Rapid estimation of compost enzymatic activity by spectral analysis method combined with machine learning

The aim of this study was to investigate the feasibility of using visible near-infrared (VisNIR) diffuse reflectance spectroscopy (DRS) as an easy, inexpensive, and rapid method to predict compost enzymatic activity, which traditionally measured by fluorescein diacetate hydrolysis (FDA-HR) assay. Compost samples representative of five different compost facilities were scanned by DRS, and the raw reflectance spectra were preprocessed using seven spectral transformations for predicting compost FDA-HR with six multivariate algorithms. Although principal component analysis for all spectral pretreatments satisfactorily identified the clusters by compost types, it could not separate different FDA contents. Furthermore, the artificial neural network multilayer perceptron (residual prediction deviation = 3.2, validation r² = 0.91 and RMSE = 13.38 μg g^?1 h^?1) outperformed other multivariate models to capture the highly non-linear relationships between compost enzymatic activity and VisNIR reflectance spectra after Savitzky–Golay first derivative pretreatment. This work demonstrates the efficiency of VisNIR DRS for predicting compost enzymatic as well as microbial activity.     

基于野外实测光谱的污灌区土壤重金属污染快速监测

利用光谱反射率反演土壤重金属含量相比传统实验室化学分析方法具有简单高效、成本低、非侵入性等优点。但光谱采集主要局限于实验室内,大大降低了监测的时效性。采集了龙口市污水灌溉区内的32个土壤样本,对9种重金属(Cr、Mn、Fe、Ni、Cu、Zn、As、Cd、Pb)进行了测定,并对土壤光谱进行了野外实测,同时对实测光谱进行光谱变换,提取了去除包络线(CR)、光谱倒数对数(lg(1/R))、一阶微分(RD1)、二阶微分(RD2)4种光谱指标,采用偏最小二乘回归方法(PLSR)建模。结果表明,污水灌溉区内存在严重的土壤重金属污染现象,As、Ni、Cd等重金属富集情况显著,与Fe、Mn呈显著相关的6种元素为Ni、Mn、As、Pb、Zn、Cr,相应的最优建模光谱指标分别为一阶微分(RD1)、二阶微分(RD2)、反射率(R)、反射率(R)、二阶微分(RD2)、去除包络线(CR),验证精度r分别为0.935 3、0.781 4、0.699 6、0.745 8、0.690 1、0.569 6。研究表明,利用野外实测光谱可以提高土壤重金属监测的时效性,预估土壤重金属含量具有一定的可行性,但针对不同的重金属应使用不同的光谱指标建模来达到最优效果。     

A pilot study on the effect of Cu,Zn, and Cd on the spectral curves and chlorophyll of wheat canopy at tiller stage

The aim of this study was to investigate canopy spectral reflectance responses to different levels of heavy metals copper (Cu), zinc (Zn), and cadmium (Cd) induced stress. Random blocks design experiment was conducted to simulate Cu, Zn, and Cd at five concentration levels. Chlorophyll and visible and near infrared canopy reflectance were measured for each treatment 56 days after seeding using chlorophyll meter SPAD-502 and spectroradiometer, respectively. New vegetative indices termed ratio of inclination angles (RIA), MERIS (the medium resolution imaging spectrometer) terrestrial chlorophyll index (MTCI), and normalized difference vegetation index (NDVI) were used to assess the response of canopy spectral reflectance to different heavy metal levels. Significant spectral variability especially in blue, red, and near infrared reflectance was observed for different heavy metal treatments. One-sample test showed significant difference for NDVI, MTCI, and RIA among five-level treatments at the 0.01 level. Correlation analysis demonstrated that the two vegetative indices (RIA and MTCI) were significantly correlated with chlorophyll meter values for Cd, Cu, and Zn treatments. However, NDVI was only significantly correlated with chlorophyll meter values for Cu and Zn treatments. Linear regression analysis also illustrated that RIA and MTCI were potential indices for predicting chlorophyll concentration with significant F test under Cu-, Zn-, and Cd-mediated stress.     

三峡库区典型消落带土壤及沉积物中溶解性有机质(DOM)的紫外-可见光谱特征

溶解性有机质(DOM)是陆地生态系统的重要组成部分,在环境地球化学中扮演着重要角色.在测定三峡库区典型消落带土壤及沉积物DOM的紫外-可见光谱基础上,通过特征参数及光谱模型来获取DOM的地化特征和信息.紫外-可见光谱特征参数表明土壤DOM的芳香性、疏水组分及腐殖化程度都明显低于上层沉积物,与高程和土地利用类型无关;而所选择的4种吸收光谱模型中以模型2为最优.同时,不同模型和不同拟合波段均对DOM光谱曲线斜率(S)有影响:S值随波长增加而减小;而不同波段S值相关性存在差异,表明各波段S值仅反映DOM部分信息,而非全部.此外,特征斜率之比(SR)反映出表层沉积物所受降解程度(光漂白或生物降解)小于土壤和往年沉积物,其芳香性更高,分子量更大,参与降解的反应活性也更大;这表明DOM越新鲜,其受到的光照和生物作用相对较小,降解潜力越大.     

利用紫外-可见吸收光谱估算三峡库区消落带水体、土壤和沉积物溶解性有机质(DOM)浓度

溶解性有机质(DOM)是陆地和水生生态系统的重要物质,有色溶解性有机质(CDOM)是可用光谱测定的DOM组分.基于2012年7月采集的土壤和沉积物样品以及2012年11月采集的水体样品,通过单波长、双波长、吸收光谱斜率(S值)和三波长模型建立CDOM与DOM浓度之间的关系.结果表明,4种模型中三波长模型效果最佳,水体数据模型的决定系数为0.788,土壤数据模型的决定系数为0.933,沉积物模型的决定系数为0.856.同时,利用2013年随机采集的32个土壤样品和36个水体样品对模型精度进行检验,结果表明,土壤DOM浓度的相对均方根误差(RRMSE)和平均相对误差(MRE)分别为16.5%和16.9%,水体分别为10.32%和9.06%,相比土壤DOM而言,三波长模型对水体DOM的预测精度更高.     

MATCHING NATURAL WATER COLORS TO MUNSELL STANDARDS1

ABSTRACT: Color of natural waters strongly influences their aesthetic appeal, suitability for recreational use, and aquatic habitat, but methods for routine specification of color have been lacking. An improved method has been developed for specifying water colors in the field. The water color, seen through an underwater viewer, is matched directly to Munsell standard patches observed simultaneously. The Munsell color-matching method was verified versus measurements of the underwater light field, made with a submersible spectroradiometer in 20 different natural waters in New Zealand (mainly lakes), which varied widely in color and other optical characteristics. Hue, the most important color attribute of natural waters, could be matched accurately; the saturation and brightness less so. Color standard patches are not available covering the full range of typically dark water colors, but fortunately, brightness can be measured with simple submersible light sensors. The Munsell color matching method seems suitable for routine water resources survey and monitoring.     

近场长周期强地震动幅值谱研究

本文研究了近场长周期强地震动加速度的傅氏谱和功率谱模型及其参数估计方法,基于强地震动参数统计上的无向性,提出了一种结合地震学模拟模型和工程学统计回归模型的综合方法。文中定义了强地震动平稳段均方根加速度的计算方法,并通过均方根加速度建立了震源力学参数和强地震动工程参数间的关系。通过对目前使用的主观持时定义的应用分析,指出主观持时定义使得估计的地震动过程的峰值加速度的变异性与实际地震动过程的峰值加速度的变异性有显著不同,明显偏小,这对于结构抗震可靠度分析可能会得到失效概度偏小的结果,应引起工程界的重视。     

Mobility of dimethoate residues from spring broccoli field

Dimethoate [O, O-dimethyl-S-(N-methylcarbamoyl-methyl) phosphorodithioate] is a broad-spectrum systemic insecticide currently used worldwide and on many vegetables in Kentucky. Dimethoate is a hydrophilic compound (log K_OW = 0.7) and has the potential of offsite movement from the application site into runoff and infiltration water. The dissipation patterns of dimethoate residues were studied on spring broccoli leaves and heads under field conditions. Following foliar application of Dimethoate 4E on broccoli foliage at the rate of 0.47 L acre^?1, dimethoate residues were monitored in soil, runoff water collected down the land slope, and in infiltration water collected from the vadose zone. The study was conducted on a Lowell silty loam soil (pH 6.9) planted with broccoli under three soil management practices: (i) soil mixed with municipal sewage sludge, (ii) soil mixed with yard waste compost, and (iii) no-mulch rototilled bare soil. The main objective of this investigation was to study the effect of mixing native soil with municipal sewage sludge or yard waste compost, having considerable amounts of organic matter, on off-site movement of dimethoate residues into runoff and infiltration water following spring rainfall. The initial deposits of dimethoate were 6.2 and 21.4 μ g g^?1 on broccoli heads and leaves, respectively. These residues dissipated rapidly and fell below the maximum residue limit of 2 μ g g^?1 on the heads and leaves after 10 and 14 d, respectively, with half-lives of 5.7 d on broccoli heads and 3.9 d on the leaves. Dimethoate residues detected in top 15 cm of soil (due to droplet drift and wash off residues from broccoli foliage) one day (d) following spraying, were 30.5 ng g^?1 dry soil in the sewage sludge treatment, and 46.1 and 134.5 ng g^?1 dry soil in the yard waste and no mulch treatments, respectively. Water infiltration was greater from yard waste compost treatment than from no mulch treatment, however concentrations of dimethoate in the vadose zone of the three soil treatments did not differ.     

The contents and forms of solid-phase species of radioactive strontium and cesium in Taiwan soils

This study was to investigate the activities and contents of ¹³⁷Cs in the profiles of selected arable and forest soils in Taiwan and various solid-phase species of ⁸⁵Sr and ¹³⁷Cs in selected arable soils in Taiwan. The gamma (γ) ray spectra of the collected soil samples and some of the soils amended with ⁸⁵Sr and ¹³⁷Cs were measured. The data indicate that the arable soils from Sanhsing series, Sanhsing Township and Chuangwei series, Chuangwei Township, Ilan County, and from Tunglochuan series, Pinglin Township, Taipei County shows significantly higher radioactivity of ¹³⁷Cs (ND − 11.0 ± 0.2 Bq kg⁻¹). Furthermore, the radioactivity of ¹³⁷Cs in the mountain soils (1.24 ± 0.07 − 42 ± 1 Bq kg⁻¹) from Yuanyang Lake Nature Preserve among Ilan, Taoyuan, and Hsinchu Counties is the highest among the investigated mountain forest soils. This may be mainly attributed to the fact that Ilan County is located in the northeastern part of Taiwan and faces the northeastern and northern seasonal winds with lots of precipitation annually from mid-autumn through mid-spring next year and is receiving greater amount of fallouts yearly. Due to longer reaction period (≥3 y) of ¹³⁷Cs with soil components, ¹³⁷Cs was mainly in the forms bound to oxides and to organic matter in the soil amended with ¹³⁷Cs and in the soil contaminated with ¹³⁷Cs. On the contrary, due to shorter reaction period (<60 d) of ⁸⁵Sr with soil components, ⁸⁵Sr was mainly in exchangeable form and partially in the forms bound to carbonates and oxides in the soils amended with ⁸⁵Sr.     

Analysis of surface and aerosol polarized reflectance for aerosol retrievals from polarized remote sensing in PRD urban region

A precise estimate of polarization induced by surface is crucial for polarized remote sensing dedicated to monitoring aerosol properties over urban area. The accurate knowledge of interaction between surface and aerosol polarized reflectance is essential for accurately achieving aerosol properties. In order to study surface and aerosol polarized reflectance for aerosol retrievals over urban area, a new airborne directional polarimetric camera (DPC) with high spatial resolution (4 m at 4000 m a.g.l) was developed. The surface polarized reflectance over distinct surface covers of urban area (forest, shrub, and soil) were studied using DPC measurements during a field campaign in the Pearl River Delta (PRD), China. The large variations were found in surface polarized reflectance of distinct urban covers due to surface type variability. For all surface types, the empirical BPDF model cannot describe accurately surface polarized reflectance at all possible illumination and observation geometries. From the quantitatively study of relationship between surface and aerosol polarized contribution to DPC measurements, we show that the polarized contributions of aerosol, which optical properties were defined by ground-based measurements, are much larger than the polarized contribution of surface, and found that the polarized contribution of surface covers increases with decreasing NDVI. The effect of polarization accuracy of measurements on aerosol retrieval was also investigated using DPC measurements, and found that 0.1% polarization accuracy of measurements can be neglected when AOD is retrieved using polarized measurements. Based on the information of effects of polarized reflectance differences between distinct surface covers and polarization accuracy of polarized measurements on retrieved aerosols over urban area, we found that the accuracy of aerosol retrieval over forest covers is higher than other surface types using polarized remote sensing.