Sample effects on species-area relationships of typical forests in karst and non-karst mixing distribution areas
Wu-Sheng XIANG,Shu-Hua LU,Shu-Jun WEN,Tao DING,Bin WANG,Ai-Ping ZHOU,Fu-Zhao HUANG,Xian-Kun LI
DOI: https://doi.org/10.11931/guihaia.gxzw201408001
2015-01-01
Abstract:The species-area relationship (SARs)is a fundamental pattern in ecological research.It is also a common tool in predicting species extinction and assessing species diversity.The species-area relationships curves describe the increase of species number with increasing scales.Various factors affect the shape of species-area curves at different scales.However,it is still not clear in selecting the best SARs model for a certain area and insufficient attention hasbeen paid to the effects of sample methods and standards on SARs.The region in north tropics of Guangxi,where the karst area and non-karst area are mixing distributed,supports very high level of biodiversity.In this region,the Nong-gang National Natural Reserve is located in the typical tropical karst terrain where forests are less disturbed and con-serve hyper diverse endemic species.The forests in the Nonggang National Natural Reserve are typical karst forests which are shaped by the distinctive geological environments.In contrast,the terrain in the Shiwandashan National Natural Reserve is non-karst forests where are mainly shaped by climatic conditions.A contrastive study of species-area relationships between the karst forests and non-karst forests in the same climatic zone and elucidate the differ-ence of species diversity may provide insight into the conservation of biodiversity in the karst area and non-karst area mixing distributed region.In this study,we established 14 plots of 400-600 ㎡ in Nonggang and 15 plots in Shiwan-dashan.These plots contain representative forests communities in the study area,for example,including Excentroden-dron tonkinense ,Cephalomappa sinensis ,Deutzianthus tonkinensis,Litsea dilleniifolia in Nonggang and Eberhard-tia aurata ,Madhuca pasquieri ,Lithocarpus pseudovestitus ,Hopea chinensis in Shiwandashan.All individuals of tree species with DBH (Diameter at breast height)greater than 2 cm were recorded and identified to species.Based on the sample data of 10 m×10 m subplots,species number was counted by a series continuous sample area.In order to de-termine the optimal SARs models and sample effects on SARs models,we constructed three SARs models(Logarithm model,Logistic model,and power model)from noncontiguous grid combinations design based on survey data of typi-cal forest communities.To assess the effects of minimum measured DBH on SARs,species numbers of individuals with DBH≥2 cm,DBH≥4 cm,DBH≥6 cm,DBH≥8 cm,DBH≥10 cm were counted separately.We used least square method and Gauss-Newton method to estimate the parameters of SARs models.Akaike Information Criterion (AIC)was used to compare the goodness-of-fit of each model.Spearman rank correlation was used to assess the effect of base sample unit and minimum measured DBH on SARs.The results showed that logistic model was the best among the three SARs models both in karst area and non-karst area.Base sample unit and minimum measured DBH both had significant effects on SARs,but the effect varied among the SARs models.We argue that logistic mod-el is a robust model to describe SARs of tree species in karst and non-karst forests in north tropics Guangxi.The re-sults also suggest that special attention must be paid to the influence of sample methods as applying SARs models to protection practices.