Figure 12.3 Hypothetical patterns of ocean circulation and heat transport during (a) an interglacial period and (a) an ice age. This schematic view represents a cross-section from the South Pole (S) to the North (N), cutting through a temperate southern region (TS), the Equator (E), and a temperate northern region (TN).

Figure 12.3 Hypothetical patterns of ocean circulation and heat transport during (a) an interglacial period and (a) an ice age. This schematic view represents a cross-section from the South Pole (S) to the North (N), cutting through a temperate southern region (TS), the Equator (E), and a temperate northern region (TN).

cal waters would be warmed more, evaporate more water, and might even become dense enough to sink of their own accord. Consequently, the mixing of the oceans would be dominated by locally delimited cells of convection, with a strong vertical component of heat transport but only a weak latitudinal component (Fig. 12.3). Because the temperate latitudes would no longer receive a heat subsidy, they would be colder, snowfalls would be heavier, and snow packs would be retained longer. Extensive glaciation and an ice age could follow.

What might cause the oceans to switch between these markedly different patterns of circulation? Clearly, powerful physical forces are at work, and it may be that the oceans would periodically make the switch irrespective of whether or not there was a biota. Consider, for example, the following scenario for shutting down the conveyor systems and plunging the climate into an ice age. Gravity acting on the markedly increased densities of water in the north Atlantic is a significant force driving the Atlantic conveyor. Presently, the waters of the north Atlantic are among the saltiest and coldest on Earth. Suppose, though, that the northern hemisphere experiences a period of warmer temperatures and increased rainfall. If these provide a sufficiently large influx of fresh water, the waters of the north Atlantic would not be so salty as they now are, they would not be so dense, and gravity could no longer act forcefully enough to make them sink. The result: no more Atlantic conveyor.

This switching between warm and cool climates may result if each of the climate patterns is self-limiting in its behavior (Fig. 12.4). If an interglacial period gets greater temperate rainfall dilution of surface oceanic water greater temperate rainfall

dilution of surface oceanic water establish conveyor shut down conveyor concentration of surface oceanic water cool temperate climate cool temperate climate greater temperate" freezing

Figure 12.4 Possible mechanisms for self-limitation of climate patterns and switching between interglacial periods (top circle) and ice ages (bottom circle).

too warm, as we have just seen, the ensuing changes could shut down the oceanic conveyors, forcing the climate back to an ice age. Conversely, if glaciers become very widespread, the consequences of long-term cold could re-establish the coldness and saltiness of waters in the north Atlantic and give the conveyor systems a push. If the push was forceful enough, it could start the conveyors moving again, ushering in an interglacial period.

It should be clear now how the biota could bring about significant changes in Earth climate, even without controlling most of the power that drives climate. If the biota could somehow bias the movement of water and energy that drives the ocean circulation, they could exert a regulatory influence by altering the duty cycle of the climate's ON-OFF controller. I confess right now that I haven't a clue how such biasing could work. But just to illustrate the point, allow me to offer one possible scenario.

Ice-nucleating bacteria (INB) are a class of microorganisms, mostly of the genus Pseudomonas, that are commonly found on the surfaces of vegetation. They have the interesting capability of altering the temperature at which ice forms from water vapor, and they play an important economic role in protecting crop plants from frost damage. When a plant dies and decomposes, some of its INB are lofted into the atmosphere, where they can serve as nuclei for the formation of cloud droplets. When they do so, they bias the tendency of cloud droplets to stay liquid or freeze. Raindrops that coalesce around INB would tend to stay liquid, but raindrops that coalesce around ordinary dust particles might be more likely to freeze. Thus, these bacteria may play a role in determining whether precipitation in a cold climate comes down as rain or snow.

Imagine now that we are in an emerging interglacial climate. As the climate warms, the extent of green vegetation on the Earth expands toward the poles. With expansion comes an increased aggregate leaf surface, which can, in turn, support larger populations of INB. If the airborne burden of INB also increases as a result, we would expect greater rainfall in the north ern latitudes. Increased rainfall, in turn, might shut down the oceanic conveyor system. In this scenario, a biological entity (INB) has biased the switch so that it may be thrown sooner, or with a smaller perturbation from the physical energy stream, than if it had not been so biased. The biasing would be physiological, in that it involves the biological manipulation of a flow of mass and energy. It also would involve positive feedback: a warming climate encourages the spread of vegetation, which promotes further climatic warming, and so forth.

Homeostasis and Symbiosis

That the biota might exert wide-ranging and subtle influences on climate is still a far cry from the central claim of Gaia: namely, that the biota regulates the climate of the Earth. To get there, two conditions must be met. First, the biota itself should have strong tendencies to self-regulation—homeostasis—and this tendency should be universal or nearly so. This conditions implies that substantial benefits accrue to homeostasis per se. Second, this tendency must involve flows of matter and energy outside the organism; the physical environment must be drawn into a physiological conspiracy, so to speak, that will confer homeostasis not only upon the organisms in the environment but upon the environment itself.

Feedback and Homeostasis

What might the benefits of homeostasis be? So far, we have been content to assume simply that homeostasis is a good thing. But why should that be? What are the real benefits organisms derive from doing all that work of homeostasis? The answer to this question lies, I believe, in an understanding of negative feedback control that goes deeper than previous chapters have gone. Up to now, I have simply declared that homeostasis is associated with negative feedback. While this often is true, it is not always true. In what follows I discuss precisely how homeostasis and negative feedback correlate, for only when this point is understood is it clear how homeostasis might be beneficial and how it might result in Gaia.

Let us start with a simple example of a negative feedback process. A driver steering an automobile keeps a point on the car's hood aligned with the stripe painted down the middle of the road. If the car deviates to one side or the other of the stripe, the driver takes corrective action and brings the car back to its proper trajectory. Good control is evident as a trajectory that deviates little from the "planned" trajectory that is indicated by the stripe.

Engineers have long been concerned with analyzing and designing controlled systems, and they have developed a set of tools, control systems theory, to help them do so. The first step is to divide the system into a series of "black boxes," conceptual devices that receive an input and produce an output. The relationship between the input and output is the black box's transfer function, usually symbolized with a The transfer function is determined by at least two values. The first value, the gain, is simply the relationship between the magnitude of the input and the magnitude of the response. High gains produce a large change in the output in response to a small change of the input. Low gains do the opposite. The second quantity, the phase, represents a time delay between a change of input to the system and the initiation of the response. The black box is "in phase" when the delay is zero (that is, the response immediately follows a change of input). The transfer function takes on a larger and larger phase delay as the time lag between input and response increases. Whenever the output of one black box serves as the input of another, this is a feedback loop.

We can use these tools to analyze steering of an automobile. We construct a system with a feedback loop between two black boxes (Fig. 12.5). One contains a transfer function which translates a deviation of the car from the stripe into a particular angle of the steering wheel. The other translates the angle of the steering wheel into a deviation from the planned trajectory. The gain is the relationship between the deviation of perceived deviation from trajectory, 8

ni\/arl angle of steering wheel, 0

Figure 12.5 The transfer functions that govern the steering of an automobile.

the car from the stripe and the degree to which the steering wheel is turned. Low gain corresponds to a slight turning of the steering wheel in response to a large deviation of the car from its intended trajectory. High gain corresponds to a large rotation of the steering wheel in response to a small deviation in the car's position.

It is easy to simulate the behavior of this simple negative feedback controller for steering, and I have reproduced the results of some of these simulations in Figure 12.6. The obvious lesson to be drawn from them is this: "successful" control of the car (that is, keeping the deviation of the car from the stripe small) results only from a particular combination of gain and phase. By tweaking the gain and phase of the model, it is actually quite simple to send it into erratic and wild behavior. It is not hard to think of real-life examples of this. A new and inexperienced driver, for example, frequently over-corrects the steering in response to a slight deviation of the car from its intended trajectory. We can simulate this by increasing the gain of the system, as is shown in Figure 12.6. The failure of control is evident. To take another example, a driver's reaction time can be impaired by various legal and illegal drugs. The result is an increased phase delay, with the driver taking corrective action only some time after the deviation is registered. Clearly, the car behaves erratically.

Feedback and Symbiosis

Let us now use our new conceptual tools to explore the matter of homeostasis in a simple cooperative assemblage of organisms, a symbiosis. A nice example is the lichen, a symbiotic association between a fungus and an alga. The fungal partner produces a substratum for the alga to live on, and some fungi form elaborate structures, called thalli, as housing. The photo-synthetic alga, in turn, provides energy that supports the fungus. The interchange of matter between the symbionts can be represented with a systems diagram—that is, with a collection of black boxes linked into a feedback loop—and we could analyze this association using control systems theory (Fig. 12.7).

Our simple model system consists of a photo-autotroph, P (the alga), and a heterotroph, H (the fun-

Deviation 0 -10

b. +10 Deviation 0

Time c.

Was this article helpful?

0 0
Oplan Termites

Oplan Termites

You Might Start Missing Your Termites After Kickin'em Out. After All, They Have Been Your Roommates For Quite A While. Enraged With How The Termites Have Eaten Up Your Antique Furniture? Can't Wait To Have Them Exterminated Completely From The Face Of The Earth? Fret Not. We Will Tell You How To Get Rid Of Them From Your House At Least. If Not From The Face The Earth.

Get My Free Ebook

Post a comment