Potter T.D., Colman B.R. (co-chief editors). The handbook of weather, climate, and water: dynamics, climate physical meteorology, weather systems, and measurements

Подождите немного. Документ загружается.

continuous time series are obtained. Examples of such networks are the U.S.

National Data Buoy Center (NDBC) buoys located around the U.S. coastline, the

Tropical Atmosphere Ocean (TAO) network in the tropical Paciﬁc Ocean, opera-

tional since the mid-1980s (McPhaden et al., 1998), and the global drifter bouys

(Swenson and Niiler, 1996; Bus hnell, 1996).

Sea surface temperatures from satellites are obtained by measuring the radiance

emitted from the sea surface at electromagnetic frequencies within spectral regions

where Earth’s atmospher e is only weakly absorbing, so-called atmospheric window

regions. The main window for SST measurements is in the infrared spectrum

between 8 and 12 mm. Clouds are opaque at infrared wavelengths, however, which

is a serious limitation for observing SST in certain regions of the world. Certain

wavelengths in the microwave spectrum do penetrate clouds, but the technology does

not allow ﬁne spatial resolution at these wavelengths. Table 1 presents the main

atmospheric window regions used for SST measurements.

For SST, the in situ and satellite systems are complementary and are used in

conjunction for the National Centers for Environmental Prediction (NCEP) SST

product, which provides global SST data at 1



latitude by 1



longitude resolution,

averaged over one week (Reynolds and Smith, 1994). The in situ SST are used to

calibrate the satellite values inferred from infrared signals in the atmospheric

window regions. Measurements are obtained at two wavelengths whose absorption

values by the intervening atmosphere are different. By differencing the two measure-

ments, the effect of atmospheric absorption can be measured and corrections

applied. This technique provides the primary atmospheric corrections (McClain

et al., 1985). The in situ SSTs from moored and drifting buoys are combined

with the satellite data after this correction for atmospheric transmission in an optimal

interpolation scheme (Barton, 1995).

TABLE 1 Spectral Regions Used for Determining SST Satellites

Frequency=Wavelength Region

Satellite Type 3.6 mm 10–11 mm 5 cm (6.6 GHz)

Polar orbiting TIROS=NOAA

(1960s onward)

Seasat, (1978)

Nimbus 7 (1978–1985)

Resolution 4 km 4 km

(1 km

maximum)

100s of km

Low orbit in

subtropical=

tropical latitudes

—

Tropical Rainfall Measuring

Mission (TRMM, 1997–)

Resolution — — 15 km

Geostationary GOES Series —

Resolution 4 km 4 km —

The satellites carrying the instruments and the typical surface resolution.

886 SATELLITE VERSUS IN SITU MEASUREMENTS AT THE AIR–SEA INTERFA CE

The basic measurement of temperature in situ is not a problem: Thermistor or

resistance thermometer units are usually reliable and the calibration remains stable,

although drift in the electronics due to seasonal changes in the mean environmental

temperature must be accounted for. Similarly, satellite infrared technology, ﬁlters,

and lenses have had a 35-year history or longer and are also very reliable. The most

difﬁcult problem for achieving stable results has been due to injections of aerosols

from volcanos (Reynolds, 1993).

The measurement problems associated with SST determination include, for the

in situ, heating of the platform by solar radiation on the buoys or ships and, in the

latter case, also the ship’s engine heating the cooling water and general heat diffusion

from the bulk of the ship. For the satellite infrared observations, effects of

atmospheric aerosols and clouds in the ﬁeld of view are persisting problem areas,

which have regional variations whose corrections have not been readily available.

Table 2 compares in situ and satellite accuracies.

As the much relied-on SST product of NCEP is produced with a ﬁt of the satellite

information to a ‘‘calibration surface’’ generated with the available in situ data,

inaccuracies develop when the sampling coverage of the in situ data is inadequate.

This is the case in large regions of the Southern Hemisphere oceans, where the ‘‘ﬁt’’

may become unreliable and can even generate errors. The presence of sea ice can

also confuse the correspondence between in situ and satellite observations

(Reynolds, 1999, personal communication).

Another difﬁculty with SST determination is that a satellite instrument senses the

radiation from the top 0.5 mm or less of the sea surface (Katsaros, 1980; Robinson

et al., 1984), while a buoy or ship measures the temperature from 0.5 to 5 m depth.

Since the oceans lose heat to the atmosphere from the sea surface, there is typically a

negative gradient in temperature toward the surface mainly conﬁned to the nontur-

bulent region nearest the interface. We refer to the layer exhibiting the temperature

difference, DT, across it as the ‘‘cool ﬁlm’’. Many processes affect the magnitude of

the DT, as illustrated in Figure 1. The DT values of the order of 0.3 to 0.7



Chave

been observed on the open ocean (Schlu

ssel et al., 1990; Wick et al., 1996).

TABLE 2 Sea Surface Temperature Uncertainties and

Absolute Accuracies

Intrinsic Errors

Random Error Absolute Error

In situ sensors

Thermistor 0.1 Of the order of 0.5

Resistance 0.1

Satellites

Polar orbiting 0.5 Of the order of 0.7

Geostationary 0.5

1 SEA SURFAC E TEMPERATURE 887

However, on a calm sea during the day, the uppermost layer can be warmer than the

subsurface layers by several degrees Celsius due to solar heating. This is not a

common occurrence but has been observed, particularly in the tropics, and even

from space (Katsaros et al., 1983). It is more likely to occur where fresh water added

by heavy rain may stabilize the surface layer (Wijesekera et al., 1999).

Estimates of temperature drop across the ‘‘cool’’ ﬁlm on the sea surface would

be required to correctly relate the buoy in situ SST to the satellite observations.

One proposed method would be to apply a simple model for the estimated DT across

the cool ﬁlm (Soloviev and Schlu

ssel, 1994).

The uncer tainties in both the satellite and surface SST measurements are of the

same order of magnitude as the typical temperature drop across the cool ﬁlm, but

since the cool ﬁlm is present over most of the ocean most of the time, it could be

argued that ignoring it introduces a bias or additional uncertainties. Most heat ﬂux

parameterizations and other uses for SST information have been based on the bulk

value of SST provided by buoys and ships, so it is wise to consider that the satellite

information represents the temperature at the interface.

Figure 1 Illustration of the ‘‘cool’’ ﬁlm and the many processes that affect the temperature

difference across it. These processes include variable shear stress on the surface, radiative,

sensible and latent heat ﬂuxes, precipitation, wave breaking, and the presence of oil ﬁlms and

slicks (after Katsaros, 1980).

888

SATELLITE VERSUS IN SITU MEASUREMENTS AT THE AIR–SEA INTERFACE

2 SURFACE WIND

Another sea surface variable observable by in situ sensors or from satelli tes is the

surface wind. The classical sea surface wind speed observations were qualitative

descriptions by mariners of the appearance of the sea surface as it was roughened by

the wind. The observations were reported via the Beaufort scale (Weather Bureau,

1948; Shaw and Simpson, 1906). This scale has since been calibrated versus direct

measurements by anemometers. Table 3 gives the Beaufort scale and conversions.

The wind direction was determined from the directions in which the shorter gravity

waves were traveling with respect to the ship’s compass.

The replacement of these qualitative observations by anemometers on ever larger

and larger ships has created much confusion and, perhaps, more erroneous data

because to provide unaffected exposure of an anemometer on a bulky, moving

ship is almost impossible due to severe distortion of the ﬂow both in speed and

direction (Yelland et al., 1998). Modern research ships carry two anemometers to

provide alter nate sensors to choose from, depending on the wind direction relative to

the ship’s heading.

Anemometers on buoys have the problem of being ‘‘pumped’’ by the rocking of

the buoy in the wave ﬁeld . The cup anemometers rectify the buoy motion, i.e., the

anemometer rotation is increased, whether the buoy is in the ‘‘to’’ or ‘‘fro’’ cycle of

its rocking, while propeller anemometers can turn in both directions and, therefore,

average out the effect of the rocking. Propeller anemometers are, therefore, used

almost exclusively on buoys (exceptions are stable buoys, such as the spar buoys,

which do not follow the waves). A directional vane also suffers errors due to buoy

motion.

The satellite method for measuring the ﬂeeting wind is either by microwave

backscatter from the sea surface by an instrument called a scatterometer (Ulaby

TABLE 3 Beaufort Scale for Wind Speed and Its Conversion to Knots

and Meters per Second (m=s)

Beaufort

Number

International

Description

Wind Speed

(knots)

Wind Speed

(m=s)

1 Light air 1–3 0.5–1.5

2 Light breeze 4–6 2–3

3 Gentle breeze 7–10 4–5

4 Moderate breeze 11–16 6–8

5 Fresh breeze 17–21 9–10

6 Strong breeze 22–27 11–13

7 Moderate or near gale 28–33 14–17

8 Gale or fresh gale 34–40 18–20

9 Strong gale 41–47 21–24

10 Whole gale or storm 48–55 25–28

11 Violent storm 58–63 29–33

12 Hurricane >64 >33

2 SURFACE WIND 889

et al., 1982) or by variations in the emissivity (for microwave radiometers).

Scatterometers measure the diffuse backscatter of a radar signal from the sea surface

with two to three antennas, or by a rotating antenna, such that the same piece of

ocean is observed several times from different look angles, e.g., the European

Remote Sensing Sa tellite, ERS, scatterometer, the NASA scatterometer NSCAT,

and QuikSCAT, so named for its short cycle from planning to launch. The difference

in the backscatter power at the different look angles allows inference on the wind

direction, as well as the wind speed. The Seasat, Nimbus 7, and SSM=I microwave

radiometers only provide the wind speed as a consequence of the increa sed emis-

sivity and presence of foam as the wind speed increases. These sensors were

pioneered in the early satellite days, but saw their main debut with the Seasat

satellite launched in 1978 (Jones et al., 1979; Lipes et al., 1979). The satellite

wind measurement techniques have in common with the qualitative Beaufort scale

that it is the effect of the wind on the sea that is being observed. The reason that this

works for the satellite radars of wavelengths between a few to tens of centimeters is

that the surface waves that interact with the electromagnetic radiation (via the Bragg

scattering mechanism) are the capillary-gravity waves, which have a short life time

and are closely linked to the wind speed (or more exactly the wind stress).

After a long hiatus of 13 years, a scatterometer was ﬁrst launched again in 1991

by the European Space Agency and in 1996 by the National Aeronautics and Space

Administration in cooperation with the Japan Space Agency. The Seasat inst ru-

ments, as well as the instruments of the 1990s, were calibrated against buoy anem-

ometer measurements (Bentamy, 1998; Graber et al., 1996), as well as against

numerical weather prediction analyses with the buoy calibration having a somewhat

more realistic wind speed range; atmospheric numerical models tend to smooth the

wind variations and, therefore, lose the extremes. However, for very low wind speeds

and for wind speeds >25 m=s, serious questions about the representativeness of the

buoy observations are also emerging (Large et al., 1995; Ben tamy et al., 1996).

The workhorse for satellite surface wind estimates from 1987 to the present has

been the microwave radiometer, the Special Sensor Microwave Imager (SSM=I),

partly for lack of a scatterometer in space. Microwave radiometry is less well adapted

to satellite wind observations due to interference by heavy clouds and rain, exactly

the condition of paramount meteorological interest. The active systems, scattero-

meters [and other radars such as altimeters and synthetic aperture radars (SARs)],

have more signal power and are, therefore, better able to penetrate clouds. A new

polarimetric microwave radiometer is expected to overcome this difﬁculty (but has

not yet been ﬂown on a satellite).

Again, sampling is the major problem for ship, buoy, and satellite systems.

Reliance on one system alone is often unsatisfactory because of the large temporal

variations of the wind. If a measurement at one speciﬁc place in the ocean is sought,

then the data from an anemometer on a moored buoy will sufﬁce, but if horiz ontal

variability of the wind is desired, a combination of satellite and observations with an

array of buoys may best serve the purpose.

Another important challenge of wind analysis lies in how to join various wind

measurements together. The surface wind is typically deﬁned to apply at 10 m height

890 SATELLITE VERSUS IN SITU MEASUREMENTS AT THE AIR–SEA INTERFA CE

above the sea (this height is an arbitrary choice and others have been in use). Ships

and buoys rarely have their anemometers at 10 m height. To make measurements

intercomparable, we convert the wind speed to a common height by applying a

correction to an individual measurement at height z to translate it vertically to the

10 m or other desired reference height. For that calculation, we must adopt an

analytical vertical wind proﬁle that depends on the atmospheric stratiﬁcation,

which is a function of the turbulence, which, in tur n, depends on the wind stress

and the buoyancy. For a neutrally stratiﬁed atmospheric boundary layer (no buoy-

ancy ﬂuxes or very strong wind shear), the proﬁle is logarithmic. For stable and

unstable stratiﬁcation (negative or positive buoyancy ﬂuxes ) caused by sensible heat

and=or water vapor ﬂux, the proﬁle differs from the logarithmic shape (Kraus and

Businger, 1994).

Methods to provide the height correction have been established (Liu et al., 1979;

Fairall et al., 1996), mostly derived from simultaneous ﬂux and proﬁle measure-

ments over land. The application of the established proﬁle shapes to the ocean have

not yet been fully proven due to the difﬁculty in measuring proﬁles from moving or

bulky platforms or the effects of the wave ﬁeld on the lowest level atmospheric

measurements for stationary towers.

The satellite measurement of the wind vector and wind speed depends on the

roughening of the sea surface by the wind, particularly by the centimeter-scale

gravity-capillary waves. These are forced by the wind stress exerted on the surface,

and the stress is directly related to the wind proﬁle. Thus, the ﬁelds of air–sea

interaction, wave generation, and electromagnetic backscatter theory are involved

in the remote measurement of surface wind. The enterprise is saved from the whole

of this complexity under most circumstances by the footprint size of the radar data

(12.5 to 50 km), which assures a certain averaging over the variability and, thereby, a

randomization of the errors. However, in regions of large wind speed gradients and

variable wave ﬁelds, this may not be true.

The measurements of SST and surface wind are now well developed and are only

being reﬁned by better sampling techniques, data recordings, and inclusion of

auxiliary measurements to improve the corrections. An example is simultaneous

measurements on recent (and future) satellites of the aerosol content in the atm os-

phere, which allows the elimination of the radiation error caused by the aerosol

via radiative transfer models. Therefore, we can now embark on producing long,

consistent time series of both SST and sea surface wind for climatological studies.

The crux is in keeping the long time series consistent with proper intercalibration

between sequential satellite sensors and when some inevitable design improvements

are made.

REFERENCES

Barton, I. J. (1995). Satellite-derived sea surface temperatures: Current status, J. Geophys. Res.

100, 8777–8790.

REFERENCES 891

Bentamy, A., Y. Quilfen, F. Gohin, N. Grima, M. Lenaour, and J. Servain (1996). Determination

and validation of average wind ﬁelds from ERS-1 scatterometer measurements, Global

Atmos. Ocean Syst. 4(1), 1–29.

Bentamy, A., N. Grima, and Y. Quilfen (1998). Validation of the gridded weekly and monthly

wind ﬁelds calculated from ERS-1 scatterometer wind observations, Global Atmos. Ocean

Syst., 5, 373–396.

Bushnell, M. H. (1996). Preliminary Results from Global Lagrangian Drifters Using GPS

Receivers, WMO=DBCP Technical Doccument No. 10, pp. 23–26.

Fairall, C. W., E. F. Bradley, J. S. Godfrey, G. A. Wick, J. B. Edson, and G. S. Young (1996).

Cool-skin and warm-layer effects on sea surface temperature, J. Geophys. Res. 101(1),

1295–1308.

Graber, H. C., N. Ebutchi, and R. Vakkayil (1996). Evaluation of ERS-1 scatterometer winds

with wind and wave ocean buoy observations, Tech. Rep., RSMAS 96-003, Div. of Applied

Marine Physics, Univ. of Miami, FL.

Jones, W. L., P. G. Black, D. M. Boggs, E. N. Bracalente, R. A. Brown, G. Dome, J. A. Ernst,

I. N. Alberstan, J. E. Overland, F. Peteherych, W. J. Pierson, F. J. Wentz, P. M. Woiceﬁyn, and

N. J. Wurtele (1979). SEASAT scatterometer: Results of the Gulf of Alaska workshop,

Science 204(4400), 1413–1415.

Katsaros, K. B. (1980). The aqueous thermal boundary layer, Boundary-Layer Meteorol. 18,

107–127.

Katsaros, K. B., A. F. G. Fiuza, F. Sousa, and V. Amann (1983). Sea surface temperature

patterns and air–sea ﬂuxes in the German Bight during MARSEN 1979, Phase 1,

J. Geophys. Res. 88(C14), 9871–9882.

Kraus, E. B., and J. A. Businger (1994). Atmosphere-Ocean Interaction, Oxford University

Press, New York.

Large, W. G., J. Morzel, and G. B. Crawford (1995). Accounting for surface wave distortion of

the marine wind proﬁle in low-level ocean storms wind measurements, J. Phys. Oceanogr.

25(11), 2959–2971.

Lipes, R. G., R. L. Bernstein, V. J. Cardone, K. B. Katsaros, F. G. Njoku, A. L. Riley, D. B. Ross,

C. T. Swift, and F. J. Wentz (1979). SEASAT scanning multichannel microwave radiometer:

Results of the Gulf of Alaska workshop, Science 204(4400), 1415–1417.

Liu, W. T., K. B. Katsaros, and J. A. Businger (1979). Bulk parameterization of air–sea

exchanges of heat and water vapor including the molecular constraints at the interface,

J. Atmos. Sci. 36(9), 1722–1735.

McClain, E. P., W. G. Pichel, and C. C. Walton (1985). Comparative performance of AVHRR-

based multichannel sea surface temperatures, J. Geophys. Res. 90, 11,587–11,601.

McPhaden, M. J., A. J. Busalacchi, R. Cheney, J.-R. Donguy, K. S. Gage, D. Halpern, M. Ji,

P. Julian, G. Meyers, G. T. Mitchum, P. P. Niiler, J. Picaut, R. W. Reynolds, N. Smith, and

K. Takeuchi (1998). The Tropical Ocean-Global Atmosphere observing system: A decade

of progress, J. Geophys. Res. 103, 14,169–14,240.

Reynolds, R. W. (1993). Impact of Mount Pinatubo aerosols on satellite-derived sea surface

temperatures, J. Climate 6(4), 768–776.

Reynolds, R. W., and T. M. Smith (1994). Improved global sea surface temperature analyses

using optimum interpolation, J. Climate 7(6), 929–948.

Robinson, I. S., N. C. Wells, and H. Charnock (1984). The sea surface thermal boundary layer

and its relevance to the measurement of sea surface temperature by airborne and spaceborne

radiometers, Int. J. Remote Sens. 5, 19–45.

892

SATELLITE VERSUS IN SITU MEASUREMENTS AT THE AIR–SEA INTERFACE

Schlu

ssel, P., W. J. Emery, H. Grassl, and T. Mammen (1990). On the bulk-skin temperature

difference and its impact on satellite remote sensing of sea surface temperature, J. Geophys.

Res. 95(C8), 13,341–13,356.

Shaw, W. N., and G. C. Simpson (1906). The Beaufort Scale of Wind Force: Report of the

Director of the Meteorological Ofﬁce upon an Inquiry into the Relation between the

Estimates of Wind-Force According to Admiral Beaufort’s Scale and the Velocities Recorded

by Anemometers Belonging to the Ofﬁce, with a Report Upon Certain Points in Connection

with the Inquiry, Darling and Son, London.

Soloviev, A. V., and P. Schlu

ssel (1994). Parameterizations of the cool skin of the ocean and of

the air-ocean gas transfer on the basis of modeling surface renewal, J. Phys. Oceanogr.

24(6), 1339–1346.

Swenson, M. S., and P. P. Niiler (1996). Statistical analysis of the surface circulation of the

California Current, J. Geophys. Res. 101(C10), 22,631–22,645.

Ulaby, F. T., R. K. Moore, and A. K. Fung (1982). Microwave Remote Sensing, Active and

Passive, Vol. II; Radar Remote Sensing and Surface Scattering and Emission Theory,

Addison-Wesley, Reading, Mass.

Weather Bureau (1948). Beaufort Scale of Wind Force, WB Form 4042A, Washington, DC.

Wick, G. A., W. J. Emery, L. H. Kantha, and P. Schlu

ssel (1996). The behavior of the bulk-skin

sea surface temperature difference under varying wind speed and heat ﬂux, J. Phys.

Oceanogr. 26(10), 1969–1988.

Wijesekera, H. W., C. A. Paulson, and A. Huyer (1999). The effect of rainfall on the surface

layer during a westerly wind burst in the western equatorial Paciﬁc, J. Phys. Oceanogr.

29(4), 612–632.

Yelland, M. J., B. I. Moat, P. K. Taylor, R. W. Pascal, J. Hutchings, and V. C. Cornell (1998).

Wind stress measurements from the open ocean corrected for airﬂow distortion by the ship,

J. Phys. Oceanogr. 28(7), 1511–1526.

REFERENCES 893

CHAPTER 48

RADAR TECHNOLOGIES IN SUPPORT

OF FORECASTING AND RESEARCH

JOSHUA WURMAN

1 INTRODUCTION

Forecasters and researchers, and the computer models that support them, require

detailed information concerning the three-dimensional state of the atmosphere and

how it evolves with time. Some forecas ting and research needs overlap; others are

particular to those respective tasks. But whether the goal is weather prediction or

scientiﬁc understanding, no tool can provide more wealth and diversity of informa-

tion than existing and emerging technologies in weather radar (Fig. 1). Weather

radars can sample large volumes of the atmosphere nearly continuously in space

and time at fairly high resolution vertically, horizontally, and temporally. Radars can

penetrate through clouds and precipitation to measure precipitation intensity, preci-

pitation type, and wind motions in many weather conditions, throughout the depth of

the atmosphere. They are a uniquely versatile tool.

A forecaster needs to know where it is raining or snowing, how hard, and

whether snow or hail is occurring. He needs to know whether the precipitation

is caused by convective systems or more stratiform uplift, whether it is moving

toward the forecast area or away, whether it is strengthening or weakening. He

needs to know if an airport runway will soon be affected by a gust front or

microburst, or if a mesocyclone will move into his forecast area, whether a

snow band will move onshore, or whether a sea b reeze will initiate storms. For

many forecasts of 6 h or less, radar is the crucial tool that provides the most up-to-

date and comprehensive information to a forecaster. Mesoscale forecasting models

Handbook of Weather, Climate, and Water: Dynamics, Climate, Physical Meteorology, Weather Systems,

and Measurements, Edited by Thomas D. Potter and Bradley R. Colman.

ISBN 0-471-21490-6 # 2003 John Wiley & Sons, Inc.

895