基于微根管技术的玉米根系生长监测

为了研究玉米根系生长规律,该文利用固城农业气象试验站内设置的大型根剖面系统,采用微根管观测法,对试验地上玉米主要生育期的根系生长动态进行定期直接跟踪监测,并以方形整段标本法作为参照标准,对试验数据采用MATLAB软件辅助图像处理和现代统计方法进行分析。结果表明：微根管法与方形整段标本法对得出的根长密度随深度增加呈递减型有较好的一致性,两者相关系数达到0.83以上,通过0.05的显著性检验;由观测数据建立的回归方程能较好地反映土壤中玉米根系生长规律,进一步表明微根管法是一种破坏性较小、可准确定位跟踪植物根系在土壤中生长动态变化的先进技术,对植物根系生长研究有较好的推广应用价值。     

近40年克里雅绿洲人口时空变化对土地利用/覆盖变化的影响

研究绿洲区域土地利用与覆盖变化及人类活动对土地利用与覆盖变化的驱动作用,对全球变化科学具有重要意义。该文通过精确空间化的1975年和2015年人口数据与土地利用与覆盖(land use and land cover,LULC)数据,运用网格单元法、线性相关、空间自相关、空间回归模型等方法,兼从全局与局部定量分析克里雅绿洲40a的人口变化与LULC变化的时空动态关系,探讨人口空间变化对LULC变化的驱动作用。结果表明:1)40 a来,克里雅绿洲的扩张表现在:人口显著增加,坤长14.277 7万人并向县、乡、镇中心集聚,建设用地与耕地向南部扩张明显,林地增加但趋于破碎化,草地与水域锐减。2)人口的空间变化是克里雅LULC变化及绿洲扩张的直接驱动力,直接表现为建设用地与耕地向南扩张,间接表现为林地变化、草地变化、水域变化。人口的空间变化对绿洲扩张的内部驱动作用,与绿洲北部恶劣的自然环境、克里雅河自南向北的流向共同作用,导致绿洲向南部扩张。3)地理加权回归模型能够反映更多的异质特征。在总体相关结果下,局部区域可能出现方向相反或大小不等的相关结果。研究可为当地政策制定提供依据。     

基于光流法与特征统计的鱼群异常行为检测

鱼类群体行为的异常检测能够为鱼类健康监控与预警提供重要的方法和手段,对研究鱼类行为的机理,提升水产养殖过程中的信息化水平具有非常重要的意义。该文通过计算机视觉和图像处理技术,基于鱼群运动特征统计方法,对鱼群异常行为检测进行研究。利用Lucas-Kanade光流法得到目标鱼群的运动矢量,并对目标运动的行为特征进行统计,得到速度与转角这2个行为特征的联合直方图与联合概率分布。最后,在联合概率分布的基础上,基于标准互信息(normalized mutual information-NMI)和局部距离异常因子(local distance-based outlier factor-LDOF)2种方法对鱼群行为进行异常检测。试验结果表明,2种异常检测方法均达到99.5%以上的准确率。     

Reducing sampler error in soil research

Most soil sampling techniques are calibrated with standards and against each other, but the error introduced by the individuals performing each technique (sampler error) is often not addressed. We determined that sampler error explains a minimum of 5.5% of the variation in soil bulk density when using a composite coring technique. This source of error is of concern because it could easily obscure the small, but significant differences anticipated with long-term (decadal) research. We suggest three methods for increasing sampling accuracy across spatial and temporal treatments: (1) use > 5 individuals to sample each treatment; (2) consider both the concentration and pool size of soil properties (e.g. nitrate, root mass, etc.); and (3) include measurements of sampler error.     

An assessment of performance of wet atmospheric deposition samplers part 2. Validation of siting criteria

The variability in performance of 4 wet/dry atmospheric deposition samplers were compared for 1 yr. Samples were collected weekly and analyzed for pH, specific conductance, common ionic chemical constituents, and sample mass. Differences in the results between collectors were interpreted in terms of violations of siting criteria. Several of the criteria used in siting collectors for the National Atmospheric Deposition Program/National Trends networks were purposely compromised to determine the effect of such criteria on the validity of samples collected at sites where the location of samplers violates accepted siting criteria. This study showed that items should be excluded within a 45° cone of the sampler and also items of sufficient bulk to disturb wind patterns should be excluded within 5 m of the sampler. The nonnormal distribution of residuals from a parametric analysis of variance of the data set necessitated the application of the nonparametric Friedman test to assess comparability of chemical deposition and volume between and within samplers. Statistically significant differences existed for most comparisons, however, the test does not permit quantification of their magnitudes, although general trends may be determined. Two methods to try to estimate the magnitude of differences are presented. Differences in analyte concentrations between samplers were small.     

Long-term trends in NADP/NTN precipitation chemistry data: Results of different statistical analyses

This paper summarizes the results of four statistical approaches for the estimation of long-term trends (1983–92) in sulfate concentration data from 90 monitoring sites across the United States. Least squares regression models and nonparametric techniques were applied to these data. Sulfate concentrations were found to be generally decreasing on the order of 0–4% at most sites. There was general agreement that trends were significant in the Great Lakes, Pacific northwest, and southwest regions. Although strengths and weaknesses are described for each approach, all of these approaches are useful for long-term trend estimation. Visualization techniques are recommended for displaying trend patterns and associated levels of statistical significance.     

多时相遥感影像检测平乐县晚稻种植面积变化

为检测中国主要的粮食作物水稻的种植面积变化,该文以广西平乐县为例,利用多时相陆地卫星专题成像仪(landsat thematic mapper)影像数据和面向对象的分类方法,提取出晚稻种植的变化区域。该文探索了变化强度计算和阈值确定的方法,并利用晚稻在不同时相的影像光谱特征变化来提取晚稻种植区域。试验结果表明:3种变化强度计算方法中,变化向量分析法对河流、滩地变化的抑制效果优于相关系数方法,而相关系数方法对山体阴影的抑制效果则优于变化向量分析法,向量相似度法对山体阴影非常敏感,对水田变化则敏感度较低;3种阈值确定方法中最小错误率方法比最大类间方差法更为精确,比双窗口变步长阈值搜寻法更为稳定。综合利用3种变化提取方法对平乐县晚稻种植面积变化进行检测,得到变化检测混淆矩阵总正确率为96.8%,稻田面积变化误差为2.85%。该方法可为作物种植面积的变化检测提供参考。     

森林生态系统涡度相关法碳通量长时间连续性缺失数据插补方法的比较

针对森林通量观测站涡度相关法碳通量观测普遍存在的长时间连续性数据缺失情景，为探究不同数据插补方法的有效性，以华北低丘山地栓皮栎人工混交林生态系统为例，以经EddyPro处理和质量控制的2017年3月1日-11月30日0.5h尺度净生态系统碳交换（NEE）数据为基准数据集，随机生成含有连续1、3、7、15和31d数据缺失的5类数据缺失集，重复10次，使用固定窗口平均昼夜变化法（MDV）、可变窗口平均昼夜变化法（MDC）、查表法（LUT）、非线性回归法（NLR）、边际分布采样法（MDS）、人工神经网络法（ANN）对缺失数据集进行插补，并将插补数据与实际观测数据进行对比，通过分析统计参数来评估不同方法的插补精度和稳定性，以评估不同方法的适用范围。结果表明：日间，当连续缺失少于15d时，ANN方法插补数据与实测数据间的R2（决定系数）相对较高，NLR方法的R2较低；LUT方法插补数据与实测数据间的相对均方根误差（RRMSE）较低，NLR方法的RRMSE较高。当缺失达到连续15d时，除NLR方法的R2显著较低（P＜0.05）外，其它方法间R2差异不显著；LUT方法的RRMSE显著（P＜0.05）较低，其它方法间RRMSE差异不显著。当缺失达到连续31d时，除NLR方法R2显著较低（P＜0.05）外，各方法间R2和RRMSE无显著差异；MDV方法的平均绝对误差（MAE）出现较多异常值，各方法间的MAE开始出现分化的趋势。随着缺失片段长度的增加，除MDV方法外，各方法的R2呈下降趋势，连续1d缺失与连续31d缺失情景下插补所得NEE与实测NEE的R2差异显著（P＜0.05）；MDV和MDS方法的RRMSE呈增大趋势，连续1d缺失与连续31d缺失情景下的RRMSE差异显著（P＜0.05），其它方法的RRMSE差异相对不显著。夜间，在各缺失情景下，ANN方法的R2较高，LUT方法的R2较低，二者之间差异显著（P＜0.05）；LUT方法的RRMSE最高，与其它方法存在显著差异（P＜0.05）。在连续缺失大于31d的情景下，各方法的RRMSE差异均不显著。除LUT方法MAE显著（P＜0.05）较高外，其它方法的MAE无明显差异。随着缺失片段长度的增加，MDC、MDS和ANN方法插补数据的R2呈下降趋势，MDV和LUT的R2始终无显著差异；各方法的RRMSE差异无显著变化。在对典型晴天0.5h尺度上NEE日变化趋势的还原方面，MDC方法性能相对较优。综上，NLR方法适用于气象数据完备、NEE数据连续缺失少于7d的情景；MDV或MDC方法适用于气象数据不可用或缺失严重、NEE数据连续缺失少于15d的情景；LUT和MDS方法则适用于气象数据缺失较少、NEE数据连续缺失少于15d的情景；ANN方法适用性相对较广，可用于气象数据缺失较少、NEE数据连续缺失长达31d的情景。     

Identifying declines in waterbirds: The effects of missing data, population variability and count period on the interpretation of long-term survey data

To manage and conserve wildlife populations effectively it is necessary to use methods that identify the often non-linear trends in populations, have an inbuilt assessment of trend quality and can analyse count data from a range of spatial scales. We present a method of trend analysis using generalised additive models. These produce smoothed indices of abundance that can be used to assess population change from one or more sites or time periods, with any number of estimates of abundance per index period. We apply this method to count data collected under the Wetland Bird Survey, a national scheme that monitors waterbirds in the United Kingdom. To highlight declining populations, ‘alerts’ were raised if the population decline was equal to or greater than 50%. Significance was determined using bootstrapped confidence intervals for analyses that included many sites, or a novel Monte-Carlo method for single site analyses. The impact of missing data, species count variability and the number of months used to calculate the population change was greater at individual sites than for national datasets, which were relatively insensitive to changes in the above parameters. For single sites it is essential that three or more counts be made per index period if reliable estimates of population change are required. We propose that the method presented could be applied to a wide range of national or other monitoring schemes for a variety of taxa.     

Effect of tillage treatments upon soil tests for soil acidity,soil phosphorus and soil potassium at three soil depths

Abstract

Soil samples were obtained at 0–3, 3–6, 6–9 and 0–9 inch depths from experimental plots receiving five tillage treatments. Each of two samplers composited approximately six one‐inch cores from each plot. Soil samples were analyzed for acidity, P and K using routine analysis procedures in the University of Illinois Soil Testing Laboratory.

Few significant differences were attributed to sampler and it was concluded that samplers using similar sampling techniques were obtaining soil samples from the same population.

No significant differences in soil acidity at different depths were observed. The different tillage methods did significantly affect soil P at the 0–3 inch depth, but had no significant effect on soil P at deeper depths. Different tillage methods also significantly affected soil K values at different depths.     

基于模糊判别成分分析法的高光谱作物信息提取与分类

东北地区是中国主要的玉米种植区,同时也是中国易发生干旱的地区,干旱常态化严重制约着该地区玉米生产的稳定发展。以辽宁省春玉米为研究对象,在明确春玉米不同发育期干旱变化特征的基础上,基于FY-3A、MODIS、春玉米发育期和土壤相对湿度观测等数据,建立春玉米干旱遥感监测指标集,构建各发育期不同土层深度的土壤相对湿度遥感监测模型,并以2000年为例开展了辽宁省春玉米干旱监测的应用研究,结果表明：1993-2012年辽宁省春玉米在各个发育期均有干旱发生,其中1999-2002年为干旱高发期,乳熟期干旱最为严重;多指数协同配合能提高遥感手段对土壤相对湿度的监测能力,其中陆表水分指数对土壤相对湿度监测能力较强,其次是水分指数;利用构建的春玉米各发育期土壤相对湿度遥感监测模型,监测2001-2004年部分发育期和土层深度的干旱状况,总体监测准确率为73.32%;实现了2000年辽宁省春玉米发育期干旱等级动态监测,所得监测结果与当年农业气象观测记录在发育阶段和空间上都有很好的一致性,遥感监测结果正确。因此,此项研究对于大范围准确跟踪监测春玉米干旱,以及提高春玉米生产的防灾减灾能力具有重要意义。     

基于变化向量分析的冬小麦长势变化监测研究

现有的农作物长势遥感监测的基本思路是利用NDVI曲线形态变化与作物苗情变化的响应关系,提取特征参数,推测作物的生长发育状况.但由于表征NDVI时间序列曲线的特征参数较多,难以对所有特征参数进行全面变化分析.本研究引进变化向量分析理论,以东部五省冬小麦为研究对象,以1999-2005年SPOT-VGT的旬最大合成NDVI数据为主要数据源,采用Savizky-Golay滤波器重构NDVI时间序列,进而构建基于变化向量分析的长势监测模型,分别对研究区的年际与年内长势变化进行时间和空间上的定量分析.研究表明,变化向量分析方法能有效地从空间域和时间域反映东部五省冬小麦长势变化规律,以单一综合性指标综合了NDVI时间序列曲线的大多数特征参数,为农作物长势遥感监测提供了一种新的研究思路.     

Evaluation of Spatial Variability of PM10 Concentrations in London

This study aims to apply principal component analysis (PCA) to identify monitoring sites with similar variations of PM₁₀ concentrations in the London Air Quality Monitoring Network. This statistical methodology was applied to hourly average concentrations measured at 27 monitoring sites during the period from January 2000 to December 2009. The monitoring sites were selected according their efficiency in the study period (greater than 75% for each year). It was observed that the hourly average PM₁₀ concentrations were decreasing along the selected period at almost all monitoring sites. PCA was performed for each year, selecting the number of principal components (PCs) that had at least 95% of the original data variance. Analysing the frequency with which each pair of monitoring sites gave a significant contribution to the same PC, nine city areas with specific PM₁₀ behaviour were identified. Thus, monitoring sites with redundant measurements during the studied period were identified, being possible to remove them to decrease the costs relative to their maintenance or replace them to increase the monitored area.     

基于 GIS 的沿海与内陆城镇用地变化差异分析——以福建省漳州市和龙岩市为例

以 1996-2005 年漳州市和龙岩市的土地利用变更调查数据和经济发展数据为依据,运用 GIS 空间分析技术,对漳州市与龙岩市城镇用地变化及驱动因素进行对比分析.结果表明,1996-2005 年漳州市城镇用地现状结构变化快于龙岩市,在变化幅度、变化速度、区域分异方面存在明显差异性.导致两市城镇用地变化的区域差异主要是在自然因素差异基础上,不同的社会经济因素及政府决策行为叠加而造成的.     

广域岩土体变化监测研究

常规岩土体变化监测方法的局限性造成采用它们来实施大型库区及周边地表形变的持续监测非常困难.为了解决大型水电站库区岩土体的高效实时变化监测这一关键问题,选取乌东德水电站库区为研究区域,根据岩土体变化类型,分别采用D-InSAR技术和假彩色合成技术对库区进行监测,提取出变动区域.最后通过2种方法的监测结果与地面验证结果的比较,验证了变化地点和岩土变动类型与分析结果相一致,达到了发现变化、监视趋势的遥感监测目的.证明了D-InSAR技术和假彩色合成技术相结合用于监测广域岩土体变化的可行性.     

GCM降尺度预测泾河流域未来降水变化

利用统计降尺度模型SDSM确定预报因子,建立预报量与预报因子之间的统计关系,用独立的观测资料验证模型,并将其应用于HadCM3在A2和B2两种排放情景下的输出生成未来气候变化情景,预测了泾河流域未来3个时段（2020s,2050s,2080s）的降水变化,以期为流域未来气候变化情景的构建及当地农业的可持续发展提供参考依据。结果表明,泾河流域未来年降水量呈不显著的减少趋势且存在一定的季节差异。夏季降水减少趋势最明显,减幅随时间的推移逐渐增大;冬季呈现微弱增加趋势;春季与秋季变化趋势类似,均有增有减,减少的时期较增加的时期多。流域降水变化的空间差异大体呈现由东南部向西北部减小的趋势。     

基于植被指数的春玉米干旱响应遥感监测

东北地区是中国主要的玉米种植区,同时也是中国易发生干旱的地区,干旱常态化严重制约着该地区玉米生产的稳定发展。以辽宁省春玉米为研究对象,在明确春玉米不同发育期干旱变化特征的基础上,基于FY-3A/MERSI、Terra/MODIS、春玉米发育期和土壤相对湿度观测等数据,建立春玉米干旱遥感监测指标集,构建各发育期不同土层深度的土壤相对湿度遥感监测模型,并以2000年为例开展了辽宁省春玉米干旱监测的应用研究,结果表明：1993—2012年辽宁省春玉米在各个发育期均有干旱发生,其中1999—2002年为干旱高发期,乳熟期干旱最为严重;多指数协同配合能提高遥感手段对土壤相对湿度的监测能力,其中陆表水分指数对土壤相对湿度监测能力较强,其次是水分指数;利用构建的春玉米各发育期土壤相对湿度遥感监测模型,监测2001—2004年部分发育期和土层深度的干旱状况,总体监测准确率为73.32%;实现了2000年辽宁省春玉米发育期干旱等级动态监测,所得监测结果与当年农业气象观测记录在发育阶段和空间上都有很好的一致性,遥感监测结果正确。因此,此项研究对于大范围准确跟踪监测春玉米干旱,以及提高春玉米生产的防灾减灾能力具有重要意义。     

Application of artificial aging techniques to samples of rum and comparison with traditionally aged rums by analysis with artificial neural nets

Artificial aging techniques were applied to samples of rum. These samples were then compared, by artificial neural nets, with traditionally aged rums. Analysis was based on the phenolic and furanic composition of each sample. There were found to be few statistical differences between samples, thus confirming the possibility of applying artificial aging techniques to obtain rum with phenolic and furanic characteristics that are similar to those of rum obtained by traditional methods.     

A comparison of water flux measurements: passive wick‐samplers versus drainage lysimeters

Quantification of soil water flow is a prerequisite to accurate prediction of solute transfer within the unsaturated zone. The monitoring of these fluxes is challenging because the results are required to answer both scientific and practical questions regarding protection of groundwater, sustainable management of agricultural, forestry, mining or set‐aside industrial areas, reducing leachate loss from landfills or explaining the fate of environmentally harmful substances. Both indirect and direct methods exist for estimating water‐flux rates and have been used with varying success. In Europe, the use of direct lysimetry methods for measuring water and solute fluxes in soils has increased in recent years. This technique ensures reliable drainage data, but requires relatively large investment and maintenance expenses. Other research groups, especially in the USA, have developed alternative techniques. In this paper we compare the functioning of a passive‐wick sampler, especially the deep‐drainage meter type (DDM), with two different types of drainage lysimeters (weighing and non‐weighing) under field conditions in Germany for the measurement period from May 2004 until April 2009. The study showed that under sandy soil conditions no significant differences occurred between the measurements from DDM and both drainage lysimeter types. Only in periods with increased precipitation was there a tendency of drainage over‐estimation by the DDM in comparison with the lysimeters tested. For longer periods, no significant differences in the amount of drainage or the pattern of drainage formation were found between weighing and non‐weighing gravitation lysimeters. The practical use of DDMs is restricted because the groundwater level must be >2 m from the soil surface. Suggestions are made for the technical improvement of the DDM as well as the testing of the device with more cohesive soils.     

农牧交错带耕地数量变化对粮食产量影响的时空关联性——以内蒙古乌兰察布市为例

[目的]分析耕地变化和粮食生产的时空动态特征及其相互关联性,为研究区耕地集约利用水平的提高与粮食产量的增加提供科学依据。[方法]利用遥感解译获取的土地利用变化数据和粮食产量等统计数据,采用地理学和经济学理论、技术与方法,分析区域耕地与粮食产量的时空动态格局,构建"粮耕弹性系数"反映农牧交错区耕地资源变化与粮食生产的互动关系,揭示耕地数量变化对粮食产量的影响机理。[结果](1)1990—2000年耕地扩张明显,研究区中部、北部增加而东部减少。粮食产量逐渐上升,四子王旗粮食产量对耕地面积具有高敏感度,其它地区粮食产量对耕地数量变化的敏感度较低。(2)2000—2013年,耕地总量明显减少,耕地利用变化主要发生在研究区中部和南部,粮食产量出现先减少后增加的趋势,乌兰察布市东部和西部地区粮食综合生产能力明显提高,粮食产量对耕地数量变化的敏感度变低。[结论]1990—2013年乌兰察布市耕地数量变化对乌兰察布市粮食产量的影响显著。全市粮食产量对耕地数量变化的敏感度有着明显的区域差异,且2000—2013年的敏感度相对1990—2000年而言较低。