引进天敌昆虫的效果评估

对我国天敌昆虫引进20多年来各阶段的引种量、引进重复率、引种国际范围、引种单位、引进种类和引进后的应用情况的变化进行了分析，结果表明，我国的年平均引种量虽然逐年有所上升，但每年基本上维持在8～10种次。2次、3次引种重复率逐年降低，至2000年2次引种率大约23％。3次引种率在10％左右。引种目标国由4个到38个，引种单位数量变化相对不大，1986年后年平均基本上维持在10个左右。应用效果分3种类型：(1)建立种群起到控害作用；(2)建立不稳定种群控害效果不明显；(3)未能建立种群。     

桃品种需冷量评价模式的探讨

 通过1986～2001年对450余份桃品种需冷量的7．2℃模式、0～7．2℃模式(不包括0℃)和犹它模式比较分析，归纳出桃品种需冷量的评价模式为：以秋季日平均温度稳定低于7．2~C的日期为需冷量测定的起点，以0～7．2℃累积低温值作为需冷量的评价标准比较适宜；犹它模式在中需冷量和长需冷量范围内能有效预测休眠的结束，而不适宜低需冷量品种的测定；7．2℃模式不适宜作为需冷量的评价模式。品种的需冷量与叶芽开放和始花期的相关系数分别为0．52和0．58，均达到极显著水平。提出了桃品种需冷量评价的系列标准参照品种。     

新疆冬小麦农田蒸散估算模型的研究

本文在田间试验资料的基础上，综合考虑了影响冬小麦农田蒸散的气象、生物学特性和土壤水分等因素，选用蒸发力、冬小麦的叶面积指数和相对有效土壤湿度建立了新疆冬小麦农田蒸散估算模型，并且检验了该模型的计算效果。     

景观生态学原理与方法在荒漠化研究中应用的理论分析

荒漠化发生发展的过程实际上是指在自然和人为因素的作用下，生态系统结构遭受破坏、功能过程受阻和演变发生异化的过程，其防治的根本措施是恢复和重建健康的生态系统。所以，景观生态学理念在荒漠化的研究和防治中有着重要的意义。本文从荒漠化与生态系统结构、功能变化、荒漠化与生物多样性、荒漠化生态系统的物质循环和能量流动以及荒漠化生态系统的稳定性等方面论述了景观生态学理念在荒漠化研究中应用的可行性，并重点分析了景观生态学关于景观格局变化的评价指标与荒漠化土地动态变化之间的关系。     

210国道榆林至陕蒙交界段建设的生态环境影响评价

在简要分析榆林至陕蒙交界段生态环境对公路及公路建设的影响的基础上 ,重点论述了公路建设在建设期和营运期对当地生态环境的影响 ,并对公路建设前后水土流失与土地沙化状况进行了预测 ,最后提出水土保持与防沙护路的方案。结论是 :只要把公路建设与生态环境建设相结合 ,重视生态环境的保护与建设 ,该公路建设和营运对生态环境的不利影响是很小的 ,而且 ,由于公路建设本身的需要所采取的防沙护路措施 ,会使公路沿线的生态环境更好     

Functional heterogeneity in resources within landscapes and herbivore population dynamics

Large mammalian herbivores are notorious for their propensity towards population irruptions and crashes, yet many herbivore populations remain relatively stable. I explore how resource heterogeneity within landscapes dampens population instability, using a metaphysiological modelling approach considering patch state distributions. Resource heterogeneity is functionally stabilizing through spreading consumption away from preferred resources before these become critically depleted. Lower-quality resources act as a buffer against starvation during critical periods of the seasonal cycle. Enriching resource quality is destabilizing, even if patch diversity is maintained, because food quantity then becomes the limitation. The potential consequences of landscape fragmentation are explored using the Serengeti ecosystem, characterised by broadscale resource gradients, as a hypothetical example. Further insights provided by the model are illustrated with specific examples concerning the effects of patch scales and waterpoint distribution. A metaphysiological modelling approach enables the basic consequences of landscape heterogeneity to be distinguished from further effects that may arise from specific patch scales and configurations, without the distracting detail of spatially explicit models.     

Using traffic flow theory to model traffic mortality in mammals

Traffic has a considerable effect on population and community dynamics through the disruption and fragmentation of habitat and traffic mortality. This paper deals with a systematic way to acquire knowledge about the probabilities of successful road crossing by mammals and what characteristics affect this traversability. We derive a model from traffic flow theory to estimate traffic mortality in mammals related to relevant road, traffic and species characteristics. The probability of successful road crossing is determined by the pavement width of the road, traffic volume, traversing speed of the mammals and their body length. We include the traversability model in a simple two-patch population model to explore the effects of these road, traffic and species characteristics on population dynamics. Analysis of the models show that, for our parameter ranges, traffic volume and traversing speed have the largest effect on traffic mortality. The population size is especially negatively affected when roads have to be crossed during the daily movements. These predictions could be useful to determine the expected effectiveness of mitigating measures relative to the current situation. Mitigating measures might alter the road and traffic characteristics. The effects of these changes on traffic mortality and population dynamics could be analysed by calculating the number of traffic victims before and after the mitigating measures. This revised version was published online in July 2006 with corrections to the Cover Date.     

甘肃省生态环境综合评价指标体系研究

甘肃省生态系统分为社会经济系统和自然生态系统两大类 ,社会经济系统指标体系由社会经济条件和人为环境压力构成 ;自然生态系统由水环境、森林生态系统、草地生态系统和土壤生态系统四个次一级系统组成 ,每个次一级生态系统由各自所属指标体系来系统评定。利用因子分析法、Delphi法和头脑风暴法经过多次综合专家意见最终确定 2 6项指标作为甘肃省生态环境质量综合评价体系。利用层次分析法 (AHP)定量各指标的权重。通过综合评价表明 ,甘南州、陇南地区、张掖地区的生态环境质量较好 ,甘肃中部广大黄土高原地区次之 ,河西走廊的武威市、金昌市和嘉峪关市环境质量最差。评价结果真实地反映了甘肃省生态环境的基本状况 ,说明该指标体系具有科学性 ,能够用于不同地区之间或同一地区不同时间的环境质量对比与变化研究。     

引诱剂在综合防治桔小实蝇作用中的评估

以使用引诱剂为主 ,坚持在杨桃园挂引诱笼 2 - 3年后 ,大量降低田间桔小实蝇BactroceradorsalisHendel雄虫数量和卵的受精率 ,配合其它的综防措施 ,防效逐年提高。三个综防点试验实践表明 :杨桃每果虫口数 ,卵孵化率及受害果率大大降低 ,防效达 92 5%。对桔小实蝇的防治初见成效。     

The effect of non-inversion tillage and light availability on dispersal and spatial growth of Cynodon dactylon

The invasive ability of Cynodon dactylon is dependent on self dispersal and on cultivation practices. Tillage can seriously change patch biomass and spatial structure, spreading vegetative propagules of the weed. The objectives of this study were: (i) to quantify the effect on non‐inversion tillage on dispersal, establishment and colonization of C. dactylon and (ii) to propose a simple model considering soil cultivation effects and light availability on spatial growth of weed patches. Two experiments were carried out, exploring different soils and environmental conditions. Spatial distribution of vegetative units differed when tillage was conducted with different non‐inversion implements and could be described by simple functions. A minimum patch biomass seems necessary before vegetative structures are vulnerable to movement by cultivation. Only a small proportion of the biomass dispersed from original patches was able to establish. However, simulation showed that the area colonized by C. dactylon mostly increased by means of tillage dispersal, both with and without crop competition, in one growing cycle. It appears sensible to consider changing cultivation practices to reduce weed dispersal and to use crop competition for light to create unsuitable habitats limiting weed colonization.