The term “climate intervention” means different things to different people. Climate intervention can be defined as the “deliberate large-scale manipulation of the planetary environment to counteract anthropogenic climate change."¹ While that is a lot to absorb, perhaps the better way to view climate intervention is to think about specific types of intervention. One type of climate intervention is rooted firmly in the world of adaptation and mitigation, where techniques are aimed at either directly, or indirectly, reducing the amount of carbon dioxide in our atmosphere. Another type of climate intervention, also called solar geoengineering, focuses on temporarily reducing the symptoms (temperature change, and everything that comes with it) rather than the cause (greenhouse gas emissions) of global warming; it is designed to invoke a cooling response in the physical climate system using radiative processes analogous to those that occur after volcanic eruptions. The goal of climate modeling research on this second type of intervention, is to fully understand any and all consequences, both regionally and globally, in response to actively intervening in the physical climate, before any action is taken (whether internationally coordinated or not). It is in this spirit that simulations of Earth’s climate are conducted with the purpose of testing the various ways humanity can cool the globe and hopefully avoid the worst consequences of anthropogenic climate change while efforts on reducing emissions and/or removing carbon dioxide from the atmosphere are developed. What might the side effects be, and where would they be felt? 

In our recent work “Atmospheric Rivers Impacting Western North American in a World with Climate Intervention'', we report on one type of climate intervention that cools the planet called “stratospheric aerosol injections” and its impact on atmospheric rivers (ARs) over Western North America.  Atmospheric rivers are long, narrow and banded regions of moisture that are akin to “rivers in the sky” and typically carry enough water to surpass the volume of the Amazon river.  ARs move water from lower to higher latitudes, and for this work, we specifically focus on the “Pineapple Express” variety that travels from near the Hawaiian Islands, across the Pacific Ocean, to ultimately make landfall over Western North America.  The picture below is an example of a simulated  Pineapple Express, produced by our Earth System Model.  We are visualizing the total amount of water present in the atmosphere, measured in kg/m² ; the deeper blue colors nicely illustrate why these phenomena are called  “rivers in the sky.” 

In the paper, we evaluate how the Pineapple Express is impacted by stratospheric aerosol injections using a state of the art Earth System Model called CESM (Community Earth System Model), a community-based climate modeling effort that incorporates many different institutions and universities around the world. Using ensemble suites to simulate the application of stratospheric aerosol injections, we can evaluate the impact this type of climate intervention has when globally cooling the Earth. This procedure essentially tries to mimic volcanic eruptions that work to reflect sunlight back into space rather than enter the troposphere and respond to elevated greenhouse gasses.  The below figure² shows the global mean temperature difference between “business as usual” scenario (RCP8.5, black) and the stratospheric aerosol injection scenario (Geoengineering, blue), compared to our current climate. 

How does this affect the Pineapple Express?  What do ARs under climate intervention look like at the end of the century? How does this compare to today as well as a “business as usual” scenario where greenhouse gasses increase unchecked?  We also compare ARs under a warming climate to those we experience today so we can put these changes into perspective. According to our climate intervention simulations, AR impacts across Western North America are not uniform, and  vary depending on latitudinal location. The atmospheric circulation that drives ARs shifts depending on climate scenario. Under stratospheric aerosol injections, this steering flow shifts such that southern California experiences an uptick in the number of ARs hitting the coastline where the Pacific Northwest sees less. For water managers, however, in addition to the number of ARs making landfall, a key concern is the amount of precipitation that ARs bring to the water table and their communities.  Information on both the amount and character of the precipitation is needed. Are the rainfall rates more intense? How much rain falls over the course of the water year?  ARs are known to produce extreme precipitation events leading to major flooding.  In winter of 2017, a series of ARs inundated northern California, one event after another, that over 45 days produced approximately 66 inches of rain at Lake Oroville, ultimately causing the collapse of the Oroville Dam on the 7th of February, 2017. The intensity and accumulation of the precipitation produced by ARs were key factors in this disaster. In this work, we evaluate AR precipitation for a globally warmed climate, the solar geoengineered climate, and present day. 

The upshot is that under stratospheric aerosol injections, simulated precipitation is generally less intense with fewer extremes and more moderate episodes. The character of AR precipitation is not much different from today. The amount of water accumulated over the course of the year is also quite similar, except for Southern California. Here, the higher number of ARs lead to more available water, produced at moderate rates. 

This could be interpreted as good news from the perspective that ARs do not change dramatically, and fewer extremes are projected to occur, however, I would advise caution when interpreting the results, simply because these are simulations of the Earth’s climate using one modeling framework. Even though our climate models do a good job at representing Earth’s climate, they are not perfect. Climate model groups such as the CESM are constantly striving to improve the models and more accurately represent physical processes, such as clouds and rain, that are historically difficult to simulate at the small scale they occur in nature. It is also very important to acknowledge that although the Pineapple Express doesn’t change too much under stratospheric aerosol injections, there are other elements of the climate system in different regions of the world, where this may not be the case. We must continue to improve the models and the representation of the physics, as well as test a wide range climate intervention designs, to fully understand the impacts and feedbacks to the climate system before any practical application is made. Studies such as this are simply another data point to add to the increasing body of literature so we can make informed, correct, and appropriate decisions regarding climate intervention. 

This work was supported by the National Center for Atmospheric Research (NCAR), which is a major facility sponsored by the National Science Foundation (NSF) under Cooperative Agreement 1852977 and by SilverLining through its Safe Climate Research Initiative. The Community Earth System Model (CESM) project is supported primarily by the National Science Foundation. Computing and data storage resources, including the Cheyenne supercomputer (doi:10.5065/D6RX99HX), were provided by the Computational and Information Systems Laboratory (CISL) at NCAR. The authors of this study, Christine Shields, Jadwiga Richter, Angeline Pendergrass, and Simone Tilmes, thank their colleagues at NCAR, Sasha Glanville, Michael Mills, and Adam Phillips for their contributions recreating supplemental figures, and processing model data. 

¹ Royal Society. (2009). Geoengineering the climate: Science, governance and uncertainty (pp. 1–98). London, UK: The Royal Society

² Tilmes, S et al. CESM1(WACCM) Stratospheric Aerosol Geoengineering Large Ensemble Project, Bulletin of the American Meteorological Society, 99(11), 2361-2371, (2018).  © American Meteorological Society. Used with permission.