from Section 2: Emergent Organizations

The Big Discovery

Back in the 1980's I was attending a mid-sized Canadian university where the policy of the biology department was that each graduate student should attend all of the seminars it hosted. The idea was to expose students to a wide range of research, thus fostering the development of a lively intellectual atmosphere within the department, but since compulsory attendance also meant that the seminar room was always packed, the atmosphere inside was often more conducive to carbon dioxide narcosis than anything else. Many a head could be seen nodding in the half-darkness as soon as the lights were turned down for the slides, and the challenge of maintaining consciousness seemed to be especially acute during ecology seminars, during which audiences were routinely subjected to seemingly endless tables and graphs displaying data concerned with things like the numbers of aphids on cottonwood trees or the growth rates of arctic lichens that may or may not have been trodden on by caribou (a kind of Canadian reindeer). The situation seldom improved when the ecologists launched into the analysis of their less than rivetting data, because the reliance on heavy-duty number-crunching and mathematical modelling that had led them to be among the first biologists to take to computers in a big way made the going rough for their more laboratory-oriented colleagues, who tended to view anything beyond simple statistics as the domain of the "soft" sciences. As it happened, my own research in genetics involved the use of analytical tools that were similar to those employed by the ecologists (statistical analysis having been invented by geneticists), and my interests also extended to theoretical and "outdoor" biology. Thus I was not so biased as some of my colleagues against the ecologists' approach to data analysis and I was also better prepared than most to follow their analytical perambulations (at least I was back then). Nevertheless, more often than not I could also be counted among those who were reduced to pinching the backs of their hands in a desperate attempt to remain within the quorum of consciousness during ecology seminars. The problem was that even when I could follow what was going on it seldom seemed worth the effort, because I was unable to see the vaguest outlines of a biosphere - or for that matter a grassland, forest or coral reef - amid the data points. In fact I came to suspect that ecology was largely concerned with what scientists disparagingly refer to as "turning the crank."

This vignette from my misspent schooldays illustrates how easy it is to get the wrong idea about research, even when you should know better. It was about as realistic for me to expect any particular ecological study to yield a unified vision of the living world or a living community as it would have been to expect my thesis on the behaviour of mutant fruit flies to present a comprehensive picture of the insect nervous system. To be fair to those who may have similar illusions, there certainly does seem to be an air of inflated expectation around ecology, which is commonly associated with expansive concepts like the balance of nature, the web of life and even the pattern which connects. However, when we take a closer look at who is actually bandying these concepts about, we often find them to be folks who are interested in ecology rather than those who actually do it for a living. The survival of the latter group depends upon their ability to produce publishable papers, defensible doctoral theses and successful grant applications, which obliges them to focus on things that can actually be studied, like the relationships between aphids and cottonwood trees, or caribou and lichens. This does not mean that the professionals never look up from their test plots and computers to consider the big picture, but their version of it differs substantially from that put forth by what I like to call "para-ecologists." For example, the amateur view of the living world still features concepts like superorganisms and successions, which were abandoned by most professional ecologists long ago.

And what has real ecology contributed to the big picture of the living world? Well, it may be a bit difficult to appreciate at first, but some recent work in ecology and related disciplines (e.g. geophysics), coupled with the results of some inadvertent if spectacular experiments, adds up to what I think may be one of the biggest discoveries in the history of biology, which is that the survival of some of the world's largest, densest, most diverse and influential living communities depends upon their ability to remain large, dense, diverse and influential. The main reason I expect it will be difficult for some readers to appreciate this Big Discovery is because we have not established several of the key concepts connected with it. For instance we have not defined what survival means for living communities, although we have recognized them as persistent patterns of interactions among constituents and between them and their environment, so it follows that in order for those patterns to persist they must regenerate (i.e. replace constituents via reproduction) and remain bioenergetically viable.

We can say that as long as a community meets these requirements it is sustainable, but does that make it containable? How do we decide where one community ends and another begins? Once again, we could jump right into sorting things in the standard way, but in keeping with our aim of allowing distinctions to emerge before we impose them, we should give living communities a chance to define themselves. This is where the Big Discovery can guide us towards an appreciation of its significance, because the communities it is concerned with are conspicuous for their influence upon their environment, both locally and on large scales, such as those associated with regional weather patterns. It stands to reason that these communities should be fairly easy to recognize - in fact they should be hard to miss.

Conspicuous Communities

Tropical rainforests certainly fall into the category of conspicuous communities. They are typically dominated by hardwood trees that reach over 30 metres in height at maturity, making the rainforest easy to distinguish from neighbouring communities like scrub forest and grassland, which are dominated by light-loving plants that have little chance of invading the damp darkness that prevails under the dense rainforest canopy. Among the shade-tolerant plants that can survive on the forest floor we find saplings of the dominant tree varieties, which bide their time until gaps in the canopy allow them to grow up through several layers containing distinctive populations of plants (e.g. epiphytes like orchids and bromeliads), animals (e.g. insects, birds, mammals), fungi and microbes. The notable exceptions to this skyward striving are the parasitic fig trees, which evade the competition for space on the forest floor by growing down from higher and brighter regions using other trees as temporary scaffolds.

As their name implies, tropical rainforests are found in equatorial regions where there is a lot of rainfall, such as between mountains and the sea or near rain-generating confluences of oceanic and atmospheric currents. This restriction does not mean, however, that rainforests and their inhabitants are simply passive beneficiaries of their environment. The first indication that such communities play an active role in their survival appears as we approach them from nearby grassland or scrub forest, where the typical diurnal (i.e. day-night) and seasonal temperature variations are measured in tens of degrees and the humidity undergoes similar fluctuations. Once we enter the rainforest the environment changes dramatically, especially with regard to fluctuations. For example in the unbroken forests of the Amazon valley, the temperatures range between 27 to 32 degrees Celsius year round and the humidity remains comfortably near saturation. This temperature stability comes mainly from the dense canopy formed by the mature trees, which protects the lower levels from the heat of the tropical sun in daytime and inhibits the re-radiation of heat back to the atmosphere at night. The canopy also contributes to the consistent humidity by shielding the lower levels from wind and torrential downpours, creating a still, humid internal environment watered by gentle rains and trickles running down the trunks of the trees. This constant flow keeps the thin, spongy rainforest soils saturated and fully expanded, maximizing their water-holding capacity.

Another aspect of the ability of rainforest communities to maintain consistent climates emerges when we approach them from above during a rainstorm. The leaves and branches that extend above and throughout the canopy catch the lion's share of rain falling on the forest and immediately present it back to the atmosphere for evaporation, while much of the water that trickles down to the forest floor is held by the spongy soil until it can be resorbed by tree roots, from where it is osmotically pumped back up to the leaves and released as a fine mist - or evapotranspired - sometimes with cloud-seeding chemicals mixed in. These various mechanisms do a very efficient job of returning rainwater to the atmosphere, where it can be taken up by clouds and dropped again; for example it has been estimated that at any given time over 50% of the rain falling in the Amazon basin has passed through the forest "rain pump." Rain recycling helps to maintain cloud cover over forests, shading them from the tropical sun, and it also allows rain clouds to travel farther inland from their origins in moist coastal regions than they would over non-forested areas.

Taken together, these observations make it reasonable to propose that tropical rainforests are larger, denser and more diverse than they would be if they had less influence over their internal and external climate. This proposition is difficult to test using standard experimental methods; for example we cannot put a forest into a growth chamber and do rain recycling experiments with it. But as luck would have it some recent experiments have provided useful insights into the growth, survival and environmental interactions of different kinds of forest communities around the world. These experiments have been performed by people who are not so much interested in studying forest communities as exploiting them, hence their manipulations have primarily involved disruptions which range in severity from the localized damage associated with road-building and the clearing of pockets for slash-and-burn agriculture to the widespread devastation that attends plantation clearing and clear-cut logging. While their performance has not always been up to the best scientific standards, these community disruption experiments have yielded a wealth of information, much of which is consistent with what we have already learned about the interactions of forests with their environment. For instance given the observed relationship between rainforest trees and local water cycles we would expect that when mature trees are cut down there will be a dramatic increase in the rate of water runoff, and that is what has been observed in patches cleared within rainforests, and indeed in most forests dominated by large trees. Increased runoff leads to the lowering of local water tables, the acceleration of nutrient leaching from soils and the rapid loss of soils due to wind and water erosion, and there can also be significant consequences downstream of the deforested area, such as flooding, silting up of rivers and the hyper-concentration of nutrients in water bodies, or eutrophication. Such "downstream effects" can add up; for example recent record-setting floods in lowland areas of China have been linked to deforestation in upland regions.

The long-term effects of localized deforestation have been observed in places where disturbed areas have been allowed to recover, and the impact varies considerably depending upon the kind of forest involved. In some cases a community much like the original one soon reappears in the cleared area, as is often the case in northern evergreen (i.e. boreal) forests where localized clearings are rapidly recolonized by plants from surrounding forests. This pattern of recovery is what we would expect from communities where intermittent catastrophes have long been a part of the life cycle, such as the forest fires that have periodically raged through boreal forests since the close of the last ice age (some boreal plant seeds actually require a good scorching before they will germinate). However, recovery from man-made disruptions like clear-cut logging tends to be slower in these communities, due to factors such as soil erosion and the loss of biomass; for example logging removes most of the material that would be recycled under normal conditions.

Similar considerations apply to the recovery of cleared patches of temperate and tropical deciduous forests, where recolonization tends to take a bit longer than in boreal forests because non-forest plants like grasses and scrub often move in first, and then gradually give way to forest plants like hardwood trees. Tropical rainforests, on the other hand, are not nearly so resilient. Their nutrient and water cycles are rapid, their topsoils are thin and their base soils are nutrient-poor when compared to other types of forest, and as we have seen rainforests also serve as climate-controlled incubators for saplings of the dominant tree varieties. This means that when a patch of rainforest is cleared and deprived of recyclable biomass by logging or burning, the resulting profound changes in nutrients, soil, water, temperature, light and other environmental factors make the cleared areas much less hospitable to rainforest plants and much more attractive to light-tolerant varieties with more modest growth requirements, such as grasses and scrub plants. Thus it is not surprising that community disruption experiments have consistently shown that when patches of rainforest are cleared by human activity they seldom recover, what happens instead is that other kinds of communities appear, such as scrub and grassland.

So much for what happens within deforested patches; what happens to the communities surrounding them? In addition to the long-term and downstream effects already mentioned, localized deforestation and road-cutting can also influence the remaining forests by creating new "edges" where sunlight, rain and wind penetrate more easily to create conditions that are different from those that prevail inside forests. As a result, edge areas tend to be colonized by non-forest plants, especially in tropical rainforests (for much the same reasons that cleared areas seldom grow back), which are also highly sensitive to another consequence of deforestation: fragmentation. Disruptions in the physical integrity of these typically large, dense and continuous communities can have a variety of consequences, such as the compromising of their ability to maintain consistent environmental conditions and the restriction of their inhabitants to smaller home ranges. Such changes can in turn lead to the loss of habitats, inbreeding and other consequences relating to another fundamental aspect of living communities that the Big Discovery is associated with: diversity.

The first step towards understanding how the diversity of communities like tropical rainforests can be affected by disruption involves gaining an appreciation of just how diverse they are. A good illustration is provided by the largest and longest-lived inhabitants of rainforests, the trees, which typically exist in populations where a variety that accounts for as much as 1% of the trunk count is considered to be "common." For example, a recent survey of a Malaysian rainforest turned up over 1000 different varieties of hardwoods growing within the same 100 acre tract. In comparison, all of the native trees in North America account for fewer than 1000 different varieties, of which only a fraction can be found growing in any particular temperate forest, where a "common" variety can account for most of the trees.

Once we appreciate how diverse rainforest communities are, it is not difficult to see how their ability to maintain high levels of diversity might be compromised. For example, if we started progressively dividing up that 100 acre tract of Malaysian rainforest by cutting it in half, then quarters and so on, and took a new survey after each division, it would not take long before we discovered that some sectors no longer contained the same diversity of tree varieties as the original forest. This sampling effect is an indication that high levels of diversity require large communities to support them - very large communities, as it turns out. For instance a recent study of rainforests in the Amazon basin found that even forest "islands" as large as 250 acres (i.e. 100 hectares) were not sufficient to nurture the seedlings required to regenerate existing populations of trees, while smaller islands were observed to lose varieties at a rapid rate (I should mention that another difference between temperate and tropical forests is that in the latter the life spans of even the largest trees are measured in decades, rather than centuries).

These observations, in combination with those already mentioned, point to some basic conclusions concerning the consequences of small and moderate disruptions in tropical rainforest communities. The first conclusion is that when essential bioenergetic and environmental interactions are interrupted - such as those associated with the cycling of nutrients and water - the directly affected areas are not only rendered non-viable, there is also little likelihood that something resembling the original community will ever re-emerge. Our second conclusion concerns rainforest communities as integral wholes, which is what the results of disruption experiments indicate that they are, because in addition to viable bioenergetic and environmental interactions the sustainability of rainforest communities - as assessed by their ability to maintain diversity - appears to depend upon the preservation of their size, density and physical integrity. One way of summarising this conclusion is to say that the sustainability of rainforests depends upon the maintenance of a critical community mass, which is associated both with the regeneration of constituents and with the ability of large, dense and diverse communities to maintain stable internal environments and participate in large-scale environmental interactions, such as the rain pump described earlier.

That last connection is actually a bit premature, because we have not yet encountered experiments where disruptions have had a clear effect on large-scale interactions between forest communities and their environments. The experiments we have covered so far would lead us to expect that such disruptions would probably require deforestation and/or forest fractionation on a massive scale and that the consequences should be spectacular. These expectations are confirmed by the results of some of the most ambitious of all community disruption experiments, such as the one that has been in progress in West Africa since the turn of the 20th century.....