Demographic Analysis of Tree Colonization in a20-Year-Old Right-of-Way

Past tree colonization dynamics of a powerline-right-of-way (ROW) corridor in the Haut-Saint-Laurent region of Quebec was studied based on the present age distribution of its tree populations. This colonization study spans 20 years, from 1977 (ROW clearance) to 1996. The sampled quadrats were classified into six vegetation types. Tree colonization dynamics were interpreted in each type, and three distinct patterns were identified. (1) Communities adapted to acidic conditions were heavily colonized by Acer rubrum, at least for the last 12 years. (2) Communities adapted to mesic or to hydric conditions were more intensely colonized in the period 1985-1987 than in the following 9 years; this past success in tree colonization may have been caused by herbicide treatments, which could have facilitated tree establishment by damaging the herbaceous and shrub vegetation. (3) Cattail, vine-raspberry, and reed-dominated communities contained few tree individuals, with almost all trees establishing between 1979 and 1990; those three vegetation types appear as the most resistant to tree invasion in the ROW studied. This study supports the need for an integrated approach in ROW vegetation management, in which the selection of vegetation treatment methods would depend on the tree colonization dynamics in each vegetation type. Minimizing disturbances inflicted on ROW herbaceous and shrub covers should be the central strategy because disturbances jeopardize natural resistance to future tree invasion, except in communities adapted to acidic conditions where the existing vegetation does not prevent invasion by A. rubrum. Many trees are surviving the successive cutting operations by producing new sprouts each time, particularly in communities adapted to mesic and hydric conditions. In these cases, mechanical cutting should be replaced by a one-time stump-killing operation, to avoid repeated and unsuccessful treatments of the same individuals over time.     

人工神经网络方法在资源与环境预测方面的应用

用人工神经网络方法对不同水域、不同环境因子之间非线性和不确定性的复杂关系进行学习训练并预测检验。结果表明：人工神经网络方法在模拟和预测方面 优于传统的统计回归模型,在资源与环境方面的应用是可行的。具有较强的模拟预测能力。与传统的回归模型相比,人工神经网络方法不要求监测数据具有很强的规律性,就可用后的网络模型对其进行预报,燕且预测相对误差均比回归模型预测相对误差要小,具有一定的实用性。两个实例的应用     

三峡茅坪溪沥青混凝土心墙堆石坝位移反演分析与安全预测

本文研究了基于综合应用人工神经网络和演化算法的位移反演分析方法,并将此方法应用于三峡茅坪溪沥青混凝土心墙堆石坝的变形反演分析中。以茅坪溪一期工程原型观测成果为依据,反演出能够正确反映坝体变形特性的邓肯-张EB模型参数,从而预测了施工期末和蓄水期末该坝的变形特性,并基于总应力法研究了心墙水力劈裂破坏发生的可能性,为茅坪溪堆石坝运行期的安全生产提供依据。研究结果表明,茅坪溪心墙堆石坝不会发生水力劈裂破坏,该坝蓄水后虽然水压上升,但心墙仍是安全的。     

多因素耦合条件下硫化矿自燃神经网络动态预测模型研究

硫化矿石自燃是多种因素、多场耦合综合作用的结果,是一典型的非线性问题。笔者应用人工神经网络技术,以Matlab软件为平台,通过现场调查和理论分析,建立了矿石含硫量、通风强度、环境温度3因素与硫化矿石自燃之间的预测模型;通过数据样本学习与部分现场监测数据相结合进行模拟,研究表明预测数据与实测结果基本吻合,误差控制在10%以内,取得了较好的效果。该研究为预防硫化矿石自燃提供一个新的思路和方法,具有一定的理论意义和应用价值。     

Human impact on beach and foredune vegetation of North Padre Island,Texas

Vehicular and pedestrian traffic decreased top and root production, percent cover, and diversity of vegetation, and modified species composition on beach and foredune areas of Padre Island National Seashore and Nueces County Park. Degradation of vegetal cover was directly related to the type and intensity of traffic; species in heavily trafficked areas represented earlier successional stages than those in comparable undisturbed areas.     

FORESTED WETLANDS IN EASTERN CONNECTICUT: THEIR TRANSITION ZONES AND DELINEATION1

ABSTRACT: The delineation of inland wetlands requires close field examination of the biological and physical gradients (transition zones) between wetlands and bordering uplands. As part of a study on the detection and delineation of inland wetlands in eastern Connecticut by remote sensing techniques, this effort was designed to investigate vegetation distribution and composition and selected physical and chemical properties of the soils of wetland to upland transition zones in deciduous wetland forests. Field research was conducted during the growing season of 1975 within a test area consisting of the 45 mi² Town of Mansfield, Connecticut. Changes in vegetation composition and structure, soil pH, and soil water content were determined along line transects extended over wetland to upland transition zones. Differences in soil pH occurred along the transects but were of such magnitude that they probably have little impact on plant distribution. There were significant changes in soil water content along the wetland to upland gradients. Discriminant analysis applied to statistical “index of abundance” data describing vegetation distribution among the various zones (wetland, transition, upland) showed which plant species best distinguish wetlands from uplands. Of the criteria studied, vegetation composition and distribution, soil water content, and relief are the most useful criteria for delineating deciduous wetland forests.     

Vegetation changes associated with barrier-dune construction on the outer banks of North Carolina

Manmade dunes are used to stabilize the barrier islands of the mid-Atlantic coast of the United States so that they are more amenable to development. Although this strategy has stabilized extensive sections of the barrier-island substrate, it has also caused significant ecological changes. The rate and pattern of the resulting vegetation changes along the Outer Banks of North Carolina were determined using sequential aerial photography. Successional trends have been altered in favor of more stable woody communities.     

Plant cover synanthropization in the pechora-Ilych biosphere reserve along an altitudinal gradient

Trends in plant cover synanthropization along a plain-foothills-mountains gradient have been revealed in the reserve. Characteristics of plant communities in anthropogenically disturbed habitats, centers of concentration of anthropophytes and apophytes, and pathways of their migration are described.     

人口与经济发展对植被的影响研究

植被是陆地表面主要的覆盖物，也是许多地球系统过程的重要变量。人类活动对植被的影响在区域尺度上是显著的。以重庆市为例，使用滑动平均和SG滤波对SPOT/VGT NDVI时间序列数据进行了平滑处理，结合GDP和人口栅格化数据，从时间序列和空间相关场两个方面分析了植被与人类活动因子之间的时空相关性。结果表明，从1998到2005年，植被的分布和生长状况并没有随GDP和人口的持续增加而增加。在时间序列上，植被覆盖与GDP和人口之间总体上呈显著的负相关关系；在空间上，这种相关关系表现出了较强的异质性。空间相关场表明：较强的负相关主要发生在相对发达的主城区周围，反映了快速的经济发展和城市化使得植被面积和生产力降低；然而在偏远的山区，由于经济落后同时植被覆盖较少，出现了正相关。