| MTP Home Page
|
This page is written for the layperson, and is intended
to introduce folks without a technical or scientific background to the
concept of microwave temperature profiling, and why it is an important
and useful technique.
There will be two discussions:
one short and one long. The short discussion will provide a top-level description
of what an MTP does without getting into all the details and technical
terms. The long discussion will go into more details and provide illustrations,
but still at the layperson level. It may be useful to read the Short Version
before the Long Version.
Links to Short
Version and Long Version
A Layperson's Guide to Remote Temperature Sounding
Short Version
MJ Mahoney
Temperature is a fundamental property of everything, big or small, and
it can tell you important information about these systems. To determine
whether or not you have a fever, you might place a thermometer under
your tongue. The thermometer directly measures your body temperature. To
measure the vertical temperature structure of the atmosphere, scientists
often rely on radiosondes which are launched from the ground and carried
aloft by a helium balloon. This is also a direct temperature measurement.
Radiosondes however are only launched twice a day, and launch sites are
typically 500 km apart. Because the atmosphere's temperature can change
significantly in 12 hours and over spatial scales much smaller than 500
km, atmospheric researchers often require a better means of measuring temperature.
This is particularly important when research instruments are carried aboard
aircraft to study problems such as the ozone holes, global warming, or
the impact of aircraft emissions on atmospheric chemisty. Research aircraft
are flying chemisty labs which measure dozens of gaseous species (e.g.,
ozone, nitrous oxide, chlorine monoxide, methane, etc.) and aerosols (liquid
or frozen particles) in the atmosphere.
To interpret what these measurements mean, it is often necessary to
know the temperature not just at the aircraft altitude -- where the chemical
measurements are made -- but above and below the aircraft as well. The
reason for this is that the vertical temperature profile provides meteorological
context for the in situ measurements. For example, it is important
to know whether the aircraft is in the troposphere (the lowest part of
the atmosphere) or in the stratosphere (the next highest region). Ozone
(O3) normally has much higher abundances in the stratosphere
than the troposphere because it is produced there. Similarly, carbon monoxide
(CO) has much higher abundances in the troposphere than in the stratosphere,
because that's where it's produced. If high O3 and low CO are
seen, one might conclude that the aircraft is in the stratosphere. However,
this condition can occur occasionly in the troposphere, and the only way
to know for sure what the real situation is is to measure where the thermal
tropopause is located, as this separates the troposphere from the stratosphere.
To do this this we need to measure an altitude temperature profile.
So how is the temperature profile of the atmosphere measured from an
aircraft? There is really only one practicable, proven way to do
this, and that is to use a microwave temperature profiler (MTP). Obviously
the MTP must make a remote temperature measurement. It does this by detecting
the naturally-occuring thermal emission of microwaves from oxygen molecules
in the atmosphere. As is explained in the detailed version of this write-up,
oxygen molecules satisfy several criteria which make such a measurement
possible. When the MTP is scanned in frequency and elevation about the
flight altitude, it measures the total emission from oxygen molecules along
the viewing direction. The temperature profile is not measured directly
however. The 20 or 30 measurements that the MTP makes in scanning from
near-zenith to near-nadir, are converted into a temperature profile by
performing a "retrieval" which takes into account where the aircraft is
flying and the time of year. You might imagine that many different combinations
of atmospheric conditions in the viewing direction might produce the same
measurement, and you would be correct! The retrieval process, however,
uses all of the measurements during a scan from zenith to nadir to figure
out which is the most likely set of temperatures to produce the measurements,
and this becomes the retrieved altitude temperature profile. In a nutshell,
that is what an MTP does.
MTPs have a many more potential applications beyond their use as a research
instrument. They are useful for example in predicting the possibility of
clear air turbulence (CAT) and icing hazard potential for commercial aircraft,
and also improving their fuel efficiency. MTPs will also be able to detect
when unstart conditions will occur is a supersonic transport (SST) fleet
so that preventative measures can be taken. A key to making these and other
applications affordable is the development of monolithic microwave integrated
circuit (MMIC) technology, which will reduce the cost of manufacture by
about an order of magnitude.
A Layperson's Guide to Remote Temperature
Sounding
Long Version
MJ Mahoney
Table of Contents
Introduction
Most people would agree that the measurement of temperature is a useful
thing to do. We like to know what the temperature is if we are planning
some outdoor activity, or what our body temperature is if we are not feeling
well. Scientists and engineers have many, many more reasons to measure
temperature, as it is a fundamental physical property of just about everything.
Since my particular research interest is the measurement of the vertical
profile of temperature through the atmosphere, and since that is what the
Microwave Temperature Profiler (MTP) Web Pages are about, the ensuing discussion
will be geared with that in mind.
How Temperature Measurements Are Made
Generally temperature measurements are made with some kind of thermometer,
and thermometers can be made in many different ways. The simplest thermometer
just makes a relative temperature measurement. For example, if you want
to check quickly whether someone has a fever, you might put your hand on
their forehead. If their forehead is warmer than your hand, heat will flow
to your hand, and you will conclude that they have a fever because they
"feel" warmer than you.. Such a thermometer is not very useful however.
What is needed is a device that can make absolute temperature measurements.
Thermometers are devices which are able to measure the average thermal
energy (in an absolute sense) of some material or medium, and assign a
temperature to it. And because the laws of physics are more simply expressed
on a temperature scale with respect absolute zero, scientists generally
measure temperature in Kelvins, rather than degrees Celcius or Fahrenheit.
Most people are familiar with in situ temperature measurements,
where the thermometer is placed in or next to the thing whose temperature
is to be determined. Often, however, this may not be possible: the medium
may be so hot that the thermometer would either melt or loose calibration,
or it may be so far away that it can't easily be reached. In these cases,
remote sensing techniques must be used. Unlike in situ
sensors, which rely on directly heating the thermometer (e.g., a mercury
bulb or a thermistor), remote sensing techniques rely on detecting
the energy radiated by the object or medium of interest, and then determining
a temperature based upon the properties of the radiation. We are all familiar,
for example, with the "heat" energy that we feel near a smoldering log.
This energy, which is invisible to the human eye, is called infrared (IR)
radiation and our skin has evolved to be able to sense it (for obvious
reasons), just like our eyes have evolved to sense visible light (also
for obvious reasons).
So
what is it about radiation that let's us measure temperature? Radiation,
or more properly electromagnetic radiation, is part of a broad energy spectrum
of naturally occuring thermal (and non-thermal) emissions which are carried
by particles called photons. Photons travel at the speed of light in a
vacuum, and their energy is characterized by either their wavelength or
their frequency. The most energetic photons have the shortest wavelengths
(highest frequencies), and include gamma rays and x-rays. In the "middle"
of the electromagnetic spectrum, at lower energy, are the ultraviolet (UV),
visible and infrared (IR) photons. Microwaves and long radio waves, the
least energetic of the photons, are at the longest wavelengths (lowest
frequencies). Now, without getting into the thermal physics of all this,
when an object is in thermodynamic equilibrium with its surroundings (that
is, when it emits and absorbs equal amounts of energy), the intensity of
its thermal emission at any frequency (or wavelength) depends on only one
thing -- it's temperaure! This remarkable fact, which physicists refer
to as the Planck Blackbody Distribution, is the basis for most remote temperature
measurements. Generally, remote temperature sensing is done in the infrared
at wavelengths between 0.7 and 20 microns. (see Infrared
Temperature Measurement Theory and Application for an introductory
discussion of IR remote sensing).
Microwave Remote Temperature Sensing
In principle, however, temperature can be measured at any wavelength of
the electromagnetic spectrum; the proper choice depends on the details
of the application. In the case of atmospheric remote temperature sensing,
advantage is taken of several properties of oxygen molecules, which comprise
23% of the mass of the Earth's atmosphere.
First, oxygen molecules radiate (and absorb) at a number of discrete
frequencies between 50 and 70 GHz (microwaves with a wavelength of ~0.5
cm). These discrete frequencies, or spectral lines, are a consequence of
rules of quantum mechanics which only allow oxygen molecules to have particular
rotational energy states. Furthermore, since the oxygen molecules are in
thermodynamic equilibrium with the local environment, this means that if
we can measure the strength of the thermal emission from the oxygen molecules,
then we can deduce the physical temperature of the molecules that produced
this emission. Microwaves also have an advantage over infrared temperature
sensing techniques in that they are only minimally affected by the presence
of clouds. In addition, the receiver and detector technology is very mature.
Second,
the oxygen absorption is strong enough that the effective distance that
emission is "seen" from is of the order of a few kilometers, depending
on the frequency and altitude, as is shown in the figure, which illustrates
the absorption in units of Nepers/km for altitudes
of 0, 10, and 20 km. On the ground the oxygen absorption is a single broad
feature because individual oxygen emission lines have been blended together
by what is known as pressure broadening. At 10 km (~33,000 feet, the approximate
altitude jet liners fly) individual absorption features (spectral
lines) are beginning to be seen as pressure broadening of the emission
lines becomes less important, and at 20 km (the approximate altitude the
highest research aircraft fly) many individual spectral lines can be seen.
Because
any oxygen emission is absorbed in proportion to its distance from the
instrument detecting it (a MTP in our case), nearby emission is absorbed
little and the absorption increases exponentially with distance (r)
in the viewing direction, which in the figure is inclined to the direction
of flight. The integrated emission can be represented by a "weighting function"
(W(r)) which characterizes the weighted mean distance of the emission.
In the special case where the temperature lapses at a constant rate with
distance, it is easy to show that the brightness temperature measured by
the radiometer is exactly equal to the physical temperature at the e-folding
(Ra)distance for the absorption. More complicated situations
require more detailed treatment.
Third, oxygen molecules are a well-mixed component of the atmosphere
at all heights, which is to say that the number of emitting molecules at
any height depends only on the pressure altitude. Water vapor, for example,
could not be used as a temperature surrogate because it's distribution
(or mixing ratio) compared to its surroundings varies
erratically. This characteristic is important because the strength of the
signal emitted by the oxygen molecules (which we want to use to determine
the temperature) depends not only on their temperature, but also on the
number of emitting molecules (as measured by the pressure), and to a much
lesser degree humidity. We must understand these dependencies to interpret
our measurements. It is clear from the figure above that absorption decreases
with altitude; the effect of temperature on absorption is opposite to pressure:
lower temperatures increase absorption. These two effects somewhat offset
each other, but pressure wins out and there is a net decrease of absorption
with altitude.
A Standard Atmosphere
The figure to the right shows how pressure and temperature vary with altitude
for a standard atmospheric model, which for the case shown represents average
conditions at mid-latitudes. Since pressure at any height is just the total
weight per unit area of all the air above, it decreases exponentially with
height (blue curve) and varies little from
the exponential shape shown. The behaviour of temperature (red
curve) is more complicated. It typically decreases with altitude
until the tropopause is reached. The rate of
temperature change with altitude (the lapse rate)
varies from about -10 K/km for dry air to about -5 K/km for moist air.
These large lapse rates are the result of convective mixing in the region
below the tropopause called the troposphere. This is where most weather
occurs. By contrast the region above the tropopause -- the stratosphere
-- is relatively benign as it is convectively stable. The temperature in
the stratosphere varies from being nearly constant (isothermal) at high
latitudes (the polar regions) to increasing slowly with altitude at low
latitudes (the tropics). The heat source in the stratosphere is the absorption
of ultraviolet radiation from the sun by ozone molecules.
The Airborne Operation of a MTP
An
MTP generally consists of two assemblies: a sensor unit (SU), which receives
and detects the signal, and a data unit (DU), which controls the SU and
records the data. In addition, on some platforms there may be a third element,
a real-time analysis computer (RAC), which analyzes the data to produce
temperature profiles and other data products in real time. The SU is shown
in the image to the right and the DU and RAC are shown in the image below.
The SU is connected to the DU with power, control, and data cables. In
addition the DU has interfaces to the aircraft navigation data bus and
the RAC, if one is present. Navigation data is needed so that information
such as altitude, pitch and roll are available for the MTP data analysis.
Aircraft altitude is needed to perform retrievals (which are altitude dependent),
while pitch and roll are needed for controlling the position of a stepper
motor which must drive a scanning mirror to predetermined elevation angles.
In
the SU image the scanning mirror is almost edge on and pointing slightly
upward and out of the page. The horn -- the equivalent of an antenna --
which receives the signal is perpendicular to the left side of the box
and faces the scanning mirror which is inclined at 45 degrees to the horn.
This mirror rotates about an axis through the horn which is nearly
normal to the flight direction (perpendicular to the side of the box with
JPL written on it) and allows viewing from near nadir (straight down) to
near zenith (straight up). At each viewing position a radio frequency signal
source, called a local oscillator (LO), is sequenced through two or more
frequencies and combined with the oxygen signal reflected off of the mirror
to produce a signal which is detected. A calibration target is the white
object behind the scanning mirror. The SU box dimensions are 14 cm W x
16 cm H x 24 cm D and it weighs ~5 kg. The DU and RAC are shown below in
a WB57F superpod. Their dimensions are 36 cm W x 10 cm H x 20 cm D and
25 cm W x 8 cm H x 35 cm D, respectively, and each unit weights ~5 kg.
Because each frequency has a different effective viewing distance, the
MTP is able to "see" to different distances by changing frequency. In addition,
because the viewing direction is also varied and because the atmospheric
opacity is temperature and pressure dependent, different effective viewing
distances are also achieved through scanning in elevation. For a two-frequency
radiometer with 10 elevation angles, each 15-second observing cycle produces
a set of 20 signals (called brightness temperatures), which are converted
through a data analysis procedure called a linear retrieval algorithm to
a profile of air temperature versus altitude, T(z).
Finally,
radiometric calibration is performed using the outside air temperature
(OAT) and a heated reference target to determine the instrument gain. However,
complete calibration of the system to include "window corrections" and
other effects, requires tedious analysis and comparison with radiosondes
near the aircraft flight path. This is probably the most important single
factor contributing to reliable calibration. For stable MTPs, such calibrations
appear to be reliable for many years. Such analysis is always performed
before MTP data are placed on mission archive computers. An example of
an altitude temperature profile (ATP) is shown to right in yellow. The
location of the tropopause is shown be the horizontal white dashed line;
the aircraft altitude is indicated by the short yellow horizontal line
at 35,000 feet (right scale), just below the tropopause. More information
on this and other MTP data products can be found here.
Current MTP Airborne Platforms
At the present time MTPs fly of three NASA research aircraft: the DC8,
ER2 and WB57 (with USAF). Click on the thumbnail image to see a full-sized
image of the aircraft.
Recent MTP Deployments
The table below summarizes six recent MTP deployments in the past 3 years.
The links in the Acronym column take you to the mission home page
where there is a discussion of the science objectives and results for each
of these missions.
| Platform |
Dates |
Acronym |
Full Mission Name |
| WB57 |
March-May 1998 |
WAM |
WB57 Aerosol Mission |
| DC8 |
October-November, 1997 |
SONEX |
SASS Ozone and Nitrogen Oxide Experiment |
| ER2 |
April - May 1997
June - July 1997
September 1997 |
POLARIS |
Photochemistry of Ozone Loss in the Arctic Region In Summer |
| DC8 |
April - May 1996 |
SUCCESS |
SUbsonic aircraft: Contrail & Cloud Effects Special Study |
| DC8 |
December 1995 - February 1996 |
TOTE/VOTE |
Tropical Ozone Transport Experiment and
Vortex Ozone Transport Experiment |
| ER2 |
May 1995
October - November 1995
January - February 1996 |
STRAT |
Stratospheric Tracers of Atmospheric Transport |
Related Links
Practical Guidelines
for Temperature Measurement
Glossary
Lapse Rate The rate at which temperature changes
with increasing altitude.
Mixing Ratio By volume, the number of molecules
of a particular specicies (e.g. water vapor) divided by the number of other
molecules in a given volume. By mass, the mass of a particular species
(e.g. water vapor) divided by the total mass of other molecules in a given
volume.
Nepers/km A unit of absorption defined such
that the reciprocal of this quantity is the e-folding distance for the
signal attenuation; that is, the distance at which the signal is reduced
by a factor of exp(-1), or 0.367 of the original signal strength. An absorption
of 3 Nepers/km would reduce a signal of strength 1 unit to 0.05 units (or
5%) in one km.
Tropopause Because of convective
mixing in the lower atmosphere (i.e., the troposphere) temperature
decreases with increasing altitude because as rising air parcels enter
regions of lower pressure, they expand and cool. This continues until a
temperature inversion (that is, a warmer temperature) is encountered. At
this point, rising air parcels are heavier than the surrounding air and
sink back down. Normally this inversion occurs at the tropopause. Above
the tropopause (in the stratosphere) the air becomes warmer because
it is being heated by the absorption of ultraviolet sunlight by ozone.
