Study On Woody Species-area Relationship Of The Deciduous Broadleaved Forest In Baotianman, Henan Province

Species area relationship (SAR) which described the number of species vary with the increasing sample area, one of the basic problems in ecology research. A lot of researches has been done on SAR, there are still some basic issues in dispute. This study intended to answer the following two questions:(1)What extent can the traditional nested sampling method (Center, Bottomleft, Bottomright, Topleft, Topright) represent the average level of SAR? (2) Does the construction methods, initial sampling location, scale have effect on the best fitting model of SAR?In this paper: we studied the influence of different sampling strategies on SAR in 1hm2 deciduous broad-leaved forest. Using repeated random sampling method and repeated nested sampling method (here 1000 times repeated) constructed SAR as approximate value of average SAR. Firstly, compared different nested sampling methods (included 5 different initial sampling locations nested sampling methods and the average of 5 sampling methods. the average of 5 sampling methods also called traditional nested sampling methods) with 1000 times repeated random sampling method and 1000 times repeated nested sampling method, to see the 5 different initial sampling locations single time nested sampling methods, traditional nested sampling methods constructed SAR represent the degreee of average community species level. The three models (logarithm model, power model, logistic model) were used to fit the SAR, to determine the best-fitted model through using the AIC (Akaike Information Criterion) method. The 95% confidence interval of parameter optimal fitting model were both repeated nested sampling method and random sampling method for 1000 times, to determine that whether the traditional nested sampling method constructed SAR can represent the true value. The three model fitted SAR by different sampling methods.Using the 95% confidence interval method of parameter optimal fitting model to determine whether the traditional nested sampling method was used to represent the average level of different scale sample of the SAR. The results showed that: 1000 times repeated nested sampling method,1000 times repeated random sampling method and the traditional nested sampling method construct SAR had no significant difference. The SAR constructed by nested sampling method with 5 different initial sampling locations compare with 1000 times nested sampling method,1000 times random sampling method almost has significant difference; the average value of the traditional nested sampling method compare with one single starting position nested sampling method to construct the SAR is more close to the average level of community SAR.Secondly, different sampling strategies, different initial sampling location, different scales have effect on the best fitting model of the SAR. Nested sampling methods and random sampling methods, the logarithm model, power model, logistic model in 1 hm2 plot of SAR, using AIC to test the three models’goodness of fit, to determine on the different scale the best fitting model of the SAR. The results shown that, for the three models using random sampling method construct SAR on the sampling scale from 4m×4m to 100m×100m, the optimal model of SAR is the logistic model all the time, power model followed; when sampling scale is less than or equal to 3m×3m the best fitting model is the power model, logistic model followed. This study suggest that, the logistic model is the most suitable model among the three model on different sclales that we slected in this study for SAR constructed by nested sampling method; for SAR constructed by random sampling method, the logistic model is the most suitable model among the three model for fitting SAR on large or medium scales, the power model is the best fitting model among the three model fitting SAR on small scales. Sampling methods, sampling location, sampling scales have different degree impacts on SAR. The forest succession stage, habitat heterogeneity, the main species distribution pattern may also have influences on SAR.