The influence of climate change on the distribution of indigenous forest in KwaZulu-Natal, South Africa

被引:203
作者
Eeley, HAC [1 ]
Lawes, MJ [1 ]
Piper, SE [1 ]
机构
[1] Univ KwaZulu Natal, Sch Bot, Forest Biodivers Programme, ZA-3209 Pietermaritzburg, South Africa
关键词
indigenous forest; distribution; climate change; bioclimatic models; South Africa;
D O I
10.1046/j.1365-2699.1999.00307.x
中图分类号
Q14 [生态学(生物生态学)];
学科分类号
071012 ; 0713 ;
摘要
Aims (1) To define the physical correlates of indigenous forest in KwaZula-Natal province and develop a model, based on climatic parameters, to predict the potential distribution of forest subtypes in the province. (2) To explore the impact of palaeoclimatic change on forest distribution, providing an insight into the regional-scale/historical forces shaping the pattern and composition of present-day forest communities. (3) To investigate potential future shifts in forest distribution associated with projected climate change. Location KwaZulu-Natal province, South Africa. Methods A BIOCLIM-type approach is adopted. Bioclimatic 'profiles' for eight different forest subtypes are defined from a series of grid overlays of current forest distribution against nineteen climatic and geographical variables, using ArcInfo GIS grid-based processing. A principal components analysis is performed on a selection of individual forests to identify those variables most significant in distinguishing different forest subtypes. Five models are developed to predict the distribution of forest subtypes from their bioclimatic profiles. Maps of the potential distribution of forest subtypes predicted by these models under current climatic conditions are produced, and model accuracy assessed. One model is applied to two palaeoclimatic scenarios, the Last Glacial Maximum (LGM) (approximate to 18,000 BP) and the Holocene altithermal (approximate to 7000 sp), and to projected future climate under a doubling in global atmospheric carbon dioxide. Results Seven variables; altitude, mean annual temperature, annual rainfall range, potential evaporation, annual temperature range, mean annual precipitation and mean winter rainfall, are most important in distinguishing different forest subtypes. Under the most accurate model, the potential present-day distribution of all forest subtypes is more extensive than is actually observed, but is supported by recent historical evidence. During the LGM, Afromontane forest occupied a much reduced and highly fragmented area in the midaltitude region currently occupied by scarp forest. During the Holocene altithermal, forest expanded in area, with a mixing of Afromontane and Indian Ocean coastal belt forest elements along the present-day scarp forest belt. Under projected climatic conditions, forest shifts in altitude and latitude and occupies an area similar to its current potential and more extensive than its actual current distribution. Main conclusions Biogeographical history and present physical diversity play a major role in the evolution and persistence of the diversity of forest in KwaZulu-Natal. It is important to adopt a long-term and regional perspective to forest ecology, biogeography, conservation and management. The area and altitudinal and latitudinal distribution of forest subtypes show considerable sensitivity to climate change. The isolation of forest by anthropogenic landscape change has limited its radiation potential and ability to track environmental change. Long-term forest preservation requires reserves in climatically stable areas, or spanning altitudinal or latitudinal gradients allowing for forest migration, along with innovative matrix management strategies. Dune, sand, swamp, riverine and lowland forest subtypes are most at risk. Scarp forests are highlighted as former refugia and important for the future conservation of forest biodiversity.
引用
收藏
页码:595 / 617
页数:23
相关论文
共 87 条
[1]  
[Anonymous], CLIM CHANG IPCC RESP
[2]  
[Anonymous], 1994, MEMOIRS BOT SURVEY S
[3]  
*ARC INF, 1996, CELL BAS MOD GRID EN
[4]  
BAKKER EMV, 1978, BIOGEOGRAPHY ECOLOGY, V31
[5]  
BAKKER EMV, 1982, PALAEOECOL AFR, V15, P77
[6]  
Bews, 1912, ANN NAT MUS, V2, P253
[7]  
Bews J.W., 1920, Annals of the Natal Museum, V4, P367
[8]  
Bews JW., 1913, ANN NAT MUS, V2, P485
[9]  
BEWS JW, 1917, ANN NAT MUS, V3, P511
[10]  
Bond W. J., 1997, P421