Feasibility of Space-Based Monitoring for Governance ofSolar Radiation Management
Preliminary analysis at Aerospace suggests that detecting tests of particle injection schemes from space will be quite challenging, especially for unannounced small-scale, localized trials with short- term observable effects.
Climate researchers and theorists have suggested that industrialized nations could combat global warming by injecting aerosols—sulfur dioxide, aluminum oxide, or manufactured nanostructured particles—into the stratosphere to actively manage the amount of solar radiation that filters through. Proposed means for lofting these aerosols into the stratosphere include large-caliber guns, rockets, balloons, tethered hoses, aircraft, and even via photophoresis (the process whereby small particles suspended in gas or liquid move away from a sufficiently intense light source).
As promising as this might appear at first glance, there are many potential downsides. The influence of aerosols and clouds is the largest source of uncertainty in climate models and forecasts, and the uncertainties and risks involved in particle injection are significant. Ideally, any experimentation with solar radiation management would be based on a global consensus regarding what strategy to pursue and how to pursue it. In reality, a single state might unilaterally attempt action. One reasonable fear is that a country may begin experimenting with solar radiation management, even at the risk of adversely affecting neighboring nations or the planet as a whole.
The Aerospace Corporation has been investigating how space-based sensors could be used to detect and track injected particles to help enforce any future international agreements concerning climate engineering and solar radiation management. Initial findings suggest that any effort to do so will be difficult. (see sidebar, Aerospace Support to Global Climate Research)
A Postulated Scenario
A country or organization developing a particle-injection capability would need to conduct extensive developmental testing. Initial tests with natural particles like sulfur dioxide or aluminum oxide might be used to evaluate various injection methods and to quantify factors such as particle aggregation, dispersion, and persistence. These early experiments would lead to larger experiments and eventually to a full-scale test deploying a huge quantity of particles. Such a full-scale test would be easy to detect—but by then, it would be too late to do anything about it. The ability to detect small precursor tests would provide the international community with more options for intervening or possibly deterring unsanctioned activities altogether.
Developing a system to detect such activities presents an enormous challenge because of the wide range of unknowns—for example, the type of material released, particle size, amount released, altitude of release, release mechanism, and area of dispersion (initial density). In addition, the process of dispersion itself is highly variable: estimates of eddy diffusivity in the stratosphere can vary by more than an order of magnitude. Requirements for data access and dissemination, redundant verification, reliability, and operational control would be similar to current systems for monitoring arms control agreements. Geopolitical constraints and possible funding mechanisms would also be important considerations.
To determine the requirements for a space-based monitoring capability, Aerospace considered, as a typical scenario, a small clandestine test involving an aircraft release of 1 to 10 metric tons of precursor gases or engineered particles. Detecting this type of unannounced test (which could be conducted anywhere in the globe) would require nearly continuous global monitoring. The aerosol cloud would not stay together more than a few hours at detectable levels, and the detection threshold would vary greatly depending on background and sensor technique. The maximum size of the cloud at detectable levels might be on the order of a few kilometers. The high wind speeds and shear prevalent in the stratosphere would transport the cloud hundreds of kilo- meters downwind while shredding it into filaments. As a rough quantitative example, 1 metric ton of sulfur released over an initial volume of 107 cubic meters is estimated to have a mean particle density of 1000 particles per cubic centimeter after 1 hour and 100 particles per cubic centimeter after 10 hours. This calculation assumes an eddy diffusivity value of 100 square meters per second horizontally and 0.1 square meters per second vertically. These concentration values would change greatly depending on the parameters and the modeling technique assumed.
The purpose of the clandestine tests would be to assess injection techniques and better understand the effectiveness of the particles in changing the albedo. Objectives would include demonstration of the delivery mechanism, observation of aerosol formation and growth rates, observation of particle dispersion characteristics, observation of vertical spreading and motion, observation of evolving particle size distribution and location, observation of particle attitude (where relevant), measurement of albedo levels, and validation of associated models.
In terms of space sensor requirements, these goals translate to an ability to quantify aerosol optical depth or extinction coefficients in the stratosphere as a function of wavelength, enabling an estimate of particle density and size distribution. Spectral information would also be used to discern particle composition. Specialized algorithms would have to be developed (most likely from ground-based reference test data) to differentiate particle shapes, orientations, and makeup. Quite a bit of uncertainty surrounds the derivation of these parameters from the directly observed radiance and backscatter measurements, so a significant research program would be needed to substantiate the baseline science and establish confidence in the retrieval methodologies.
The ability to accurately determine the altitude of an aerosol layer would be critical for determining its origin—but not sufficient in itself. For example, with the exception of volcanic aerosols and some phenomena associated with specific polar regions and seasons, natural clouds generally do not extend into the stratosphere. Thus, an aerosol cloud in the stratosphere could be a good indication of human intervention. However, at higher latitudes, jet aircraft do fly above the tropopause—so in these regions, it may be difficult to distinguish normal jet contrails and cirrus clouds from a particle injection test. Also, because observed instantaneous aerosol optical depth values can change by a factor of two or more from day to day, only very large spikes in sensor measurements would indicate intentional particle injections.
The required sensor revisit rate, spatial resolution, and measurement accuracy all depend upon the type of sensor, the dispersal rate, and other characteristics of the aerosol tests, especially during the first minutes to hours after injection. Other critical parameters to monitor (in addition to ambient conditions) are particle size distribution and spatial distribution as the plume spreads out.
The types of space-based sensors that would be most effective in detecting intentionally injected aerosols are passive multispectral imagers, both reflective and emissive, and active laser-based sensors or lidars. These two types of sensors have complementary advantages and deficiencies and would need to be used in combination to be most effective.
For sensors with nadir-viewing geometry, such as NASA’s Moderate Resolution Imaging Spectroradiometer (MODIS), the combination of background clutter and relatively short column depth makes it difficult to detect and characterize aerosol concentrations with low optical depths (i.e., less than or equal to 0.1–0.3). Even thin high cirrus clouds, consisting of rather large ice crystals, are difficult to detect or measure with these instruments.
Solar occultation sensors (which view the atmosphere tangentially against the backdrop of the sun as it sets or rises) are significantly more sensitive to small aerosol concentrations as a result of very long viewing paths and related factors; however, viewing is limited to times and regions correlating to occultation events, resulting in spotty coverage for any given orbital pass. In addition, the sensing geometry results in poor horizontal resolution and geolocation capability.
Active lidar sensors, such as the Cloud-Aerosol Lidar with Orthogonal Polarization instrument (CALIOP) on NASA’s Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) spacecraft, can detect aerosol layers with higher sensitivity than the nadir-looking passive sensors and provide accurate aerosol heights and horizontal positions. In particular, the low background density in the stratosphere (less than 10 particles per cubic centimeter at 20 kilometers) means that even fairly diffuse particles can be detected with lidar.
One of the challenges in detecting injection tests lies in distinguishing intentionally injected particles from naturally occurring particles. There may be some peculiarities with regard to spectral region, polarization, or geometrical behavior that would allow for their differentiation. For instance, nonspherical particles tend to depolarize the scattered photons from a polarized light source. Thus, if the scattered signals are resolved polarimetrically, lidar sensors can provide some information regarding the shape of the aerosols.
The main disadvantages of using lidar sensors are the small field-of-view and relatively high-power requirements. For example, CALIOP’s ground swath is only 100 meters wide, resulting in a 16-day revisit time—far too long for a single spacecraft to accomplish an effective monitoring mission.
Increasing the footprint of an orbiting lidar sensor would probably entail an increase in laser power, allowing the beam to be either spread out or split into multiple spots while maintaining sufficient power density for high sensitivity. While high-power solid-state and fiber lasers have been demonstrated on the ground, considerable development will be required to qualify any of these to meet the challenging requirements for use in space.
In light of these considerations, an appropriate suite of sensors might include visible and thermal multispectral imagers; a long-wave (5–12 micron) hyperspectral imager for chemical resolution and detection; a passive solar-occultation spectrograph; and a multiwavelength, polarization-sensitive, wide-swath (about 10 × 0.5 kilometer) lidar system.
Detection of a particle injection test would require extensive analysis of the temporally and spatially colocated passive multispectral sensor data and lidar data. However, even with advanced spacecraft-based sensor systems, detection of the small tests would be difficult, given the background noise and infrequent revisit rate of a single spacecraft. A large constellation of spacecraft would reduce the revisit time, but the huge cost of such a system weighed against the benefit of quickly detecting a small particle-injection test would make it impractical. Given the high level of uncertainty and the lack of a background reference data set, the job of detecting, identifying, and monitoring such tests for treaty purposes will need to be shared and cross-checked by numerous assets. (see sidebar, Pros and Cons of Solar Radiation Management)
Deterrence of unsanctioned and clandestine solar radiation management activities will require monitoring systems that can reliably detect early test phases involving relatively small amounts of particles. Doing so from space will be very challenging. Indeed, future treaty negotiations may need to consider alternative methods of monitoring such activities. As with monitoring nuclear tests, detecting clandestine particle-injection experiments and development activities will require a combination of techniques on the ground and in space. Still, given the strong need for improved understanding of the role of aerosols in the stratosphere, as well as the need to monitor volcano dust for airline safety, the impetus may exist for the development of a multifunction system of space-based sensors.
The authors would like to thank Carl Rice and Richard Walterscheid for their helpful advice related to this article.
An earlier version of this article appeared in the proceedings of the American Institute of Aeronautics and Astronautics Space 2010 Conference and Exposition.