安徽钻孔体应变与降雨、地下水位关系的研究

利用合肥台、黄山台钻孔体应变2013～2015年观测资料日均值与降雨引起的地下水位变化进行相关分析,同时考虑体应变变化的滞后性,使用Matlab计算水位和体应变不同滞后时间的相关系数,确定滞后天数;扣除滞后天数,再进行回归拟合分析,消除水位干扰。鉴于水位对体应变干扰的复杂性,在运用长周期数据时,以月为窗长进行回归分析,能较好地消除水位对体应变的干扰,利于体应变的趋势性判断。     

海南岛逐日太阳总辐射长期演变特征与关联因子

为了解海南岛太阳辐射资源的时空变化规律及变化原因,以海口（1961―2013 年）和三亚（1993―2013年）2 个气象站点长时间序列逐日太阳总辐射、日照时数、气温等观测资料为基础,利用气候倾向率、Mann-Kendall突变检验方法分析了全年、雨季、旱季日均太阳总辐射的长期演变特征,利用多重线性回归分析了辐射变化的主要关联气象因素。结果表明：1961―2013 年期间海口全年、雨季、旱季日均太阳总辐射气候倾向率分别为0.021、0.004、0.038 MJ/m²,1993―2013 年三亚全年、雨季、旱季日均太阳总辐射气候倾向率分别为-0.68、-0.658、-0.714MJ/m²。海口日均太阳总辐射变化呈波动趋势,全年、雨季日均太阳总辐射显著的“变暗期”分别出现在1975―1992、1975―1990 年;海口旱季日均太阳总辐射呈先降后升的趋势,1967―2003 年为下降期,2003 年之后呈现上升趋势,显著的“变暗期”为1995、1988―1989 年。1993 年以来,三亚全年、雨季、旱季日均太阳总辐射变化都呈下降趋势,显著的“变暗期”分别出现在1999―2013（全年）、2000―2003 和2009―2011 年（雨季）、1996―2001 和2004―2013 年（旱季）,突变年份分别为1997、1995、1995 年。三亚站点太阳辐射资源较海口站点丰富;日照时数与海口全年、雨季、旱季和三亚全年、雨季日均太阳总辐射变化关联度最强,而日平均气温则与三亚旱季日均总辐射变化关联度最强。     

灰线性加权非等距GM(1,1)形变预测模型

结合加权非等距GM(1,1)模型与线性回归理论,构建灰线性加权非等距GM(1,1)预测模型,并给出对模型预测精度起决定性作用的灰指数v和参数m的优化方法。与加权非等距GM(1,1)模型和线性回归预测模型相比,灰线性加权非等距GM(1,1)预测模型的精度更高,预测有效时间更长,模型的稳定性更好。优化v和m后,灰线性加权非等距GM(1,1)预测模型的实用性、稳定性进一步提高。     

基于PCA和BP神经网络的径流预测

径流预测为流域水资源的合理开发利用与统筹配置提供依据。运用多元线性回归、主成分回归、BP神经网络及主成分分析和BP神经网络相结合的方法,对新疆呼图壁河流域石门水文站2009-2011年各月径流量进行预测,并采用相关系数、确定性系数及均方根误差对各模型预测精度进行比较。结果表明：（1）神经网络等智能算法具有高速寻优的能力,对短时间尺度的月径流量的预测结果较好;（2）主成分回归等常规算法能充分反映出某地区径流的年际的稳定性,对全年径流总量的模拟精度较高;（3）主成分分析和BP神经网络相结合的方法,提高了神经网络的收敛速度,同时降低了局部极值的影响,优于简单的BP神经网络,适用于呼图壁河月径流量预测。     

基于断裂力学的岩石切削数值分析探讨

为研究岩石切削过程中刀具的力学特性以及刀具平均峰值作用力的预测,基于断裂力学和有限元理论建立了岩石切削的数值模型。模拟分析岩石对刀具的反作用过程,得到不同力学特性岩石和刀具作用条件下刀具所受的反作用力,结果表明,冲击速度在一定范围内对刀具平均峰值作用力影响较小;岩石体积呈阶跃性递减,刀具作用力大小与崩落岩石碎片体积变化一致。对模拟、实验以及已有理论计算得到的刀具平均峰值作用力进行线性回归分析,研究结果表明,数值方法计算结果与实验结果、Goktan[1]理论计算结果的线性相关系数分别为0.812、0.930,相关性较好,表明所建数值模型用于模拟岩石切削是正确和可靠的。     

Modeling patch characteristics of farmland loss for site assessment in urban fringe of Beijing, China

Farmland protection and delimitation in the urban fringe considers not only natural factors but also the spatial characters and site factors. Taking Daxing District, Beijing in China as a case study, this paper used landscape ecology and power-law methods to analyze and evaluate farmland loss during the period of 2004-2007 based on the interpretation results of SPOT5 remote sensing images in 2004 and 2007. At the patch level, we selected four landscape indices, namely patch size, shape index, the nearest neighbor distance between farmland and construction land (including residential land and other construction land), and cropping type, to evaluate the risk of farmland loss and establish a farmland site analysis indicator system. The results showed that patch size and shape index have a significant positive correlation with farmland loss, whereas the distance to construction land has a clear negative correlation with farmland loss. As regards cropping type, fallow farmland is much easier for non-agricultural use than cultivated farmland. The relative transition ratio among vegetable land, fallow farmland and cultivated farmland is 1︰5.6︰1. The patch size of lost farmland follows a power-law distribution, indicating that not only small parcels but also large parcels can be lost. Patch size less than 4 ha or more than 15 ha is in high loss risk, between 4 ha and 10 ha in medium loss risk, and larger than 10 ha and less than 15 ha in low risk. Farmland with a more regular shape has a higher likelihood of loss. Patch shape index less than 2.0 is in high loss risk, between 2.0 and 3.0 in medium loss risk, and larger than 3.0 in low risk. Construction land has a varying impact on farmland loss, the residential land effected distance is 1000 m, and that of the other construction land is 2000 m. This analysis showed the relationships between site factors and farmland loss, and the analysis framework can provide support and reference for farmland protection and delimitation of prime farmland in China.     

A time-dependent stochastic grassland fire ignition probability model for Hulun Buir Grassland of China

Grassland fire is one of the most important disturbance factors in the natural ecosystems.This paper focuses on the spatial distribution of long-term grassland fire patterns in the Hulun Buir Grassland located in the northeast of Inner Mongolia Autonomous Region in China.The density or ratio of ignition can reflect the relationship between grassland fire and different ignition factors.Based on the relationship between the density or ratio of ignition in different range of each ignition factor and grassland fire events,an ignition probability model was developed by using binary logistic regression function and its overall accuracy averaged up to 81.7%.Meanwhile it was found that daily relative humidity,daily temperature,elevation,vegetation type,distance to county-level road,distance to town are more important determinants of spatial distribution of fire ignitions.Using Monte Carlo method,we developed a time-dependent stochastic ignition probability model based on the distribution of inter-annual daily relative humidity and daily temperature.Through this model,it is possible to estimate the spatial patterns of ignition probability for grassland fire,which will be helpful to the quantitative evaluation of grassland fire risk and its management in the future.     

Spatial evolution and locational determinants of high-tech industries in Beijing

Using datasets on high-tech industries in Beijing as empirical studies, this paper attempts to interpret spatial shift of high-tech manufacturing firms and to examine the main determinants that have had the greatest effect on this spatial evolution. We aimed at merging these two aspects by using firm level databases in 1996 and 2010. To explain spatial change of the high-tech firms in Beijing, the Kernel density estimation method was used for hotspot analysis and detection by comparing their locations in 1996 and 2010, through which spatial features and their temporal changes could be approximately plotted. Furthermore, to provide quantitative results, Ripley′s K-function was used as an instrument to reveal spatial shift and the dispersion distance of high-tech manufacturing firms in Beijing. By employing a negative binominal regression model, we evaluated the main determinants that have significantly affected the spatial evolution of high-tech manufacturing firms and compared differential influence of these locational factors on overall high-tech firms and each sub-sectors. The empirical analysis shows that high-tech industries in Beijing, in general, have evident agglomeration characteristics, and that the hotspot has shifted from the central city to suburban areas. In combination with the Ripley index, this study concludes that high-tech firms are now more scattered in metropolitan areas of Beijing as compared with 1996. The results of regression model indicate that the firms′ locational decisions are significantly influenced by the spatial planning and regulation policies of the municipal government. In addition, market processes involving transportation accessibility and agglomeration economy have been found to be important in explaining the dynamics of locational variation of high-tech manufacturing firms in Beijing. Research into how markets and the government interact to determine the location of high-tech manufacturing production will be helpful for policymakers to enact effective policies toward a more efficient urban spatial structure.     

基于分式析因及最佳子集回归的多种污染物复合污染特征——阿特拉津与多种污染物在松花江沉积物上的吸附效应

为揭示农药（晶体乐果、甲霜灵、阿特拉津、马拉硫磷、扑草净）与重金属（铜、锌、铅、镉、镍）10种污染物共存时的复合污染规律,以阿特拉津为目标污染物,采用分辨度为Ⅲ的2^10-6分式析因实验设计,并用完全折叠实验和区组设计进行辅助,通过多元线性回归的建模方法筛选并建立了以阿特拉津为因变量、10种污染物因子主效应和二阶交互效应为自变量的多元线性回归吸附模型。研究表明：共存体系中因子主效应项铅、铜、锌、镉、镍和农药晶体乐果显著抑制（显著性水平α=0.05）阿特拉津在沉积物上的吸附,与阿特拉津产生竞争吸附效应,因重金属主效应产生的拮抗作用占总拮抗作用的63.5%,其拮抗作用从大到小为镍、镉、晶体乐果、锌、铅、铜;在以镍为关键自变量的二阶交互效应项中,因二阶交互效应项产生的协同或拮抗作用占阿特拉津吸附能力的35.3%,其中,铅、镉和晶体乐果抑制镍与阿特拉津的竞争吸附,与阿特拉津在沉积物上的吸附发生协同作用,协同作用从大到小为镉×镍、铅×镍、镍×晶体乐果;而镍作为调节自变量的二阶交互效应项镍阿特拉津会促进阿特拉津的吸附,说明多种污染物共存体系中重金属镍对阿特拉津在沉积物上吸附影响作用的复杂性和重要性。