by Pavel KRESTOV, Dr. Sc. (Biol.), Biological and Pedological Institute, the Far Eastern Branch of the Russian Academy of Sciences (Vladivostok)

Man is prone to look into the future, and numerous scenarios of global climatic warming or cooling draw by no means an optimistic picture of the natural environment in the centuries to come. In the meantime, there have always been organisms capable of surviving in the most severe periods of global history, and therefore their record is of great value. Despite the existing climate, plant communities and whole ecosystems keep developing-those quite common some time ago and now using unique properties of their habitats or refugia (refuges). Plants and animals inhabiting such areas have retained special adaptation mechanisms not manifest under present-day conditions. Will their hour come?

ICE AGE: A "WHITE SILENCE"?

Vast glaciations of the Late Pleistocene on the European and North American continents were associated with climatic conditions fatal to life. During the Pleistocene maximum (18-20,000 years ago) an ice sheet up to 3,000 m thick covered almost the whole territory of Canada, thousands of kilometers of "white silence"... Only rocks of the Coast Range cut across a dead expanses in a strip of the far west of North America connecting the Cordilleras, covered with forest vegetation exuberant for that period, with "frozen" but not ice-bound North Alaska, Chukotka and now extinct Beringia.

At the same time extremely severe climatic conditions in Asia made rank vegetation growth virtually impossible, as proved by polynological spectra and remains of insects home to cold desert ecosystems. The obtained paleodata gave birth to a special term for the dominant native biome, i.e. "tundra steppe". However, Pleistocene megafauna findings broke harmony of the concept. Many a finding made it possible to suggest that mammoths, wooly rhinoceros, bisons, horses and beasts of prey associated with them were numerous even during the Pleistocene maximum. How could the low-yielding tundra steppes feed these animals and sustain the composite and diverse foodchains? The ventricles of big plant-eating mammals buried by Pleistocene mud streams were filled with succulent grass, not stiff Arctic shrub. Consequently, at least grassland ecosystems were widespread.

Studies of the post-Pleistocene history of the plant cover led to an unexpected conclusion that the tree vegetation whose existence on the American continent north of the glacier sheet was traditionally excluded, actually spread there in no time after the glacier's retreat and formation of narrow and ice-free corridors. Doubts in regard to the "islands of life" amid the "white silence" were fully dispelled.

An outstanding explorer of the Arctic, the Swedish botanist and phytogeographer Eric Hulten already in the 1920s and 1930s postulated, contrary to the widespread idea that the zone of the Late Pleistocene glacia-tions was absolutely no good for plants, that refugia* keep up even under the most extreme conditions, and he substantiated it from phytographic positions.

The biota of refugia not only concentrated the gene pool contrasting with the background gene pool but was also a source of species dispersion with the change in habitation conditions. This phenomenon is an essential factor of plant cover growth in the Early Holocene. Do we have the right to think of it as a historic past and forget it when forecasting the future? Where are such refuges in the modern biosphere? Who "hides" in them? How will their inhabitants respond to future fluctuations in climate?

WHY ARE REFUGIA STILL THERE?

As shown by biogeographic data, a substantial share of biota diversity on a regional level falls not on zonal ecosystems occupying sometimes vast expanses, but on local areas representing certain geomorphological, geo-chemical or locally climatic anomalies, such as calcareous and ultrabasic rocks, sand deposits, rock exposure, etc. As a rule, local biotas differ greatly from background regional biotas and are characterized by high endemism.

The phytogeographical and floristic literature places high emphasis on relict species hailing from the flora of past geological epochs. However, things do not make any headway after a list (often rather speculative) indicating the time of origin has been drawn. The classical research works on relicts of North Asia by Boris Yurtsev, Yuri Kozhevnikov and Andrei Khokhryakov, all Drs. Sc. (Biol.), carried out at the Komarov Botanical Institute of the RAS (St. Petersburg) and the Institute of Biological Problems of the North, the Far Eastern Branch of the RAS in the 1960s-1980s, are rather an exception to the rule.

Meanwhile, even if having information on the relict diversity and localization of modern refugia, we have but a scant knowledge on the adaptation of communities formed there to habitation conditions predominant some time ago. Plant species developing in a quite different climate for a long time with a set of characters, which enabled them to survive far back in the past, are unable to dominate mostly due to constraints imposed by zonal biotas. However, with the next climatic change their characters, useless today, can become manifest again.

Climate is one of the main factors of biota transformation in ecosystems on a global level. Vegetation responds to climate changes in the first place, and the flora (regional phytome) responds slower. Climatic cooling moves inevitably a majority of organisms closer to the limits of their ecological tolerance, and there comes a time when temperatures become too low and the growing period excessively dry for the further existence of species. If it is a local habitat, plants will die and so will a whole ecosystem where they were primary producers. But when events occur in a landscape, some of its habitats can make up for the deficit of a climatic resource: if it is too cold, there is a warm sunny southern slope for the plants; if it is scorching dry, there are depressions, the ephemerous acqueous runoffs accumulating moisture; if soils are deeply frozen, there are areas in which snow is accumulated in wintertime. Species extinct everywhere have a chance to survive in such refuges.

The so-called competitive exclusion is another reason for species extinction from habitat. If an environment becomes comfortable for the development of a certain species, the only way to preserve its viable populations is to find specific conditions mitigating the action of aggressive "neighbors". Most often these are special substrates, unfavorable for the development of dominant species in an ecosystem (exposure of ultrabasic and calcareous rocks, highly drained soils, permafrost or areas with soils thawing late in summer).

Sometimes animals tip the scales in the struggle between potential predominants of the plant cover in

* See: V. Orlov, "The First Model of Phyletic Evolution", Science in Russia, No. 4, 2009. -Ed.

favor of a weaker competitor. Ecosystems formed in the tundra around the colonies of burrowing mammals* (e.g. gophers or ground squirrels), actually become "self-made" oases (constructions made by ground squirrels and other rodents are sometimes even more impressive than "creations" of some city designers), providing good survival chances in the tundra for more thermophilic grasses in need of good mineral nutrition. In their turn, such grasses are a valuable feed for rodents.

Thus, refugia enable species populations and sometimes even whole biotic communities to survive and develop for a long time in isolation from aggressive environments.

SUCCESSORS OF ARCTO-TERTIARY COMMUNITIES

Not glaciers or even low temperatures were the factors which inflicted the greatest harm on the Arcto-Tertiary flora of Asia. The heaviest aridization of the climate in the Pleistocene and especially during the Late Pleistocene maximum ousted the thermophilic and moisture-dependent vegetation from this continent, while sustaining other composite ecosystems. In the temperate latitudes a mosaic of dry meadow-like communities appeared, one that might have no analogs at present, while the savanna-like vegetation with the predominance of grasses came into being in the subtropics. Mesophytic complexes could not survive, what with the harsh water shortage; but cold-resistant xerophytes, mostly of the Central Asian origin, were widespread.

These processes led to significant rearrangements of the plant cover both in the northern and southern latitudes, which created in a sense a "zero reference point" in the Late Pleistocene for the modern landscape. However, some species and associated vegetation complexes, which have ancestry in the Tertiary period, survived in refugia and quickly dispersed under favorable conditions. Localizations of such shelters are now difficult to identify, as the environment existing there for thousands of years leveled out over vast areas. Nonetheless, information on them is very important both for a reconstruction of the Holocene scenario and for studying the history of many species dominating in modern ecosystems.

At latitude 22-25° North the dipterocarp** and palm flora disappears practically synchronously from the plant cover in Asia's continental regions, and the tropic of flora is quickly supplanted by subtropical vegetation, enriched in temperate subtropical, and then by lukewarm climate species. On islands (on Taiwan and Ryukyu Archipelago) the above change in the proportion of species is slower mainly due to the lack of seasonal differentiation of the oceanic climate of both temperature and transpiration-related nature. For example, on Ryukyu Archipelago the evergreen lukewarm climate vegetation occupies a zonal position, while insolation habitats serve as a refugium for tropic floristic complexes with such most striking representatives as palms (Satakentia liukiuensis, Arenga tremula) and aroid*** (Epiptremium pinnatum, Alocasia odora).

At latitude ~35° North the evergreen broad-leaved species "fall out", which corresponds to the zero isotherm of absolute minimal temperatures in Asia. The main habitats of the moderate-cool zone are occupied by communities of the crenate beech (Fagus cre-nata) in Japan and the Mongolian oak (Quercus mon-golica) on the Korean Peninsula. On sea islands evergreen representatives of the Cyclobanopsis and Castano-psis genera of the beech family and Myrica from wax myrtles reach 38° North (the southern coast of the Honshu Island), where they are in an original seaside refugium. Here the mitigating action of oceanic air

* See: A. Ivanov, "Zoogenic Landscape Complexes", Science in Russia, No. 3, 2009. -Ed.

** A major part of species of the Dipterocarpaceae family, evergreen giant (up to 60-75 m high) trees with an upright columnar trunk; occur in India, on the Sri Lanka Island, Andaman Islands and Seychelles, in Indochina, New Guinea and in the tropics. -Ed.

*** Representatives of the aroid family, i.e ground, fenny and acquatic grasses with lengthened rootstock, occur mostly in tropical countries (lianas, epiphytes) and often reaching giant sizes there. -Ed.

The south-eastern coast of Hokkaido is optimal for the beech FAGUS CRENATA, but it does not grow there. Photo, A. Omelko.

masses rednesses the zonal deficit of heat and the negative temperature periods.

At latitudes 38-40° North representatives of the following generu disappear from the Asian moderate continental flora: Alangium (alang family), Celtis (elm), Lindera (laurel), Magnolia (magnolia), Toxicodendron (cashew), etc. The vegetation of this belt is represented by broad-leaved forests, and in zonal habitats aggregations of mixed broad-leaved and cedar forests replace communities of the Mongolian oak. And again, in the island sector of East Asia under oceanic climate conditions thermophilic representatives of the temperate flora, including the above-said Cirsidophyllum, Magnolia, Toxicodendron, reach latitudes 44-45° North as far as the dark coniferous taiga.

Another important climatic refugium is represented by the eastern macroslope of Sikhote Alin which, collecting a maximal amount of precipitations in the southern part of the Russian continental Far East, is characterized by a great number of foggy days. Here, in isolation from the main parts of areals, the following representatives of the Japanese flora have found the shelter: short-fruited rhododendron (Rhododendron brachycarpum), the primrose (Primula jezoana), the savoyed holly (Ilex rugosa), and also a number of species, which have the nearest relations on the Japanese Archipelago, such as, for example, the Oplopa-naxelatus (devil's-club), the systematically close O.japoni-ca (Japanese devil's-club), probably surviving Pleistocene events as part of the Far East fir and spruce taiga.

Ranking the Far East broad-leaved and cedar forests among direct successors to the Turgay vegetation will makes sense if we consider the eastern Sikhote-Alin as a refugium for a complex of mesophilic species having ancestry in the Pliocene. The existence of thermophilic and moisture-dependent plants there allows us to suggest that this region was one of the key points of dispersal of the abnormal (broadleaf wood) flora which formed later into a unique biome known as "the Ussuri taiga" occupying now the territory of the Primorye Maritime Region and the Amur River Basin.

ROLE OF THE INSULAR AND COASTAL ECOSYSTEMS

As said above, thermophilic and moisture-dependent floristic complexes move northward through island arcs and coastal mountainous territories.

Due to the relatively high humidity in the Late Pleistocene marked by aridization of the climate, the oceanic

sector of North-East Asia was probably the most important refugium for the flora formed on the eastern margin of the continent in Pliocene. Our research proved, that the amphi-Pacific* species like the Parageum calthifoli-um of the rose family, the Langsdorf violet (Viola langs-dorfii), the Unalaska arnica (Arnica unalaschcensis) of the aster family, the Aleutian phyllodoce (Phyllodoce aleuti-ca) of the heather family, etc. are now distributed in a rather remarkable manner: Japan-Kuriles-Aleutians, i.e. their areal is a thin strip of land along the North American coast down to California. This pattern testifies to the degradation of the erstwhile much wider areal due to lack of moisture during the Pleistocene maximum. In honor of Eric Hulten, the author of the pioneering works on this region, his colleague, the Japanese geobotanist Misao Tatewaki, named the present propagation area of these plants Hultenia.

The main floristic complexes of the Kamchatka Peninsula with its boreal suboceanic climate also survived the severe period in refugia and formed an original vegetation cover. One of the most essential factors of differentiation of the latter is depth of a mantle of snow which protects soils from freezing but melts for a long time (2-3 weeks); that is why the vegetative season decreases. High air temperatures, in combination with cold soils and high humidity in early summer, sustain the development of unique tall grasses of the Far East.

Together with the zonal vegetation of Kamchatka, the Ajan spruce (Picea jezoensis) and graceful fir (Abies gracilis) survived the Pleistocene maximum. No direct paleobotanical evidence of their existence as a part of the local flora in that period is found yet, but the genetic heterogeneity of their populations and genetic variations from the nearest relations of these species on the continent and islands are indicative of their early isolation on the peninsula.

Migration processes in the Holocene were an essential factor for the formation of the present-day flora of the region. However, its current habitus was determined by Pleistocene refugia of the moderate and boreal vegetation of eastern Asian origin. In the southern part of Sakhalin scores of species of the Japanese and Japanese-Manchu floras are found, such as the ursine angelica (Angelica ursine), the umbrella leaf (Diphylleia grayana), the savoyed holly (Ilex rugosa), the Kalopanax septem-

* Amphi-Pacific species occur on the western and eastern coasts of the Pacific Ocean, not in continental inland regions.-Auth.

The PARAGEUM CALTHIFOLIUM and some other species are indicative of phytogeographic links between Asia and North America in the Pleistocene. Their habitat is known as Hultenia. Photo, P. Krestov.

lobum, etc. Although the cold boreal climate is not favorable for this vegetation complex, deep snow and unfrozen soils help nemoral elements to survive.

MAMMOTH-POPULATED LANDSCAPE

According to paleoreconstructions, the ice-free expanses of North-East Asia and Alaska in the Late Pleistocene were under by grass tundras and dry scattered tundra-like clusters in convex ground features as well as under cotton grass, sedgy and osier communities in concave ground features with maldrainage. Were mammoths an element of such ecosystems? The director of the North-East Biological Station of the Pacific Institute of Geography of the Far Eastern Branch of the Russian Academy of Sciences Sergey Zimov *, the mastermind of the "Pleistocene Park" project, does not deny this possibility. However, while trying to reconstruct the habitus of the ancient flora from the present one, we encounter many problems. The first and perhaps the most important thing here is the absence of a great number of big grass-eating mammals at present.

The modern boreal and arctic ecosystems of North-East Asia are a product of the longtime development of vegetation on permafrost with as good as no animals of this kind.

Depth of the active layer of soil (one that thaws out in the vegetation period), which depends on the climate and, in particular, on the amount of solar heat, is a significant, if not the main, factor conditioning the flora in permafrost regions. Meanwhile, cold-resistant and hardy plants like the lichen, moss and small shrubs form a compact ground cover, thus isolating the soil from solar heat and becoming a kind of thermos for permafrost rocks. Only the active destruction of the moss-lichen cover by animals could provide for a deeper warming of the surface soil as well as faster decomposition of organic remains and their incorporation in the topsoil.

In a very cold and extremely dry climate the plant cover was not uniform: the semiarid ecosystems of microelevations alternated with relatively humid and lush grass-shrub depressions, which favored a very wide occurrence of frigostable and drought-enduring plants formed in inland continental regions over a long time.

The thermophilic and moisture-loving vegetation complexes occurring on the whole territory of Asia in

* See: S. Zimov, "Mammoth Steppes and Future Climate", Science in Russia, No. 5, 2007. -Ed.

the Tertiary period have completely degenerated by now. Since water was much in deficit, some representatives of the arcto-tertiary flora could survive in moist coastal mountain systems where an oceanic warming effect was pronounced.

Today the available paleobotanical evidence on the survival of tertiary flora successors in such refugia has been confirmed by molecular genetics data. Vladimir Poten-ko, Dr. Sc. (Biol.) (Gomel State Medical University, Belarus), and Svetlana Semerikova, Cand. Sc. (Biol.) (Institute of Plant and Animal Ecology, the Ural Branch of the Russian Academy of Science), have demonstrated that populations of the Ajan spruce and graceful fir isolated on the Kamchatka Peninsula date from the Pre-Pleistocene time. Similar results have been obtained also for representatives of the lukewarm climate vegetation on the Honshu Island.

While the savanna type flora dominated at the same latitudes in the continental regions of East Asia, this island was covered with dark coniferous forests, and representatives of the above species of the Castanopsis and Machylus genera and evergreen species of the Quercus genus persereved in coastal mountain refugia.

Larch forests covering the boundless Siberian expanses may look dull and monotonous. However, phytogeographical data show that the species composition of their Pacific communities is tenfold as diverse as that of their continental analogs. Molecular-genetic studies carried out recently at the Institute of Plant and Animal Ecology by Vladimir Semerikov, Dr. Sc. (Biol.), and colleagues have demonstrated that the vast larch-tree populations argue in favor of the existence of larch refugia during the Pleistocene maximum on the coast of the Sea of Okhotsk, on the Kamchatka Peninsula, on the Sakhalin Island and in the south of the Russian Far East.

PLEISTOCENE TRACES IN THE PLANT COVER

In a period from 10 to 6,000 years ago the flora of North Asia underwent radical changes. Of particular note are larch forests which spreading most in the north of Siberia and Far East, advanced as far as the coast of the Arctic Ocean. Meanwhile the Pleistocene megafau-na disappeared, and people settled for and wide in the region. Mankind in the meantime reached quite a new technological level thanks to the better efficiency of hunting weapons.  Different cause-effect versions of these events have been discussed; however it is obvious that since that time the role of animals in northern ecosystems has essentially decreased.

The propagation of cold and drought-resistant species in climatic conditions favorable for the heat- and moisture-loving flora is indicative of Pleistocene maximum traces in the present day plant cover. Such relicts are typical of all vegetation zones of Asia.

Taiwan, situated as it is on the border of tropics and subtropics and being small in size, has a pronounced altitudinal zonality of the plant cover. The upper forest-land border here lies 3,500 m above sea level and even higher. Being a part of the continental land in the Pleistocene, this island felt the effect of a dry and arid continental climate as seen in the presence of cold-resistant Pleistocene relicts.

Migrations of that time resulted also in a cold-resistant complex of dark coniferous forests of the Morrison spruce (Picea morissonicola) and the Kawakami fir (Abies kawakamii) associated with the following herbaceous plants growing in temperate latitudes: the Alpian circaea (Circaea alpina), the Ito wintergreen (Gaultheria itoana), stiff club-moss (Lycopodium annotinum), etc. The taiga forms a pronounced upper belt of vegetation, and its lower border is at a height of 2,500 m.

Communities of the Taiwan beech (Fagus hayatae) are another essential Pleistocene relict on Taiwan. Its prop-

agation here is bounded by the Hsuyeshan Range, which is 18 km long, and by the dominating elevations of 1,300-2,000 m above sea level. The Taiwan beech is largely isolated from other representatives of its genus, and their nearest relations, the Japanese beech (Fagus japonica) and the crenate beech (F. crenata), represent the zonal vegetation in the northern half of the Honshu Island. The beech is tightly associated with its suite of species systematically close to Japanese ones: the Japanese maple (Acer palmatum), the Dendropanax pelluci-dopunctata, the Taiwan rhododendron (Rhododendron formosanum).

The most noticeable Pleistocene relict complexes representing the drought-resistant vegetation of Japan are located in the central part of the Honshu Island between massive mountain systems of the Japanese Alps. The existence of a refugium here is possible due to more continental conditions in the inland regions of the island. It is an oak complex closely related to contemporary oak forests of the south of the Russian Far East and including as dominants the Mongolian oak (Quercus mongoli-ca s. str.), the Daurian birch (Betula davurica) and the Schmidt birch (B. schmidtii) and their suite known from the Russian Sikhote-Alin: the Keiske sagebrush (Artemisia keiskeana), the sprout-bearing sagebrush (A. stolonifera), the lanceolate sedge (Carex lanceolata), the asperous aster (Doellingeria scabra), the bicolor les-pedeza (Lespedeza bicolor), and also a number of species systematically close to the Far Eastern ones, such as the Japanese atractylodes (Atractylodes japonica) of the aster (Compositae) family.

The steppificated communities beautifully described by Boris Yurtsev, Dr. Sc. (Biol.), are of special importance for an understanding of the Pleistocene history of North Asia. Their propagation over vast expanses from the Amur River basin up to the coast of the Arctic Ocean in the north is limited to small edaphic (soil-related) refugia: the steep slopes of southern expositions, the sand deposits of sea and lake shores and river banks. A specific set of species is typical of them, such as the pectinate wheat grass (Agropyron cristatum), the pasture sagebrush (Artemisia frigida), the cleistrogene squarrose (Cleistogenes squarrosa), the sheep's fescue grass (Festuca ovina), the pectinate June grass (Koeleria cristata), the Krylov feather grass (Stipa krylovii)... But in the north steppificated communities are characterized by numerous endemics confined to a certain type of habitat.

The current approaches to a predictive modeling of plant cover development are based mostly on environmental factors as described by a number of variables characterizing the climate, insolation level, soil conditions, geomorphology, geology and the ongoing natural processes. At the same time, one of the crucial formative factors of the plant cover is in its history. But how can we measure history? How shall we formalize such kind of information for mathematical models? These questions are still open. One of the thinnest leads to the perennial question, "What next?", is perhaps anchored in the genome of plants growing in refugia.