In general, however, would you expect such a simple (even if it takes multiple factors into account) model to explain the limits of a species’ geographic range?? Probably not as there are some important caveats to the acceptance of the niche as the main factor determining a species’ range:
(a) Habitat quality. Not all habitats within a certain area will have the same character in terms of the environmental factors important in determining a species’ range. This variability in habitat quality often results in individual populations being described as "source" or "sink" populations.
Source populations are those where environmental conditions are such that the birth rate exceeds the death rate (i.e. the area within the ellipse in the figure above) and such populations produce "surplus" individuals that emigrate to other habitats. Sink populations are those where the environmental conditions are so unfavourable that death rate exceeds birth rate. These populations, therefore, are only sustained by immigration from source populations. In other words, certain localities that are part of a species’ range may be sink habitats where the occurrence of that species is maintained by immigration from other areas. Often such habitats are found at the "leading edge" or periphery of a species’ range.
(b) Some favourable areas may not be occupied. This can occur because the habitat is too isolated from the main body of the species’ range and that species has limited dispersal abilities (e.g. think of islands, many freshwater fish in B.C. are found only on the mainland and not on Vancouver Island or the Queen Charlottes Islands (despite the presence on these islands of suitable freshwater habitats) because the intervening saltwater is a (lethal) dispersal barrier). Also, historical factors can prevent the occupancy of suitable habitat (e.g. dispersal barriers developed before a species could arrive).
(c) Metapopulation structure. A metapopulation can be described as "a population consisting of a set of subpopulations that are linked by a cycle of alternating immigration, extinction, and recolonization". This may occur because the habitat is "patchy" and isolated by areas of unsuitable habitat or owing to random factors or environmental variability that can cause extinction of local subpopulations that are subsequently re-established by emigration from other subpopulations. In other words, some habitats may be occupied only intermittently owing to metapopulation dynamics.
An example of a metapopulation in a butterfly (Melitaea cinxia) in southwestern Finland. The meadow habitat is shown by the outline shading, suitable meadow patches by the small circles, those occupied during one summer are shown as small and large black circles, those unoccupied by light small circles. See Saccheri et al. 1998 (below) for more details.
The bottom line with respect to (a) – (c) is that the fundamental niche concept can not really take these factors into account very well and, therefore, is an imperfect description of the geographic range of a species (although it does contribute understanding!).
(2) Disturbances.
Disturbances (e.g. catastrophic environmental events like floods, hurricanes, volcanic eruptions, fires) can strongly influence the geographic range of a species. For example, in modern times, forest fire suppression has lead to restrictions in the geographic range of native grasses in many parts of the world. Such grasslands were promoted by a natural cycle of fire and regeneration that kept shrub and woody species "at bay". Such processes have played a major role of so-called "desertification" –replacement of grasslands by shrubs and woody species. A good recent example of hurricanes and distribution can be found in Calsbeek and Smith (2003).
(3) Competition. Organisms that share requirements for growth and survival (e.g. a food resource) often compete for such resources which can lead to declines in population growth for both species. If niche overlap is too great, one taxon can be excluded from otherwise suitable habitats by the presence of a close competitor. For example, one hypothesis for why characins (a very diverse group of carnivorous, small-bodied fishes) have not invaded North America to any great degree (see map below) is because NA is home to an ecologically similar group of fishes known as the Cyprinidae (minnows). Perhaps competitive exclusion plays a role in limiting characins to SA (mostly) and cyprinds to NA?
Figure. World distrubution of the Characidae (upper) and Cyprinidae (lower). Is the absence of cyprinds in South America and of characids in North America (mostly) caused by competitive exclusion?
(4) Predation: Distributions of predators may be influenced by the geographic range of their prey species. This can also be said for the distributions of parasites and their hosts. Evidence in favour of this idea would include the complete coincidence of predator/prey (host/parasite) ranges. There are, however, few good examples were distributions are so tightly coupled. Ehrlich (1965), however, described how the checkerspot butterfly (Euphydryas editha) was restricted to a very localized area in California where its host plant, Plantago hookeriana, was restricted. Similarly, parasitic pinworms appear to be restricted tightly by the geographic range of their primate hosts. In this case, the evolutionary relationships of the different species of worms is mirrored by that of their hosts (i.e. divergence of the hosts promoted divergence of the parasitic worms, the latter being restricted in distribution by their hosts).
Other examples include the fate of prey items when important predators are introduced by humans (on purpose and by mistake). For instance, distribution of the lake trout (Salvelinus namaycush) in parts of western Canada (e.g. the lower Yukon River) was thought to be limited, in part, by parasitism by the Pacific lamprey (Entosphenus tridentatus). Some evidence in favour of this idea was revealed when related parasitic lamprey gained access to the upper Great Lakes (upstream of Niagara Falls) with the construction of the Welland Canal earlier this century. The sea lamprey (Petromyzon marinus) gained access to the upper Great Lakes through the canal and decimated the native lake trout, causing local extinctions and threatening the existence of various native fish in the Great Lakes. The brown tree snake, Boiga irregularis, snake was introduced inadvertently to the island of Guam in the south Pacific and has caused extinction of most of the native bird fauna there (up to 10 –12 species, Savidge 1987). The Nile perch (Lates niloticus) was introduced into Lake Victoria in east Africa and has caused the extinction of many of the native cichlid fish species.
Figure: Spread of the brown tree snake on Guam (arrows) and subsequent declines in numbers of native birds across years (from Savidge 1987).
(5) Mutualism. This is a third kind of interspecific interaction and one in which both species benefit from the interaction. In such cases, if the mutualism is very specific, the distribution of one species limits the distribution of the other. Few cases of such "obligate" mutualism exist in nature. The red clover, however, did poorly in New Zealand until it’s pollinator (the bumble bee) was subsequently introduced. Most cases of mutualism, however, involve multiple possible species that can take place in the interaction. For instance, there are 8 or so species of hummingbirds in North America which use over 100 species of flowers as a food source. Even though the interaction between these birds and the flowers is obviously mutualistic, there is little evidence of correspondence of geographic ranges of particular species of hummingbirds and plant species. A good exception is the tight, mutalistic interaction between Clark's nutcracker and the whitebark pine trees. The distribution of the pine (especially) is dependent on the dispersal of seeds by nutcrackers which feed on the cones (and hence release the seeds). On the other hand, the cones/seeds provide an excellent food source for the nutcracker. Both species are found with widely overlapping ranges in the Rocky Mountain sub-alpine habitats.
(6) Historical factors. Sometimes particular events in the history of a taxon can have profound effects on the geographic distribution. For instance, the timing of geo/climatological events (e.g. mountain building, development of land bridges, seawater barriers, changes in climate) can influence the possibility of dispersal of that taxon to a geographic area that is suitable (but unoccupied) in terms of its physiological/ecological character. Recall the lack of many freshwater fishes species on Vancouver Island compared to the mainland of B.C. (about 16 on Vancouver Island versus 60 or so on the mainland). Also, note the way in which the distribution of the white sucker (see figure at top of this page) is limited to the headwaters of some large rivers west of the continental divide (e.g. Fraser and Skeena rivers), but not others (e.g. the Columbia River). After the glaciers receded most recently (about 10,000 years ago) the Rocky Mountains were a formidable dispersal barrier to these suckers dispersing from refuge areas in eastern North America. Watershed exchanges between the Peace River (flows east through the Rocky Mountain trench) and the Fraser and Skeena rivers permitted the colonization of these areas by the white sucker. No such "historical accident’’ involved the Columbia River which could explain the lack of the species in that major westward-flowing river even though suitable habitats exist there. It is also suspected that competition with the ecologically simliar largescale sucker (Catostomus macrocheilus), which has a distribution largely limited to west of the continental divide has limited further westward expansion of the geographic range of the white sucker.
(iii) Gene flow, adaptation, and the geographic range
Can the geographic range of a species evolve? Local adaptation is a process whereby traits that increase the survival or reproductive success of individuals within a particular environment increase in frequency in that environment. Local adaptation is driven by natural selection and requires that there be genetic variation for the trait to increase in frequency across generations (i.e. for it to evolve). There is plenty of evidence for local adaptation in natural populations (just look in any standard evolution text or reference). For example, Lewontin and Birch (1966) documented a southward range extension for the fruitfly, Dacus tryoni, into colder climates in Australia. They also demonstrated adaptation by peripheral populations to colder temperatures that appear to have facilitated the range expansion. If such adaptation is common, then why haven’t all species expanded their ranges through local adaptation in peripheral populations? Three possible reasons have been suggested to constrain such adaptation and limit the expansion of geographic ranges: (1) evolutionary constraint, (2) gene flow from central populations, and (3) evolutionary tradeoffs.
Evolutionary constraints simply means that the taxon in question may not have the required "evolutionary potential" to facilitate adaptation. Some freshwater fish, for example, simply don’t have the biochemical or physiological requirements for living in saltwater (i.e. the alleles for saltwater tolerance don’t exist within that taxon and only multiple mutations – rare events – could introduce such potential). Another example is terrestrial animals. They simply don’t have the required traits for living in aquatic environments. Of course, marine mammals did evolve such traits, but it took millions of years for their terrestrial ancestors to adapt to marine conditions and they lost the capacity to live on land!
Gene flow from central populations. Gene flow is the movement of alleles from one population into the other (as animals move and reproduce from one population into another). Gene flow tends to make populations more similar in genetic traits (e.g. those that might be related to environmental adaptation) while differential selection in contrasting populations would tend to make populations different from one another. Imagine a central population that has a high frequency of an allele, let’s call it A, at a gene that codes for thermal tolerance. The A allele is selected for in cold environments, and its alternative form, let’s call it a, is selected for in warm environments. A group of individuals (with both alleles present in the group) founds a peripheral population in a warm environment (relative to the source population) and selection favours the a allele which increases in frequency (the population "becomes better adapted") increasing population growth and increasing the range of the taxon in question. Now, imagine gene flow occurs also (as individuals migrate from the cold to the warm habitat) which results in a continual inflow of the A allele which does poorly in the new environment. If gene flow is high enough, the new colony does poorly in terms of population growth (death rates exceed birth rates) and the colony can only be sustained by continual immigration from the source population (i.e. it is, in effect a sink population). In this case, the chances of this species expanding its range are much poorer and gene flow from the central population may limit further range expansion.
In sum, the potential for local adaptation (and possible range expansion) will reflect a balance between gene flow and selection in the new habitat (high gene flow relative to selection will constrain adaptation, low gene flow relative to selection will promote adaptation). There isn’t any good evidence of a gene flow-selection balance limiting geographic range expansion on a broad, biogeographic scale, but a good example on a local scale involves shading pattern, bird predation, and habitat in water snakes.
See King and Lawson (1997) below for further info.
Fig. Gene flow (mediated by wind direction) and tolerance to mine tailings in plants.
Fig. Schematic of the balance between gene flow and selection between two habitats that can constrain local adaptation in the new habitat and thus limit the evolution of range expansion. The specific dynamics will depend on the relative strengths of gene flow and selection as well as the dominance relationships between alleles (don't worry about that!).
Evolutionary tradeoffs. Simply stated, adaptation in a new environment may be costly and involve trade-offs, i.e. adaptation for thermal tolerance (as an example) may come at a cost that involves reduced adaptation for performance in salinity tolerance, reproductive rate, or feeding behaviour. Such tradeoffs, therefore, may compromise the ability of populations to expand their range. Again, adaptation to a marine existence by ancestors of today's marine mammals likely involved tradeoffs in terms of their abilities, for instance, to move on land.
References:
1. Brown, J.H., G.C. Stevens, and D.M. Kaufman. 1996. The geographic range: size, shape, boundaries, and internal structure. Ann. Rev. Ecol. Syst. 27:597-623.
2. Calsbeek, R. and T.B. Smith. 2003. Ocean currents mediate evolution in island lizards. Nature 426: 552-555.
3. Ehrlich, P.R. 1965. The population biology of the butterfly Euphydryas editha. II. The structure of the Jaspar Ridge colony. Evolution 19: 327-336.
4. King, R.B. and R. Lawson. 1997. Microevolution in island water snakes. Bioscience 47: 279-286.
5. Kirkpatrick, M. and N. Barton. 1997. Evolution of a species’ range. Am. Naturalist 150:1-23. (Theoretical treatment)
6. Lewontin, R.C. and L.C. Brich. 1966. Hybridization as a source of variation for adaptation to new environments. Evolution 20: 315-336.
7. Lindsey, C.C. 1964. Problems in the zoogeography of the lake trout, Salvelinus namaycush. J. Fish. Res. Bd. Canada 21: 977-994.
8. Nelson, J.S. 1968. Hybridization and isolating mechanisms between Catostomus commersoni and C. macrocheilus (Pisces: Catostomidae). J. Fish. Res. Bd. Canada 25: 101-150 (see pp. 113-115).
9. Saccheri, I. And 5 co-authors. 1998. Inbreeding and extinction in a butterfly metapopulation. Nature 392: 491-494.
10. Savidge, J.A. 1987. Extinction of an island forest avifauna by an introduced snake. Ecology 68: 660-668.