基于无人机可见光影像对喀斯特地区植被信息提取与覆盖度研究

为准确快速获取喀斯特石漠化地区植被和植被覆盖度信息，本研究利用四旋翼无人机采集喀斯特石漠化区域的可见光影像，选择过绿指数（EXG，excess green）、可见光波段差异植被指数（VDVI，visible-band difference vegetation indx）、红绿蓝植被指数（RGBRI，red green and blue vegetation index）、过绿减过红指数（ExG-ExR，excess green-excess red），利用植被指数时序图交点法和样本统计法思想，运用阈值分割法进行植被和植被覆盖度的信息提取，并以监督分类得到的植被和植被覆盖度信息为真实值，进行精度验证。结果表明，在贵州省关岭贞丰花江喀斯特石漠化地区，对于植被信息提取，ExG-ExR的精度最高，总体精度为95.56%，Kappa系数为0.919；ExG-ExR得到的植被覆盖度精度最好，为99.174%，均方根误差RMSE为0.097，R²为0.977。由此可见，在喀斯特地区利用植被指数时序图交点法和样本统计法思想，适合该地区的植被信息提取和植被覆盖度的提取，具有较高的精度。     

KM数学模型在预测、控制森林传染病中的新应用

采用传染病数学模型对我国松林松材线虫病疫情进行模拟,针对森林染病的特点,通过对数学模型阈值的分析研究,得出对染病类松树采取隔离措施是控制疫情发展的一种方法,但实施起来并非易事,因而本文提出了清除染病松树的切实可行措施,详细分析了移除倍率与染病松树总量最大值之间的关系,并对采取清除措施后疫情的发展进行了预测,提出了控制疫情发展的条件.     

Ecoenzymatic stoichiometry at the extremes: How microbes cope in an ultra-oligotrophic desert soil

Arid ecosystems are characterized by stressful conditions of low energy and nutrient availability for soil microorganisms. It has been observed that the ecoenzymes needed for the transformation of organic compounds into assimilable products show similar scaling relationships in different habitats (logarithmic C:N:P scaling ratios ∼1:1:1). In this study in Cuatro Ciénegas Basin (CCB) in the Chihuahuan desert of México, we report among the lowest ecoenzymatic activities yet quantified in soil. Nevertheless, activities for both organic N and organic P acquisition enzymes scale with C acquisition with a slope of ∼1.0, indicating that the soil microbial communities of this ultra-oligotrophic desert ecosystem follow the global ecoenzymatic stoichiometry patterns. CCB soil microbial communities were co-limited by C and either by N or P but this co-limitation played out differently in different parts of the CCB as indicated by microbial ecoenzymatic shift to allocate more resources to acquire and immobilize the scarcer nutrient. By extending ecoenzymatic analyses to these ultra-oligotrophic soils, our findings support the broad utility of the approach in illuminating how microbes acquire limiting resources in arid ecosystems.     

Linear–threshold animal model for birth weight, gestation length and calving ease in United Kingdom Limousin beef cattle data

Breeding value evaluation for UK Limousin beef cattle data was carried out by multiple-trait linear–threshold animal model with variance components assumed to be known. Polychotomous calving ease with five categories was analysed with two continuous traits: birth weight and gestation length. Field data consisted of 220,799 animals with observations with every possible combination of traits, and 270,035 animals in the pedigree. The threshold model was solved either with Newton Raphson or Expectation Maximisation algorithm, and solutions were compared to evaluation by a linear model with original and normalised scores. There were insignificant differences in solutions between the two algorithms for threshold model analyses. Furthermore, solutions of the continuous traits were similar by the threshold and linear models. For the categorical trait, correlations for random effects from the threshold and linear models were high. In case of normalised scores (original scores case in brackets) correlations with solutions from the threshold and linear model were 0.97 (0.94) and 0.97 (0.93) for direct and maternal genetic effects and 0.95 (0.89) for permanent maternal effects. Even so, at least one third of the top 1% ranking of bulls differed between the linear and the threshold models. Predictive abilities as correlations between estimated breeding values and pedigree indices were almost equal between the linear and threshold models for both continuous and categorical traits. In conclusion, despite the higher computational demand, the linear–threshold animal model can be seen worthwhile in the genetic evaluation of the national UK beef cattle data set.     

节水灌溉的土壤水分控制标准问题研究

农田土壤水分散失是在二个表面上进行的，一是土壤表面称之为蒸发，另外是作物表面称之为蒸腾。两者均与土壤水分呈线性相关，即随着土壤水分增大而增大。但土壤水分与光合速率关系研究表明，当土壤水分低于６５～６９％田持时，随着土壤水分增大光合速率增大，若土壤水分高于６５～６９％田持时，随着土壤水分增大光合速率降低。光合作用对土壤水分有一阈值反应。这一阈值可考虑为节水灌溉的田间土壤水分控制标准     

Threshold levels of cadmium for soil respiration and growth of spring wheat (Triticum aestivum L.), and difficulties with their determination

Summary Inconsistent results were obtained from comparative studies on the impact of increasing Cd contamination in three soils on growth of spring-wheat plantlets and soil respiration. With identical soil Cd loads, plant growth was increasingly inhibited in the following sequence: Neutral sandy hortisol (pH 7.0) < phaeosem (pH 6.9) < acidic cambisol (pH 5.6), suggesting a strong dependence on pH. In contrast, oxidation of a glucose-glutamate mixture by these soils was increasingly inhibited in the sequence: Acidic cambisol < neutral sandy hortisol < phaeosem. Inhibition of plant growth was correlated with the extractability of Cd from soils by 0.1 M CaCl₂. However, comparison of dose-response curves with dose-extractability and dose-uptake curves suggested the presence of a soil factor that modified plant uptake of available Cd. This factor, possibly the concentration of antagonistic cations, was apparently also active within the plants. The inconsistency in the responses of plant growth and of soil respiration with respect to the soil Cd load was ascribed to microbiological soil properties exceeding the importance of Cd availability. Relatively high in vitro sensitivity of prokaryotes and their biochemical interdependence together with relatively high diversities of streptomycetes and fungi were paralleled by a relatively strong inhibition of soil respiration.     

浅谈转基因产品阈值管理

"阈值管理"是指对所有认为有风险的事物(如混杂率、异交率、阳性率、致死率、危害率等)给出一个容许的"度",并以这个"度"为依据进行管理。实施阈值管理是保障生物安全并推动转基因技术健康发展的需要。文章分析了阈值的含义、设置原则以及当前阈值管理存在的问题和发展趋势。     

西苕溪流域主要经济林土壤磷素流失风险研究

经济林土壤磷素积累与潜在流失风险的研究对流域内磷素管理和面源污染控制十分必要。通过采样调查和室内分析研究了西苕溪流域主要经济林土壤测试磷的状况及磷素流失的潜在风险,调查采集了西苕溪流域安吉段主要经济林(毛竹、白茶、板栗)土壤样品105个,探讨了土壤理化性质对土壤磷素流失的影响以及土壤有效磷的控制阈值。结果表明,土壤水溶性磷(WEP)与土壤有机碳(SOC)、全磷(TP)呈极显著正相关(P<0.01),模拟酸雨浸提磷(SARP)也与土壤SOC、TP呈极显著正相关(P<0.01),土壤WEP、SARP与pH呈极显著负相关(P<0.01),但决定系数R2分别仅有0.187~0.280,影响相对较小。土壤WEP、SARP与有效磷(Bray 1-P)的关系可用分段线性回归方程描述(P<0.01),R2分别可达0.992、0.991,估算得出,与WEP、SARP相对应的土壤Bray 1-P的阈值分别为93.63、87.68 mg×kg^-1,后者较前者降低了5.95 mg×kg^-1。此外,土壤Bray 1-P含量超过40 mg×kg^-1、低于5 mg×kg^-1的样品占比分别可达17.14%、38.01%,缺磷与磷过度积累现象并存。土壤磷素的流失风险主要受土壤Bray 1-P、TP、SOC、pH等因素的影响,其中Bray 1-P是最重要的影响因子。酸雨会加大土壤磷素流失的潜在风险,作为酸雨频发区域的西苕溪流域,土壤有效磷水平的限制应更为严格。     

小波变换的模极大值在图像边缘检测中的应用研究

边缘检测在图像处理中有着重要的作用。依据canny算子的核心思想,运用小波分析技术,提出了一种基于小波变换的模极大值边缘检测算法。仿真结果表明：该算法能提取图像较弱的边缘,有较好的去噪效果,且边缘有较强的连续性,优于传统的边缘检测算子。     

关于土壤水分对风蚀起动风速影响研究的现状与问题

在土壤风蚀研究中,Bagnold建立的适用于干沉积物的起动风速与沉积物粒径的关系已被广泛接受。然而,沉积物中的水分亦是影响沙粒起动风速的重要因素,因为在湿润状态下,水分子与沉积物颗粒之间的张力增大了颗粒间的粘聚力,从而增大了起动风速。准确地估算土壤风蚀量或地表沙尘释放量及其所造成的危害需要建立沉积物起动风速与其含水率之间的关系,但这方面的研究目前远不能满足实际需要。本文较为全面地综述了近50年来有关沉积物起动风速与含水量关系方面的研究成果,将已有的研究成果归纳为半定量描述、经验模型和理论模型等三大类,分别介绍了代表性成果。通过对比发现,各家研究结论分歧较多,除了定性意义上较为相似外,定量关系五花八门,差异很大。作者将造成差异的原因归结为:问题本身的复杂性导致理论研究使用了过多的假设与简化、对颗粒之间的粘聚力与水分的关系缺乏清晰的认识、颗粒起动的判别标准与确定方法不同、实验过程中土壤水分的加入与确定方法以及实验土壤的颗粒特征和其他物理化学性质等不同。因此,这一领域的研究尚有很大的探索空间。