Feedback Mechanisms in Climate

"Sometimes you're the windshield,
Sometimes you're the bug...".

M.C. Carpenter, "The Bug"

We wish to learn:

What is a positive versus negative feedback mechanism?
How might clouds impact climate change?
What role do aerosols play in climate change?
What role can humans play in modifying climate change?
How do climate models work and what are their predictions for the future?

Introduction To Global Change I - Lecture Notes			Format for Printing
Feedback Mechanisms	Effect of Aerosols	Biological Response	Summary	Self-Test

1. Feedback Mechanisms

Any change in the environment leading to additional and enhanced changes in that system is the result of a positive feedback mechanism. Alternatively, if a change in the environment leads to a compensating process that mitigates the change it is a negative feedback mechanism. In climate change discussions the focus is on the atmospheric radiation field as a forcing of the climate system (radiative forcing). Currently the discussion concentrate on the radiative forcing associated with the steadily increasing concentrations of different gases in the atmosphere - the so-called greenhouse gases : CO₂, CH₄, N₂O, CFC-gases etc. Other changes in the environment can also lead to changes in the radiative budget, like deforestation, changes in land use and air pollution (ozone, SO₄-aerosols, CONTRAILS, É). Also non-anthropogenic changes are important in disturbing the radiative balance: fluctuations in the Solar output and volcanic activity.

Important positive feedback mechanisms include:

the ice-albedo mechanism
lower tropospheric water vapor content
ocean warming.

Negative feedback mechanisms include

black body radiation
cloud mechanisms

Clouds are important factors in the climate system. They are known to have a negative impact on the surface temperatures in the present climate system. However, a changing climate may involve changes in the clouds with both positive and negative effects on the radiative balance. Currently it is not known whether the total effect of these changes will be a negative or positive feedback.

Figure 1. The positive feedback of increased CO₂ to warmer air temperatures, to warmer ocean temperatures, to a reduced ability for CO₂ removal from the atmosphere that further increases CO₂ concentration increase.

Positive Feedback

Ocean warming provides a good example of a potential positive feedback mechanism. The oceans are an important sink for CO₂ through absorption of the gas into the water surface. As CO₂ increases it increases the warming potential of the atmosphere. If air temperatures warm it should warm the oceans. The ability of the ocean to remove CO₂ from the atmosphere decreases with increasing temperature. Hence increasing CO₂ in the atmosphere could have effects that exacerbate the increase in CO₂ in the atmosphere.

Similar examples can be drawn for warmer air temperatures increasing the rate of glacier and sea ice melting. As the ice melts it changes the surface characteristics of the surface as the underlying ocean or land will have a lower albedo than the ice and hence an enhanced ability to absorb solar radiation.

Likewise, the increase in temperature would permit more water vapor to be stored in the atmosphere. The amount of water vapor the atmosphere can hold increases exponentially with temperature so increases in temperature can yield increases in atmospheric water vapor. The increased water vapor, as a greenhouse gas, enhances the greenhouse effect and could lead to further warming as long as this positive feedback isn't modified by an increase in cloudcover that could lead to a negative feedback (bee below).

Figure 2. The negative feedback of increased CO₂ to warmer air temperatures, to more water vapor content for the atmosphere, to increased cloud cover that increases albedo and reduces the incoming solar radiation, hence leading to a self-limiting cooling.

Negative Feedback

A good example of a negative feedback mechanism will be if the increase in temperature increases the amount of cloud cover. The increased cloud thickness or extent could reduce incoming solar radiation and limit warming.

On the other hand it is not obvious that if additional cloud cover happens at what latitudes and at what times might it occur. Also it is not obvious what types of clouds might be generated. Thick low clouds would have a stronger ability to block sunlight than extensive high (cirrus) type clouds.

Low clouds tend to cool, high clouds tend to warm. High clouds tend to have lower albedo and reflect less sunlight back to space than low clouds. Clouds are generally good absorbers of infrared, but high clouds have colder tops than low clouds, so they emit less infrared spaceward. To further complicate matters, cloud properties may change with a changing climate, and human-made aerosols may confound the effect of greenhouse gas forcing on clouds. With fixed clouds and sea ice, models would all report climate sensitivities between 2°C and 3°C for a CO₂ doubling. Depending on whether and how cloud cover changes, the cloud feedback could almost halve or almost double the warming.

Figure 3. This image shows the aerosol index from the TOMS satellite indicating the location of UV-absorbing tropospheric aerosols. Click on the image for an enhanced view.

2. Effects of Aerosols

Aerosols are tiny (0.001 to 10 µm) airborne particles. In the troposphere, the lower about 10 to 15 km of our atmosphere, human-made aerosols have greatly increased since about 1850. They present a large source of uncertainty in assessing human influences on climate. `Fine' aerosol particles with sizes between about 0.1 and 1 µm can influence climate in two ways. Under clear skies they scatter and absorb solar radiation; some of the scattered sunlight goes back to space (the direct effect). They also can act as cloud condensation nuclei, they may enhance reflectivity and lifetime of clouds (indirect effect).

Sulfur dioxide from fossil fuel burning, yielding sulfate particles after oxidation, is presently the largest source of fine human-made aerosols. Another large source is organic and elemental carbon from burning of tropical forests and savannas. Globally averaged, fine human-made tropospheric aerosols may currently cancel about 50 % of the warming effect of human-made greenhouse gases. So far, though, the uncertainty range is large, stretching from roughly 10 to 100%. Moreover, global averages are misleading. Even if the global averages of aerosol and greenhouse gas forcing cancel, their different distributions may cause climatic changes. With life spans of up to over 100 years, human-made greenhouse gases are fairly evenly distributed. Most tropospheric aerosols are washed out after about a week, they are unevenly distributed. Human-made sulfate aerosols occur mainly down- wind of northern industrialized areas. Most biomass smoke rises from tropical land areas during the dry season.

Cutting back sulfur dioxide emissions or biomass burning reduces the aerosol load quickly, leaving over the more long-lived greenhouse gases. By the way, roughly one third of the tropospheric sulfate load has natural precursors, mainly oceanic dimethyl sulfide (DMS) and volcanic sulfur dioxide. Violent volcanic eruptions, like Pinatubo 1991, give rise to stratospheric sulfate aerosols which, being more long-lived than their tropospheric cousins, tend to warm the stratosphere and to cool the troposphere and surface for a few years.

'Coarse' aerosols with particle sizes between 1 and 10 micrometers include mineral dust raised by wind blowing over dry soils. Human influences like over-cultivation and soil erosion may have up to doubled the flux of mineral dust. Mineral dust is most abundant over North Africa, the Arabian Sea, and South Asia. It scatters sunlight and absorbs outgoing terrestrial infrared. These two effects appear largely to cancel at the top of the atmosphere, thus mineral dust presumably has little effect on earth's radiation balance. However, it may regionally cool the surface and warm the atmosphere, which in turn may affect atmospheric circulation. Pinning down aerosol effects more precisely will be tough. Aerosols are hard to measure. Size, shape, composition and regional distribution of the particles vary. So do their effects on climate. Aerosols can cause not just local but also distant responses, because heat or rather, in the case of many aerosols, coolness is transported by the atmosphere and ocean. Assessing the climatic effects of aerosols involves modeling of regional climates and of clouds, both of which are not yet very reliable.

3. Biological Response

Some facts

Except for atmospheric CO₂, carbon reservoirs and natural fluxes are hard to measure. Their estimates vary somewhat across the literature. Carbon enters and leaves the atmosphere largely as CO₂. Other fluxes involve various carbon compounds. The above irreverently lumps land animals with soils and detritus, and it omits many other details as well. For instance, both volcanic CO₂ and CO₂ removal via silicate weathering are in the order of 0.1 GtC/year and play a role on geologic time scales only. Here's what is estimated:

Gt	is a gigatonne = 10⁹ metric tonnes, the mass of one cubic kilometer of water
1 GtC	corresponds to ~3.67 Gt CO₂
2.12 GtC (or ~7.8 Gt CO₂)	corresponds to 1 ppmv CO₂ in the atmosphere. ppmv = parts per million by volume

Carbon Reserves

Carbon Reservoirs (GtC)

LAND		AIR		OCEAN
Terrestrial Vegetation	600	Atmosphere	750	Surface Ocean	1000
Soil & Detritus	1600			Marine Biota	3
				Dissolved Organic Carbon	700
				Deep Ocean	38,000

CO₂ uptake by land plants through photosynthesis is roughly balanced by plant and soil respiration. Depending on whether photosynthesis exceeds or falls below respiration, the net result is CO₂ drawdown or CO₂ release. Today, photosynthesis is probably slightly ahead. In future, climatic changes or rising CO₂ level may trigger feedbacks that curb or speed up the rise of atmospheric CO₂. To name a few:

CO₂ fertilization should promote photosynthesis and draw down some CO₂, as long as respiration doesn't catch up.
Warming may stimulate or slow down either photosynthesis or respiration, depending, among others, on soil moisture.
The mix of species in ecosystems is likely to shift, which in turn may affect atmospheric CO₂. Dieback of vegetation can release CO₂.

The overall effect of these and other feedbacks is hard to tell. Ecosystem models tentatively suggest that carbon storage in vegetation and soils may eventually win out. Temporarily, however, carbon may be released, especially if large and rapid changes should cause forests to die back.

Turning to the ocean, a sea surface warming of 1°C may increase atmospheric CO₂ by up to 10 ppmv through degassing. More importantly, marine life, in spite of its low biomass, takes up and releases about 50 Gt of carbon annually. Marine biological production occurs largely in the sunlit surface and is thought to be limited mostly by nitrogen. Surface nutrient supplies are replenished primarily through transport from deeper ocean layers. (In the open ocean, iron can be limiting; it enters the ocean mainly in airborne dust and via rivers.) The export of organic carbon from the surface to deeper ocean layers, known as the biological pump, is not or little affected by CO₂ availability, but it may be affected by changes in temperature, cloud cover, ocean currents, nutrients availability, or ultraviolet radiation. These and other marine biological processes are complex. Researchers cannot yet say how they will respond to disturbances. It has been estimated that, with no biological pump, pre-industrial atmospheric CO₂ would have been 450 instead of 280 ppmv, whereas a marine life seizing all available surface nutrients could have lowered this to 160 ppmv. On the other hand, preliminary results suggest that changes in the biological pump may affect atmospheric CO₂ only by 10s rather than 100s of ppmv.

Back to the land, spreading of boreal forest into tundra may lead to warmer winters. Trees protrude above the snow-covered ground, they reflect less sunlight back to space than snow-covered tundra. During and after deglaciation, the expansion of boreal forests amplified the warming of northern land areas. The reverse process, displacement of boreal forest by tundra, probably played a role in the onset of the last glaciation. For another example, rising CO₂ tends to improve the water-use efficiency of vegetation. Plants may then release less water vapor to the ambient air. Regionally, this may warm the surface and affect precipitation and soil moisture.

Living in the United States we are responsible for more than our fair share of greenhouse gas emissions. Check your own Personal CO₂ Calculation to get an idea of how your carbon emissions compare to others. Human response may be the least well understood feedback mechanism. On one hand we may respond by demanding ever more energy to cool our homes on warmer days, plus the exponential growth in population will produce even more demands for energy even if consumption stabilizes. On the other hand improved education, technology and political pressures may at least reduce the rate of growth in carbon emissions.

The breadth of feedback mechanisms in the climate system is enormous. Figure 4 tries to articulate some of the feedback mechanisms without quantifying their importance. The bottom line is our attempts to forecast future climates holds considerable uncertainty. Nonetheless it should be understood that the uncertainty could go either way and our estimates of temperature rise could be overestimated ...or underestimated.

Figure 4. A concept map of potential feedback mechanisms in climate change. Click on the map for a larger view.

4. Summary

1. Feedback mechanisms exist that can potentially exacerbate or mollify temperature increases.

2. There are a number of reasons to believe that temperature increases will enhance processes that either increase CO₂ concentrations or the absorption of incoming solar radiation.

3. Many of the feedback mechanisms are poorly understood, so there is ample uncertainty in our estimates of future climates, nonetheless it is just as possible we are underestimating future temperature rises as overestimating them.

Take the Self-Test for this lecture.