SCoPEx: Stratospheric Controlled Perturbation Experiment

SCoPEx is a scientific experiment to advance understanding of stratospheric aerosols that could be relevant to solar geoengineering. It aims to improve the fidelity of simulations (computer models) of solar geoengineering by providing modelers with experimental results vital to addressing specific science questions. Such simulations are the primary tool for estimating the risks and benefits of solar geoengineering, but current limitations may make the simulations look too good. SCoPEx will make quantitative measurements of aspects of the aerosol microphysics and atmospheric chemistry that are currently highly uncertain in the simulations. It is not a test of solar geoengineering per se. Instead, it will observe how particles interact with one another, with the background stratospheric air, and with solar and infrared radiation. Improved understanding of these processes will help answer applied questions such as, is it possible to find aerosols that can reduce or eliminate ozone loss, without increasing other physical risks?

At the heart of SCoPEx is a scientific balloon, fitted with repurposed off-the-shelf airboat propellers. The repurposed propellers serve two functions. First, the propeller wake forms a well mixed volume (roughly 1 km long and 100 meters in diameter) that serves as an experimental ‘beaker’ in which we can add gasses or particles. Second, the propellers allow us to reposition the gondola to different locations within the volume to measure the properties of the perturbed air. The payload can achieve speeds of a few meters per second (walking speed) relative to the surrounding air, generally for about ten minutes at a time.

The advantage of the SCoPEx propelled balloon is that it allows us to create a small controlled volume of stratospheric air and observe its evolution for (we hope) over 24 hours. Hence the acronym, Stratospheric Controlled Perturbation Experiment. If we used an aircraft instead of a balloon, we would not be able to use such a small perturbed volume nor would we be able to observe it for such long durations.

SCoPEx builds on four decades of research on the environmental chemistry of the ozone layer in the Anderson/Keith/Keutsch groups. SCoPEx will use or adapt many of the high-performance sensors and flight-system engineering experience developed for this ozone research. Analyzing these experiments will improve our knowledge beyond what is currently available within computer models or is measurable with confidence under laboratory conditions.


This FAQ aims to answer some basic questions about the SCoPEx experiment. We will update this FAQ periodically. For a more in-depth overview, see our 2014 publication.

What is the experiment?

We plan to use a high-altitude balloon to lift an instrument package approximately 20 km into the atmosphere. Once it is in place, a very small amount of material (100 g to 2 kg) will be released to create a perturbed air mass roughly one kilometer long and one hundred meters in diameter. We will then use the same balloon to measure resulting changes in the perturbed air mass including changes in aerosol density, atmospheric chemistry, and light scattering.

Why conduct the experiment?

This experiment will help us learn more about the efficacy and risks of solar geoengineering. Computer modeling and laboratory work tell us some very useful things about solar geoengineering, but as with all other aspects of environmental science, computer models ultimately rest on observations of the real environment. Measuring the ways that aerosols alter stratospheric chemistry can, for example, improve the ability of global models to predict how large-scale geoengineering could possibly disrupt stratospheric ozone. Outdoor experiments can provide in situ perspective that is impossible to obtain in the laboratory and SCoPEx can help us validate important model parameters that have yet to been tested against measurements. Learn more here.

What material will be released?

In the future, if a science flight is approved by the independent Advisory Committee, we plan to release calcium carbonate, a common mineral dust. We may also release other materials such as sulfates in response to evolving scientific interests.

For more background, sulfate aerosol (chemically sulfuric acid) is one of the most studied materials for stratospheric aerosol geoengineering because it already exists naturally in the stratosphere. This means that researchers have some level of understanding of its potential effects even though there are still many uncertainties. However, this also means we know that sulfate aerosol, despite its potential benefits, has two major first order stratospheric impacts: ozone destruction and stratospheric heating. The dangers of ozone destruction are fairly well documented, but stratospheric heating is a poorly understood risk because we don’t yet understand how it could change the dynamics of the stratosphere (the motion of the stratosphere). Materials that could therefore reduce these first order risks will reduce the undesirable stratospheric perturbations, which will in turn reduce any further risks that would result in the troposphere and at the Earth’s surface due to the complex coupling of the Earth system. Harvard researchers have analyzed a range of alternative materials, including diamond, and have found that calcium carbonate could be promising. Early research suggests that it has near-ideal optical properties, meaning that for a given amount of reflected sunlight it would absorb far less radiation than sulfate aerosols, causing significantly less stratospheric heating; and it has the potential to greatly reduce the activation of ozone-depleting halogen species compared to sulfate aerosol, meaning that it could reduce ozone loss. Yet, calcium carbonate does not exist naturally in the stratosphere even though it is non-toxic and earth abundant. Therefore, though we can almost certainly expect that calcium carbonate will not have the stratospheric reactivity of sulfate, the actual stratospheric reactivity is not known, which means laboratory and outdoor studies are needed.

Moreover, despite the fact that recent laboratory results suggest that calcium carbonate could reduce ozone loss compared to sulfate, many uncertainties remain that could affect this hypothesis, e.g., some potentially important heterogenous reactions have not been studied (HOCl+ClONO2 for example) and conditions have not explored polar vortex conditions sufficiently. With SCoPEx the latter conditions will not be explored but the conditions will include all species that occur in the mid-latitude stratosphere, not just those studied in our recent laboratory work.

Is this material dangerous?

The test will pose no significant hazard to people or the environment. Calcium carbonate is a nontoxic chemical commonly found in nature, for example as limestone, and sub-micron precipitated calcium carbonate particles like the ones we will use are a common additive to consumer products such as paper and toothpaste. In general, the amount of materials to be released (less than 2 kilograms for calcium carbonate) will be very small compared to other routine releases of material into the stratosphere by aircraft, rockets, or routine balloon flights. For example, the release of experimental materials will be small compared to the release of the iron filling ballast that are commonly released to control the altitude of stratospheric balloons. Additionally, if we test sulfate in this experiment, the amount we would use would be less than the amount released during a one minute of flight of a typical commercial aircraft. Aircraft release sulfates due to residual sulfur content of aviation fuel.

Do other environmental science experiments release materials outdoors?

Yes. A number of environmental science experiments release or have released materials outdoors to create controlled perturbation for the same essential reason as we plan to do in SCoPEx—to directly control an experimental variable, which is crucial to scientific understanding. Examples of experiments include Free-Air Carbon Dioxide Enrichment (FACE) experiments, which release ozone and carbon dioxide (CO2) in the air for long durations to understand the impacts of climate and air pollution on crops and natural ecosystems; or Dispersion of Air Pollution and its Penetration into the Local Environment (DAPPLE) experiments, which have released sulfur hexafluoride (SF6) and perfluoromethylcyclohexane into urban air to study the transport of air pollutants. These experiments differ in various ways, e.g., they have not released material into the stratosphere (the upper atmosphere); they are listed here to merely show that there are environmental science experiments that release materials outdoors.

Are there other risks?

As with any aircraft flight, when flying a balloon there is a possibility of malfunction and need for early and safe flight termination. The flight will therefore undergo standard environmental health and safety reviews, such as those administered by the FAA. The balloon flight provider will also engage in this process. We were working with Raven Aerostar, but due to scheduling constraints, we are now seeking to identify a new partner. We will update this page once we select a new balloon flight provider.

What is the location and timing of the first flight?

We formally asked the independent SCoPEx Advisory Committee to review our plans for a proposed platform test in Sweden in June 2021. In March 2021, the Advisory Committee recommended that we suspend the platform test until a more thorough societal engagement process can be conducted to address issues related to solar geoengineering research in Sweden.

This balloon flight would be managed by Swedish Space Corporation (SSC), flying out of Esrange Space Center in Kiruna, Sweden. This test would not be the experiment itself, but rather a test of the SCoPEx platform without the release of any particles. Specifically, we would like to review the gondola’s horizontal and vertical control using the winch system and propellers as well as the power, data, navigation, and communication systems. An aerosol injection/release system would not be part of this test flight and no aerosols would be released. Still, the team will not proceed with this flight without a formal recommendation from the Advisory Committee to Harvard leadership authorizing the flight.

You can request notification about the results of the societal engagement process and any upcoming decisions by the Advisory Committee by signing up to this SCoPEx science email list. You can also receive updates on the governance of the experiment by signing up to this separate SCoPEx governance mailing list, which is managed independently by the experiment's Advisory Committee.

Why is a platform test needed before a science flight?

A platform test is needed before the science experiment because SCoPEx will use a new flight platform that has not flown before. In other words, there are significant technical challenges in developing it as an operational vehicle independent of the challenges of the actual solar geoengineering experiment. For example, only a few propelled balloon systems have flown in the stratosphere to date. Balloon operators have told us about a few other simple gondolas with propellers that are analogous to SCoPEx, but we do not have access to the details of those systems. We therefore need to test the new platform and concept of operations before conducting the science flight so we can examine its novel capabilities, e.g., verify its operation, controls, and communications. Yet, this platform will not carry systems for releasing particles.

The timing of the science flights will be contingent on the results from this proposed platform test. It’s possible that we will need additional platform flights if we do not resolve sufficient engineering issues on this flight, which might occur if we have a major equipment or concept of operations failure. The timing of the science flights will also depend, of course, on the Advisory Committee’s review and authorization and on other relevant regulatory bodies. Additionally, the location of the science flights is undetermined at this time because it will depend on the evolving balloon launch industry, scientific objectives, and input from the Advisory Committee and other stakeholders.

How often are stratospheric balloons flown?

Because this is not the science flight, the platform test is a rather standard stratospheric balloon flight when viewed in the context of other balloon flights. In 2019, for example, estimates suggest there were more than 300 stratospheric balloon flights over the course of the year. This platform test is interesting to us as engineers and scientists because we are hoping it will provide information about our equipment for the science flight, but stratospheric balloons are flown regularly. For example, Google launched at least 35 balloons in 2020 as it sought to build a new layer of connectivity technology in the stratosphere to expand internet access worldwide, and NASA usually launches 10 or more balloons each year as it carries out scientific and technological investigations, including fundamental scientific discoveries that contribute to our understanding of the Earth, the solar system, and the universe. Swedish Space Corporation, for example, has flown balloons in partnership with NASA, CNES (the French space agency), and STRATOS (the Canadian space agency balloon program).

Why was Sweden chosen as a location for the platform test?

We chose to partner with Swedish Space Corporation and fly in Sweden because of their availability for summer 2021, promising flight trajectories, and significant experience launching scientific balloons. We also looked at several US balloon operators, but because of COVID-19 and other logistical and scheduling challenges, there were no US based options that could provide a 2021 early-summer launch with a landing on land, and that had already secured launch equipment. (Landing on land is important because we need to recover and reuse our platform equipment. And launch equipment, such as a crane or bucket loader and balloon spool, is needed to safely get the payload off the ground. Indeed, US based Raven Aerostar had to cancel our prior agreement many months ago because they could no longer obtain launch services, as we noted on our website at the time.)

We also needed to make sure we identified a partner who could successfully manage the needs of the SCoPEx balloon, as flight safety is of utmost importance to us. The experimental design for future SCoPEx science flights requires a balloon that will fly a 600 kg payload near an altitude of 20 km. This is a lower altitude than many other stratospheric balloon missions, and launch equipment for this weight class and balloon size is not readily available. We also required that the launch site have relatively low winds, the ability to fly a 4-6 hour flight, and the capability to land on land, as noted above. Swedish Space Corporation performed a trajectory analysis based on wind data from 2018 and 2019 and showed favorable trajectories with suitable landings for the entire summer season (April 15-September 15). For example, these projected landings are in extremely low population areas, which is important from a flight safety perspective.

Swedish Space Corporation is a global provider of advanced space services and has been launching scientific balloons for over 40 years. They have been launching balloons out of Esrange Space Center in Kiruna, Sweden since 1974 and they have also provided launches from outside Esrange since 2018. They will be an exceptional partner for SCoPEx. And we hope that the significant partnerships between Swedish Space Corporations and various US balloon operations could lead to enhanced international collaboration and future launches both from Esrange and within the US. Moreover, and perhaps most importantly, working in Sweden will enable us to increase international scientific collaboration around the SCoPEx experiment, which is critical to us. We have already begun to reach out to Swedish scientists, and we are in discussion with German scientists who have now expressed some interest in collaborating on this flight. We are looking forward to forming new partnerships that will enhance the diversity of perspectives studying SCoPEx and improve the science.

Will SCoPEx test geoengineering itself?

This is an experiment not a test. A test could make sense late in the development of an engineering system when design and development have proceeded far enough that it could be useful to test whether some part of the system works as designed. That's not our goal. This is a science experiment that will (we hope) improve knowledge of some aspects of stratospheric aerosol physics and chemistry relevant to solar geoengineering. This knowledge will improve large-scale models (which are all ultimately dependent on physical observations) that will in turn improve estimates of the overall efficacy and risks of solar geoengineering. This may seem like an idle distinction, but it matters. We are not, for example, testing whether it's possible to scatter sunlight back to space, because there is no meaningful scientific uncertainty about that question.

Will SCoPEx develop hardware for geoengineering deployment?

We are not conducting SCoPEx to develop hardware that can be used for deployment. In fact, this is one of the reasons why we chose to loft the particles using a balloon rather than an aircraft (since aircraft are more likely to be used for deployment). Overall, the purpose of SCoPEx is NOT to advance our understanding of the aircraft or other platforms for deployment of solar geoengineering. It aims to reduce the uncertainty around specific science questions by making quantitative measurements of some of the aerosol microphysics and atmospheric chemistry required for estimating the risks and benefits of solar geoengineering in large atmospheric models.

Who is providing the funding?

We think it's essential that SCoPEx is transparent about all funding sources and fundraising principles. Experimental hardware and operations are funded from internal Harvard research funds provided to Professors David Keith and Frank Keutsch. Additional research funding is provided by Harvard’s Solar Geoengineering Research Program (SGRP). All donations to SGRP are philanthropic. Also, in addition to Harvard’s standard funding policies, SGRP follows two further policies:

1) We do not accept anonymous donations.

2) We do not accept donations from corporations, foundations, or individuals if the majority of their current profits or wealth come from the fossil fuel industry unless they can clearly demonstrate that they do not have a conflict of interest and present a strong track record of supporting efforts to address climate change.

The Advisory Committee has been carrying out a financial review of the experiment. You can learn more about our funding sources and policies by reading the details on their website.

How will intellectual property be managed?

One of SGRP’s core principles is to operate in a way that is open access across all activities. As we list publicly on our website, we aim to provide “full transparency with open-access publications and liberal data sharing,” and we “discourage patents and any form of IP protection.”

If it were possible, SGRP would actually forbid patenting for any solar geoengineering related technologies it supported, but there is not a legal way to do so. Harvard owns the intellectual property arising from research conducted using university resources, based on Harvard’s IP Policy and the individual Participation Agreements faculty and researchers sign. But as a practical matter Harvard would not file to protect or enforce intellectual property against the wishes of the contributing faculty member. Because of this, key SCoPEx personnel have personally committed to not file for patents associated with SCoPEx, including Frank Keutsch, David Keith, Norton Allen, Martin Breitenlechner, John Dykema, Mike Greenberg, Michael Litchfield, Terry Martin, Marco Rivero, and Yomay Shyur. Moreover, neither SGRP nor its donors can make any claim on the intellectual property related to the experiment or other research endeavors.

As it relates to activities outside of Harvard, we cannot prevent third-party contractors from filing for patents since they (not we) own the technology that they create. Importantly, however, we have not and do not expect to contract with a third-party vendor for work that could result in a patent of a core piece of solar geoengineering technology. For example, in the case of SCoPEx, any hardware that the balloon vendor develops will not be core to solar geoengineering. It may, for example, be useful to a range of stratospheric balloon flights, including those unrelated to solar geoengineering experiments (if, of course, any new technology is developed at all), but it will not be specific or central to solar geoengineering. This is largely because stratospheric solar geoengineering would most likely be deployed by aircraft, not balloons, if deployed at all.

The Advisory Committee included a series of questions on intellectual property in their financial review of the experiment. You can learn more about their questions and our detailed responses by visiting their website.

Does SCoPEx violate the Convention on Biological Diversity?

SCoPEx does not violate the Convention on Biological Diversity (CBD).

The Conference of the Parties to the CBD adopted a decision that includes a section on climate related geoengineering. It states, "that no climate-related geo-engineering activities that may affect biodiversity take place, until there is an adequate scientific basis on which to justify such activities and appropriate consideration of the associated risks for the environment and biodiversity and associated social, economic and cultural impacts, with the exception of small scale scientific research studies that would be conducted in a controlled setting in accordance with Article 3 of the Convention, and only if they are justified by the need to gather specific scientific data and are subject to a thorough prior assessment of the potential impacts on the environment."

SCoPEx would not affect biodiversity because it would pose no significant hazard to people or the environment, as noted above.

How will the experiment be governed?

The SCoPEx team seeks to perform the experiments in a manner that exemplifies good governance by developing and implementing norms, mechanisms, and practices that can serve as useful templates for possible future solar geoengineering field experiments. Initial oversight of environmental, health, and safety issues is being managed by responsible entities from Harvard University and Swedish Space Corporation. Scientific peer review and broader research governance matters are being overseen by an independent Advisory Committee. The purpose of the Advisory Committee is to provide advice on the research and governance of SCoPEx, operating independently from the research team. You can learn more here.

Key Personnel

Stonington Professor of Engineering and Atmospheric Science, Harvard John A. Paulson School of Engineering and Applied Sciences

Professor of Chemistry and Chemical Biology, Harvard University

Principal Investigator of SCoPEx

Gordon McKay Professor of Applied Physics, Harvard John A. Paulson School of Engineering and Applied Sciences

Professor of Public Policy, Harvard Kennedy School

Michael Litchfield

Senior Engineering Lead for Climate Research

Postdoctoral Fellow, Harvard John A. Paulson School of Engineering and Applied Sciences

Project Scientist, Harvard John A. Paulson School of Engineering and Applied Sciences

Managing Director, Harvard's Solar Geoengineering Research Program

Governance Manager for SCoPEx Team

Learn More

If you are interested in receiving more information on SCoPEx, please fill out this form.