (OB2), and yd 31 (OB3).
 
 
 
 
 
 

Figure 3. Meteorological conditions during the Khromov winter cruise. The panels show from the top wind speed (m/s), wind direction in degrees relative to E, air temperature and SST (red) (^OC), relative humidity (%), barometric pressure (mb), and incident shortwave and longwave (red) radiation (W/m²). Two cold air outbreaks cause the minimum air temperatures observed on yd 68 (OB4) and yd 72 (OB5).
 
 

Figure 4. Surface fluxes during the Revelle winter cruise. From the top, the panels show the wind stress magnitude (N/m²), the net heat flux Q_NET, and the heat flux components Q_SW, Q_LW, Q_SEN, and Q_LAT, all in W/m². Positive flux is heat transfer into the ocean.
 
 

Figure 5. Surface fluxes during the Khromov winter cruise. From the top, the panels show the wind stress magnitude (N/m²), the net heat flux Q_NET, and the heat flux components Q_SW, Q_LW, Q_SEN, and Q_LAT, all in W/m². Positive flux is heat transfer into the ocean.
 
 

The mean net surface heat loss during the Revelle cruise was 353 W/m², with 84% due to the large sensible and latent losses (Table 2). The three cold air outbreaks caused maximum sensible and latent losses of 300 W/m² each, and sustained combined sensible and latent losses of 400 W/m². The tendency for increased cloudiness and reduced insolation during cold air outbreaks was offset by reduced longwave heat loss, with the result that the dominant variability in net heat flux was controlled by the combined sensible and latent components.
 
 

Table 2. Mean conditions during the Revelle and Khromov cruises. The record mean and standard deviation (in parenthesis) for each variable are given.
 
 

 

Variable	R/V Revelle	R/V Khromov
WS (m/s)	9.8 (3.7)
Stress (N/m2)	0.22 (0.16)
Ta (OC)	-0.7 (4.4)
RH (%)	67 (10)
BP (mb)	1022.3 (4.9)
SST (OC)	8.19 (3.17)
QNET (W/m2)	-353.1 (179.4)
QSW (W/m2)	+ 53.3 (103.5)
QLW (W/m2)	-108.9 (30.3)
QSEN (W/m2)	-138.5 (83.4)
QLAT (W/m2)	-158.9 (65.9)

Two cold air outbreaks occurred during the Khromov cruise; OB4 (yd 66-71) and OB5 (yd 72-75). The ship was sampling near the Russian coast north of the subpolar front during OB4, so that while the air temperature drop to -10 ^OC was similar to that during OB2, the ocean surface temperature was within a degree of 0 ^OC, resulting in sensible and latent heat losses during OB4 that were roughly one half that observed during OB2. The Khromov cruise occurred later in the solar cycle, with a significant increase in shortwave heat gain. The resulting mean surface heat loss was only 84 W/m², with the mean sensible and latent loss roughly one half that found during the earlier Revelle cruise.
 
 
 
 

BOUYANCY FLUX COMPARISON
 
 

One objective of the Revelle SeaSoar cruise was to obtain high-resolution hydrographic measurements across the subpolar SST front during a cold air outbreak. The SeaSoar was towed at roughly 4 m/s along a series of north-south legs across the front (Figure 1). Turbulence data collected with the WHOI sonic system on legs 12 and 14 allow a comparison of the buoyancy flux computed using the bulk aerodynamic and direct covariance methods during the end of outbreak OB2 (yd 27-28). On both legs the ship steamed northward at night across the subpolar SST front into a relatively strong and steady wind with little change in air temperature and relative humidity (Figure 6). The resulting net heat loss is controlled by the quick increase in sensible and latent cooling and buoyancy loss due to the sharp increase in SST across the front (in the southward downwind direction). The excellent agreement in buoyancy fluxes computed using the two independent measurements and approaches illustrates the rapid adjustment in the surface fluxes across the SST front, and supports the use of the bulk aerodynamic method for estimating air-sea fluxes during winter in the JES.
 
 

Figure 6. Meteorological conditions and surface fluxes during legs 12 and

14 on the Revelle cruise. The buoyancy flux was computed using a) the basic meteorological data and the TOGA/COARE bulk aerodynamic formulation and b) the sonic turbulence data and the direct covariance method.
 
 
 
 

SUMMARY
 
 

High quality meteorological measurements were made on two oceanographic research cruises in the Japan/East Sea during winter 2000. These measurements allow estimation of the surface momentum and heat fluxes. Cold air outbreaks can cause sensible and latent surface heat losses that increase in magnitude on the warm (southern) side of the subpolar SST front due in part to the enhanced air-sea temperature difference. A preliminary comparison of the cross-front change in buoyancy flux computed using bulk aerodynamic and direct covariance methods shows excellent agreement, demonstrating the rapid change in surface fluxes on scales of a few km's across the front.
 
 

ACKNOWLEDGEMENTS
 
 

The authors want to thank the captains and crews of the R/V Revelle and R/V Khromov for their patience and skill in helping collect meteorological data during the winter 2000 cruises. G. Allsup, D. Hosom, B. Way, and R. Payne constructed the prototype ASIMET longwave radiometer used in this program. D. Alden, W. Sutherland and his shipboard technical support group, and J. Ware provided addition help with the instrumentation and fieldwork, and R. Payne helped with initial data processing and sensor evaluation. M. Caruso and A-M. Michael helped with poster presentation. The Office of Naval Research funded this research.
 
 

References
 
 

Beardsley, R., E. Dever, S. Lentz, and J. Dean, 1998. Surface heat flux variability over the northern California shelf. J. Geophys. Res., 103, 21,553-21,586.
 
 

Edson, J.B., A.A. Hinton, K.E. Prada, J.E. Hare, and C.W. Fairall, 1998. Direct covariance flux estimates from mobile platforms at sea. J. Atmos. Oceanic Tech., 15, 547-562.
 
 

Fairall, C., E. Bradley, D. Rogers, J. Edson, and G. Young, 1996. Bulk parameterization of air-sea fluxes for Tropical Ocean-Global Atmospheric Coupled Ocean-Atmospheric Response Experiment. J. Geophys. Res., 101, 3747-3764.
 
 

Pawlowicz, R., R. Beardsley, S. Lentz, E. Dever, and A. Anis, 2001. The Air-Sea Toolbox: Boundary-layer parameterization for everyone. EOS, (in press).