Modeling high-frequency position data of large herbivores with a phase-state model

Understanding the rules and factors that drive the foraging behavior of large herbivores is important to describe their interaction with the landscape at various spatial scales. Some unresolved questions refer to landscape-behavioral interactions that result in oriented or random search in seasonally changing landscapes. Remotely sensed position data indicate that herbivores select local patches of heterogeneous landscapes depending on a complex host of dynamically varying animal and environmental conditions. Since foraging paths consist in successions of relatively short steps, increasing the frequency at which position information is acquired would contribute to entangle the mechanisms resulting in herbivores’ foraging paths. We addressed the question whether herbivores would obtain information at a patch scale that would modify their distribution at a landscape scale based on directed movement or navigation ability. We considered a set of 100,000 high-frequency (1 min intervals) position data of several free-ranging sheep (Ovis aries) at a seasonal-varying range (Patagonian Monte, Argentina) and observed their movements at landscape and at single vegetation patch scales. At a landscape scale, we inspected the spatial co-variation of seasonally varying forage offer and ewes’ movement speeds. At a patch scale, we developed a phase-state (P-S) model of movement cycles based on the occurrence of behavioral phases along foraging paths, and fitted it to the observed daily time series of ewes’ movement speeds. Ewes were preferentially distributed in areas with high forage offer during periods of low forage availability and the reverse occurred during the season of high forage availability. Parameters of the model of activity cycles amenable to control by ewes (duration of speed phases, time elapsed between speed cycles) did not covariate with forage offer, but varied significantly among ewes. The shape (kurtosis) parameter of the model of movement cycles, one which is unlikely under ewes’ control, co-varied significantly with spatial forage offer but did not differ among ewes. We conclude that ewes allocated foraging time along a series of similar movement efforts irrespective of forage availability at small patches. Average forage scarcity at multi-patch level increases the ratio of searching to feeding time. This results in apparent selective time allocation to richer forage areas but does not imply evidence for oriented movement at a landscape scale. We advance a behavioral-based definition of forage patches and discuss its implications in developing foraging theory and models. The P-S model applied to high-frequency position data of large herbivores substantially improves the interpretation of the factors controlling their time allocation in space with respect to previous models of herbivore spatial behavior by discriminating among behavioral-based and environmentally induced components of their movements.     

Understanding Interaction Effects of Climate Change and Fire Management on Bird Distributions through Combined Process and Habitat Models

Abstract: Avian conservation efforts must account for changes in vegetation composition and structure associated with climate change. We modeled vegetation change and the probability of occurrence of birds to project changes in winter bird distributions associated with climate change and fire management in the northern Chihuahuan Desert (southwestern U.S.A.). We simulated vegetation change in a process‐based model (Landscape and Fire Simulator) in which anticipated climate change was associated with doubling of current atmospheric carbon dioxide over the next 50 years. We estimated the relative probability of bird occurrence on the basis of statistical models derived from field observations of birds and data on vegetation type, topography, and roads. We selected 3 focal species, Scaled Quail (Callipepla squamata), Loggerhead Shrike (Lanius ludovicianus), and Rock Wren (Salpinctes obsoletus), that had a range of probabilities of occurrence for our study area. Our simulations projected increases in relative probability of bird occurrence in shrubland and decreases in grassland and Yucca spp. and ocotillo (Fouquieria splendens) vegetation. Generally, the relative probability of occurrence of all 3 species was highest in shrubland because leaf‐area index values were lower in shrubland. This high probability of occurrence likely is related to the species’ use of open vegetation for foraging. Fire suppression had little effect on projected vegetation composition because as climate changed there was less fuel and burned area. Our results show that if future water limits on plant type are considered, models that incorporate spatial data may suggest how and where different species of birds may respond to vegetation changes.     

Modelling the abundance of three key plant species in New Zealand hill-pasture using a decision tree approach

Decision tree models were developed to investigate and predict the relative abundance of three key pasture plants [ryegrass (Lolium perenne), browntop (Agrostis capillaris), and white clover (Trifolium repens)] with integration of a geographical information system (GIS) in a naturalised hill-pasture in the North Island, New Zealand, and were compared with regression models with respect to model fit and predictive accuracy. The results indicated that the decision tree models had a better model fit in terms of average squared error (ASE) and a higher percentage of adequately predicted cases in model validation than the corresponding regression models. These decision tree models clearly revealed the relative importance of environmental and management variables in influencing the abundance of these three species. Hill slope was the most significant environmental factor influencing the abundance of ryegrass while soil Olsen P and annual P fertilizer input were the most significant factors influencing the abundance of browntop, and white clover, respectively. Soil Olsen P of approximately 10 μg/g, or a slope of about 10.5° was critical points where the competition between ryegrass and browntop tended to come to an equilibrium. Integrating the decision tree models with a GIS in this study not only facilitated the model development and analyses, but also provided a useful decision support tool in pasture management such as in assisting precision fertilizer placement. The insights obtained from the decision tree models also have important implications for pasture management, for example, it is important to maintain a soil Olsen P higher than 10 μg/g in order to keep the dominance of ryegrass in the hill-pasture.     

浅谈GAM与GM（1,1）模型

对ＧＡＭ水环境预测模型提出了四点不同看法，与有关作者商榷，指出ＧＭ（１，１）模型与ＧＡＭ模型二者并无优劣之差，精度亦相当。     

生态缺水条件下流域生态经济系统安全性分析

生态用水和社会经济用水的竞争状况使得研究水资源短缺地区的流域生态经济系统安全状况、有效进行生态缺水条件下流域生态经济系统的调控十分必要.从宏观角度出发,首先分析了流域生态经济系统演化的机制和影响因素,在此基础上,通过建立指标体系,提出了构建流域主要指标为状态变量的生态经济系统演化模型,并利用模型进行系统安全性分析.以潮白河流域为例,构建了人口数量、人均GDP、生态系统指数以及生态用水比例为状态变量的生态经济系统演化模型.模型模拟结果表明,如果不加以调控,该流域将无法提供生态经济系统的安全保障.提出了提高生态用水比例、控制经济增长速度和积极进行生态建设、控制水土流失的对策,为实现潮白河流域的生态经济系统安全保障服务.     

基于人工神经网络理论的船舶动力装置安全综合评价模型

为提高船舶动力装置安全综合评价水平，依据人工神经网络的基本原理，分析了BP神经网络模型的基本原理和优化策略，将其应用于船舶动力装置的安全评价之中。提出了基于神经网络理论的船舶动力装置综合安全评价模型及实现方法，并以实例论证了此方法的可行性。     

海岸带含水层咸淡水界面随潮汐波动的数值模拟

在分析、研究了滨海地带含水层和不透水层的水文地质特性及含水层内的咸淡水渗流运动特征的基础上,运用数学推理的手法,建立了一种较为简单适用的模拟海岸带含水层咸淡水界面和天然地下水面变动规律的二维数学模型,通过该模型的计算值与实验值的比较,证明该模型的计算结果与实验结果有着非常好的拟合度,即该模型能较客观地揭示海水入侵引起的咸淡水界面的变化规律.其后,运用该模型系统的探讨了含水层以下为非平坦的不透水层的滨海地带天然地下水面、咸淡水界面伴随着潮汐的波动而变化的规律.即伴随着潮汐的波动,一方面天然地下水面和咸淡水界面与潮汐具有相似的振动波形,波动的幅度随离海岸距离的增加而减小.在海岸附近咸淡水界面的振幅大于天然地下水面的振幅,而在离开海岸一定距离后天然地下水面的振幅超过咸淡水界面的振幅.总体上天然地下水面的振幅呈负指数衰减,而咸淡水界面的振幅几乎呈直线衰减(其衰减直线的倾角在135～150 °之间变动),且潮汐波动对天然地下水面影响的范围远大于其对咸淡水界面的影响;另一方面天然地下水面和咸淡水界面波动的振幅及它们之间的相位差的大小还与天然地下水面的水力坡度、含水层的渗透系数、有效孔隙率、不透水层形状及其变动的幅度有关,但它们之间的相位差的大小与潮汐波动的幅度无关,其振幅与潮汐的振幅成比例增减.当不透水层的形状一定时,不论天然地下水面的水力坡度、含水层的渗透系数、有效孔隙率及潮汐波动的幅度怎么变动,天然地下水面、咸淡水界面振动的相位差的变化趋势及其峰值出现的位置几乎不变,即不透水层的形状决定着天然地下水面、咸淡水界面振动的相位差的变化趋势,而其大小与含水层的水文地质参数(水力坡度、渗透系数及有效孔隙率)密切相关.此外,当天然地下水面和咸淡水界面的振动存在相位和振幅的较大差异时,可以断定含水层以下存在非平坦的不透水层,且这种差异越大,不透水层凸凹不平的程度越高.     

基于GM(1,1)的残差修正模型的电梯故障率预测

为研究某城市某品牌电梯故障率发展趋势,建立了该城市该品牌电梯故障率的GM(1,1)灰色预测模型,并对所建模型进行了数据检验,检验结果表明该预测模型的预测精度波动较大。为了提高GM(1,1)灰色预测模型的预测精度,利用对模型进行数据检验时得到的残差序列,建立GM(1,1)灰色预测模型的残差修正模型,利用该残差修正模型对原预测模型进行修正。利用经残差修正模型修正后的故障率预测模型对该城市A品牌电梯的故障率进行预测,结果表明:1)残差修正模型对原模型修正后的相对误差与修正前相比有升也有降,但精度有所提高且趋于稳定,表明残差修正模型有利于提高预测精度;2)利用所建立的故障率预测模型求得的预测故障率与实际故障率相比,相对误差不超过8.010%,表明该故障率预测模型的预测精度较高;3)修正模型预测值表明,在现有状态下该城市A品牌电梯的故障率呈上升趋势,应加强该品牌电梯的检维修与管理。     

中国虚拟水贸易的测算及影响因素研究

本文测算2005年、2007年与2010年3个年度中国虚拟水进出口贸易量,采用投入产出及IO-SDA等方法分解分析其变化的影响因素。研究结果显示部门间用水系数差距都非常大,直接用水系数最大的农业部门857.9 t/万元(3年平均值,下同),与直接用水系数最小的部门差距达600倍之多。完全用水系数最大的也为农业部门1 078.7 t/万元,与完全用水系数最小的部门亦达113倍。各部门的完全用水系数和直接用水系数也存在较大不同,有些部门完全用水系数和直接用水系数差距很大,如服装皮革羽绒及其制品业部门差距达115倍,食品制造及烟草加工业部门差距达618.5 t/万元;而有些部门差距非常小,如农业部门差距仅仅0.24倍,煤炭开采和洗选业部门差距仅为8.2 t/万元。虚拟水出口量和进口量分别从2005年的1 774.3亿t和1 378.7亿t增加到2010年的2 254.1亿t和1 836.5亿t,且出口量一直大于进口量。影响因素分解显示,在此期间,进出口规模的增加是虚拟水贸易量增加的主要影响因素,其中出口规模合计为103 698.8×10~6t,进口规模合计为80 401.2×10~6t。进出口贸易结构的影响在不同部门间存在较大的差异。真实用水系数和真实中间投入技术是虚拟水贸易变化的主要负向因素,其中真实用水系数对虚拟水出口与进口的影响分别为-36 793.2×10~6t和-34 668.3×10~6t,真实中间投入技术对虚拟水出口与进口的影响分别为-1 719.7×10~6t和-1 660.8×10~6t,表明在此期间用水效率和中间投入效率都得到了改善。另外,直接用水系数和中间投入技术分解出的结构效应都不明显,中间投入进口比率效应也不明显。     

现代能源生态系统建设:一种理论探讨

在现代社会发展中,能源资源开发和利用所产生的巨大极化效应使整个人类可持续发展面临越来越严重的挑战。论文应用工业生态学的观点对能源系统的建设进行重新认识,提出了“能源生态系统”的概念;阐述了能源生态系统的3个组成部分:内生系统,外生系统,共生系统;并探讨了其空间结构的类型划分和基本形态。论文建立的现代能源生态建设基本框架,为进一步开展此方面的理论探讨提供有益借鉴。