集成灰色分析和元胞自动机用于景观动态模拟

针对景观生态系统\"复杂性\"的特点,采用灰色局势确定邻域转换规则,构建元胞自动机,用于模拟和分析景观生态动态变化.在信息不完全的情况下,提高了元胞自动机景观生态动态模型的可靠性和可行性.以中国云南省怒江流域中段作为实例,确定影响元胞转换规则的三因素主要有元胞邻域的聚集程度,土地适宜程度和人类活动影响程度,根据景观这一类复杂系统的动态变化特征其权重随地理空间位置和时间而变化,同时用Monte Carlo法考虑了模拟时转换的随机性.通过计算显示模拟未来发展的情景总的趋势是符合实际的,不但模拟了微观的景观单元的自组织机制,而且在一定程度上反映了宏观的社会经济因素影响,因而更具有针对性、典型性及准确性.     

基于元胞自动机的城市空间动态模拟

城市空间动态的模拟与预测可以为城市可持续发展规划与管理提供重要的参考依据。SLEUTH元胞自动机模型在城市空间模拟中较强的适用性和可移植性,该模型通过对历史数据的蒙特卡洛迭代自动寻找城市增长误差最小的参数组合,解决了传统元胞自动机模型中转换规则不易确定的问题。以武汉市为研究案例,运用SLEUTH模型进行了城市空间动态模拟与情景预测。2007年至2011年的城市空间模拟结果显示,模拟结果与实际历史数据可以获得良好的空间匹配度,Lee-Sallee形状指数均在0.6以上,显示SLEUTH元胞自动机模型经过本地化校正后具有较强的适用性和满意的模拟精度。进而,设置了现状趋势、基本保护、严格保护等3种情景对武汉2025年城市空间动态进行了预测,结果显示,各情景模式下城市居住用地均明显增长,农业用地、林地、水域等均有所减少;现状趋势情景和基本保护情景下农田、林地、水域减少的幅度较大,会加剧区域的生境破碎、耕地功能下降、水资源匮乏、湖滨湿地萎缩等生态问题,说明这两种情景不能有效满足城市生态系统健康和可持续发展的需要。严格保护情景下,城市居住用地扩张的程度得到了明显的控制,水域和林地得到了有效的保护,对于重要的自然生态系统组分保护及其服务能力维持可以起到显著作用。     

元胞自动机模拟物种生态策略选择的研究

本文在100×100的二维网格平面上赋予元胞自动机L(生命长度)与P(繁殖期)两种属性,在空间以及捕食两种压力的共同作用下模拟物种对于生态策略选择,时间标度T以单位时间作为步长。当只给予生存空间压力的情况下,固定P=0．5,L初始值20,模拟生态系统在T=120时达到稳定,L平均值稳定在2.1。对于以上系统,在不同的生殖时间P=0.7,0.5,0.3,0.1下,物种对空间的占领速度的百分数变化,在T=120,145,220,450附近,空间占有率分别稳定于60％,55％,67％,77％左右,而L值稳定于3．1,2．5, 4．6,6．0附近。当使L与P双参数自由变化,发现P平均值稳定于2．4,L平均值稳定于0．6。当给与生存压力与捕食关系双重压力下,发现L平均值变化为S型曲线。     

基于元胞自动机的城市生态安全格局模拟——以东莞市为例

以城市化快速发展的东莞市为研究对象,选取景观生态安全程度为评价标准,获取生态安全的元胞1 km×1 km范围内的城市化元胞数,将其嵌入元胞自动机(CA)转换规则中作为约束条件,控制城市发展,建立生态安全的城市CA,模拟生态安全的城市发展形态.结果表明: 东莞市综合景观生态安全指数从1988年的0.497降低到2005年的0.395,景观尺度上的生态安全下降.利用生态安全的CA模拟的2005年东莞市生态安全指数由实际的0.395增加到0.479,模拟的城市景观生态压力减小、生态安全状态和综合景观生态安全程度提高.CA可以作为探索生态安全城市研究的有效工具.     

基于元胞自动机的喀斯特石漠化格局模拟研究

西南喀斯特山区的石漠化问题是目前我国西部地区最为突出的地域环境问题之一,其迅速发展已经严重影响到当地人们的生产生活。以导致石漠化发生发展的自然、人文因子为切入点,通过模拟影响喀斯特系统地表覆被变化的基本生态过程(如植物定居、植物死亡、水蚀风蚀引起的土地退化以及岩石成土过程等),利用随机元胞自动机具有的简单邻域规则产生复杂空间格局的特点,使喀斯特系统地表覆被植被-裸土-裸岩状态在一定概率下发生状态转换,并结合RS和GIS技术,构建了简单、有效的喀斯特石漠化模拟及预测模型(KarstCA)。以典型喀斯特石漠化地区关岭县为研究区,在自然、人文驱动因素共同影响以及只考虑自然驱动因素情景下,KarstCA模型模拟的研究区2007年石漠化空间分布格局的差异主要分布在中部和南部,其主要是不同空间范围上人类活动作用方式和强度差异所致。在16a中(1992—2007年)喀斯特地区地物(植被-裸土-裸岩)丰度变化成非线性关系,当植被覆达到54%以上并继续增加时,裸岩发展趋势与之呈明显的负相关(P<0.01)。在模拟期内人类活动对研究区石漠化的发展起到抑制作用,人类活动的正效应(植树造林等)与负效应(乱砍滥伐、过度放牧等)在一定程度上抵消了植被总面积的剧烈变化趋势。将地表过程耦合进元胞自动机模型,突破了以往该类研究只通过概率考虑状态转换,而对其机理认识的不足;同时本研究考虑了自然、人文驱动因素在不同空间尺度上作用于石漠化现象的复杂性,对于探索这些因素是如何作用于地表过程及其贡献率等研究具有一定的参考价值。     

基于元胞自动机的生物性病原物类疾病传播模型

生物性病原物类疾病是单一植物群落中常见的一种传染病,它主要依靠风、雨和昆虫等外力传播病菌,传播过程中具有一定的潜伏性,元胞自动机是一种在时间和空间上离散,以每个个体状态的同步更新为前提来讨论系统整体性质的模型,本文利用其特点,将整个森林中的树木分为健康、潜伏期、发病期三类,以树木的营养状况和被感染时间长度为辅助变量,在考虑到树木得病后被治愈的情况后,建立了基于元胞自动机的树林生物性病原物类疾病传播模型．模拟结果表明,在不同的初始条件下,得病树所占的比例由疾病传播和被治愈的相对强度确定．疾病每次传播时,从爆发到被控制的过程具有相似的特点：随着恢复强度从小到大,初始阶段,被感染的树木迅速增加,接着进入传播速度相对缓慢的稳定过程,最终如果恢复机制足够强,则树林中的不健康树木会全部消失．     

基于多智能体与元胞自动机的上海城市扩展动态模拟

利用元胞自动机模型和多智能体模型相结合的方法,在GIS技术手段的支持下构建了一个能够兼顾两种模型优点的城市扩展动态模型,并以上海市为实证对象,模拟了上海市2005年的城市扩展动态,分别预测了2010年和2020年上海城市扩展的动态演化结果。在元胞自动机模型中定义城市系统中的各种自然、社会和交通等要素,在智能体模型中定义政府和居民智能体的行为。模型将二者结合起来,模拟上海中心城区、城郊区及外围区的城市用地扩展模式。对模型模拟的上海2005年土地利用状态和实际土地利用状态进行验证,Kappa系数的平均值达到0.75以上,说明模型具有较高的可信度。对预测出2010年和2020年上海市土地利用状态分析表明,城镇用地以向东部和南部扩张最为明显。     

“生态位”元胞自动机在土地可持续规划模型中的应用

快速城市化带来了一系列的环境生态问题。有必要把生态学的概念引进城市规划中,以减少城市发展带来的弊端。提出了基于“生态位”的元胞自动机（CA）的新模型,并将其应用在土地利用规划中。探讨了如何通过“生态位”元胞自动机和GIS的结合进行城市土地可持续利用的规划。该模型可方便地探索不同土地利用政策下城市土地利用发展情景,能够为城市规划提供有用的决策支持。旨在探索通过模拟的手段对城市土地利用进行合理的规划。将该模型应用于快速发展的广州市,并取得了较有意义的结果。     

利用元胞自动机研究一类捕食食饵模型中的斑块扩散现象

利用概率元胞自动机模型对空间隐式的、食饵具Allee效应的一类捕食食饵模型进行模拟,发现随着相关参数的变化,种群的空间扩散前沿由连续的扩散波逐渐转变为一种相互隔离的斑块向外扩散,这种斑块扩散现象与以往的扩散模式有所不同。研究结果表明：(1)在斑块扩散的情况下,相关参数的微小变化会导致种群灭绝或者形成连续的扩散波,即斑块扩散发生在种群趋于灭绝和连续扩散之间;(2)当种群的空间扩散方式为斑块扩散时,种群的扩散速度会变慢,与其他扩散方式下的速度有着明显的区别。该研究结果对生物入侵控制和外来物种监测有重要的启发和指导作用。     

江河源区高寒草甸退化序列上\"秃斑\" 连通效应的元胞自动机模拟

高寒草甸草毡层\"秃斑\"是高寒草甸退化进程中最活跃的表征,明确其在退化序列上的演变时空规律是揭示草地退化动力学机制的关键之一。采用实地调研结合元胞自动机模拟的方法,对高寒草甸退化序列上秃斑格局动态进行了模拟,是对退化序列上秃斑的连续动态图谱表达,建立了高寒草甸由初步退化到系统崩溃的符合实际的草地秃斑连通的图谱序列,模拟吻合度达93.9%。通过对图谱序列关系的分析表明,秃斑连通进程分为3个阶段:低速连通期——第0—2年、跃变期——第2—7年、连通不可逆转期——第7—9年;最大秃斑面积、最大秃斑面积-秃斑总面积比的跃变过程,是草地退化等级的质变过程;通过对跃变期和连通不可逆转期相应的草地生态与恢复性能的对比分析,确定了连通阈值为54.5%;秃斑连通的过程伴随着临界阈现象的发生。由于连通不可逆转期\"黑土滩\"形成过程的不可逆性及形成后的巨大危害性,连通阈值的确定将为高寒草甸生态系统安全预警及其退化恢复治理提供依据。     

Ecological risk assessment of genetically modified crops based on cellular automata modeling

The assessment of ecological risk in genetically modified (GM) biological systems is critically important for decision-making and public acceptance. Cellular automata (CA) provide a potential modeling and simulation framework for representing relationships and interspecies interactions both temporally and spatially. In this paper, a simple subsystem contains only four species: crop, target pest, non-target pest and enemy insect, and a three layer arrangement of L × L stochastic cellular automata with a periodic boundary were established. The simulation of this simplified system showed abundant and sufficient complexity in population assembly and densities, suggesting a prospective application in ecological risk assessment of GM crops.     

Advantages of using different data sources in assessment of landscape change and its effect on visual scale

This paper presents a data-flexible indicator framework for analysis of visual landscape character; the VisuLands framework. The theory-based framework encompasses currently used indicators for visual assessment based on four different data sources: land cover data, aerial photographs, landscape photographs and field observations. This paper presents a study applying the VisuLands framework in analysis of landscape change and its effect on visual scale in a landscape in Southeast Sweden. The paper provides a critical assessment of the pros and cons of the approach. It identifies the advantages and disadvantages of using different data sources as well as the applicability and sensitivity of existing indicators in detecting visible landscape change. The results show that while some of the VisuLands indicators are relatively easily applied, others are more complex and demanding in terms of interpretation. The flexibility of the VisuLands framework makes it applicable and user-friendly as it helps meet the requirements and restrictions of the users. The assessment has shown that the different data sources complement each other and that applying indicators using various data sources, when available, will enhance the comprehensiveness of visual landscape assessment. The experience of this study is that the VisuLands framework is a useful tool in landscape analysis, monitoring and planning, which provides a repeatable, systematic and transparent approach with strong links to landscape theory.     

基于CA-Markov模型和多目标优化的大连市土地利用格局

运用马尔科夫转移矩阵、多准则评估以及元胞自动机耦合的CA-Markov模型,基于1990、2000、2010年土地利用图、地形因子和地理要素等,模拟大连市未来土地利用景观格局变化趋势.基于大连市土地利用结构、经济社会、自然环境等特点,结合大连市城市总体规划和土地利用规划,建立经济、社会、环境3个目标导向的模糊多目标优化模型,优化配置了大连市未来土地利用格局.结果表明:1990-2010年,大连的快速发展呈现出建设用地持续扩张而耕地、林地面积缩小的特点,以现有城市化速度发展,到2020年大连市景观格局土地覆盖将发生很大变化,景观破碎化程度将加剧.优化调整土地利用数量结构,能满足未来大连可持续发展的要求.     

基于NDVI的不同火烧强度下大兴安岭林火迹地森林景观恢复

利用LANDSAT TM影像,通过分类、提取森林景观类型及NDVI值,在较大尺度上探讨了火烧区火烧强度与森林景观格局、功能恢复的关系。结果表明,火烧区森林总体恢复情况较好。恢复状况与火烧强度具有明显的相关性。火烧强度越高,恢复状况越差。重度火烧区的针叶林景观所占比重低且生长状况较差;沼泽面积高于未火烧对照区,这一现象应引起足够重视,特别是在全球变化气温升高的背景下,应防止寒温带针叶林的退化以及林地沼泽化。在三种主要森林类型(针叶林、阔叶林、针阔叶混交林)中,针阔叶混交林是生长状况最好的,标志着火烧迹地正由演替的初期阶段向中期阶段过渡。     

Life Cycle Impact assessment of land use based on the hemeroby concept

The impact category ‘land use’ describes in the Life Cycle Assessment (LCA) methodology the environmental impacts of occupying, reshaping and managing land for human purposes. Land use can either be the long-term use of land (e.g. for arable farming) or changing the type of land use (e.g. from natural to urban area). The impact category ‘land use’ comprises those environmental consequences, which impact the environment due to the land use itself, for instance through the reduction of landscape elements, the planting of monocultures or artificial vegetation, or the sealing of surfaces. Important environmental consequences of land use are the decreasing availability of habitats and the decreasing diversity of wildlife species. The assessment of the environmental impacts of land use within LCA studies is the objective of this paper. Land use leads to a degradation of the naturalness of the area utilised. In this respect the naturalness of any area can be defined as the sum of land actually not influenced by humans and the remaining naturalness of land under use. To determine the remaining naturalness of land under use, this study suggests applying the Hemeroby concept. “Hemeroby is a measure for the human influence on ecosystems” (Kowarik 1999). The Hemeroby level of an area describes the intensity of land use and can therefore be used to characterise different types of land use. Characterization factors are proposed, which allow calculating the degradation of the naturalness of an area due to a specific type of land use. Since the resource ‘nature/naturalness’ is on a larger geographical scale by far not homogeneous, the assessment of land use needs to be regionalised. Therefore, the impact category ‘land use’ has been subdivided into the impact sub-categories ‘land use in European biogeographic regions’. Following the general LCA framework, normalization values for the impact sub-categories are calculated in order to facilitate the evaluation of the characterization results with regard to their share in a reference value. Weighting factors, which enable an aggregation of the results of the different land use sub-categories and make them comparable to other impact categories (e.g. climate change or acidification) are suggested based on the assumption that the current land use pattern in the European biogeographic regions is acceptable.     

Structure regulation of land use and landscape pattern changes based on matter element analysis in a small watershed in Sichuan,China

The agricultural land suitability assessment based on the natural characteristics of land blocks is tra-ditionally the basis for agricultural and/or land use plan-ning.The assessment,however,is static and cannot be incorporated with potential land use changes.Recently,a dynamic approach,i.e.,matter element analysis,has been effectively applied for land use planning.In the pre-sent study,based on matter element analysis,we estab-lished a matter element analysis model for land use suitability assessment using the suitability grade of land use,evaluation indicators and their characteristic values as matter elements,and also using sutra field,controlled field,weight value and correlation degree from field sur-vey,as well as an expert system.This model was applied to the structure regulation of the land use in the small Yuejiagou watershed of the Langzhong Municipality in Sichuan Province,China.Results show that the propor-tion among agriculture,forestry and animal husbandry tended to rationalize land use.The economic and eco-logical indices of the lands were increased from 1529.8 to 1719.99 and from 1460.94 to 1758.21 after the regu-lation,respectively.The regulation also caused changes in landscape patterns as follows:The indices of diversity and evenness were increased from 1.3028 to 2.0920 and from 0.6108 to 0.8463,or by 60.58% and 38.56%,respect-ively.However,the indices of dominance and contagion were decreased from 0.6431 to 0.2106 and from 0.7467 to 0.7125,respectively.This indicated that the land use in the small watershed was rational.The spatial distribution of patches tended to be uniform.The patch congregation was gradually dispersed.This study verified that the mat-ter element analysis approach can not only overcome the factitious influences and improve the precision of land assessment,but also can be used for the structure regu-lation of specific patches.     

Structure regulation of land use and landscape pattern changes based on matter element analysis in a small watershed in Sichuan, China

The agricultural land suitability assessment based on the natural characteristics of land blocks is traditionally the basis for agricultural and/or land use planning. The assessment, however, is static and cannot be incorporated with potential land use changes. Recently, a dynamic approach, i.e., matter element analysis, has been effectively applied for land use planning. In the present study, based on matter element analysis, we established a matter element analysis model for land use suitability assessment using the suitability grade of land use, evaluation indicators and their characteristic values as matter elements, and also using sutra field, controlled field, weight value and correlation degree from field survey, as well as an expert system. This model was applied to the structure regulation of the land use in the small Yuejiagou watershed of the Langzhong Municipality in Sichuan Province, China. Results show that the proportion among agriculture, forestry and animal husbandry tended to rationalize land use. The economic and ecological indices of the lands were increased from 1529.8 to 1719.99 and from 1460.94 to 1758.21 after the regulation, respectively. The regulation also caused changes in landscape patterns as follows: The indices of diversity and evenness were increased from 1.3028 to 2.0920 and from 0.6108 to 0.8463, or by 60.58% and 38.56%, respectively. However, the indices of dominance and contagion were decreased from 0.6431 to 0.2106 and from 0.7467 to 0.7125, respectively. This indicated that the land use in the small watershed was rational. The spatial distribution of patches tended to be uniform. The patch congregation was gradually dispersed. This study verified that the matter element analysis approach can not only overcome the factitious influences and improve the precision of land assessment, but also can be used for the structure regulation of specific patches. __________ Translated from Acta Ecologica Sinica, 2006, 26(7): 2093–2100 [译自: 生态学报]     

基于RS和GIS的科尔沁沙地南缘章古台地区土地利用变化及其驱动力

利用遥感影像,借助于GIS技术以及数理统计方法,对章古台地区近20年间土地利用变化及驱动力进行了分析。结果表明:1988-2009年间,章古台地区土地利用构成以耕地和林地为主;土地利用类型变化以林地、耕地和建设用地面积的增加和草地、水域及未利用土地面积的减少为特征,其中,草地、建设用地和未利用土地的年变化率大于其他土地利用类型;土地利用程度综合指数由1988年的235.34增至2009年的251.65;土地类型间的转化以草地、未利用土地向林地、耕地转化和耕地与林地间转化为主,占总转化面积的95.58%。章古台地区土地利用变化受人口、政策和自然因素的综合影响,虽然自然因素是决定土地利用变化的重要因素,但在20年的时间尺度上,人口增长和政策引导则是加剧这种变化的主导因素。     

Rapid assessment of rivers using macroinvertebrates: A procedure based on habitat-specific sampling,family level identification and a biotic index

Abstract This paper describes a simple and inexpensive procedure for the rapid biological assessment of water quality in rivers and streams in eastern Australia. The procedure involves the standardized collection of samples of 100 macroinvertebrates from defined habitat types within a water body. Specimens are identified to family level only and a biotic index is calculated. Proposed future testing and evaluation are described, and the limitations of the rapid approach are discussed.     

Modelling and optimization of fermentation factors for enhancement of alkaline protease production by isolated Bacillus circulans using feed-forward neural network and genetic algorithm

Aim: Modelling and optimization of fermentation factors and evaluation for enhanced alkaline protease production by Bacillus circulans. Methods and Results: A hybrid system of feed‐forward neural network (FFNN) and genetic algorithm (GA) was used to optimize the fermentation conditions to enhance the alkaline protease production by B. circulans. Different microbial metabolism regulating fermentation factors (incubation temperature, medium pH, inoculum level, medium volume, carbon and nitrogen sources) were used to construct a ‘6‐13‐1’ topology of the FFNN for identifying the nonlinear relationship between fermentation factors and enzyme yield. FFNN predicted values were further optimized for alkaline protease production using GA. The overall mean absolute predictive error and the mean square errors were observed to be 0·0048, 27·9, 0·001128 and 22·45 U ml^?1 for training and testing, respectively. The goodness of the neural network prediction (coefficient of R²) was found to be 0·9993. Conclusions: Four different optimum fermentation conditions revealed maximum enzyme production out of 500 simulated data. Concentration‐dependent carbon and nitrogen sources, showed major impact on bacterial metabolism mediated alkaline protease production. Improved enzyme yield could be achieved by this microbial strain in wide nutrient concentration range and each selected factor concentration depends on rest of the factors concentration. The usage of FFNN–GA hybrid methodology has resulted in a significant improvement (>2·5‐fold) in the alkaline protease yield. Significance and Impact of the Study: The present study helps to optimize enzyme production and its regulation pattern by combinatorial influence of different fermentation factors. Further, the information obtained in this study signifies its importance during scale‐up studies.