| Complicated forest structure means steady and long ecological functions. Plant diversity is the key factor for forest biodiversity, it is also the measurement for forest community structure. Besides, the competition between plants has an important influence on forest community structure and its succession process. Therefore, the study on forest structure, diversity and competition has important significance for secondary forest management and its ecological function recovery. While, accurate analysis methods are the key precondition to cognize forest structure, diversity and competition. There are lots of approaches to describe forest structure, diversity and competition at present. However, those approaches still need to be improved. The common feature of the approaches described forest structure is that they only can be depicted unilateral structure, namely, they cannot simultaneously provide more than three aspects of stand structure informations. To solve such problem, trivariate distribution of spatial structural parameters that based on the relationships of the nearest four neighbour woods was proposed in this paper. It contained three spatial structural aspects, namely, trivariate of dominance-mingling-uniform. This method was used to study the differences between near to primeval forest and secondary forest, and also used to study the structure dynamics of structure-based forest management and control plots. The common feature of methods described tree size diversity is that they cannot correctly express the tree size diversity of different types of forests. To solve such difficult problem, this contribution proposed three quantitative measure indexes based on differentiation of diameter distribution—the tree differentiation of Simpson(DT), the tree differentiation of Shannon(HT) and the mean of tree differentiation(?T), then simulated and experimental data were used to study tree size diversity. The tree height has limited further application of the well competition index based on intersection angle; because of tree height is not easy to measure and usually with large error. This paper have discussed the feasibility of two empirical calculation ways that using DBH directly instead of measured tree height(CID) or the estimated tree height obtained by the diameter-height curve model to instead of measured tree height(CID-H) in different stands.Attempting to give the best empirical calculation way of forest competition index based on intersection angle. Besides, this paper used improved methods to study dynamics of forests under different forest management measures. Fourteen permanent fixed standard plots were established in the Quercus aliena var. acuteserrata secondary forests on northern face of Qinling Mountains, Northwest China PR. Each plot was labeled by a specific number. Two plots with length 70 m × 70 m have been treated as control(GSC1) and another on as the demonstration of structure-based forest management [SBFM(GSC2)] after the plots being built, respectively. Twelve plots with length 28.28 m × 28.28 m with 4 treatments(CT: control; LT: light thinning; MT: medium thinning; and HT: heavy thinning) by 0%, 20%, 30% and 40% of basal area removal, respectively. Those plots were reexamined after five or six years. The dynamics characteristics of the different plots were analyzed in terms of species composition, diversity, population competition and stand structure. The main results are as follows:(1) The trivariate distribution of spatial structural parameters was showed in this paper. The bivariate distribution of the spatial structural parameters included 5 × 5 × 5 = 125 structure combinations, those combinations take full advantages of the dominance, species mingling and uniform angel index frequency distribution when they are used to express the spatial structure characters of forest stands. The trivariate distribution simultaneously quantified three aspects of stand structures from different way. The trivariate distribution further refined the spatial structure of the stand, and provides more direct and useful heterogeneity of spatial structure informations than bivariate and univariate distributions of spatial structure parameters can. The trivariate distribution of near to primeval forest showed that most trees in the natural Quercus aliena var. acuteserrata forest were highly mixed by species and randomly distributed, trees with different advantages were equally distributed. While, the trivariate distribution of secondary forest showed that most trees were randomly distributed and slightly mixed by species, trees with different advantages were unequally distributed. Otherwise, on the stand level, the dynamics trivariate distribution of CK plot GSC1 showed that litter change was found during six years, while, SBFM plot GSC2 changed a lot. On the Quercus aliena var.acuteserratapopulation level, the dynamics trivariate distribution of CK plot GSC1 showed that litter change was found during six years, while SBFM plot GSC2 changed a lot.(2) It was considered that differentiationmean value(?T) as the best method to express diameter size diversity. Deeper structure information could be provided by tree size differentiation that considering tree sites. The methods based on differentiation distribution of Simpson(DT), Shannon(HT) and its mean value(?T) can correctly express the tree size diversity of different types of forests.It was better for?T to discriminate the differences between different types’ stands than DT and HT can, due to its ability to estimate continuous variable, reflect the differential degree between tree and its nearest neighboring tree, and prone to explain its biological meaning. Therefore, it was considered that the?T as the best method to express tree size diversity.(3) It was feasible to use DBH directly instead of measured tree height to calculate the competition index based on intersection angle. Forest competition index based on intersection angle(CIH) is concise, could express the over shading and the lateral extruding from the competitive trees. Empirical calculation that using DBH instead of measured tree(CID) could express calculated results by measured tree height(CIH) more than 95%, both of them can express that different size of tree was suffered different competition pressure. Although the value of CIH<CID, trends in both of their results were consistent. Therefore, using DBH directly instead of measured tree height is the optimal experience calculation way for forest competition index based on intersection angle.(4) Both of GSC1 and GSC2 are middle-mature stands. After six years in two plots, the shrub species diversity, competition pressure of Quercus aliena var. acuteserratapopulation, mean value of stand uniform angle index was declined, the mean value of stand mingling and tree species diversity was increased, the value of tree size diversity?T remained, the diameter distribution was changed from inverted J curve to single-peaked curve. SBFM promoted tree growth in GSC2; it showed that the PAI of individual tree DBH, height, volume and stand basal area, volume was greater than GSC1 over 7.9%、17.4%、26.5%、82.8% and 38.4%, respectively. Beyond that, SBFM improved seedling regeneration and promoted more trees grow into higher diameter classes.(5) All of thinning plots and its control plots are middle-young stands. After five years in all plots, the shrub species diversity, value of tree size diversity?T, competition pressure of Quercus aliena var. acuteserratapopulation and seedling regeneration was declined, while, the mean value of stand uniform angle index and mingling was increased, the diameter distribution was reminded multi-peaked curve. Compared to control plots, thinning management promoted tree growth; it showed that the PAI of stand basal area was greastest in LT, over than control 109.3%; the PAI of individual tree DBH and volume was greatest in MT, over than control 56.0% and 45.1%, respectively; the PAI of individual height and stand volume was greatest in HT, over than control 18.1% and 66.9%, respectively. In addition to this, thinning also promoted more trees grow into higher diameter classes.The improved methods that trivariate distribution of structural parameters, differentiationmean value(?T) and competition index based on intersection angle calculated by DBH play important role in evaluating forest management effects. Besides, SBFM not only promoted tree growth, but also improved seedling regeneration and protected biodiversity. |
PDF Full Text Request