A microwave radiometer (MWR) is a radiometer that measures energy emitted at millimetre-to-centimetre wavelengths (frequencies of 1–1000 GHz) known as microwaves. Microwave radiometers are very sensitive receivers designed to measure thermal electromagnetic radiation emitted by atmospheric gases. They are usually equipped with multiple receiving channels in order to derive the characteristic emission spectrum of the atmosphere or extraterrestrial objects. Microwave radiometers are utilized in a variety of environmental and engineering applications, including weather forecasting, climate monitoring, radio astronomy and radio propagation studies.
Using the microwave spectral range between 1 and 300 GHz provides complementary information to the visible and infrared spectral range. Most importantly, the atmosphere and also vegetation is semi-transparent in the microwave spectral range. This means its components like dry gases, water vapor, or hydrometeors interact with microwave radiation but overall even the cloudy atmosphere is not completely opaque in this frequency range.
For weather and climate monitoring, microwave radiometers are operated from space as well as from the ground. As remote sensing instruments, they are designed to operate continuously and autonomously often in combination with other atmospheric remote sensors like for example cloud radars and lidars. They allow to derive important meteorological quantities such as vertical temperature and humidity profile, columnar water vapor amount, or columnar liquid water path with a high temporal resolution in the order of seconds to minutes under nearly all weather conditions.
First developments of microwave radiometer were dedicated to the measurement of radiation of extraterrestrial origin in the 1930s and 1940s. The most common form of microwave radiometer was introduced by Robert Dicke in 1946 in the Radiation Laboratory of Massachusetts Institute of Technology to better determine the temperature of the microwave background radiation. This first radiometer worked at a wavelength 1.25 cm and was operated at the Massachusetts Institute of Technology. Dicke also first discovered weak atmospheric absorption in the MW using three different radiometers (at wavelengths of 1.0, 1.25 and 1.5 cm).
Soon after satellites were first used for observing the atmosphere, MW radiometers became part of their instrumentation. In 1962 the Mariner-2 mission was launched by NASA in order to investigate the surface of Venus including a radiometer for water vapor and temperature observations. In following years a wide variety of microwave radiometers were tested on satellites. The launch of the Scanning Multichannel Microwave Radiometer in 1978 became an important milestone in the history of radiometry. It was the first time a conically scanning radiometer was used in space; it was launched into space on board the NASA Nimbus satellite. The launch of this mission gave the opportunity to image the Earth at a constant angle of incidence that is important as surface emissivity is angular dependent. In the beginning of 1980, new multi-frequency, dual-polarization radiometric instruments were developed. Two spacecraft were launched which carried instruments of this type: Nimbus-7 and Seasat. The Nimbus-7 mission results allowed to globally monitor the state of ocean surface as well as surface covered by snow and glaciers. Today, microwave instruments like the Advanced Microwave Sounding Unit (AMSU) and the Special Sensor Microwave Imager / Sounder (SSMIS) are widely used on different satellites.
Ground-based radiometers for the determination of temperature profiles were first explored in the 1960s and have since improved in terms of reduced noise and the ability to run unattended 24/7 within worldwide observational networks. Review articles, and a detailed online handbook  are available.
Principle of operation
Solids, liquids (e.g. the Earth's surface, ocean, sea ice, snow, vegetation) but also gases emit and absorb microwave radiation. Traditionally, the amount of radiation a microwave radiometer receives is expressed as the equivalent blackbody temperature also called brightness temperature. In the microwave range several atmospheric gases exhibit rotational lines. They provide specific absorption features shown at a figure on the right which allow to derive information about their abundance and vertical structure. Examples for such absorption features are the oxygen absorption complex (caused by magnetic dipole transitions) around 60 GHz which is used to derive temperature profiles or the water vapor absorption line around 22.235 GHz (dipole rotational transition) which is used to observe the vertical profile of humidity. Other significant absorption lines are found at 118.75 GHz (oxygen absorption) and at 183.31 GHz (water vapor absorption, used for water vapor profiling under dry conditions or from satellites). Weak absorption features due to ozone are also used for stratospheric ozone density and temperature profiling.
Besides the distinct absorption features of molecular transition lines, there are also non-resonant contributions by hydrometeors (liquid drops and frozen particles). Liquid water emission increases with frequency, hence, measuring at two frequencies, typically one close to the water absorption line (22.235 GHz) and one in the nearby window region (typically 31 GHz) dominated by liquid absorption provides information on both the columnar amount of water vapor and the columnar amount of liquid water separately (two-channel radiometer). The so-called „water vapor continuum“ arises from the contribution of far away water vapor lines.
Larger rain drops as well as larger frozen hydrometeors (snow, graupel, hail) also scatter microwave radiation especially at higher frequencies (>90 GHz). These scattering effects can be used to distinguish between rain and cloud water content exploiting polarized measurements  but also to constrain the columnar amount of snow and ice particles from space  and from the ground.
A microwave radiometer consists of an antenna system, microwave radio-frequency components (front-end) and a back-end for signal processing at intermediate frequencies. The atmospheric signal is very weak and the signal needs to be amplified by around 80 dB. Therefore, heterodyne techniques are often used to convert the signal down to lower frequencies that allow the use of commercial amplifiers and signal processing. Increasingly low noise amplifiers become available at higher frequencies, i.e. up to 100 GHz, making heteorodyne techniques obsolete. Thermal stabilization is highly important to avoid receiver drifts.
Usually ground-based radiometers are also equipped with environmental sensors (rain, temperature, humidity) and GPS receivers (time and location reference). The antenna itself often measures through a window made of foam which is transparent in the microwave spectrum in order to keep the antenna clean of dust, liquid water and ice. Often, also a heated blower system is attached the radiometer which helps to keep the window free of liquid drops or dew (strong emitters in the MW) but also free of ice and snow.
As it seen from the figure attached after being received at the antenna the radiofrequency signal is downconverted to the intermediate frequency with the help of a stable local oscillator signal. After amplification with a Low Noise Amplifier and band pass filtering the signal can be detected in full power mode, by splitting or splitting it into multiple frequency bands with a spectrometer. For high-frequency calibrations a Dicke switch is used here.
The calibration of microwave radiometer sets the basis for accurate measured brightness temperatures and therefore, for accurate retrieved atmospheric parameters as temperature profiles, integrated water vapor and liquid water path. The simplest version of a calibration is a so-called „hot-cold“ calibration using two reference blackbodies at known, but different, „hot“ and „cold“ temperatures, i.e. assuming a linear relation between input power and output voltage of the detector. Knowing the physical temperatures of the references, their brightness temperatures can be calculated and directly related to detected voltages of the radiometer, hence, the linear relationship between brightness temperatures and voltages can be obtained.
The temperatures of the calibration targets should be chosen such that they span the full measurement range. Ground-based radiometers usually use an ambient temperature target as „hot“ reference. As a cold target one can use either a liquid nitrogen cooled blackbody (77 K) or a zenith clear sky TB that was obtained indirectly from radiative transfer theory. Satellites use a heated target as „hot“ reference and the cosmic background radiation as „cold“ reference. To increase the accuracy and stability of MWR calibrations further calibration targets, such as internal noise sources, or Dicke switches can be used.
Retrieval of temperature and water vapor profiles
The retrieval of physical quantities using microwave radiometry (e.g. temperature or water vapor profiles) is not straightforward and comprehensive retrieval algorithms (using inversion techniques like optimal estimation approach) have been developed.
Temperature profiles are obtained by measuring along the oxygen absorption complex at 60 GHz. The emission at any altitude is proportional to the temperature and density of oxygen. As oxygen is homogeneously distributed within the atmosphere and around the globe, the brightness temperature signals can be used to derive the temperature profile. Signals at the center of the absorption complex are dominated by the atmosphere closest to the radiometer (when ground-based). Moving into the window region, the signal is a superposition from close and far regions of the atmosphere. The combination of several channels contains therefore information about the vertical temperature distribution. A similar approach is used to derive vertical profiles of water vapor utilizing its absorption line at 22.235 GHz.
Weather satellite instrumentation
Microwave instruments are flown on several polar orbiting satellites for Earth observation and operational meteorology as well as part of extraterrestrial missions. One distinguishes between imaging instruments that are used with conical scanning for remote sensing of the Earth surface, e.g. AMSR, SSMI, WINDSAT, and sounding instruments that are operated in cross-track mode, e.g. AMSU/MHS. The first type uses lower frequencies (1–100 GHz) in atmospheric windows to observe sea-surface salinity, soil moisture, sea-surface temperature, wind speed over ocean, precipitation and snow. The second type is used to measure along absorption lines to retrieve temperature and humidity profile. Furthermore, limb sounders, e.g., MLS, are used to retrieve trace gas profiles in the upper atmosphere.
Other examples of microwave radiometers on meteorological satellites include the Special Sensor Microwave/Imager, Scanning Multichannel Microwave Radiometer, WindSat, Microwave Sounding Unit and Microwave Humidity Sounder. The Microwave Imaging Radiometer with Aperture Synthesis is an interferometer/imaging radiometer capable of resolving soil moisture and salinity over small regions of surface.
By the 2010s four microwave radiometers have been flown on interplanetary spacecraft. The first was Mariner 2, which used a microwave instrument to determine the high surface temperature of Venus was coming from the surface not higher up in the atmosphere. There are/were also radiometers on the Juno Jupiter probe, the Rosetta comet probe, and Cassini-Huygens.
The Juno probe, launched in 2011, is characterizing the atmosphere of Jupiter using a microwave radiometer suite. The Microwave Radiometer (MWR) instrument on Juno has several antennas observing in several different microwave wavelengths to penetrate the top cloud layer of the planet, and detect features, temperatures, and chemical abundances there.
Microwave Radiometer Ground-Based Networks
MWRnet is a network established in 2009 of scientists working with ground-based microwave radiometers. MWRnet aims to facilitate the exchange of information in the MWR user community fostering the participation to coordinated international projects. In the long run, MWRnet’s mission aims at setting up operational software, quality control procedures, data formats, etc. similar to other successful networks such as EARLINET, AERONET, CWINDE.
- Microwave Remote Sensing—Active and Passive”. By F. T. Ulaby. R. K. Moore and A. K. Fung. (Reading, Massachusetts: Addison-Wesley, 1981 and 1982.) Volume I: Microwave Remote Sensing Fundamentals and Radiometry.
- Thermal Microwave Radiation: Applications for Remote Sensing, C. Matzler, 2006, The Institution of Engineering and Technology, London, Chapter 1.
- Westwater, Edgeworth Rupert, 1970: Ground-Based Determination of Temperature Profiles by Microwaves. PH.D. Thesis, UNIVERSITY OF COLORADO AT BOULDER, Source: Dissertation Abstracts International, Volume: 32-02, Section: B, page: 1134.
- Dicke, R.H. (1946). "The measurement of thermal radiation at microwave frequencies" (PDF). Review of Scientific Instruments. AIP. 17 (7): 268–275. Bibcode:1946RScI...17..268D. doi:10.1063/1.1770483. PMID 20991753. Archived from the original (PDF) on 2012-04-03.
- Passive Microwave Remote Sensing of the Earth, Physical Foundations, Eugene A. Sharkov, Springer-Praxis Books in Geophysical Sciences, Chapter 14: Passive microwave space missions
- "MWRnet - An International Network of Ground-based Microwave Radiometers". cetemps.aquila.infn.it.
- Westwater, E.R., C. Mätzler, S. Crewell (2004) A review of surface-based microwave and millimeter-wave radiometric remote sensing of the troposphere. Radio Science Bulletin, No. 3010, September 2004, 59–80
- Westwater, E. R., S. Crewell, C. Mätzler, and D. Cimini, 2006: Principles of Surface-based Microwave and Millimeter wave Radiometric Remote Sensing of the Troposphere, Quaderni Della Societa Italiana di Elettromagnetismo, 1(3), September 2005, 50–90.
- "Microwave radiometer - EG-CLIMET". cfa.aquila.infn.it.
- Czekala et al. (2001), Discrimination of cloud and rain liquid water path by groundbased polarized microwave radiometry, Geophy. Res. Lett., DOI: 10.1029/2000GL012247
- Bennartz, R., and P. Bauer (2003), Sensitivity of microwave radiances at 85–183 GHz to precipitating ice particles, Radio Sci., 38(4), 8075, doi:10.1029/2002RS002626.
- Kneifel et al. (2010), Snow scattering signals in ground-based passive microwave radiometer measurements, J. Geophys. Res., DOI: 10.1029/2010JD013856
- "Juno at Jupiter: The Juno microwave radiometer (MWR) - IEEE Conference Publication". ieeexplore.ieee.org.
- "Archived copy". Archived from the original on 2016-11-30. Retrieved 2017-02-03.CS1 maint: Archived copy as title (link)
- "Science Overview". Jet Propulsion Laboratory. Retrieved: 21 December 2016.