Airborne Bacteria Discredit Climate Modeling Dogma

The formation of low-level clouds—clouds that have a cooling effect on Earth's climate—has vexed climate scientists for years. Current climate models treat cloud cover simplistically and make the assumption that cloud cover decreases as temperatures rise. New data from a cloud sampling experiment indicates that biological material—bacteria, spores and plant material—may account for 1/3 of the airborne material involved in cloud formation. Furthermore, biological material can form clouds at much warmer temperatures than mineral dust. These new discoveries indicate that modelers have the effects of temperature on low cloud cover backwards, placing all model predictions in doubt.

A team led by Kim Prather, of Scripps Institution of Oceanography and the University of California San Diego, used an aircraft with a specially designed lab instrument called a mass spectrometer to identify the particles that ice crystals and water droplets form around. Ice crystals, sampled from clouds and quickly analyzed while in flight, showed bits of biological material along with other aerosols. “The key to cloud formation is these little seeds that feed the clouds,” Prather said in an interview. “We are basically trying to understand what is forming clouds.”

Aerosols, including dust, soot, salt from ocean spray and organic materials, form the seeds clouds grow from. Around these tiny particles, water and ice in the atmosphere condense and grow, a process called nucleation. Scientists are trying to understand how clouds form, because clouds play a critical role by both cooling the atmosphere and affecting regional precipitation.

“By sampling clouds in real time from an aircraft, these investigators were able to get information about ice particles in clouds at an unprecedented level of detail,” Anne-Marie Schmoltner, of the National Science Foundation's Division of Atmospheric sciences, said in a statement. “By determining the chemical composition of the very cores of individual ice particles, they discovered that both mineral dust and, surprisingly, biological particles play a major role in the formation of clouds.”

Sampling clouds above Wyoming, the researchers found that biological matter accounted for 33% of the particles in ice crystals, and mineral dust accounted for 50%, some of it from as far away as Asia. The findings suggest biological particles that get swept up in dust storms help induce cloud formation. While it has long been known that microorganisms or parts of them get airborne and travel great distances, this study, reported in Nature Geoscience online, is the first to measure their participation in cloud ice formation.

Credit: Andrew J. Heymsfield, NCAR

In climate change science, which derives many of its projections from computer simulations of climate phenomena, the actions of aerosols on clouds represent what scientists consider the greatest uncertainty in modeling predictions for the future. Because General Circulation Models (GCM) use a very coarse grid—on the order of 100 km square—the effect of cloud formation processes cannot be accurately modeled, even if there were enough spare computer cycles to do so. Instead, simple average values for cloud cover effects are used to calculate the impact of clouds on temperature, and that is where the faulty assumptions have been applied.

According to Dr. Roy Spenser, meteorologist and former Senior Scientist for Climate Studies at NASA’s Marshall Space Flight Center, the warming in models is known to be mostly controlled by low and middle level clouds. He explains in a posting on his blog:

The main reason the models produce so much warming depends upon uncertain assumptions regarding how clouds will respond to warming. Low and middle-level clouds provide a ‘sun shade’ for the Earth, and the climate models predict that those clouds will dissipate with warming, thereby letting more sunlight in and making the warming worse.

The important participation of biological and mineral aerosols in cloud formation provides a viable explanation for increased, not decreased cloud formation as Earth heats up. As I reported in an earlier post, dust and material from North Africa seems to have a significant impact on the tropical Atlantic. In other places, climate change is also being blamed for increase in airborne particulates.

An obvious cause would be desertification, long touted to be an effect of global warming. But there are subtler linkages as well. Extended growing seasons, also being attributed to global warming, give plants more opportunity to contribute to airborne material. Drought, leading to more natural fires, would contribute carbon based aerosols while stressed plant populations could be more susceptible to bacterial infection, increasing the amount of airborne bacteria.

This implies a negative feedback loop: global warming increases aerosols, which increases low and medium level cloud cover, cooling the climate. To again quote from Dr. Spenser: “If feedbacks end up being negative... then extra CO2 will have caused even less warming, which means that there is even more room for natural cloud variability to explain the warming experienced in the last 50 to 100 years.”

Cloud fraction January 2008. The cloud fraction expresses how much of the Earth is covered by clouds. Cloud cover plays an important role in regulating the amount of energy that reaches Earth from the Sun as well as the amount of energy that Earth reflects and emits back into space. Source NASA.

This is not the first indication that bacteria may help regulate climate. Professor Brent Christner of Louisiana State University, with colleagues in Montana and France, reported evidence in the journal Science that “rain-making” bacteria are widely distributed in the atmosphere. “My colleague David Sands from Montana State University proposed the concept of 'bioprecipitation' over 25 years ago and few scientists took it seriously, but evidence is beginning to accumulate that supports this idea,” said Christner.

There has been speculation that biological particles could factor heavily into the precipitation cycle, affecting climate, agricultural productivity and even global warming for some time now. This is because it is well established that biological particles can promote nucleation and at higher temperatures than particles of mineral dust. Many ski resorts use a commercially available freeze-dried preparation of ice-nucleating bacteria to make snow when the temperature is just a few degrees below freezing.

“The role that biological particles play in atmospheric processes has been largely overlooked. However, we have found biological ice nuclei in precipitation samples from Antarctica to Louisiana - they're ubiquitous. Our results provide an impetus for atmospheric scientists to start thinking about the role these particles play in precipitation,” said Christner. Non-biological particles are good at collecting water at temperatures below about 14°F (-10°C), biological particles seem to be the main active nuclei above that temperature, according to Christner's findings. Unfortunately, this knowledge isn't incorporated into climate models.

“Our results provide an impetus for atmospheric scientists to start thinking about the role these particles play in precipitation,” says Prof Christner. “It clearly demonstrates that we are just beginning to understand the intricate interplay between the planet's climate and biosphere.”

Pseudomonas syringae cells trapped within an ice crystal lattice. Photo credit: Shawn Doyle and Brent Christner

Even prior to the discovery of the bacteria climate link, aerosols' role in cloud formation had been inadequately portrayed in climate models. Joyce E. Penner, a leading atmospheric scientist at the University of Michigan, presented a talk on the topic “Aerosol-Cloud Interactions and Climate Projections,” during a panel at a meeting of the American Association for the Advancement of Science in San Francisco on Feb. 17, 2007. Using known temperature data starting in 1850, two different climate models were compared. One had low climate sensitivity and small amounts of aerosols and the other high climate sensitivity and high amounts of aerosols. Penner's group showed that both models follow almost identical predictive paths in the past, but diverge significantly when predicting the temperature in the future.

Further, Penner's presentation discussed the predictive capability of three other climate models: the US NCAR-Oslo model, a French model and a Japanese model. Tests showed that large differences and significant changes in results occurred, especially when having the models predict both aerosols and their cloud effects using assumed historical aerosol levels. It is no secret to savvy climate modelers that their models are not capable of accurately predicting future climate change. “We know that aerosol effects on clouds need to be included in climate models,” Penner said.

These are just the latest revelations about the shaky foundation climate models are built on. Missing factors and erroneous feedback loops highlight how arbitrary decisions, which happen to reinforce the modeler's desired result, can undermine a model's veracity. Nobody denies that atmospheric CO2 increases environmental heating. The problem is that, relative to other forcings, this is a small effect that is quickly overwhelmed by other factors. A slight change in cloud cover can completely nullify any effect of CO2, and we have no way of accurately predicting cloud cover—certainly not with the models currently available.

Answers to fill the gaps in our spotty knowledge can only be found in one place—Earth's physical environment. Still the allure of computer models keeps climate scientists gazing at display screens instead of the real world. Dr. Prather, Dr. Christner and their respective teams are to be commended for taking what is becoming a radical step in climate science—actually performing experiments and observing nature.

Be safe, enjoy the interglacial and stay skeptical.


Doug, what kind of plane was used to gather the cloud data? My guess would be a cessna, but I'd be interested to know for sure.
Thanks Doug! Great post.