An Overview of TOCS

31 October 1996

JUSTOS PROJECT -- Japan-U.S. Tropical Ocean Study
SCIENCE OBJECTIVES FOR JAMSTEC CRUISE

Japan Project Leader: Kunio Yoneyama (JAMSTEC)
U.S. Project Leaders: David Parsons (NCAR), and R. Micheal Reynolds (BNL)

1. INTRODUCTION:

The scientific focus of this cruise is influenced by a number of factors including the extensive experience of JAMSTEC in the Tropical Western Pacific (TWP), the ongoing research by these investigators on the TOGA COARE data set, the highlighting of outstanding scientific problems over the TWP by participants at the recent meeting of the ARM Ocean Working Group, and the session at the recent TOGA COARE air-sea flux workshop that outlined serious errors in the rawinsonde measured low-level humidities.

This plan is meant to provide a broad outline of the scientific goals for the winter 1997 TOCS cruise by the JAMSTEC R/V Kaiyo. This vessel is shown in Fig. 1. The cruise tracks are shown in Fig. 2 with the vessel departing from Majuro on 26 January 1997, in port in Truk from 11-13 February 1997 and arriving in Palau on 2 March 1997. The experiment is meant to provide a supplement to the upper air sounding and oceanographic experiments conducted by JAMSTEC during these cruises. This experiment is also meant to serve as a test of a possible long-term collaboration between JAMSTEC and U.S. scientists interested in future field studies into ocean-atmosphere interactions.

2. SCIENTIFIC QUESTIONS:

a. Investigations Into Spatial Variability in SSTs

A better understanding of the spatial variations in the oceanic heat budget is needed. Aside from the well known meridional variation, we need to better understand both standing and random variability in solar radiation and all climatic variables such as rainfall, cloud fraction and type, and atmospheric and oceanic structures.

For example, recent analysis of data taken over the TWP during TOGA COARE shows large local departures in the sea surface temperatures (SST) measurements taken by cruising vessels such as the results presented by Lukas and Soloviev (1996). These SST departures of up to several degrees are significant in magnitude when compared against meridional and temporal variations over the warm pool. The cause of these variations (i.e., pools of rain water, cloud shading, ocean circulations) and their possible impacts on the oceanic and lower atmospheric heat budgets and on cloud formation are still relatively unknown. We suspect that these spatial variations will have a significant impact on the magnitude of the surface fluxes predicted by global climate models (GCMs) particularly in light winds.

Unfortunately, these relationships are difficult to explore with current data sets since (1) most of the research vessels during TOGA COARE that carried extensive atmospheric instrumentation were usually stationary requiring that temporal statistics be converted into spatial variations, (2) aircraft observations which are well suited to examining spatial variations do not reveal much about the upper ocean other than measurements of the SST, and (3) the SST measurements on cruising vessels are often placed at depths of 1-3 m below the surface which according to Lukas and Soloviev (1996) may cause the magnitude of the SST variability to be underestimated.

The continuous near surface measurements of SST during this cruise will provide measurements that can be used to provide an estimate of how the SST varies locally during a variety of conditions (i.e., diurnal vs. nocturnal, suppressed vs. disturbed, calm vs. moderate winds). The concurrent atmospheric measurements will allow us to investigate the causes and impacts of these spatial SST variations. For example, the lidar measurements will allow us to investigate whether cloud shading during diurnal heating is important for causing some of these variations, while salinity measurements will address the residual role of precipitation. In addition, the surface meteorological measurements will allow us to investigate the impact of the SST variations on the atmospheric boundary layer. We believe this work has a number of implications including (1) improved treatment of the subgrid scale variations in SST on surface fluxes, (2) improved treatment in GCMs of how clouds interact to cool or warm the SST, and (3) providing data for verification of the behavior of high resolution coupled models.

b. Lower Atmosphere Mixed Layer Physics, the Upper Ocean and the Diurnal Cycle of Clouds and Convection

The diurnal cycle of convection over tropical oceans has long been a subject of intense interest as summarized recently in Randall et al. (1991) and Janowiak et al. (1994). To summarize these studies, at many locations, a weak early morning maximum in deep convection tends to occur over the tropical oceans. Radiative differences in heating/cooling profiles between cloudy and cloud-free air have been proposed to account for this difference. At other locations, a late afternoon to early evening convective maximum has been observed with the impacts of nearby land masses suspected as possible causes. Other studies and recent work based on the TOGA COARE radar data sets seem to indicate that different cloud populations have different diurnal cycles. For example, an early evening precipitation maximum apart from land effects is clearly evident during dry intrusion conditions as discussed by Parsons and Yoneyama (1996).

Linking these diurnal patterns to observed atmospheric stability changes in the mixed layer has been far more difficult due to (1) sounding frequencies insufficient for resolving the diurnal cycle, (2) a small diurnal signal, and (3) sonde accuracy limitations and errors. These errors will be discussed in detail in the next section. At this time it is sufficient to note that the errors are large and the data sets are generally uncorrected. For some errors a correction is applied but correction is as large or larger than the diurnal signal so that the extent to which the diurnal signal is known depends directly on the accuracy of the correction procedure.

Hence it is not surprising that a confusing picture of the diurnal structure of the lower mixed layer has emerged as at times the mixed layer becomes more stable during midday heating when one would expect instability from diurnally-driven variations in the SST. Unaccounted for terms in the radiation budget, such as long-wave heating and/or short wave absorption by the moist mixed layer are suspected. The increased stability and concomitant suppressed fluxes of heat and vapor result in a rectified flux signal which biases climatic means. Until the diurnal signal in the mixed layer is clarified, there is little hope for understanding the different diurnal modes of clouds and convection that occur over the warm pool.

Through our measurements we hope to better understand the links between the upper ocean, the atmospheric mixed layer and the factors that control the diurnal variation of clouds and convection including variations in the dynamic and kinetic controls on deep convection. The special measurements most relevant to this portion of the cruise are calibrated rawinsonde launches at 3-h intervals, high resolution measurements by the Scanning Aerosol Backscatter Lidar (SABL) that monitor atmospheric mixed layer changes and the formation of non-precipitating clouds, profiler and RASS measurements that monitor the mixed layer winds and stability, the SST and bulk flux measurements. Special emphasis will also be placed on periods of light winds where strong diurnal changes occur in the upper ocean and during intrusions of dry air from higher latitudes into the tropics (e.g., Yoneyama and Fujitani 1995; Mapes and Zuidema 1996).

c. Improved Characterization of Water Vapor in the Lower Atmosphere

In addition to sonde accuracy impacting the diurnal signal, an accurate determination of water vapor in the boundary layer is crucial to many of the scientific issues related to understanding the behavior of the air-ocean system over the tropical western Pacific. These issues were discussed in depth at the recent meeting at Woods Hole of the Air-Sea Flux Working Group but range from GCM predictions of clouds, deep convection and surface fluxes, to accurate prediction of radiation profiles, and to fundamental understanding of what processes regulate deep convection in the tropical ocean-atmosphere system (Mapes 1996).

Unfortunately, the characterization of humidity in the tropical boundary layer, even during special projects like TOGA COARE, is typically lacking. A recent analysis tool is to compare mixed layer humidities to surface sites. Examination of even the mean differences between these fields for the TOGA COARE sounding data sets reveal strong suspected errors in humidity as evidenced from large site-to-site differences that appear to be related to the sonde vendor (Table 1). The consensus at the Woods Hole meeting was that examination of time series at individual sites indicates sharp temporal variations perhaps related to different sounding batches and more scatter about the mean than expected from atmospheric considerations (Fig. 3). These problems are evident in the mean mixed layer profile and the cause of these errors is uncertain. Additional errors from known sources such as solar heating of the sensor arm and launches from enclosed containers have been found to produce additional errors in the lower mixed layer. These known errors can be partially corrected (e.g., Cole and Miller 1995). Previous tropical experiments seem to suffer from some of the same accuracy problems as suggested by recent reanalysis of the GATE data.

At the Woods Hole meeting, a procedure was proposed by E. Zipser (Texas AM) to correct these site-to-site and intrasite problems based on assuming some difference between the surface layer and the mixed layer. This gradient can be estimated from surface layer theory for idealized conditions. The occasional tether sonde or kite operations during this cruise will allow a quantitative check of this proposed correction. Also the special calibration procedures used will allow sonde-to-sonde differences to be addressed through post-calibration procedures. The high resolution lidar measurements will also allow the corrected humidity fields to be compared against cloud base measurements as a further check on sonde humidity accuracies.

d. Some Additional Issues

The first Atmospheric Radiation and Cloud Stations (ARCS), began data collection on Manus Island, PNG, in September 1996. A major question concerning these sophisticated systems, will be island effects: How well do measurements from small islands represent conditions in the surrounding ocean? The data sets from these cruises will be compared to the ARCS stations and data taken during the recent Combined Sensor Cruise (CSP). Some of the ARCS measurements can be extrapolated to the surrounding ocean, either directly or with the help of models. However, the ability of the ARCS to represent oceanic features must be confirmed and complemented by comparison measurements.

The meteorological and oceanic issues investigated by this cruise include extending the long-term climatological monitoring of the western Pacific atmosphere-ocean system (i.e., in different portions of the ENSO cycle, different times of year). One aspect of the monitoring will be to extend the climatology of dry intrusions that account for a significant portion of the variation in long wave radiation observed at the surface. Other questions concerning the upper ocean mixed layer include how the behavior of SST varies under different cloud conditions. Another issue is how convection outflows impact currents in the upper ocean on short time scales and how these events in aggregate combine to produce wind-driven events ranging up in scale to ENSO events.

3. INSTRUMENTATION

The cruise by the R/V Kaiyo provides an excellent opportunity to address the previously described scientific issues. For example, the ship will be touching in near Manus so there is another opportunity to make comparison measurements with one of the ARCS. The current vessel is a SWATH-type design, very stable in all weather conditions. The basic instrumentation strategy is a ship-borne Integrated Sounding System (ISS) (Parsons, et al., 1994).

The ship is a state-of-the-art oceanographic vessel with CTD measurements, acoustic doppler current profiler (ADCP) and various water sampling systems. JAMSTEC will also provide a Vaisala rawinsonde sounding system and an optical rain gauge. Equipment will be obtained for the tether sonde or simple kite operations during periods of light winds and a standard surface system that measures wind, temperature and humidity. A 915 MHz wind profiler with RASS capabilities will also be sent consisting of materials from C. Fairall's group at NOAA/ETL and NCAR. Other measurements include the Scanning Aerosol Backscatter Lidar (SABL) from NCAR's Remote Sensing Facility. This lidar system has excellent resolution (scale of several meters), operates at two wavelengths and can see thin tropical cirrus.

Brookhaven National Laboratory will provide its Portable Radiation Sensor (PRP) for making direct and diffuse short and long wave radiation measurements. Sensors in the PRP will include

The standard measurement interval will be 1 sample/sec. Each minute the average standard deviation, minimum, and maximum will be computed and stored. Wind speed and direction, compass heading, GPS location, GPS ship speed, and course will be recorded each second and one-minute averages will be computed during analysis. All measurements will allow for the computation of the surface energy flux using the bulk transfer formulae (see Fairall, et al., 1996).

Turbulence flux measurements had been considered, but the vessel has a blunt bow which creates considerable turbulence for eddy correlation measurements of fluxes. Oceanographic measurements will be provided by JAMSTEC.

4. OPERATIONS

The primary activity of the R/V Kaiyo cruise will be the maintenance to selected TAO buoys in the western Pacific. It will be advantageous to make as many CTD and bio-optical stations (if available) as time allows. As we are primarily interested in depths less than 100-200 m, casts can be accomplished in minimal time (typically ½ hour each). With the selected instrumentation and the current cruise plans for the Kaiyo we feel that we can combine the continuous oceanic and atmospheric measurements to help answer the scientific questions outlined above. For example, the PRP radiation observations could be combined with the lidar, profiler satellite, and cloud picture data to get some idea of the impact of different cloud types on the surface energy budget.

Some additions to the continuous measurements will also be mentioned here. For example, the humidity problems with the TOGA COARE rawinsondes lead us to conclude that some calibration of all sondes must be accomplished prior to launch. We will calibrate against a source with a ~1% accuracy. Validation of the launch container or open air launches are necessary. To investigate the diurnal signal, soundings will need to be launched at 3-h intervals. To qualitatively investigate the impact of the upper ocean variation on the atmosphere we will need to continuously monitor the SST. The radiation skin temperature will be monitored with the IRT and where possible we will tow a special SST(1 cm) sensor. Measurements of temperature and humidity in the surface layer will be made from the ship's tower and we will fly a tethered sonde through the mixed layer (~400m). Salinity, and current cloud measurements could be used to better understand the reasons for this upper ocean variation.

5. References

Cole, H.L. and E. Miller, 1995: A correction for low-level radiosonde temperature and relative humidity measurements. Ninth AMS Symposium on Meteorological Observations and Instrumentation, Charlotte, NC, March 1995, 32-36.

Fairall, C.W., E.F. Bradley, D.P. Rogers, J.B. Edson and G.S. Young, 1996: Bulk parameterization of air-sea fluxes for TOGA COARE. In press, J. Geophys Res.

Janowiak, J. E., P.A. Arkin, and M. Morrissey, 1994: An examination of the diurnal cycle in oceanic tropical rainfall using satellite and in situ data. Mon. Wea. Rev., 122, 2296-2311.

Lukas, R. and A. Soloviev, 1996: Small-scale spatial variability in near-surface temperature and salinity in the Western Pacific warm pool. Eighth AMS Conference on Air-Sea Interaction and Conference on the Global Ocean-Atmosphere-Land System (GOALS), Atlanta, GA, January 1996, J-46-J-49.

Mapes, B.E., 1996: What controls large-scale variations in deep convection. Submitted to ECMWF Workshop on Convective Parameterization.

Mapes, B.E., and P. Zuidema, 1996: Radiative-dynamical consequences of dry tongues in the tropical troposphere. J. Atmos. Sci., 50, 2026-2037.

Parsons, D., W. Dabberdt, H. Cole, T. Hock, C. Martin, A-L Barrett, E. Miller, M. Spowart, M. Howard, W. Ecklund, D. Carter, K. Gage, and J. Wilson, 1994: The Integrated Sounding System: Description and preliminary observations from TOGA COARE. Bull Amer. Meteor. Soc., 75, 553-567.

Parsons, D.B. and K. Yoneyama, 1996: Behavior of the boundary layer and convection over the Tropical Western Pacific during dry intrusion. To be submitted to J. Atmos. Sci.

Randall, D.A., Harshvardhan, and D.A. Dazlich, 1991: Diurnal variability of the hydrologic cycle in a general circulation model. J. Atmos. Sci., 48, 40-62.

Document Maintenance: Bob Rilling, / NCAR Atmospheric Technology Division
Created: 4-Mar-1997
Last modified: Wed Apr 29 09:06:26 1998