Spectrum Management for Science in the 21st Century (2010)

Natural radio emissions from objects as diverse as hurricanes and distant galaxies are employed by scientists to study the objects themselves. This information is vital for analyzing and forecasting weather and climate and for understanding the distant cosmos. The geophysical studies use remote sensing satellites and ground-based instruments, and the radio astronomy observations employ large, ground-based radio telescopes, or antennas. The techniques used by the two groups are fundamentally similar. The variations in the strength and polarization of the radio signals with direction, frequency, and time are measured with receivers of ever-increasing sensitivity and sophistication. The detailed techniques must be different because satellites pass quickly over any given spot on Earth, whereas a radio telescope can track a given source in the sky for hours. The two groups of users, working within the Earth Exploration-Satellite Service (EESS) and the Radio Astronomy Service (RAS), use many of the same frequency bands, and they have many interference problems in common.

The launch of the first U.S. Earth remote sensing satellite, TIROS-1, in 1960, ushered in an era of unprecedented scientific understanding of the planet as a complex system of systems. For the first time humanity was provided the opportunity to visualize and understand the interactions of many of Earth’s constituent processes. This global view was only made possible by the development of advanced sensors that were able to take advantage of the new perspective offered by satellites. Chief among the new classes of sensors used in the nearly five decades since TIROS has been the passive microwave radiometer, which holds the unique advantage over optical and infrared systems of being able to probe through clouds. These sensors

are passive in that they do not transmit any signals or communications; they only receive naturally occurring signals. Passive instruments are thus much different from active devices, with which people are most familiar. Active devices include cellular telephones, wireless Internet networks, garage-door openers—anything that emits a signal, whether purposefully or not. Owing to the unique purposes for which passive instruments are used, these instruments also have unique designs and needs, which are discussed in detail in this report. These passive devices are used by Earth scientists for economically and scientifically important observations of Earth’s environment and by astronomers to observe the vast reaches of the cosmos beyond this planet.

The use of passive microwave sensors allows rainfall, clouds, ocean surface winds, temperature, and moisture distributions—the primary variables of meteorology—to be quantified over the globe, under clear and cloudy conditions and during both day and night. Data from passive microwave sensors are now a vital component in the complex calculations used for weather prediction.

In addition to enabling new understanding of atmospheric processes, passive microwave observations¹ have brought about a new understanding of Earth’s surface processes. The distinct microwave signatures produced by water in its various phases (liquid, ice, vapor) permit all-weather measurements of environmental variables such as snowpack depth, soil moisture, sea ice extent, sea surface temperature, and sea surface salinity. These and other related passive microwave observables, including biomass and vegetation water content, are becoming increasingly important as drivers of industry and agriculture, particularly as global resources of freshwater, arable land, and fisheries are further stretched to satisfy an increasingly large and demanding global population. The role played by passive microwave observations from space as well as from surface-based and airborne platforms in the management of these resources and the understanding of their interactions with other natural systems cannot be overstated. Currently, 21 satellites carrying passive microwave sensors are orbiting Earth.

During roughly the same era in which Earth remote sensing was developed—but preceding it by about a decade—passive radio observations were used to study the makeup of the cosmos under what is now the discipline of radio astronomy. Radio astronomy is a young science, about 60 years old, but it has contributed enormously

to the understanding of the universe. It provided the first view of the cosmos outside the optical band and revealed an extraordinary variety of remarkable objects and phenomena that had never been suspected, including pulsars (spinning condensed stars acting as rotating radio beacons), the cosmic microwave background radiation (showing the universe to have started in an initial explosion—the Big Bang—and to have been expanding ever since), gigantic molecular clouds where new stars are being born, and active galactic nuclei (in which reside giant black holes fed by a disk of gas and dust and from which emanates an enormously energetic pair of jets, going far across intergalactic space). In the near future, radio astronomical observations will provide insights about the events that occurred around the time that the first stars were born—known as the epoch of reionization.

Radio sciences have a strong practical importance. Accurate weather forecasting is vital for many activities critical to human health, welfare, and security—including agriculture, transportation, military defense, and the mitigation of damage from extreme weather events such as hurricanes, wildfire, and drought. These applications and others are discussed in detail in Chapter 2 of this report. The long-term monitoring of Earth is vital for climate assessment and prediction and is a major aid in the understanding of climate change—one of the most important problems currently facing humanity. The proper management of the environment both now and for decades to come will be contingent on remotely sensed data.

On a larger scale, radio astronomy opened people’s view of the cosmos to take in the violent, enormously complex universe that it is now known to be. The shift in thought engendered by radio astronomy is analogous to that caused 500 years ago when it was first recognized that Earth orbits the Sun: the notion of Earth’s special location in the universe disappeared, and people began seeing their world as an element of the cosmos rather than as its center. But radio astronomy also has practical applications in technology that support today’s development of our information infrastructure, as discussed in Chapter 3. These include technical developments in high-performance antennas, sensitive radio receivers, electronics, computing, signal processing, and scientific education.

Unfortunately, human-made radio frequency interference (RFI) can make the radio science measurements more difficult, and in some cases it can render them unusable. This problem is introduced in §1.3 and §1.4 below and discussed in depth in Chapter 4.

Radio astronomy and passive Earth remote sensing both rely on detecting, recording, and interpreting weak natural radio frequency emissions. These emissions are radiated by all absorptive bodies: for example, forests, clouds, the Sun, and galaxies. The detailed features of the radiation provide information on the

temperature, density, composition, motion, and other characteristics of the object or medium being observed. Earth studies and astronomical studies cover most of the radio spectrum, ranging from about 15 MHz (the lower limit for the radio transparency of Earth’s ionosphere), to the current limits of radio technology at many hundreds of gigahertz. The highest radio frequencies (at 1 THz and above) merge with infrared radiation, and some studies require continuous measurements from the radio into the infrared bands, and even on to optical bands or beyond.

Natural radio emissions are generated by a variety of mechanisms. All matter emits “thermal noise,” with a characteristic frequency spectrum that depends on its temperature. While hot objects, such as stars, emit mainly in the infrared, optical, and ultraviolet portions of the electromagnetic spectrum, cold gas, dust between the stars, and materials on Earth such as water, soil, and atmospheric gases with smaller temperatures (of a few hundred kelvin and below) emit mostly in the radio wave and submillimeter-wave portions of the spectrum (see Figure 1.1).

Radio radiation is also emitted from atoms and molecules when they move from one quantum state to another. This line radiation appears at characteristic

FIGURE 1.1 The characteristics of the electromagnetic spectrum. See Figures 1.2 and 1.3 in this chapter for a more detailed picture of the atmospheric penetration of electromagnetic waves. Image from NASA, Science Program for NASA’s Astronomy and Physics Division, Washington, D.C., 2006.

frequencies determined by the particular quantum transition of the atomic or molecular species in question (Figures 1.2 and 1.3). The measurement of radiation at and near these transition frequencies is extremely important for both Earth science and radio astronomy. In Earth remote sensing, line radiation spectra can be used to obtain temperature and humidity profiles in the atmosphere from the surface of Earth on up into the mesosphere. Observations for such profiling purposes occur near the centers of atmospheric absorption lines and within the immediately adjacent “wings” on either side of the line centers. In radio astronomy, a proper interpretation of line radiation provides information on the composition, density, and temperature of the material under study. Radio astronomers are interested in frequency bands where an interesting atomic or molecular transition occurs and where Earth’s atmosphere is particularly transparent. Spectra from these bands are often used to derive the motions in cosmic clouds, or in galaxies. Because different molecules—for example CO and HCN—have different excitation conditions, the study of several molecules (or several lines from one molecule) can give three-dimensional information about the cloud. This is analogous to the way that atmospheric profiles are found in Earth science measurements.

FIGURE 1.2 The opacity of Earth’s atmosphere in the radio range of frequencies from 1 to 1000 GHz for six scenarios. Image courtesy of A.J. Gasiewski, University of Colorado.

FIGURE 1.3 The transmission spectrum of Earth’s atmosphere in the radio range of frequencies from 1 to 1000 GHz at Mauna Kea, Hawaii, a very dry site at an altitude of approximately 14,000 ft. Such high, dry sites are drier than Scenario E as given in Figure 1.2, making them suitable for astronomical observations above 200 GHz. Note that the water vapor line at 22 GHz (see Figure 1.2) causes negligible loss in transmission at this site, but the lines at 556 and 752 GHz are so strong—even on the high mountaintop—that the atmospheric transmission is essentially zero, and no astronomical observations can be made from 520 to 580 GHz and 730 to 780 GHz. Image courtesy of L. Ziurys, University of Arizona.

For many nongaseous materials on Earth (such as liquid water, ice, soil, snow, and vegetation), the radiation is broadened by the strong interaction of closely spaced molecules into a continuum that exhibits a slow spectral variation over a wide range of frequencies. Continuum radiation spectra can also occur when the scale of surface roughness or feature size (i.e., raindrop or cloud-particle diameter) is comparable to or smaller than a wavelength of the radiation. In Earth remote sensing, continuum radiation spectra are measured at a variety of microwave frequencies. Optimal frequencies for measuring continuum radiation are between the major transition frequencies for oxygen and water vapor (see Figures 1.2 and 1.3). In these “windows,” the ability to probe through the atmosphere is maximized, thus making the continuum radiation easy to observe. Similar frequencies are used in radio astronomy for continuum measurements.

The Doppler effect, in which motion of the emitter toward or away from the observer shifts the received frequency, provides a means of determining the motion of the material. Doppler-shifted radiation enables the measurement of the rotation of matter in spiral galaxies and that of the motion of air in the upper atmosphere. The expansion of the universe leads to a similar shift in the frequencies of spectral

lines that increase as the distance to the source increases. This effect, known as the cosmological redshift, allows distances to faraway galaxies to be accurately measured. (See Box 3.2, “Redshift,” in Chapter 3 of this report.)

Another widespread emission mechanism is synchrotron radiation, generated by the acceleration of electrons in a magnetized plasma. Our Milky Way Galaxy is suffused with a dilute plasma that emits synchrotron radiation at frequencies of about 1 GHz and below. Over a much wider frequency range, this radiation is also associated with some of the most powerful events in the universe: pulsars, supernovae, gamma-ray bursts, and quasars, in which matter falling into a giant black hole radiates a prodigious amount of radiation; and radio galaxies, in which jets and giant cocoons of plasma ejected from a galaxy nucleus extend well outside the host galaxy. Synchrotron and thermal radiation are emitted across broad frequency bands and with a characteristic spectral shape. Their measurement is often not restricted to any one particular frequency, although when the band shape needs to be defined, many samples of the spectrum at well-separated frequencies are needed. The spectral lines from quantum transitions, however, must be measured at their specific natural frequencies with allowance for Doppler shifts.

With the threat of climate change and related environmental changes over the next several decades, the need for environmental information for critical policy and economic decision making will increase. The ability of passive microwave Earth remote sensing to study water in various phases, at both continuum and spectral line frequencies, means that these instruments will be increasingly used to provide key information. Whether obtained for use in day-to-day weather forecasting operations or for long-term climate studies, passive microwave measurements of Earth represent one of the most important scientific uses of the radio spectrum.

A number of contemporary problems in physics also require radio astronomers’ continued ability to observe the cosmos. For instance, studies at radio frequencies provide the only way to investigate the epoch of reionization that occurred when ultraviolet radiation from the first stars ionized intergalactic space, bringing the universe to the state in which it exists today. Radio astronomy also provides the only way to study large numbers of pulsars to determine if Einstein’s theory of general relativity is actually correct in the universe’s most extreme conditions.

Finally, the use of passive radio techniques to observe the Sun provides the prospect of monitoring our own star for subtle changes in emission characteristics that may lead to geomagnetic disturbances. Such disturbances regularly affect the operation of satellite communications, navigation systems, and terrestrial power grids. Just as Earth environmental information is expected to grow in societal importance in the decades ahead, so is space environmental information.

Because the total radio spectrum is a limited resource, competition exists among its various users. In the United States, spectrum use is regulated by the National Telecommunications and Information Administration (NTIA) for federal government users and by the Federal Communications Commission (FCC) for all others. The regulation of spectrum use includes the assignment of frequency bands, the specification of maximum allowed power levels, and other specific conditions (“footnotes”) regarding potential interference with other users. The regulations identify the uses as “services” (see §4.1 in Chapter 4), and this report focuses primarily on the Radio Astronomy Service and the Earth Exploration-Satellite Service.

The U.S. regulatory system, as well as the International Telecommunication Union (ITU) system, was organized prior to 1950, when there were far fewer uses for the radio spectrum than there are now.² As new technologies have been developed, the FCC has allocated new bands for them. During the past two decades, however, the pace of development in radio communications has begun to strain the regulatory system, with the biggest problem being a lack of unallocated spectrum available for new technologies.

There are two fundamentally different categories of spectrum users. One category consists of active users—those who transmit radio signals to achieve their ends, which may be voice or data communications, radar surveillance, or even Earth remote sensing using radars or other transmitters. As a group, active users need ever-increasing amounts of spectrum for the ever-increasing uses that are invented; telecommunications companies (in particular) pay large sums of money to obtain the rights to use it. The other category consists of passive users, such as those in radio astronomy and passive remote sensing, who operate in receive-only mode. These users also need increasing amounts of spectrum to obtain the increased sensitivity required for new studies and services.

The uses and desires of these two communities of spectrum users are asymmetric, because the passive services do not transmit any radio signals. Accordingly, active users can interfere with passive users but not vice versa. Since the passive services can operate in any spectral band, they can face radio frequency interference from active services over much of the spectrum. This can include (at times) interference in the bands allocated on an exclusive, primary basis to the passive services. Such interference is discussed in Chapters 2, 3, and 4. See Box 1.1 for more information.

To be more specific, three types of unwanted emissions that cause interference are defined in the FCC regulations: “spurious” emission, “out-of-band” emission, and emissions in adjacent channels. See the glossary in this report for greater detail. Loosely speaking, spurious emissions come from a transmitter emitting at frequencies outside its assigned band and are caused, for example, by nonlinearities that generate harmonics. Out-of-band emissions are emissions at neighboring frequencies that are spread into adjacent bands by the modulation process. Both can interfere with radio astronomy and Earth remote sensing observations. The generation of small spurious and out-of-band signals is virtually inevitable due to the technological limitations in transmitter electronics, but the actual levels emitted can be controlled at the transmitter and kept to

within allowable limits. Emission in adjacent channels can create a “blocking” interference within a receiver. However, this particular occurrence is a result of the technical limitations of the receiver, not the transmitter. The allowed emission levels within these categories are defined by regulatory agencies (the FCC and the NTIA) and through international treaties, as discussed in Chapter 4. In the vast majority of cases, both spurious and out-of-band RFI is inadvertent, that is, unintentional. Nonetheless, such emissions are prohibited if they rise above the allowed level in a protected band. However, in cases where Earth remote sensing or radio astronomy observations must be made in bands where no primary allocation for these uses exists, there is no recourse with respect to the problems and data outages caused by RFI.

Finding: Owing to their receive-only nature, the passive Earth Exploration-Satellite Service and Radio Astronomy Service, operating from 10 MHz to 3 THz, are incapable of interfering with other services.

Users of the RAS and EESS go to considerable effort to mitigate the effects of RFI. Such effort includes careful attention to the design of the receivers to block authorized transmissions that are nearby (both in terms of geography and in terms of frequency), excision techniques (in time and frequency) to eliminate unwanted signals, and, now in development, advanced processing techniques to recognize RFI and either excise it or subtract it. These “unilateral” techniques are all expensive to do on a regular basis. Furthermore, there are fundamental limitations on their ability to distinguish natural thermal noise (the desired signal) from an efficiently modulated communications or radar signal (the RFI) in which power is uniformly distributed across the allocated band.

Cooperative interference mitigation involves cooperative use of spectrum, with RAS and EESS users coordinating their observations to take advantage of the large amounts of unused spectrum at any time and location. Cooperative mitigation techniques hold great promise, but are untested and would require new spectrum use policies and practices so that these techniques could develop. Both unilateral interface mitigation and cooperative interference mitigation are discussed in Chapter 4.

Another major mitigation cost is incurred up front when an observatory is located in a remote area to escape RFI. Current interest in locating receiver arrays to the far side of the Moon is perhaps the most extreme example of this type of cost. While this strategy can be useful for radio astronomy, Earth remote sensing satellites observe the entire Earth over the course of each day and therefore could not avail themselves of a similar advantage. See Box 1.2 for more information.

The goal of this report is to highlight the importance of the passive uses of the radio spectrum, to identify issues that threaten the ability of the science services to provide benefits to society, and to recommend steps for the mitigation or elimination of these threats while recognizing the importance of the other services. Chapters 2 and 3 discuss the knowledge gained from and benefits to society produced by the EESS and RAS, respectively, as well as current and future spectrum requirements for maintaining progress. Chapter 4 discusses current trends in spectrum use and technology that shape the environment in which the EESS and RAS operate, as

well as methods for mitigating the impact of interference. Finally, Chapter 5 provides the committee’s recommendations for continuing to enable passive scientific uses of the radio spectrum. The committee’s findings are presented throughout the report and in Chapter 5, and an acronym list and glossary of terms are provided in the appendixes, in addition to other useful supplementary material.