Search

Aim The assumedly anomalous occurrence of savannas and forest–savanna mosaics in the Gran Sabana – a neotropical region under a climate more suitable for tropical rain forests – has been attributed to a variety of historical, climatic, and anthropogenic factors. This paper describes a previously undocumented shift in vegetation and climate that occurred during the early Holocene, and evaluates its significance for the understanding of the origin of the Gran Sabana vegetation. Location A treeless savanna locality of the Gran Sabana (4°30′–6°45′ N and 60°34′–62°50′ W), in the Venezuelan Guayana of northern South America, at the headwaters of the Caroní river, one of the major tributaries of the Orinoco river. Methods Pollen and charcoal analysis of a previously dated peat section spanning from about the Pleistocene/Holocene boundary until the present. Results Mesothermic cloud forests dominated by Catostemma (Bombacaceae) occupied the site around the Pleistocene/Holocene boundary. During the early Holocene, a progressive but relatively rapid trend towards savanna vegetation occurred, and eventually the former cloud forests were replaced by a treeless savanna. Some time after the establishment of savannas, a marked increase in charcoal particles indicates the occurrence of the first local fires. Main conclusions The occurrence of cloud forests at the Pleistocene/Holocene boundary contradicts the historical hypothesis according to which the Gran Sabana is a relict of the hypothetical widespread savannas that have been assumed to have dominated the region during the last glaciation. The first local fires recorded in the Holocene were on savanna vegetation, which is against the hypothesis of fire as the triggering factor for the establishment of these savannas. Climate change, in the form of global warming and a persistently drier climate, emerges as the most probable cause for the forest–savanna turnover. [ABSTRACT FROM AUTHOR]

Aim Within the past few decades, seven of the 25 historically described populations of American pikas ( Ochotona princeps) in the Great Basin of arid western North America appear to have become extinct. In this paper, the prehistoric record for pikas in the Great Basin is used to place these losses in deeper historical context. Location The Great Basin, or area of internal drainage, of the western United States. Methods The location, elevation, and age of all reported prehistoric Great Basin specimens of American pikas were extracted from the literature. Elevations of extinct pika populations were arrayed through time, and latitudes and longitudes of those populations used to determine changing distances of those populations from the nearest extant populations. Results The average elevation of now-extinct Great Basin pika populations during the late Wisconsinan ( c. 40,000–10,000 radiocarbon years ago) and early Holocene ( c. 10,000–7500 years ago) was 1750 m. During the hot and dry middle Holocene ( c. 7500–4500 years ago), the average elevation of these populations rose 435 m, to 2168 m. All prehistorically known late Holocene ( c. 4500–200 years ago) populations in the Great Basin are from mountain ranges that currently support populations of this animal, but historic period losses have caused the average elevation of pika populations to rise an additional 152 m. The total elevational increase, from the late Wisconsinan and early Holocene to today, has been 783 m. As lower elevation pika populations were lost, their distribution increasingly came to resemble its modern form. During the late Wisconsinan, now-extinct pika populations were located an average of 170 km from the nearest extant population. By the late Holocene, this distance had declined to 30 km. Main conclusions Prehistoric alterations in the distribution of pika population in the Great Basin were driven by climate change and attendant impacts on vegetation. Today, Great Basin pikas contend with both climate change and anthropogenic impacts and thus may be on the brink of extinction. [ABSTRACT FROM AUTHOR]

The vegetation of subantarctic Campbell Island consists mainly of lowland Dracophyllum scrub and upland tussock grassland and tundra. Soon after the island was discovered in 1810, occupation by sealers and whalers led to localized burning and cutting of scrub. Further burning and cutting took place as a result of sheep farming between 1894 and 1931. Since the earliest photographs and vegetation descriptions of the island in the late nineteenth century, scrub cover has expanded markedly. Also, since the 1960s, the island has become warmer and drier. The aim of this paper is to quantify scrub changes as depicted in photo-sequences from the island, and to establish if the spread of scrub is a result of the cessation of anthropogenic activities or a response to global warming. Campbell Island, an isolated island 600 km south of New Zealand mainland (52°33.7′S, 169°09′E). Changes in scrub cover in the island were established through analysis of 33 photographic sequences in the island, the earliest photos dating back to 1888. Visual estimates were made of percentage woody cover in matching areas of repeat photographs, and of changes in scrub distribution with altitude and landscape type. Dracophyllum scrub was restricted at the time of the first photographs (1888) relative to its present extent, and earlier written observations suggest that there had been little change from the 1840s. Scrub cover has subsequently increased in most areas. Most of the spread occurred after the cessation of farming. Change has mostly consisted of a thickening and expansion of pre-existing scrub patches. There is no indication that the upper elevational limit of scrub has increased. Initial reduction of scrub was probably due to anthropogenic fire in the early nineteenth century, although it is possible that less favourable climates had also restricted its distribution. Failure of the scrub to regenerate significantly between 1840 and c. 1895 may also have been a consequence of generally cooler, wetter climates at this time. Rapid scrub expansion began between the end of farming and burning in 1931 and 1988 when the remaining feral sheep were removed. Sheep grazing in the absence of fire reduced competition, and encouraged regeneration and growth of woody scrub. Spread was probably assisted by a pronounced shift to warmer, drier climates in the second half of the twentieth century. Upper elevational limits of scrub have not increased, suggesting that factors other than summer temperature are controlling scrub line in this superoceanic environment. [ABSTRACT FROM AUTHOR]