The paper integrates islands, climate change, and invasive biota. Islands have always fascinated biologists, and research on islands has revealed key aspects of biological and ecological theory, such as evolution, speciation, biogeography, and ecosystem dynamics. Climate change currently holds the interest of all humankind. Scientists, politicians, and the public are asking questions like: Is it really happening? Has it happened before? What will its effects be—how soon will we be able to detect them, can we predict them? The biology and ecology of invasive organisms is another important and very topical field of study worldwide and my paper gives examples of how climate change and invasive organisms interact to influence the biota and ecology of Marion Island.

No—although the Marion Island climate data set was one of the first to be analyzed to show how intense warming is occurring in the sub-Antarctic, that was done 14 years ago. The climate analysis in the paper is only a more up-to-date one. Most of the biological and ecological responses to climate change and alien biota have been reported elsewhere—the paper is more a review. Its main purpose is to show that sub-Antarctic islands, which (unlike continental Antarctica or the maritime Antarctic) have never really been "on the map" in terms of playing a prominent role in international biological studies, have much to offer in furthering our understanding of how organisms and ecosystems respond to climate change.

Marion and Prince Edward Islands are situated in the sub-Antarctic region about 2100 km southeast of Cape Town. Their climate is cold and wet; annual mean temperature is 5.6°C and mean annual precipitation (mainly rain rather than snow) is 2326 mm per year. The high precipitation is associated with much cloudiness so that on average there is less than 4 hours of sunshine per day. Winds are fierce and gales occur on about 100 days per year.

However, the islands’ climate is changing. Annual mean air temperature on Marion Island has increased by 0.04°C per year since 1969. Annual precipitation at Marion Island has decreased since the mid 1960s, so that the 1990s was the driest of the five decades that precipitation has been measured at the island. The annual total sunshine hours showed an average increase of 3.3 hours per year between 1951 and 2002.

Hence, Marion Island (and almost certainly also Prince Edward Island, which is only 22 km away but for which there is no climate data) is becoming warmer, drier, and sunnier. These climatic changes have implications for the islands’ plants and animals, which have evolved under the cool, humid conditions typical of sub-Antarctic islands. To understand these implications it is necessary to realize that there are differences in the way ecosystems develop on oceanic islands compared with on continents. At the heart of this difference is that continental ecosystems have a large pool of species to draw on, whereas oceanic islands are remote, initially have no species on them, and their ecosystems depend on whatever plants, animals, and microbes can reach them by long-distance dispersal. Marion and Prince Edward Islands, for instance, originated as undersea volcanoes only about a half-million years ago, long after Africa split off from Antarctica and migrated north to its current position. The islands were thus never connected to, or even near, any continent. The pool of plant and animal species on the islands is made up a relatively few species that managed to somehow reach the islands over long distances across the ocean. Many important functional groups of plants and animals are not adapted to such long-distance dispersal and so do not occur on the islands naturally (i.e., they are not indigenous to the islands). By functional group is meant a group of species that "do the same sort of thing" ecologically. Examples of such groups that do not occur on Marion or Prince Edward Island are terrestrial mammalian herbivores and carnivores such as the cud-chewing buck and deer and the predatory cats and dogs that are so characteristic of the ecology of continents, especially Africa. Less conspicuous (but ecologically extremely important) animals are also absent; for example there are no frogs, reptiles, rodents, or rabbits. Even the insect fauna of the islands is species-poor, although in some habitats the few insect species that do occur may be present in high numbers. Overall, then, like all oceanic islands, biodiversity (species-richness) at Marion and Prince Edward Islands is low.

This low biodiversity, especially the absence of important functional groups such as herbivores and carnivores, affects ecological functioning at the islands. Ecological functioning has to do with the flow of energy and the cycling of nutrients in the ecosystem. Energy reaches the island as sunlight, which is fixed by plants and results in their growth. The amount of vegetation growth over a year is termed the annual primary production. Since there is no dry season and no really bitterly cold weather to stop plant growth, annual primary production is high on the island. The vegetation needs to take up a substantial amount of nutrients to support that growth. Climatic warming might be expected to increase productivity and the demand for nutrients even further. But will it? In most ecosystems, herbivores eat the produced plant material and excrete a portion of the nutrients—they thus play an important role in recycling nutrients, making them available for re-use by the plants. Without herbivores (even the insects feed mainly on plant litter and microorganisms) the nutrients taken up by the growing plants remain trapped in the plant material, which dies to form plant litter. Decomposition of the litter releases the nutrients in a form that can be taken up by plants again. Hence, nutrient recycling on the island occurs mainly through decomposition, rather than grazing. However, decomposition (and hence nutrient release) occurs only slowly in the cold, wet island soils, unless assisted in some way. Insects (especially the larvae of moths, weevils, and flies), snails, and earthworms provide that assistance, by feeding on plant litter. The litter is partly broken down in the animal’s gut and egested in a form that is easier for soil microorganisms such as bacteria and fungi to break down further. Insects and earthworms thus play an important role in the processes of decomposition and nutrient cycling; in fact, it has been estimated that they make available up to 88% of the nutrients required by some vegetation types on the island.

Since the insects and earthworms are cold-blooded, their activity is strongly temperature-dependent. Increasing temperature, as is happening at the islands, should thus result in enhanced rates of litter consumption and hence of nutrient release, which will allow the potential for increased primary production due to elevated temperature to be realized. But will this happen?

House mice (Mus musculus, the same species that can be caught in most houses the world over) occur on Marion Island, having been introduced there by sealers. The mice feed mainly on adults and larvae of moths, weevils, and flies, and on earthworms and snails, the very same invertebrates that are so important in driving ecosystem functioning on the island. It is estimated that that mice annually consume between 1 and 6 times the average population size of their invertebrate prey species. Such high consumption rates have a particularly severe effect on the island’s invertebrates, as is clear from comparisons of the invertebrate populations of Marion Island with nearby Prince Edward Island, where mice do not occur. There are big differences in the size, structure, and composition of the insect populations of the two islands, and also in the maximum body size attained by the various species. Mice are even preventing speciation of weevils on Marion Island by feeding preferentially on adult weevils of a specific size.

This predation by mice on soil invertebrates is also affecting other components of the Marion Island biota. For example, the lesser sheathbill (Chionis minor) is the only non-migratory bird species on the island and relies on soil macroinvertebrates as food in winter. Between the mid-1970s and mid-1990s the sheathbill population on Marion Island decreased by 23% whereas the one on Prince Edward Island did not change. Another example is the sedge, Uncinia compacta, which rarely manages to set seed on Marion Island since mice remove the seed long before it ripens. As a result, the sedge is a much more important component of the vegetation on Prince Edward Island than it is on Marion Island.

More insidious, but probably even more profound than these direct effects of house mice on the island's biota, is their influence on ecosystem function through their removal of the cardinal agents of energy flow and nutrient cycling at the island. For instance, it is estimated that predation by mice on moth larvae alone prevents the consumption of 1000 kg plant litter per hectare per year, a decrease of about 40% compared with what would be consumed in the absence of mice. There is evidence that the island’s mouse population has increased since it was first studied in 1979/80, possibly due to climatic warming and/or the fact that the island’s feral population of domestic cats was eradicated in the early 1990s. An increasing mouse population translates into greater predation pressure on the invertebrates and hence in lower rates of nutrient cycling. This will result in less nutrients being available to the plants which will result in a lower primary production and, more importantly, the plant material that is produced will have a low nutrient quality. This will lead to low-quality litter which decomposes even more slowly. The net result will be that peat will accumulate faster than it does at present. Peat accumulation is a major driving force for ecological succession on the island. By "ecological succession" is meant the process by which one type of plant community replaces another. Most of the habitats on the islands are very wet—the water table is close to or even at the surface, as in bogs and mires. As peat accumulates it raises the whole community above the watertable and allows the development of a drier type of community.

House mice are only one example, albeit a striking one, of the far-reaching effects that an introduced organism can have on the biology and ecology of Marion Island. Climatic warming can be expected to increase the ease with which the island can be invaded by alien species and there is already evidence of this. The Kerguelen cabbage (Pringlea antiscorbutica) is found only on four sub-Antarctic island groups, including the Prince Edward Islands. The species is one of the last—and perhaps the only—remaining relic of a once extensive circum-Antarctic flora and has a special place in sub-Antarctic folklore since it was used by sailors, sealers, and whalers to prevent scurvy. On Marion Island its distribution and abundance has declined alarmingly over the past 20 years for several reasons, all to do with invasive alien biota. The European slug (Deroceras caruanae) was introduced to the island in the mid 1960s and increased in abundance and distribution in the 1990s. The Kerguelen cabbage is one of its preferred food items there. The Diamondback cabbage moth (Plutella xylostella L,), a major pest of crucifers (cabbages, cauliflower, brussel sprouts), is a recent arrival on the island (first discovered in 1986). It severely affects P. antiscorbutica, which is its only host plant there. Botryotinia fuckeliana, the fungus that causes grey mould rot in crucifers and other vegetable crops, has also reached the island through vegetables sent as food for the island personnel (a practice that has been discontinued). Many stands of the cabbage have been infected by the fungus, whole plants collapsing into a black slimy residue.

The study of oceanic islands has been cardinal in developing and testing key aspects of biological and ecological theory, such as evolution, speciation, biogeography, and ecosystem dynamics. The account given above suggests that sub-Antarctic islands have particularly much to offer in furthering our understanding of a topical and important theme in biology worldwide—how organisms and ecosystems respond to climatic change. Global warming is especially intense in the sub-Antarctic region and because of their isolation, species-poor biota, and harsh environments; sub-Antarctic island ecosystems are relatively simple and sensitive to perturbations. They are therefore ideal "ecological laboratories" for studying how organisms, ecological processes, and whole ecosystems respond to a changing climate.

The results of the first South African biological research expedition to Marion and Prince Edward Islands were published as a book in 1971, when I was studying for a B.Sc. Honors degree. In that study I was exposed to the classic accounts of the biological exploration of islands by the greats such as Darwin, Wallace, van Steenis, and a host of others, and also to MacArthur and Wilson’s theory of island biogeography, which had been published only a few years before. Those works dealt with places that seemed exotic and inaccessible, certainly for a young scientist with no influence. When I read the book on the Marion and Prince Edward Islands expedition I realized that the opportunity was right on my doorstep. I made inquiries as to whether there were plans for further expeditions to the islands and the result has been a wonderful career in which I have been fortunate to have visited the islands about 40 times.