毛竹林冰雪灾害与环境、生物学特性和管理的关系

选取浙江临安双石村和奉化石门镇两个不同受灾性质的样点,运用NMS、CCA和Monte Carlo分析了冰雪气候过程中主要影响毛竹林灾情的环境条件、自身生物学特征参数和经营管理手段.结果表明:双石(SJ组)与石门(SII组)两地样地的NMS聚类结果的最终压力值为7.7345>5,最终不稳定性值为0.00009＜0.0001,表明2地样地之间存在明显的不同,并于143个重复时找到最佳维数为3维,依据FB、DS和GR可分成2组,SI组总体环境解释量为68.4%,SII组总体解释量为95.6%,SI组样地受到雪压的影响较大,其外部因素和条件比受到冰害的SII组样地而言更为复杂;Monte Carlo测试表明SI和SII的实际数据和模拟数据的特征值有区别(p＜0.01),而且受灾参数与冰雪灾害有相关性.受灾参数的CCA结果得到了较好的解释,钩梢对翻蔸、折断、破裂数量都有较大影响;立竹密度显著影响立竹形态结构的发育,影响立竹对冰雪灾害的敏感性;竹子的削尖度大大低于乔木(平均值为0.0078＜0.01),且尖削度是一个较好地体现毛竹林冰雪灾害程度的参考参数.冰雪灾害受损程度和胸径之间存在一定的关联,密度和削尖度之间存在临界点.因此,竹林管理中要注意合理采伐和控制削尖度.数据没有显示由于坡向的不同所引起的对毛竹林生态系统的影响;从三维图释可以看出,不同系列的变量反应了弯曲、开裂、翻蔸和开裂等不同受灾现象,因此在竹林管理中应区别对待.CCA与NMS的分析表明各种类型的干扰对竹林生态系统的影响其实是一种多因素的综合性作用.

The Relationship among Environment,Management,Biological Characteristics and Snow/Ice Disaster in Phyllostachys edulis forest

The stand condition and growth analysis characteristics of P.edulis were measured and the management type was quantified.Nonmetric multidimensional scaling(NMS)was used to analyze the orientation of total sample plots and canonical correspondence analysis(CCA)was applied to analyze the multifactor environmental variances;meanwhile,Monte Carlo test was applied to check up the results of above methods.Community separation was clearly shown by NMS ordination with a final stress of 7.734 5>5 and a final instability of 0.000 09<0.000 1.A three-dimensional solution was found after 143 iterations and total sample plots were divided into two parts,i.e.,Shuangshi group(SI)and Shimen(SII),which were depended on lowest height of living branch,stand density and gradient.Monte Carlo test proved that all three axes had a smaller stress than a random solution.CCA showed that 48.1% variance was explained by the first axes and 68.4% explained cumulatively in SI,60.5% variance was explained by the first axes and 95.6% explained cumulatively in SII respectively.So snow disaster was more complex than ice disaster.Furthermore,the correlations between sample plots and axis responded to the percentage of variance explained by each axe.Monte Carlo test showed that there was difference between eigenvalue of real data and simulated data(P<0.01)as same as species-environmental correlation(P=0.0350<0.05)of SI and SII respectively,what indicated that events were not at random,but did represent the relationship between environment variables and biological information.CCA implied that obtruncation influenced the quantity of uprooting,breakage and split,but it was not shown in the analysis of ordination.And stand density affected the morphological structure and sensitivity responding to snow/ice of P.heterocycla.Data did not show that the difference of slope direction,maybe,for the havoc in the investigation area both on south and north slope,but indicated that correlation between damage degree of snow/ice disaster and diameter at breast height existed,and there was one critical value between density and taper.So selective cutting should be appended on the list of management to achieve reasonable stand density and better taper of bamboo which was much less than arbor(average value=0.007 8<0.01)implying that it was easy to fall under snow/ice pressure.Properties of taper played an important role in bamboo forest management.Different phenomena like bending,breakage,uprooting and split were reflected by different series of variances showed by 3-dimension illustration,so it should be treated variously on forest management.As a whole,the nonlinear nature of energy and water exchanges between the land surface and the atmosphere makes the aggregation of parameters a challenge for analysis,so CCA and NMS indicated that effects of all kinds of disturbance on bamboo ecosystem were a kind of multifactor integration function same to respond of plastical behavior of bamboo forest ecosystem.