2.0 Overview »
Learning Objectives

At the end of this chapter, you should understand and be able to describe:

• Why remote sensing is important in the tropics
• Several tropical applications of ground-based radar
• The advantages and limitations of airborne and spaceborne radar
• Several tropical meteorology applications of satellite radar and microwave remote sensing
• The benefits and weaknesses of satellite estimates of water vapor content
• How GPS satellite signals are used to derive temperature and humidity profiles and how this benefits tropical meteorology
• The benefits and weaknesses of satellite precipitation estimates
• How lightning is detected by satellite
• The benefits and weaknesses of satellite wind estimation
• Why microwave sensors are useful for identifying surface moisture
• How vegetation and other land use/land cover changes are monitored by satellite
• How meteorologically important features, such as cloud properties, are monitored with satellite imagery
• How satellites are used for air quality assessment, such as dispersion of volcanic ash, chemical pollutants, dust, and smoke

2.1 Introduction to Remote Sensing »
2.1.1 Why Remote Sensing in the Tropics?

Surface and radiosonde observations are sparse in the tropics, especially when compared with the Northern Hemisphere (Fig. 2.1). Primary among the surface data-sparse regions are the tropical oceans; the Pacific Ocean alone occupies about one half of the circumference of the equator. In a few regions, the surface network and regularity of reports have diminished over the past two decades.¹ Given these conditions, remote sensing is the primary, and sometimes only, means of regular observations for most of the tropics. The World Meteorological Organization (WMO) provides current information on the global system of observations under its World Weather Watch program.

Since 1964, satellite sensors have provided routine observations for weather analysis.² Satellites and aircraft remote sensors have provided a wealth of information on hurricanes, the most hazardous tropical weather system. Satellite-based climate studies have produced significant information about clouds and precipitation, large-scale circulations, air-sea interaction, air chemistry, and land surface changes. The assimilation of satellite data into numerical prediction models has improved model performance.^3,4,5

WMO Global Observing System, http://www.wmo.int/pages/prog/www/OSY/GOS.html
Current global data coverage, http://www.ecmwf.int/products/forecasts/d/charts/monitoring/coverage/
Joint Center for Satellite Data Assimilation,http://www.jcsda.noaa.gov/
(NOAA, NASA, US Navy, and US Air Force Partnership)

2.1 Introduction to Remote Sensing »
2.1.2 Basics of Remote Sensing by Radar and Satellite

Remote sensing applications in weather and climate are possible because of the variety of ways in which the atmosphere and other earth systems interact with the electromagnetic spectrum (EM) (Fig. 2.2). For example, snow scatters visible light, water vapor absorbs infrared (IR) radiation, and hail scatters microwave radiation. The portions of the EM spectrum with little absorption by the atmosphere are referred to as "windows" and are used to observe surface properties. The atmosphere is most transparent in the visible and the microwave parts of the spectrum and opaque in the IR except for a small window close to 10 μm. Satellite sensors measure energy from particular wavelengths, which are referred to as "channels" and numbered in increasing order from shortwave to longwave. The visible, IR, and microwave wavelengths are used most often in meteorology.

2.1 Introduction to Remote Sensing »
2.1.2 Basics of Remote Sensing by Radar and Satellite »
2.1.2.1 The Basis of Weather Radar

Doppler Radar animation.

The ability of objects to scatter radiation is the basis for the radar applications in meteorology. Weather radars operate by transmitting high-frequency microwave (mm-cm scale) pulses to the atmosphere and measuring the "backscatter" or echoed pulses to the radar (Fig. 2.3). The returned signal is interpreted to determine where it is precipitating. Typically the radar scans as the antenna is raised through higher and higher preset angles to provide a three-dimensional look at the atmosphere. Note that weather radar is also used to track insects, migrating birds, and dust storms.

2.1 Introduction to Remote Sensing »
2.1.2 Basics of Remote Sensing by Radar and Satellite »
2.1.2.2 The Global Satellite Observation System

Satellites are the primary means of obtaining global-scale observations (Fig. 2.4). They are grouped according to their orbiting patterns:

• geostationary, which orbit at about 35,800 km above the equator and move at the velocity of the earth's rotation;
• polar-orbiting, which orbit around the poles at approximately 850 km above the surface, and
• research and development, which orbit between certain latitudes a few hundred km above the surface.

Operational satellites are dedicated to weather analysis and forecasting. Data are available in a timely manner for immediate use in hazard mitigation, commerce, security, and daily life. Emphasis is placed on ensuring that these services are not interrupted. Research satellite missions are dedicated to particular scientific goals, such as the measurement of ozone in order to document and understand climate change or the measurement of precipitation to better understand the global energy and water cycle. Many research satellites are designed to have a limited lifetime although some have lasted beyond the expected mission duration and some have been applied to operations. For example, the Tropical Rainfall Measurement Mission (TRMM) satellite data have been used extensively for monitoring hurricane intensity and position. With the growing recognition of the need for long-term climate analysis, operational data are being archived and used as climate data record.

Geostationary Satellites

Geostationary satellites sensors are most useful for tracking atmospheric features over great distances because of their high temporal resolution (15 - 30 minute intervals, or better) and hemispheric field of view (Fig. 2.5). However, the orbital distance of the satellites means that their spatial resolution is less than optimal for the identification of features smaller than 1 km.

Polar Orbiting Satellites

Polar-orbiting environmental satellites (POES) provide superior spatial resolution over a small field of view (on the order of 100 to 1000 km). For example, the NOAA POES carry the Advanced Very High Resolution Radiometer (AVHRR), which has 1-km resolution at six channels.

However, they view any given location at most twice per day. For the tropics, a single polar-orbiting satellite has many gaps in coverage during a given day (illustrated in the animation of Figure 2.6). With the proliferation of sequential polar-orbiting satellites, including the European Space Agency's MetOp, Joint Polar Satellite System (JPSS), and Global Precipitation Mission (GPM), we can expect more frequent observations of the tropics at high resolution.

To learn more about polar-orbiting products, access the COMET Satellite Meteorology module series at https://www.meted.ucar.edu/training_detail.php?page=1&topic=12&language=1&orderBy=publishDateDesc.

Fig. 2.6. 24-hour coverage of the EUMETSAT polar-orbiting satellite, MetOp.

Satellite sensors use either active or passive sensing techniques. Active sensors send signals toward the earth's surface and measure the signal return (like a radar). Passive sensors detect naturally occurring radiation. While passive sensors receive information from layers, active sensors have the advantage of measuring radiation from discrete levels and producing, for example, vertical precipitation and clouds profiles. Active sensors have higher spatial resolution than passive sensors.

You can review basic satellite remote sensing in the COMET modules Remote Sensing Using Satellites (http://www.comet.ucar.edu/nsflab/web/index.htm) and Feature Identification from Environmental Satellites (http://www.meted.ucar.edu/npoess/nrlsat/). A brief description of orbital and scanning geometry of meteorological satellites is presented as a special section.

Comprehensive Large Array-data Stewardship System (CLASS), http://www.class.noaa.gov/
NOAA Geostationary Satellite Server, http://www.goes.noaa.gov/

Rotating globe movie of current weather, http://www.ssec.wisc.edu/data/globe/spincg.html

2.1 Introduction to Remote Sensing »
2.1.2 Basics of Remote Sensing by Radar and Satellite »
Box 2-1 Radiative Transfer

This section provides a brief summary of the properties of electromagnetic radiation and how it interacts with the atmosphere. Electromagnetic waves are described by their:

• Wavelength, λ, the distance between crests of an electric or magnetic field (SI unit, m),
• Wavenumber, κ, the reciprocal of the wavelength, (SI unit, m^-1),
• Frequency, ν, the rate of oscillation of radiation at a point, (SI unit, Hz), where c is the speed of light in a vacuum, 2.99792458 x 10⁸ m s^-1.

A useful concept in radiation is a blackbody, an object which absorbs all incident radiation and emits the maximum amount of energy at all wavelengths. The energy per photon emitted is

E = hν, where h is Planck's constant (6.625 x 10^-34 J s) and ν is the frequency (s^-1)

Radiance is the energy per unit time/wavelength/solid angle/area. The radiance emitted by a blackbody is expressed by Planck's Law, and measured in units of W m^-2 sr^-1,

                        (B2-1.1)

where λ = wavelength, T = temperature (K),  c₁ = 1.1910439 × 10^-16 W m^-2 sr^-1, and  c₂ = 1.438769 × 10^-2 m K.

The brightness temperature, T, is determined by inverting the Planck function. At microwave wavelengths (mm, cm) and for earth and atmospheric temperature range, radiance to temperature is a simple proportion.

As illustrated below (Fig. 2B1.1), energy on the EM spectrum increases from right to left, so that visible wavelengths (shortwave) have more energy than infrared wavelengths (longwave).

Fig. 2B1.1. (left) Radiance for solar and thermal blackbodies and (right) the same curves annotated with the wavelengths of peak emission for solar (Vis) and Thermal (LWIR) and overlapping wavelengths (NearIR and SWIR).

The energy emitted per unit area (from all wavelengths and represented by the area under the blackbody curve) is related to the absolute temperature through the Stefan-Boltzmann Law,

                                                (B2-1.2)

where the Stefan-Boltzmann constant, σ = 5.67 x 10-8 W m^-2 K^-4.

Wien's Displacement Law tells us that the wavelength of maximum blackbody emission is inversely proportional to its temperature, i.e., the hotter the object, the shorter the peak wavelength at which it emits:

                                                (B2-1.3)

A beam of radiation passing through the atmosphere can be changed by absorption, emission, scattering out of the beam and scattering into the beam of radiation. For example, atmospheric gases absorb solar radiation and reduce the amount of solar radiation at the surface (Fig. 2B1.2).

Fig. 2B1.2. Solar spectral irradiance with color shaded areas marking the absorption by atmospheric gases.

The rate of change of radiance, L, over distance, s, is a sum of the extinction (absorption and/or scattering) and sources (emission and/or scattering into the beam), and may be written as the radiative transfer equation:

                                                (B2-1.4)

where k_λ is the extinction cross-section (in units of area per mass) for wavelength, λ, ρ is the density of the medium, and j_λ is the source function coefficient.

Remote sensors measure radiation transmitted through and by the atmosphere. The most fundamental unit measured by satellites is monochromatic radiance. An electromagnetic signal is recorded by a detector, usually a radiometer, after it interacts with target molecules, particulates, or surfaces. If Τ and S denote the target and signal, we can represent their relationship by:

Remote sensors measure radiation transmitted through and by the atmosphere. The most fundamental unit measured by satellites is monochromatic radiance. An electromagnetic signal is recorded by a detector, usually a radiometer, after it interacts with target molecules, particulates, or surfaces. If Τ and S denote the target and signal, we can represent their relationship by:

                                    (B2-1.5)

where F is a function that governs radiative transfer.

Although radiation is characterized by a specific wavelength and its sensitivity to some physical aspect of the transmitting medium, the inversion solution may not be unique. A combination of unknown parameters can lead to the same radiance measurement. This non-uniqueness creates an inherent uncertainty in remotely-sensed measurements.

2.1 Introduction to Remote Sensing »
2.1.2 Basics of Remote Sensing by Radar and Satellite »
Box 2-2 Orbital and Scanning Geometry

Geostationary satellites are, in essence, motionless above a point on the equator, usually referred to as the subsatellite point. With an orbit of about 35,800 km, geostationary satellites have a hemispheric field of view. Polar orbiting satellites at 850 km have a smaller field of view with higher spatial resolution.

The figure (Fig. 2B2.1, left) shows the true relative distances of geostationary and polar orbiting satellites. From geostationary altitude, the entire Earth disk only subtends an angle of 17.4 degrees. A typical polar orbiting satellite sees a relatively small portion of the globe at any one time.

As polar-orbiting satellites move around the globe, their instruments take measurements along their track. Instruments scan across their orbit track, scan conically, or scan like a push broom. Below is an example of a polar-orbiting satellite conical scanning geometry (Fig. 2B2.1).

Fig. 2B2.1. (left) Geometry of geostationary and polar orbiting satellite systems and (right) the scan geometry of the SSM/I polar orbiting satellite.

Animations of Satellite Scanning Geometry,
Conical,  Cross-track,  Pushbroom

2.2 Weather Radar in the Tropics »
2.2.1 Ground-based Weather Radar

Hurricane Donna was one of the early tropical cyclones observed by radar (Fig. 2.7a). Hurricane precipitation features can now be examined in greater detail with Doppler radar (Fig. 2.7b), which measures the position and radial velocity of objects.

Doppler radars are so named because these radar measurements use the "Doppler shift", the apparent shift in the frequency and wavelength of a wave perceived by an observer moving relative to the source of the wave, hypothesized by Johann Christian Andreas Doppler. The radial velocity of targets can be calculated based on the phase shift between the transmitted pulse and the received backscatter. A positive phase shift indicates movement towards the radar, while a negative shift indicates movement away from the radar. A positive to negative shift signifies rotation, which facilitates the detection of tornadoes.

Weather radar is less able to track precipitation in complex terrain because the radar beam can be blocked by high terrain. Since the radar scans at an angle and the earth's surface is curved, radar measures the top portions of the precipitation system at far distances. Note that radar detection is limited to a few hundred kilometers. You can learn more about radar operations from the United States National Weather Service radar information services.

The assimilation of radar data into numerical weather prediction (NWP) models has been shown to improve the prediction of thunderstorm structure and the amount of precipitation expected within 0-6 hours.^7,8 For more information on the assimilation of radar data into NWP models, see the lecture by Dr. Juanzhen Sun at http://meted.ucar.edu/AMS_Radar05/index.htm.

United States National Weather Service Doppler Radar information,
http://www.srh.noaa.gov/jetstream/doppler/doppler_intro.htm
Tropical Warm Pool International Cloud Experiment (TWPICE),
http://www.arm.gov/campaigns/twp2006twp-ice
History of Operational Use of Weather Radar by US Weather Services: Part I: The Pre-NEXRAD Era,
AMS Journal Article Online

2.2 Weather Radar in the Tropics »
2.2.2 Airborne Doppler Radar

Doppler radars have been flying on research aircraft since the first prototype was tested in 1982.⁹ Airborne Doppler radars are advantageous for studying individual storm cells and mesoscale weather phenomena because:

• They have very high spatial and temporal resolution. For example, the dual-beam Electra Doppler Radar (ELDORA), developed jointly by NCAR and the French government, is noted for its horizontal sampling resolution of about 0.4 km,^10,11 which is an order of magnitude better than most ground-based radars
• They operate in or near the phenomena of interest

Airborne Doppler radars scan towards the front and the rear of the aircraft, which yields two wind components for each location in the atmosphere (Fig. 2.9). When the principles of conservation of momentum and mass are applied to those data, a three-dimensional view of the atmosphere can be produced. Some limitations of airborne Doppler radar are: flight legs must be relatively straight, and the accuracy of radial velocity is within 1.5 m s^-1 due to aircraft motion (corrections must be made for the movement of the aircraft).^12,13 Aircraft operations are expensive and can, therefore, only provide limited spatial and temporal coverage.

Compare the reflectivity image in Fig. 2.10 (below) with the radar image of Hurricane Donna in Fig. 2.7a. What features can you see now that were not observed in 1960?

The most commonly used airborne Doppler platforms and systems

NOAA P-3, http://www.eol.ucar.edu/projects/vortex95/wp3.html
NCAR ELDORA, http://www.eol.ucar.edu/rsf/eldora/eldora.html
University of Wyoming, http://www-das.uwyo.edu/wcr/
NASA EDOP, http://ghrc.nsstc.nasa.gov/camex3/instruments/edop.html

2.2 Weather Radar in the Tropics »
2.2.3 Satellite-based Precipitation Radar

TRMM, Joint mission of NASA, http://trmm.gsfc.nasa.gov/
Japanese National Space Development Agency (NASDA),
http://www.eorc.nasda.go.jp/TRMM/index_e.htm

Examples of TRMM PR reflectivity images, Tropical Cyclone Boloeste, 3 Feb 2006,
http://trmm.gsfc.nasa.gov/trmm_rain/Events/SIND/SIND.2006-2-3T2225ZBOLOETSE_small.qt

2.2 Weather Radar in the Tropics »
2.2.4 Wind Profilers and Boundary Layer Applications

Wind profilers operate on the same principle as Doppler radar except that they point in a vertical direction from the surface (Fig. 2.12a). Wind profilers provide high-resolution measurements of boundary layer properties, such as wind velocity, temperature, and moisture. Those measurements allow forecasters to identify low-level wind shear, which is hazardous to aviation operations. Wind profiler products show convergence zones or low-level moisture boundaries that can lead to severe weather. For example, a time-height profile of wind velocity can be used to track the timing and depth of the sea-breeze front over the west-central Florida coast (Fig. 2.12b), a common cause of strong thunderstorms.

An array of very high frequency (VHF) wind profilers called the Trans Pacific Profiler Network contributes to the climatology of wind velocity and the natural variability of winds in the equatorial Pacific. The first of these profilers was set up in 1984. Information from the network has been useful for ENSO studies as well as for understanding local island effects on the large-scale winds. A 915 MHz (33 cm) wind profiler was developed by NOAA scientists especially for the tropics.¹⁸ That profiler typically measures 3 to 6 km above the surface except during precipitation when it is able to observe higher altitudes because of its sensitivity to hydrometeors. The profiler has been used to classify tropical precipitating cloud systems into convective or stratiform precipitation. Wind profilers (especially VHF profilers) have provided the best measurements of vertical air motion within convective storms because the air motion and hydrometeor fall velocities can be distinguished from one another. For an in-depth review of wind profiler applications, see Dr. Paul Neiman's lecture at http://meted.ucar.edu/AMS_Radar05/.

A 1998 review of US wind profilers, http://www.ofcm.gov/r14/pdf/profilers.pdf

2.2 Weather Radar in the Tropics »
2.2.5 Upper-tropospheric Radar Applications

The tropical tropopause and lower stratosphere are observed at high resolution using the Equatorial Atmosphere Radar (EAR), a VHF wind profiler (Fig. 2.13a). The EAR has operated at the equator in Indonesia since July 2001. Its purpose is to observe winds and turbulence between 1.5 km and 20 km (troposphere and lower-stratosphere) and irregularities in the ionosphere above 90 km. It has high time and space resolution (1.5 mins and 150 m, respectively). EAR observations of wind velocity help to identify wave features associated with troposphere-stratosphere exchange (Fig. 2.13b)¹⁹ in one of the most convectively active regions of the world.

The EAR is part of a growing regional network of radars and other instruments aimed at improving our knowledge of how cumulus convection and equatorial waves affect weather and climate, regionally and globally.

Equatorial Atmosphere Radar,
http://www.rish.kyoto-u.ac.jp/ear/index-e.html
http://www.rish.kyoto-u.ac.jp/English/EAR/index.html
Atmospheric dynamics in and near the tropical tropopause,
http://kagi.coe21.kyoto-u.ac.jp/en/tidbit/tidbit41.html
Circulation and clouds in the tropics, http://kagi.coe21.kyoto-u.ac.jp/en/tidbit/tidbit38.html

2.3 Satellite Detection of Water Vapor »
2.3.2 Satellite Microwave Estimates of Water Vapor

Early satellite estimates of near-surface humidity were based on empirical relationships between monthly mean precipitable water and in situ observations of specific humidity.^20,21 Now estimates of specific humidity are made directly from the satellite microwave sensors such as the Defense Meteorological Satellite Program (DMSP) Special Sensor Microwave Imager (SSM/I) observations and the NOAA Advanced Microwave Sounding Unit B (AMSU-B). Figure 2.15a shows an example of surface specific humidity as measured by the SSM/I and AMSU-B instruments.²² Estimates of total column water vapor can be estimated from newer satellite sensors such as the NASA Advanced Scanning Microwave Radiometer Earth observing system (AMSR-E), which was launched in 2002 (Fig. 2.15b).

DMSP SSM/I, http://www.ncdc.noaa.gov/ol/satellite/ssmi/ssmiproducts.html
AMSU, Cooperative Institute for Research in the Atmosphere, CIRA, http://amsu.cira.colostate.edu/
AMSR-E, a cooperative effort between NASA and the Japanese Space Agency (JAXA), http://weather.msfc.nasa.gov/AMSR/

2.4 Satellite Soundings

With the scarcity of radiosonde observations in the tropics (Fig. 2.16), satellite soundings are the primary means of observing how moisture is distributed through the tropical troposphere. The first attempts to profile the atmosphere from space used geostationary satellite data to estimate precipitable water and total column water vapor. New observations have taken advantage of the signals from the Global Positioning System (GPS) network of defense satellites. The assimilation of satellite soundings can significantly improve the forecasts of temperature and moisture variable; however, the model performance can still depend on the quality of the satellite retrievals.²³

2.4 Satellite Soundings »
2.4.1 GOES, POES, and DMSP Soundings

GOES, POES, and DMSP Special Sensor Microwave Imager/Sounder (SSMIS) satellite sensors are able to measure temperature and moisture changes with height because atmospheric gases absorb certain wavelengths of outgoing radiation and that absorption occurs preferentially at different heights. A weighting function indicates the relative contributions to the outgoing radiance from various levels of the atmosphere and thereby determines the layer of the atmosphere that is sensed for a given wavelength or spectral band.

Sounders from the past two decades used about 20 spectral bands, while current sounders are hyperspectral i.e., they use an order of magnitude more bands. For example, the polar-orbiting NASA Atmospheric Infrared Sounder (AIRS) uses 65 spectral radiances for temperature, 42 for water vapor, 26 for ozone, and 23 for surface temperature. Hyperspectral sounders provide profiles of about 1 K/1-2 km depth. The next generation of geostationary satellites will have a hyperspectral IR sounder. Polar-orbiting sounders have horizontal resolutions on the order of 1-10 km across swaths, which are on the order of 1000 km (Fig. 2.17). Geostationary sounders provide continuous observations on a hemispheric scale with an approximately 8 km horizontal resolution. Geostationary satellite images also provide estimates of wind velocity around the globe (Fig. 2.17). Geostationary satellites also provide continuous estimated winds derived from water vapor cloud tracks (Fig. 2.17).

Description of satellite sounding instruments and radiometers:
http://cimss.ssec.wisc.edu/itwg/sssp/satellite_programs/instruments.html

2.4 Satellite Soundings »
2.4.1 GOES, POES, and DMSP Soundings »
Box 2-3 Basis of Satellite Sounding Retrievals

Fig. 2B3.1. Outgoing long wave radiance curves for terrestrial temperatures and the absorption by atmospheric gases at different wavelengths or bands.

Satellite remote sensing is optimized for different parts of the spectrum based on wavelength, amount of energy, and altitude of the peak transmission or absorption by different gases and aerosols. Emission and absorption can only occur at discrete wavelengths related to the molecular structure of gases or aerosols. Within the molecules, various possible energy transitions are associated with a series of signature wavelengths. The outgoing radiance (Fig. 2B3.1) is marked by absorption bands.

By selecting different sounding wavelengths, the observed radiances can be used to infer temperature and humidity profiles as well as cloud top pressures.

The emission source has to be a relatively abundant gas of known and uniform distribution. Carbon dioxide is one of the gases that occurs in uniform abundance below about 100 km. The choice of sounding wavenumbers requires understanding of weighting functions, which indicate the relative radiance contributed from various levels of the atmosphere. For instance, the 13.3 micron carbon dioxide weighting function shows that most of the energy emitted at this wavelength originates near the low-middle troposphere, as illustrated below (Fig. 2B3.2).

Fig. 2B3.2. (left) Schematic of radiance emitted by carbon dioxide at 13.3 μm and (right) the corresponding weighting function for that wavelength.

In general, discrete sounder bands are chosen so that the atmosphere becomes progressively more opaque from one spectral band to the next. With increasing opacity, the sensed signal comes from higher and higher in the atmosphere.

2.4 Satellite Soundings »
2.4.1 GOES, POES, and DMSP Soundings »
2.4.1.1 Sounder versus Radiosonde

Satellite sounders provide layer mean temperature and moisture, as opposed to radiosondes, which provide level-specific temperature and pressure. Satellite sounders smooth vertical features, but capture the mean vertical profile very well. An example of each is presented in Figure 2.18. In particular, the GOES sounder:

• Cannot sense through clouds, which is a disadvantage compared with radiosondes.
• Provides 10-km average measurements in the horizontal while radiosonde observations are single-point measurements that are far apart (Fig. 2.16).
• Provides a relatively high, hourly temporal resolution compared to the conventional 12-hour radiosonde observations. The timely information from the GOES sounder can improve mesoscale analysis and nowcasting of severe weather if used to diagnose the pre-convective environment (for example, moisture gradients and boundary layer cold pools) before significant cloud formation.

The COMET module, Introduction to Using the GOES Sounder, explains the scientific basis for sounding retrievals and sample applications.

2.4 Satellite Soundings »
2.4.1 GOES, POES, and DMSP Soundings »
2.4.1.2 Satellite-derived Winds

Wind velocity is derived from automatic tracking of water vapor features in the mid-upper troposphere and cloud elements in the lower troposphere^24,25 (Fig. 2.19). The latter is limited to areas that are free of thick clouds. The assignment of heights is one of the main limitations to the accuracy of feature-tracked winds.

Cooperative Institute for Meteorological Satellite Studies (CIMSS)
Tropical Products, including cloud motion vectors, http://tropic.ssec.wisc.edu/

2.4 Satellite Soundings »
2.4.2 Soundings from GPS Radio Occultation

Radio occultation is the technique by which satellite receivers intercept signals from GPS and infer the deviations in the signal's path caused by temperature and moisture gradients (Fig. 2.20). The technique was first used in the late 1960s to study the atmospheres of distant planets.²⁶

The MetOp satellite carries the first instrument, Global Navigation Satellite System Receiver and Atmospheric Sounding (GRAS), to use radio occultation for the operational use of GPS satellites for atmospheric sounding. Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) consists of six satellites launched in 2006 as part of a joint Taiwan-US project to obtain profiles of atmospheric temperature and moisture at high spatial and temporal resolution using GPS satellites.

The spatial resolution of COSMIC soundings is a special boon to the tropics where radiosonde profiles are few and far between (Fig. 2.21). The assimilation of GPS soundings into numerical prediction models has improved hurricane forecasts as demonstrated in simulations on the COSMIC website, http://www.cosmic.ucar.edu/. You can learn more about radio occultation in the COSMIC webcast, http://www.meted.ucar.edu/COSMIC/.

Real-time Information from COSMIC, http://www.cosmic.ucar.edu/

2.4 Satellite Soundings »
2.4.3 Surface-based GPS Meteorology

Surface-based GPS soundings are derived from the measurement of the signal propagation delay caused by the atmosphere above the receiver. The surface network of GPS receivers (Fig. 2.22) provides autonomous, frequent, and accurate observation of water vapor content at the surface. The ability of this technique to provide highly accurate measurements of atmospheric water vapor content has been established since 1992.²⁷ While the vertical and temporal resolution of the data is very good, its horizontal resolution is poor because of the relatively small number of stations operating globally. Therefore, surface GPS measurements can be applied effectively to the analysis of the passage of frontal boundaries or time series at a point but must be used in combination with other data for regional or continental-scale analysis.

Real Time Integrated Atmospheric Water Vapor from GPS, http://www.suominet.ucar.edu

2.5 Satellite Detection of Clouds and Precipitation »
2.5.1 Observing Tropical Clouds

We use cloud patterns to find weather phenomena, such as the ITCZ, tropical cyclones, sea breezes, and tornadoes. Clouds cover almost two-thirds of the earth's surface²⁸ and help to regulate the radiation balance of the earth-atmosphere system.²⁹ Satellite remote sensing is the primary method of observing and documenting global cloud radiative properties. A variety of new satellite sensors also help us understand more about cloud structure, forms, and microphysical properties.

2.5 Satellite Detection of Clouds and Precipitation »
2.5.1 Observing Tropical Clouds »
2.5.1.1 Three-dimensional (3-D) Cloud Structure

The 3-D nature of clouds is evident from visible satellite images in which protruding clouds create shadows, especially at low sun angles (Fig. 2.23a). Cloud top heights can be estimated from satellite measurements of IR radiance emitted from the cloud top, which is a function of the absolute temperature. Generally, in the tropics warmer cloud tops are lower and colder cloud tops are higher (Fig. 2.23b). During the most recent decade, new satellite sensors with simultaneous, multi-angle observations have provided 3-D views of clouds.³⁰

Multi-angle Imaging Spectro-Radiometer (MISR)

The Multi-angle Imaging Spectro-Radiometer (MISR), which was launched on the NASA Terra satellite in 1999, views clouds from nine different angles (for example, Fig. 2.24) and locates them in three dimensions. It scans at blue, green, red, and near-IR wavelengths (0.446, 0.558, 0.672, 0.866 µm, respectively), has spatial resolution of 275 m across a 360-km swath, and achieves global coverage about once every 9 days at the equator.

Animation showing clouds over Florida,
http://eoimages.gsfc.nasa.gov/ve/169/misr_florida_short_small.mov

As the angle of the view changes, the clouds appear to move, an effect known as parallax. (You can experience it by placing a finger in front of your face and looking at it with one eye at a time. Your finger will appear to move.)

Let us see how this effect is put to work in satellite meteorology. Click on the animation link above to see a small movie of clouds over Florida or link to a larger file of the same movie at NASA's Visible Earth. The greater apparent motion of the cirrus and cumulus tells us that the cirrus is higher. Cloud location in 3-D can be combined with other information, such as the amount of sunlight reflected, to identify different cloud types. Subsequently, it may be possible to understand how each type of cloud affects the global radiation balance.

MISR, http://misr.jpl.nasa.gov/Mission/misrInstrument/
IR satellite images archive for southern Africa (current to past six months),
http://saws.satreponline.org/archive.php

2.5 Satellite Detection of Clouds and Precipitation »
2.5.1 Observing Tropical Clouds »
2.5.1.2 Cloud Forms

Satellite images allow us to distinguish among cloud forms or morphology. For example, cellular boundary layer clouds are easily identified in geostationary visible (1-km resolution) or true color higher resolution images (Fig. 2.25).

Cellular convection occurs when a layer of fluid is warmed at the base or cooled at the top. Open cells indicate air that is sinking in the middle and rising on the edges. Closed cells indicate that air is rising in the middle and sinking on the edges. Both of these cloud forms occur under synoptically stable conditions. Hexagonal cells are typically observed where the air-sea temperature difference is more than 3°C and the wind speeds are less than 7 m s^-1. The association of cloud forms with particular boundary layer conditions was first described by Woodcook,³¹ who observed the soaring patterns of birds over the tropical ocean.

Operational geostationary IR and visible images provide the most frequent views of cloud forms and types.

2.5 Satellite Detection of Clouds and Precipitation »
2.5.1 Observing Tropical Clouds »
2.5.1.3 Microphysical Parameters of Clouds

CloudSat, launched in 2006, is the first millimeter-wavelength cloud radar in space. It detects smaller liquid drops and ice than weather radar and, thus, provides more information about the cloud mass.³² CloudSat measurements provide a means for inferring cloud properties such as particle concentrations, cloud liquid water, and precipitation intensity, which can be used to understand climate variability.

In Figure 2.26, we see the CloudSat profile through Tropical Storm Ernesto. A broad area of high reflectivity extends south of the mountains of the Dominican Republic. Red and orange areas indicate the presence of large amounts of cloud water and/or ice while blue areas above indicate cloud ice. Wavy blue lines along the bottom of the cloud mass indicate intense rainfall. The top-down satellite-IR view misses two small thunderstorms beneath the cirrus anvil. The role of orographic lift in producing large amounts of cloud water is indicated by high reflectivity along the mountain peaks. At the same altitude over the ocean, reflectivity values are mostly lower. You can learn more about CloudSat in Focus Section 1.

2.5 Satellite Detection of Clouds and Precipitation »
2.5.2 IR Estimates of Precipitation

Infrared (IR) emissions are most useful for estimating precipitation from convective clouds because higher, and thus colder, cloud tops correlate with higher precipitation rates (Fig 2.27). To estimate precipitation, IR temperatures are averaged over various areas and times. Those averages are then compared with precipitation measurements to arrive at an operational temperature-precipitation correlation. One example of this is the GOES Precipitation Index (GPI), which uses 235K as the IR temperature with the best correlation to average precipitation for areas spanning 50-250 km over 3-24 hours.

The primary advantage of IR-based techniques is the high temporal frequency of images, for example, up to 15 minutes for GOES and Meteosat Second Generation (MSG) geostationary satellites. IR-only techniques are at a disadvantage compared with radar because the lower-resolution satellite IR cannot detect convective-scale structure and rain from warm clouds. In addition, thick cirrus and convective precipitation appear similar. This means that satellite IR-only techniques tend to underestimate precipitation early in the lifecycle of convective systems when warm rain processes dominate, and overestimate precipitation in the decaying stages when cold cirrus is common. The GPI has a large bias over equatorial Africa and Indonesia,³³ where virga may be interpreted as surface rainfall, and over high mountains, where snow may be interpreted as precipitation (Fig. 2.28).

2.5 Satellite Detection of Clouds and Precipitation »
2.5.3 Microwave Observations of Clouds and Precipitation

Starting with the DMSP SSM/I in 1987, microwave satellite sensors have fundamentally changed how we discern cloud properties and measure precipitation from satellites because they directly detect precipitation in clouds—an advantage over IR-techniques, which have difficulty distinguishing between precipitating and non-precipitating clouds. The SSM/I sensors measure the microwave scattering and emission signatures of liquid water or ice particles.

Microwave detection of precipitation uses several channels, which are usually described by their frequency in gigahertz (GHz). Some channels are located in window regions where the atmospheric gases absorb very little radiation (Fig. 2.29). These windows allow the satellite sensors to "see" the surface, even through clouds. Both the window and high absorption regions are used to derive various products for identifying surface and cloud properties.

Special Sensor Microwave Imager (SSM/I)

For the SSM/I, the brightness temperatures at 37 and 85 GHz (Fig. 2.29) are used to infer the quantity of liquid water and ice in a column, which correlates well to surface precipitation. Both scattering and emission-based methods are used to identify rainfall. Emission techniques work well over the ocean where the background emissivity from the ocean surface is low and uniform. Over land, the background emissivity is higher and more variable, making it difficult to distinguish raindrops. The SSM/I 85 GHz channel is strongly scattered by ice and is used to define rain intensity. Therefore, precipitation-retrievals over land rely upon a measure of scattering.

The SSM/Ic bias is largest in the Northern Hemisphere in winter, when the precipitation estimate is lower than gauges³³ (Fig. 2.30). Contributing factors may be sampling gaps or incorrect removal of data due to surface snow and ice cover. Over Indonesia, biases may be due to the abundance of high clouds. Like the GPI, the SSM/Ic has a large bias over equatorial Africa where hydrometeors detected by the satellite do not reach the surface.

Advanced Microwave Sounding Unit (AMSU)

The AMSU includes channels at 89 and 150 GHz that are well-suited for detecting ice particles. The AMSU precipitation product is based on the scattering of microwaves by ice particles. The equations used to determine the AMSU-derived precipitation allow for more accurate retrievals over land. This contrasts with the earlier SSM/I emission algorithms, which are more suitable for rainfall detection in marine environments.

TRMM Microwave Imager (TMI)

The TRMM Microwave Imager, or TMI, measures the intensity of radiation at 10, 19, 22, 37, and 85 GHz. The four higher frequencies are similar to the SSM/I. The 10 GHz channel was designed to provide a more linear response for the high rainfall rates common in tropical rainfall. By virtue of being in a lower orbit, the TMI provides better resolution than the SSM/I or the AMSU (Fig. 2.31). The TMI will eventually be replaced by even better instruments on the GPM satellites.

Advanced Scanning Microwave Radiometer (AMSR-E)

The NASA AMSR-E is a recent addition to the sensors used to measure precipitation (Fig. 2.32). The frequencies of the AMSR are 6.9, 10.7, 18.7, 23.8, 36.5, and 89.0 (horizontal and vertical polarization), and 50.3 and 52.8 GHz (vertical polarization only). The AMSR-E sensor is similar to the SSM/I and TMI. Over the ocean, the AMSR-E can "see" through smaller cloud particles to measure the microwave emission from the larger raindrops. Over land, the AMSR-E can measure the scattering effects of large ice particles which later melt to form raindrops. These measurements are converted to rain rates using a cloud model.

Unique characteristics of microwave identification of clouds and precipitation include:

At 85-91 GHz

• Deep convection appear relatively cold
• Water clouds and moist air masses have warm brightness temperatures over water surfaces
• Imagery can penetrate thin cirrus canopies and reveal internal storm structure
• Imagery is able to distinguish deep convection, but can not always see low-level circulations when associated primarily with low-level water clouds
• Spatial resolution higher than for imagery at lower microwave frequencies

At 37 GHz

• Water clouds and precipitating clouds appear warm against a relatively cold ocean
• Upwelling radiation is largely unaffected by ice particles, which allows imagery to highlight low-level cloud features
• Imagery resolves details missed by 85-91 GHz
• Imagery shows regions of low-level clouds and rain

More information is available in the COMET modules, Polar Satellite Products for the Operational Forecaster: Microwave Analysis of Tropical Cyclones at http://meted.ucar.edu/npoess/tc_analysis/ and Microwave Remote Sensing: Clouds, Precipitation, and Water Vapor, at http://www.meted.ucar.edu/npoess/microwave_topics/clouds_precip_water_vapor/.

US Naval Research Laboratory Satellite-derived Precipitation Products,
http://www.nrlmry.navy.mil/sat-bin/rain.cgi
NOAA Climate Prediction Center, CMORPH,
http://www.cpc.ncep.noaa.gov/products/janowiak/cmorph.html

2.5 Satellite Detection of Clouds and Precipitation »
2.5.4 Satellite Observation of Tropical Cyclone

One of the first and most important applications of satellite observations in the tropics has been the estimation of tropical cyclone position and intensity. Since the earliest images, the spiraling tropical cyclones have been among the most easily identified features. In addition to identifying tropical cyclone features, enhanced IR images are used to estimate intensity. A special enhancement known as the BD curve is used to identify the cold cloud tops associated with intense convection. Dvorak³⁴ developed a technique, which has since been updated and automated,³⁵ that associates intensity of cyclones with the temperature of the eye cloud organization, and surrounding environment. Figure 2.33 shows changes in the cloud organization, central area of the cyclone, and rain bands as Tropical Cyclone Indlala became more intense. During 13 and 14 March, the formation of a small eye (with warmer temperatures relative to the eyewall) and the expanded area of very cold cloud in bands around the eye indicate the increased intensity.

Microwave sensors have improved the detection of internal cyclone structure, such as the location of the eye, because those wavelengths are sensed through high clouds that sometimes obscure the eye in IR images. For example in Figure 2.34, it is difficult to identify the eye of Tropical Cyclone Indlala from the IR and visible images. However, the eye shows up prominently in the 85 GHz microwave images. Details of the Dvorak technique and its adaptations, as well as other tools used for the assessment of tropical cyclone intensity will be covered in the chapter on tropical cyclones.

CIMSS, Advanced Dvorak Technique, http://cimss.ssec.wisc.edu/tropic/adt/
Tutorial on the use of satellite data to define tropical cyclone structure,
http://severe.worldweather.org/iwtc/pptFile/Topic1/Topic1a/IWTC6-rz.ppt

2.6. Lightning Detection from Space

Lightning, a deadly hazard, is ubiquitous across the tropics, especially the tropical continents, where the highest rates occur.³⁶ Satellite sensors produce a regional to global-scale view of lightning activity,³⁷ unlike most ground-based detectors, which are limited to land areas and detect only cloud-to-ground lightning. Lightning variability is an indicator of thunderstorm severity, the microphysical properties of clouds, and variability in the natural production of nitrogen compounds.

The first satellite lightning sensor was the Optical Transient Detector (OTD),³⁷ which observed lightning globally from a low-earth-orbit satellite between 1995 and 2000. The Global Hydrology Resource Center (GHRC) Lightning Imaging Sensor (LIS),³⁸ an improved version of the OTD, on the TRMM satellite, has observed lightning in the tropics since 1997. Annual average flash rates derived from the LIS and OTD show that central Africa has the highest density of lightning flashes (Fig. 2.35). The Fast Onorbit Recording of Transient Events (FORTE)³⁹ instrument has been observing lightning globally since 1997 and the DMSP Operational Linescan System (OLS) detects nighttime global lightning.

The LIS has a small, solid-state camera with special filters that admit only the peak optical wavelength emitted by lightning. It maintains a 90% detection efficiency by separating background emissions from weak lightning signals. The camera views a 600 km-wide swath, and each LIS pixel covers 5-10 km on the ground, enabling it to resolve storm cells that might have lightning. The OTD operated by detecting the momentary changes in an optical scene that occur with lightning. The sensor viewed a 1300 swath with a spatial resolution of 10 km and had a 40-65% detection rate.³⁷

The OLS has two telescopes and a photomultiplier tube that allows nighttime visible imaging across a 1080 km swath. The OLS produces nighttime views of cities, fires, some clouds, and lightning. Lightning appears as horizontal streaks in the smoothed 2.7 x 2.7 km OLS images (Fig. 2.36). For more information about the OLS and new instruments for nighttime lightning detection access http://www.meted.ucar.edu/npoess/viirs/.

Lightning detection by FORTE is based on the electromagnetic radio frequency energy that a lightning flash produces. Each phase of a lightning flash radiates in a specific frequency band that is recorded by a VHF instrument. Two optical instruments are used to confirm the occurrence of lightning.

The space shuttle cameras have also served as lightning observers from space. Video recordings from space shuttle missions provided confirmation of upper atmospheric optical flashes (called sprites, blue-jets, starters, and ELVES) that had been reported by pilots and ground-observers years earlier. Space observations of electrical pulses in the stratosphere and ionosphere above thunderstorms confirmed early sightings.

The next series of GOES will carry lightning sensors to aid short-term weather analysis and forecasting on a global scale.

GHCC: Observing Lighting from Space, http://thunder.nsstc.nasa.gov/
Lightning: GHCC Space Research and Observations, http://thunder.msfc.nasa.gov/research.html
Lightning Datasets and Browse Images, http://thunder.nsstc.nasa.gov/data/
Los Alamos National Laboratory, http://www.forte.lanl.gov/index.shtml
Geostationary Lightning Mapper (GLM),
http://ams.confex.com/ams/88Annual/techprogram/paper_129399.htm

2.7 Scatterometry »
2.7.1 Surface Wind Velocity

Ocean surface wind observations are essential for short-term forecasting, climatology, and oceanography. Microwave satellite sensors are the primary means of observing the global ocean surface because they are sensitive to small-scale roughness (e.g., foam, breaking waves) on the ocean surface caused by near surface winds. The SeaWinds scatterometer on the QuikSCAT satellite, launched by NASA in June 1999, provides high resolution ocean surface wind vectors. QuikSCAT views an 1800 km-wide swath of the surface, which means that a given location is viewed at most twice per day. Wind measurements are retrieved on a 25 km x 25 km spatial scale.

The scatterometer infers near surface wind velocity by sending pulses of microwave energy to the ocean surface and measuring the backscatter from small-scale waves (Fig. 2.37). Scatterometer-derived winds can be used to track atmospheric weather systems such as tropical waves.

Can you find the wave in Fig. 2.37b?

Feedback:

The wave is evident as an inverted "V" in the wind field

SeaWinds real-time data, http://manati.orbit.nesdis.noaa.gov/quikscat/
NASA JPL Scatterometer, http://winds.jpl.nasa.gov/
US Navy NRL WindSat, http://www.nrl.navy.mil/WindSat/

2.7 Scatterometry »
2.7.2 Precipitation Estimates from Scatterometry

Satellite observations are sometimes used in ways unanticipated by the instrument developers. The rainfall attenuation that is a disadvantage for wind retrieval has become valuable for rainfall measurement. Recent studies have shown that the backscatter from the rain volume in the atmosphere can be used to estimate rain rates.⁴¹ Using a technique known as "differential reflectivity," the scatterometer is able to discriminate between small spherical rain drops and large, oblate (flattened) raindrops. At rain rates greater than 5 mm hr^-1 (when the rain volume has more large, oblate drops), the horizontal polarization of the radar signal exceeds the vertical polarization. The dual polarization of the scatterometer enables the measurement of rain rate (Fig 2.38).

2.8 Satellite Detection of Dust

While we have primarily focused on detecting clouds and precipitation, it is also vital to track the opposite extreme-dry air. The dry, hot air and dust of the Sahara feeds the far-reaching Saharan Air Layer (SAL), which has a great impact on the weather and climate of the tropics.⁴² The SAL has large amounts of mineral dust and strong winds (~10-25 m s^-1). The endless supply of dust has adverse socio-economic impacts on health, agriculture, and marine life.

Special products derived from geostationary satellite sensors (GOES and Meteosat) are used to track dust and other aerosols (airborne particles). The detection of dust from satellites uses the differencing of IR channels (Fig. 2.39), mid-level water vapor, and multi-spectral true color imagery.

A dramatic outbreak of Sahara dust, which reduced visibility across northern Africa, the Canary Islands, and the Mediterranean in March 2004, could be tracked using these satellite products (Fig. 2.40).

The SAL inhibits the intensification of tropical cyclones.⁴² The lifecycle of Hurricane Erin illustrates the impact of the SAL on hurricane intensification (Fig. 2.41). Dry air and enhanced vertical wind shear from the strong winds in the SAL kept the tropical cyclone weak. Erin strengthened rapidly over the western Atlantic after escaping the SAL.

False color enhancements can be used to identify dust over land and water (Fig. 2.42). The COMET webcast, Dust Enhancement Techniques Using MODIS and SeaWiFs, http://www.meted.ucar.edu/npoess/dust_enh/, has more information about dust enhancement on satellite images.

Since June 2006, the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) mission has provided vertical profiles of dust and other aerosols. Its mission is to help scientists understand the impact of clouds and aerosols on Earth's weather, climate, and air quality. The satellite carries a lidar, an active sensor that is like radar, except that it emits short pulses of green and IR light. The light pulses are reflected from cloud and aerosol particles thereby producing a vertical profile of the atmosphere. Each lidar sample profile is 300-feet long. The profiles are collected and streamed together to create a vertical slice of our atmosphere along the satellite orbit. The satellite also carries passive IR and visible imaging systems. CALIPSO and CloudSat observations are taken just 15 seconds apart, which enables scientist to study how aerosols and clouds interact.

CIMSS SAL Analysis, http://tropic.ssec.wisc.edu/
CALIPSO, Joint mission between NASA and the French Space Agency (CNES),
http://www.nasa.gov/mission_pages/calipso/main/index.html

CALIPSO Animation,
http://www.nasa.gov/mpg/133356main_calipso%20science_NASA%20WebV_1.mpg

2.9 Monitoring Earth's Surface from Space »
2.9.1 Surface Water: Sea Surface, Soil Wetness, Floods

Animation of SST anomalies 2002-2003 El Niño,
http://www.ssmi.com/ocean_data/video/el_nino_2002_2003/amsre_sst_anomaly.mpg

Satellite remote sensing of the Earth's surface is based on two critical properties of microwave radiation, polarization and an electrical property known as the dielectric effect (a dielectric has low electrical conductivity). Water surfaces are strong polarizers of microwave radiation, which leads to strong reductions in emissivity and brightness temperature. Anomalous changes in the properties of the water surface, such as its density and temperature, can be monitored from space. This includes the sea surface temperature (SST) anomalies associated with El Niño (Fig. 2.43). Land surfaces are generally weak polarizers, making emissivity and brightness temperature significantly greater and resulting in a strong contrast with water.

Learn about polarization from COMET's Polarization of Electromagnetic Radiation.

The amount of microwave radiation emitted by the soil diminishes significantly as soil moisture increases.⁴³ Very dry soil will appear relatively bright/warmer to a microwave sensor, while saturated soil will appear relatively dark/colder (Fig. 2.44).

The dielectric effect strongly influences microwave radiation interactions with the earth and atmosphere. Most gases and dry land-based materials have a weak dielectric effect, while wet soil, moist vegetation, and water surfaces have a strong effect. These traits aid satellite monitoring of soil moisture (e.g., Fig. 2.45).

TRMM TMI and PR-derived products are used to monitor flood potential in the tropics. A combination of visible and IR channels can be used to monitor flooding. Figure 2.46 shows the flooding of the Zambeze River (southeast Africa) between January and February 2001.

Real-time monitoring by US Drought Information Center, http://www.drought.noaa.gov/
TRMM Flood Potential, http://trmm.gsfc.nasa.gov/publications_dir/potential_flood.html
Dartmouth Flood Observatory, http://www.dartmouth.edu/~floods/

2.9 Monitoring Earth's Surface from Space »
2.9.2 Vegetation

Various vegetation indices and satellite image enhancements (e.g., Fig. 2.47) have been developed to detect changes in vegetation cover and levels of plant stress - information that can aid in famine mitigation.

• The Leaf Area Index (LAI) corresponds to the green leaf area. Values higher than 1 indicates multiple, overlapping layers of leaves.
• The Fraction of Photosynthetically Active Radiation (FPAR) measures how much sunlight the leaves are absorbing.

NASA Visible Earth website, http://visibleearth.nasa.gov/ Search on "vegetation".

2.9 Monitoring Earth's Surface from Space »
2.9.3 Biomass Burning

Detection of biomass burning from satellite (Fig. 2.48) has become increasingly sophisticated and has been used for tracing chemicals, forecasting aviation hazards, and documenting deforestation. Satellite sensors also monitor burn scars (Fig. 2.48c) as those areas can form caked surfaces through which water does not infiltrate easily. The result is that when heavy rainfall occurs, most of the water remains on the surface and causes flash floods.

The Wildfire Automated Biomass Burning Algorithm (WFABBA) is currently generating half-hourly fire data for the Western Hemisphere in real-time.

Wildfire Automated Biomass Burning Algorithm, http://cimss.ssec.wisc.edu/goes/burn/wfabba.html

2.9 Monitoring Earth's Surface from Space »
2.9.4 Land Use, Land Cover, and Other Surface Changes at High Resolution

Surface-atmosphere feedbacks affect weather and climate. For example, urbanization affects weather by creating convergence zones for thunderstorm initiation, and changes the regional energy balance and climate. High-resolution land-use/land-cover data are especially critical to the performance of mesoscale models and boundary layer turbulence models.

One of the sensors used to monitor land-use/land-cover is the Synthetic Aperture Radar (SAR).⁴⁴ SAR is suitable for observing the surface because it uses microwave technology that is largely unaffected by clouds and time of day. SAR works like other radars except that it has very fine resolution in the azimuthal direction. Finely resolved images are normally achieved by using a large antenna to focus the transmitted and received energy into a sharp beam. However, an antenna suitable for observing Earth's surface would be impractical. A large antenna can be synthesized by combining signals from an object along a radar flight track and processing the signals as if obtained simultaneously from a single large antenna. The distance over which the signals are collected is known as the synthetic aperture.

SAR was first used on the NASA Seasat satellite in 1978.⁴⁵ SARs on space shuttles produced high-resolution global maps of topography, land use, and land surface changes (Fig. 2.49). SARs on current satellites include the Canadian RADARSAT-1 and European Envisat Advanced Synthetic Aperture Radar (ASAR), and the Japanese Phased Array type L-band Synthetic Aperture Radar (PALSAR) whose properties are summarized in (Table 2).

These satellite SARs can monitor land surface changes at a given location on the timescale of weeks, which provides information about the impact of river flooding, burn scars from biomass burning, and so on.

In addition to monitoring land surface changes, satellite SARs measurements of the ocean surface have aided tropical cyclone science. Envisat-ASAR is sensitive to surface roughness and can resolve surface heights down to sub-millimeter scales- a useful capability as sea-surface height anomalies on the mm-cm scales are associated with enhanced tropical cyclone ocean heat potential and intensity changes.⁴⁶ Satellite SAR observations of ocean surface circulations are also useful for estimating surface wind velocity and identifying wind streaks in tropical cyclones.⁴⁷

Prior to SAR, high-resolution surface images were produced mainly by Landsat sensors, which use visible plus reflected and thermal IR bands. Landsat images cover hundreds of kilometers and resolve features on the order of 10s of meters. Commercial and military reconnaissance satellites also produce high-resolution surface images.

What is imaging radar? http://southport.jpl.nasa.gov/desc/imagingradarv3.html
RADARSAT, http://www.asc-csa.gc.ca/eng/satellites/radarsat1/
Envisat-ASAR, http://envisat.esa.int/instruments/asar/descr/charact.html
PALSAR, http://www.eorc.jaxa.jp/ALOS/about/palsar.htm
Landsat, http://landsat.usgs.gov/

2.10 Monitoring Air Chemistry from Satellites

Surface monitoring of chemical pollutants is non-existent in much of the tropics, where biomass burning (for example, forest and grassland fires) is the primary source of tropospheric pollution. Satellite observations are essential for understanding the complex interactions of chemistry with weather, and climate. Recent satellite missions have provided new information on the sources, amounts, and transport of ozone (O₃), carbon monoxide (CO), and other pollutants.

For example, the Measurements Of Pollution In The Troposphere (MOPITT) instrument on the EOS Terra satellite detects CO from IR emissions and methane, and CO from reflected sunlight. Its spatial resolution is 22 km. It is able to distinguish among sources of pollution. High levels of CO have been recorded over central Africa, southeast Asia, and surrounding oceans^48,49 (e.g., Fig. 2.50). Elevated levels of CO are produced when grassland, farmland, and forests are burned to prepare land for agriculture.

The AURA satellite, launched in Aug. 2004, has four instruments mainly observing the upper troposphere, stratosphere, and mesosphere. The Ozone Monitoring Instrument (OMI) is the most valuable for tropical pollution climate studies. It continues 34 years of ozone monitoring that began with the Backscatter Ultraviolet Detector (BUV) in 1970 and the Total Ozone Mapping Spectrometer (TOMS) in 1978. The OMI measures solar reflected and backscattered light in the ultraviolet and visible parts of the spectrum (Fig. 2.2a). Those data are also used to estimate the amount of UV reaching the surface, which helps forecasters decide when to warn the public about excessive UV radiation.

The OMI will also map sulphur dioxide and, in combination with other instruments, monitor tropospheric ozone and nitrogen dioxide. The Tropospheric Emission Spectrometer (TES) measures lower tropospheric pollution associated with air quality. By combining atmospheric models with AURA satellite measurements of chemical pollutants, scientists are able to show the link between biomass burning and regions of elevated ozone amounts (Fig. 2.51).

MOPITT, http://mopitt.eos.ucar.edu/mopitt/
EOS AURA, http://aura.gsfc.nasa.gov/

2.11 Remote Sensing of Volcanoes

Volcanic eruptions create numerous natural hazards⁵⁰ and many of the world's active volcanoes threaten population centers in the tropics. Wind-driven ash (tephra) threatens far larger areas than any other volcanic hazard. Satellites are practical tools for monitoring the diffusion of ash and other volcanic debris⁵⁰ (Fig. 2.52), which pose a particular danger to aviation. Ash can also aggregate into large particles that fall quickly and cause tremendous harm to life at the surface.

Volcanic ash clouds are commonly identified by an abnormal negative temperature difference between the 11 and 12 µm IR channels, referred to as a "split-window" or "reverse absorption" technique. In the resulting image, volcanic ash is distinguished from meteorological clouds (Fig. 2.52b). This technique is standard and generally reliable for operational forecasting but it can produce false alarms because ice-topped clouds have a similar IR temperature difference. In the tropics, high water vapor content and the presence of large amounts of cloud ice makes the technique less effective. Therefore, other techniques are applied. True color images, in which visible and near IR channels are combined, can also detect ash (Fig. 2.52a). Some true color images can be deceptive in that they identify denser, lower debris rather than the fine ash that is dangerous for aviation.

Volcanic clouds contain gaseous pollutants and silicate particles (ash), which scatter microwave radiation. They can be tracked using radar⁵¹ and satellite microwave sensors.⁵² Noxious gases from eruptions are monitored not only by air chemistry sensors, such as OMI and TOMS, but also by geostationary satellites (Fig. 2.53). Since 2003, the Atmospheric Infrared Sounder (AIRS) has provided images of volcanic sulphur dioxide and aerosols. Distinctive thermal features of volcanoes are also identified by the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) on the NASA Terra satellite. Volcanic topography has been measured in fine detail (1 m-100 m scale) by synthetic aperture radars on aircraft, satellite, and space shuttle missions.

Volcanic Emissions from OMI and TOMS, http://so2.gsfc.nasa.gov/
Remote Sensing of volcanoes, http://www.geo.mtu.edu/~raman/Remote.html
GOES volcano enhancement, http://rsd.gsfc.nasa.gov/goes/text/montserrat.html

Focus Areas »
Focus 1: Satellite Cloud and Precipitation Radars »
2F1.1 CloudSat

CloudSat,³² launched in 2006, carries the first satellite-based millimeter wavelength cloud radar. It is the world's most sensitive cloud-profiling radar (Fig. 2F1.1), more than 1000 times more sensitive than current weather radars. It collects data about the vertical structure of clouds, including the quantities of liquid water and ice, and how clouds affect the amount of sunlight and terrestrial radiation that passes through the atmosphere. The poor representation of clouds and their impacts are sources of error in climate models. CloudSat measurements should increase understanding of cloud processes and reduce errors in prediction of both climate and weather.

Cloudsat-animation (mpg)

CloudSat images are also providing valuable information on the vertical structure of hurricanes (Fig. 2F1.2). From an IR perspective, Hurricane Gordon appears to be fairly symmetric, but the CloudSat profile reveals a dramatic contrast between the weak northern and strong southern parts of the eyewall. Ileana has stronger and symmetric eyewalls that extend up to 16 km compared with the asymmetric and weaker convection in Gordon, where the eyewall reaches only to 14 km. Comparing the profiles of Ileana and Gordon, we see a tilted southern eye wall of Tropical Storm Gordon. When tropical cyclones are being sheared aloft, they weaken, and the vertical structure of the clouds can indicate that shearing. The CloudSat profile also shows the variations in the outer rainbands of Ileana. The south side of the cyclone has a nearly continuous cloud while the north side has a gap between the outer rainbands. From the cloud profile, we can distinguish between cirrus and deep convective clouds, which cannot be done with images of cloud tops only. Notwithstanding these examples, CloudSat rarely gets a pass right through the eye of a tropical cyclone, and the latency of its products is 3-6 hours, so it is not useful for real-time analysis of tropical cyclones.

Let us take another look at the CloudSat profile of Hurricane Ileana. Do you notice the heavy line near the bottom of the cloud profile in Figure 2F1.3? What does that line represent?

Fig. 2F1.3. CloudSat Profile of Hurricane Ileana.

Feedback:

The line represents the reflectivity of the ocean surface.

CloudSat Home, http://cloudsat.atmos.colostate.edu/
Cloudsat Data Browser,
http://meteo04.chpc.utah.edu/cgi-bin/cloudsat/browser/wellcome.pl

2F1.2 TRMM Precipitation Radar

http://trmm.gsfc.nasa.gov/trmm_rain/Events/O_Orbit_new.mov

TRMM Precipitation Radar (PR), launched in 1997, is the first space-based precipitation radar.⁵³ It has a spatial resolution of 4 km horizontally, and 250 m vertically. The TRMM PR starting swath of 220 km was increased to 247 km (Fig. 2F1.6) when the satellite was boosted to a higher orbit in 2001.⁵⁴ As an active sensor, the TRMM-PR has superior spatial resolution to the TMI, a passive microwave sensor. After validation tests in which the rainfall measurements of the PR were found to be comparable with those of ground-based radars⁵⁵ (Fig. 2F1.7), the TRMM PR has been used for several applications.

The PR has provided more accurate information about the amount and type of precipitation across the tropics,⁵⁵ the diurnal cycle of precipitation,⁵⁶ tropical cyclone structure and intensity (Fig. 2F1.8), precipitation variability, cloud microphysics, the structure and organization of mesoscale convection, and heat and moisture budgets.

Latent heat, the heat released or absorbed when water changes phase, is a critical component of the global energy cycle. For example, large quantities of latent heat are released in hurricanes, especially in the eyewall⁵⁷ (Fig. 2F1.9).

Data from the PR have been used to more accurately document the diurnal cycle of rainfall,⁵⁶ which is currently forecasted poorly by models. Numerical prediction models, including hurricane prediction models, have shown improvements when data from the PR are assimilated.^58,59 New flood potential products have been derived from PR and TMI measurements.

TRMM NASA Home, http://trmm.gsfc.nasa.gov/
TRMM JAXA Home, http://www.eorc.nasda.go.jp/TRMM/index_e.htm
TRMM Educational Resources, http://trmm.gsfc.nasa.gov/education_dir/education.html
TRMM Data Products, http://trmm.gsfc.nasa.gov/data_dir/data.html
TRMM Flood Potential, http://trmm.gsfc.nasa.gov/publications_dir/potential_flood.html

Current Week of TRMM Rainfall, http://trmm.gsfc.nasa.gov/trmm_rain/Events/wksmall.qt
Week of Rain Accumulation, http://trmm.gsfc.nasa.gov/trmm_rain/Events/acummulate_small.qt

Focus 2: JPSS, GPM, and Tropical Precipitation Estimates

Microwave sensors and radars on various orbiting platforms provide several views of tropical rainfall, atmospheric humidity, and land and ocean surface properties that help us to understand tropical weather and climate. New instruments scheduled for the Joint Polar Satellite System (JPSS), and Global Precipitation Measurement (GPM) missions promise more complete temporal coverage and advanced sensors.

2F2.1 JPSS

Movie of JPSS Three-Orbit System.

The JPSS constellation will work in partnership with MetOp as illustrated in Fig. 2F2.1. The JPSS primary mission areas are: atmosphere, land, ocean, space environment, and climate. The enhanced capabilities of JPSS systems include higher spatial resolution, multi-spectral imagers, hyperspectral sounder, large improvement in the timeliness of data availability (95% of global data will be available within the half hour), and repeat local coverage every four hours.

The current system of geostationary IR sensors, operational polar-orbiting microwave sensors, and research microwave instruments are being used to develop prototype blended precipitation measurements. The frequently available (15-30 mins. or greater) geostationary IR is calibrated by measurements from satellite microwave sensors and TRMM radar. Precipitation products are being tested by NASA, NOAA, and the US Naval Research Laboratory (NRL) (Fig. 2F2. 2).

View animation.

View animation.

View animation.

View animation.

Environmental Satellite Resource Center, (ESRC), http://www.meted.ucar.edu/esrc

2F2.2 GPM

The Global Precipitation Mission (GPM) will extend the TRMM mission by providing coverage at higher latitudes. It will be capable of measuring rain rates from 0.25 to 100 mm hr^-1. GPM will aim for three-hourly revisits over 80% of the globe with the goal of making data available to users within three hours of observation time.

The GPM will consist of a core satellite with dual-frequency precipitation radar (Fig. 2F2.3) and microwave instruments and a constellation of polar-orbiting satellites whose precipitation estimates can be calibrated against those of the core satellite (Fig. 2F2. 4).

GPM Home, http://pmm.nasa.gov/GPM

GPM Era Satellite Coverage, http://pmm.nasa.gov/multimedia/videos/GPM

GPM Multimedia Resources, http://pmm.nasa.gov/multimedia

Operational Focus

Weather Radar

Satellite Water Vapor

Satellite Soundings

Clouds and Precipitation Analysis from Satellite Imagery

Satellite-derived Winds

Air Quality and Hazard Assessment Applications of Satellite Imagery

Surface Moisture Detection from Satellites

Lightning detection by satellites

Summary

Remote sensing in the tropics is essential to weather analysis, numerical weather prediction, climate studies, and mitigation of weather-related hazards. During the past four decades, remote sensing from satellites and radars has vastly increased our knowledge of the tropical atmosphere and surface properties. The advent of high-resolution microwave sensors brought fundamental new information on the water cycle in the tropics. New airborne radars have provided detailed information about the inner core of tropical cyclones - information that is valuable for understanding intensity changes. Microwave sensors and radars on various orbiting platforms provide several views of tropical rainfall, atmospheric humidity, and land and ocean surface properties that help us to understand tropical weather and climate. With the next generation of satellite sensors and radars, we expect to learn even more about tropical meteorology.

Status of Current and Future Environmental Satellites,
http://www.wmo.int/pages/prog/www/OSY/GOS.html

Appendix A: Tropical Radar Images

Appendix B: Satellite Microwave Sensors and Measurement Capabilities

Questions for Review

1. List at least three applications of radar in tropical meteorology.
2. Discuss the advantages of spaceborne radars compared with ground-based radars.
3. Describe the advantages of using geostationary satellite sensors for tropical weather and climate analyses.
4. Describe the advantages of using polar-orbiting satellite sensors for tropical weather and climate analyses.
5. List three types of satellite remote sensing systems that produce images of clouds. Include several advantages and disadvantages of each system.
6. How are satellites used for air quality hazard assessment, such as dispersion of volcanic ash, chemical pollutants, dust, and smoke?
7. Describe three types of sensors that are used for air quality study and list their advantages and limitations.
8. Why are microwave sensors useful for identifying surface moisture?
9. How do microwave sensors differentiate between moist and dry soils?
10. Describe how vertical temperature and moisture profiles are derived from satellite sensors.
11. What are some advantages and limitations of satellite-based soundings of temperature and moisture?
12. Satellite wind observations are available over the tropical oceans, formerly data-sparse regions, yet limitations remain. How is wind velocity derived?
13. What are the limitations of satellite observation of winds?
14. Lightning, a ubiquitous hazard in the tropics, will be detected by the next generation of geostationary satellite sensors. How is lightning detected from satellites?
15. What types of variables are measured by satellite sensors used for lightning detection?

QUIZ

You may also take a quiz and email your results to your instructor.

Brief Biographies

Radar Meteorology: Dr. David Atlas

Dr. David Atlas is recognized as one of the founding fathers of radar meteorology. According to the AMS, Atlas "solved many puzzles and invented numerous techniques that transformed a fledgling application into a vital scientific and operational tool." He served as one of the first radar meteorologists with the U.S. Army Air Corps during World War II. He received his B. S. in Meteorology from New York University (NYU) in 1946. The following year, he invented the iso echo contouring concept which quantified reflectivity on cathode ray tubes - a tool used on ground and aircraft radars for decades. He led the Weather Radar Branch of the U.S. Air Force Cambridge Research Laboratory (AFCRL) for 18 years. He earned his doctorate from the Massachusetts Institute of Technology (MIT) in 1955. While at the AFCRL, he studied the application of Doppler radar to wind measurements. He and Roger Lhermitte captured audio recording of the shift in frequency upwind and downwind of the radar as they observed the corresponding velocity-azimuth image in 1957. From 1966 to 1972, he was a professor in the Department of Meteorology at the University of Chicago, where he led a collaborative effort to create the first mobile Doppler radar system for use in field programs around the country. The CHicago ILLinois (CHILL) radar, is still operating as a research and educational facility at Colorado State University. In 1972, he became the Director of the NCAR Atmospheric Technology Division and spearheaded a period of major advancement in weather radar use, including the study of severe weather using Doppler radar. He also headed the National Hail Research Experiment at NCAR. He was the establishing director of the NASA Laboratory for Atmospheric Sciences at the Goddard Space Flight Center in 1977. He directed the development of space-based instruments for observing the earth systems until his retirement in 1994. Although retired, he continues his research as a NASA Distinguished Visiting Scientist.

He is the recipient of several awards including: Member of the U.S. National Academy of Engineering, Fellow of the American Geophysical Society, the UK Royal Meteorological Society, from whom he received the Symonds Gold Medal and the AMS. He served as AMS President in 1975 and received the Rossby Medal, the highest honor of the AMS, in 1996.

Radar Meteorology: Dr. Louis Battan (1923 - 1986)

Dr. Louis Battan was a pioneer in cloud physics and radar meteorology. Along with colleague and close friend, Dr. David Atlas, he underwent rigorous training in radar engineering and meteorology in the U.S. Army Air Corps, at Harvard University, and MIT. He received his B.S. from NYU in 1946 and then moved to the University of Chicago where he obtained his M.S. and Ph.D. in 1953. During the Thunderstorm Project (1946-48), he used radar analysis to show precipitation initiation from coalescence in midlatitude convective clouds. Dr. Battan, along with Drs. Roscoe Braham Jr. and Horace Byers, conducted one of the first randomized experiments on cloud modification by the artificial nucleation of cumulus clouds. After obtaining his Ph.D. in 1953, he remained at Chicago until 1958. After that, he became a professor in the Department of Atmospheric Sciences and Institute for Atmospheric Physics, University of Arizona in 1958 and served as its director from 1973 to 1982. There, he conducted research on clouds, precipitation processes, lightning, and radar relationships. He led the development of the first 3-cm Doppler radar to measure vertical motion and particle sizes in thunderstorms in 1964. He was the AMS President from 1966-67 and served on numerous national and international committees including the U.S. President's National Advisory Committee on Oceans and Atmosphere in 1978. He was instrumental in the founding of NCAR. He received many awards including the AMS Meisinger Award in 1962 and the AMS Half Century Award in 1975.

He was a prolific writer whose repertoire includes one of the first textbooks on radar meteorology in 1959 and "Radar Observation of the Atmosphere" in 1973, which became the reference text on the subject. He authored 16 books and more than 100 articles. Dr. Battan died in 1986. His contribution to meteorological education, through publications written in an accessible and informative style, has been honored by the AMS with the establishment of two annual "Louis J. Battan Author's Awards".

Satellite Meteorology: Dr. Verner Suomi (1916-1995)

Dr. Suomi is known as the "Father of Satellite Meteorology". Viewers of satellite images should be aware that these images are the product of the innovative mind of Dr. Suomi. His most famous invention is the spin-scan camera which, in 1966, provided the first continuous observations from geostationary orbit. He was the founding director of the Space Science and Engineering Center (SSEC) and the Cooperative Institute for Meteorological Satellite Studies (CIMSS) at the University of Wisconsin-Madison in 1965, where he was also a professor of atmospheric and oceanic sciences and soil science. Dr. Suomi also directed the development of the McIDAS software to display, analyze, and integrate satellite images with other data. He has received numerous awards including the AMS Meisinger Award for aerological research achievements in 1961, the National Science Medal in 1977, and the WMO International Meteorological Organization Prize for pioneering contributions as father of weather satellites in 1993. After his death in 1995, Dr. Suomi was commemorated by his students, colleagues, and friends from the University of Wisconsin and elsewhere in:

Verner Suomi, A man for all seasons. University of Wisconsin-Madison. SSEC Publication No.98.03.SI. [Available from the Schwerdtfeger Library, 1225 W. Dayton St, Madison. WI 53706]

The forty-year anniversary of the first Earth-observing geostationary satellite was celebrated in December 2006. Highlights are featured on the website of the SSEC at: http://www.ssec.wisc.edu/media/features/dec26_06.htm.

References