Modelling Multi-Ion Transport in Saturated Soil and Parameter Estimation. II: Experimental

Multi-ion miscible displacement experiments were conducted

1 This paper is based on the M.S. thesis of the senior author. Chen Wenlin is now Research Associate at the Institute of Geography of the Academia Sinica, Beijing 100012, P.R. China.

with saturated soil columns under steady-state flow conditions. The experiments were carried out with four ionic species (Ca² Na⁺, SO₄2 and CI and the columns were 30 cm long. The experimental data were used to evaluate the multi-ion convective-dispersive miscible displacement model, developed in Part I of this study. Some parameters needed for this model were estimated with simplex and nonlinear least-square fitting methods. It was found that the model described the observed break through curves satisfactorily, when the soil matrix-solute interaction was described as an exchange process instead of an adsorption process.     

基于GLUE的土壤溶质运移参数反演及不确定性

溶质运移模型参数的识别结果常存在较高的不确定性,制约了模型的实际应用。以土壤中Cu2+运移过程为例,采用广义似然不确定性估计(Generalized likelihood uncertainty estimation,GLUE)并引入最大似然值(MaximumNash-Sutcliffe,MNS)等三种定量指标,探讨了数值反演估计弥散系数等参数的不确定性。结果表明,非线性最小二乘法(Nonlinear least squares,NLLS)得到的唯一"最优"参数组合对Cu2+出流曲线拟合效果很好(R20.937),但因"异参同效",无法刻画预测结果的不确定性。GLUE则可明确溶质运移参数及其响应界面的不确定性,MNS对应的参数组合对Cu2+出流曲线拟合R20.937,效果与NLLS的拟合结果高度一致。GLUE计算的95%置信区间覆盖了80%以上的观测点(NLLS为46.3%),其反演参数的取值范围也远大于NLLS的结果。在模型参数及响应界面不确定性分析两方面GLUE方法均优于NLLS方法。     

Tracking the Transport of Silver Nanoparticles in Soil: a Saturated Column Experiment

Silver nanoparticles (AgNPs) can enter the environment when released from products containing them. As AgNPs enter soil, they are often retained in the soil profile and/or leached to the groundwater. This research assessed the transport of AgNPs in their “particle form” through the soil profile using a series of columns. Three soil types were put into soil columns: LSH (loam with high organic matter (OM)), LSL (loam with low OM), and Sand (no OM). The results showed that AgNP transport and retention in soil as well as particle size changes are affected by soil organic matter (OM) and the cation exchange capacity (CEC) of soil. OM affected the transport and retention of AgNPs. This was evident in the LSH columns where the OM concentration was the highest and the AgNP content the lowest in the soil layers and in the effluent water. The highest transported AgNP content was detected in the Sand columns where OM was the lowest. CEC had an impact on the particle size of the AgNPs that were retained in the soil layers. This was clear in columns packed with high CEC-containing soils (LSL and LSH) where AgNP particle size decreased more substantially than in the columns packed with sand. However, the decrease in AgNP sizes in the effluent water was less than the decrease in particle size of AgNPs transported through but retained in the soil. This means that the AgNPs that reached the effluent were transported directly from the first layer through the soil macropores. This work highlights the ability to track AgNPs at low concentrations (50 μg kg^?1) and monitor the changes in particle size potential as the particles leach through soil all of which increases our knowledge about AgNP transport mechanisms in porous media.     

基于数据同化的土壤水力参数反演方法：研究进展与展望

土壤水力参数及其非均质性刻画关乎到诸多土壤与地下水等领域的量化模拟研究问题。受限于时间和采样成本,传统的直接测定方法并不能很好地解决这个问题。随着物联网技术的发展,与土壤水运动有关的一些状态表征量（如含水量和水头）已经能够通过传感器实时获得。如何充分融合这些观测数据信息,反演出土壤水力参数是当前的一个研究热点。数据同化方法能够通过融合观测数据与模型预测值信息,实现对模型参数的反演估计。本文系统分析了土壤水力参数不确定性的来源及测定方法,阐述了常用数据同化方法的基础理论及其在土壤水力参数反演方面的应用,并从计算效率和反演精度两方面着重论述了数据同化方法的最新前沿进展,最后探讨了数据同化方法未来的发展方向。研究表明：数据同化方法能够突破传统测定方法的限制,用于土壤水力参数及其非均质性刻画。尽管如此,由于土壤非饱和流模型的强非线性以及原位观测数据的相对稀缺性等问题的存在,当前数据同化方法的计算效率和反演精度还有待进一步提升。未来可从发展监督式降维方法、多源多尺度数据融合以及耦合物理规律的机器学习等方面深化土壤水力参数反演方法研究,这有利于农业土水管理、污染防治和修复等工作的合理开展。     

生物膜和土壤矿物对人工纳米颗粒在饱和多孔介质传输的影响

本文探究了人工纳米颗粒(NPs)在饱和多孔介质中的传输规律和影响机制，重点阐明生物膜和土壤矿物对纳米颗粒传输的影响及其机制。结果表明，在干净的石英砂介质中，不同种类和粒径NPs的传输效率具有明显差异，不同种类NPs传输效率表现为ZnO> CeO2>Fe2O3；对于报告粒径在20～100 nm之间的CeO2 NPs，粒径的增加有助于其在多孔介质的传输。溶液离子强度的增大会降低Zn O NPs的传输，而NPs浓度的增大不利于其在饱和多孔介质的传输，传统的DLVO理论能够很好地解释NPs在无涂层饱和多孔介质中的传输。生物膜和土壤矿物均能抑制纳米颗粒在饱和多孔介质的传输，其主要通过对纳米颗粒的吸附和异质聚集作用影响纳米颗粒的传输，非DLVO相互作用以及介质涂层的表面特性对增强纳米颗粒沉积有很大贡献。     

土壤胶体对砂质土柱中草萘胺迁移的影响

采用室内砂柱模拟试验方法研究了红壤和黑土两种土壤胶体对草萘胺在砂质土柱中迁移的影响。结果表明,整个淋滤过程中红壤和黑土胶体的回收率比较高,分别达到66．76％和74．75％;胶体促进了草萘胺在砂质土柱中的迁移,与对照相比（无胶体）,加入红壤和黑土胶体后草萘胺的回收率分别提高了9．82％和10．63％。但是吸附在胶体上迁移的草萘胺数量比较少,只占到迁移总量的1．15％（红壤）和1．68％（黑土）,物理非平衡机制可能在迁移过程中起支配作用。     

Reactive Transport Parameter Estimation and Global Sensitivity Analysis Using Sparse Polynomial Chaos Expansion

We present in this paper a new strategy based on the use of polynomial chaos expansion (PCE) for both global sensitivity analysis and parameter optimization. To limit the number of evaluations of the direct model, we develop a simple and efficient procedure to construct a sparse PCE where only coefficients that have a significant contribution to the variance of the model are retained. Parameter estimation is performed using an adaptive procedure where the intervals of variation of the parameters are progressively reduced using information from sensitivity analysis calculated using the sparse PCE. The strategy is shown to be effective for the parameter estimation of two reactive transport problems: a synthetic reactive transport problem involving the Freundlich sorption isotherm and a field experiment of Valocchi et al. (Water Resources Research 17:1517?C1527, 1981) involving nonlinear ion exchange reactions.     

生物炭颗粒在饱和多孔介质中的迁移与滞留

生物炭在实际生产实践中具有许多潜在的农业和环境效益,因此受到了越来越多的关注。生物炭在多孔介质中的迁移不仅会影响它在土壤中的归趋,还可能会影响微生物群体及土壤有机质的动态变化以及被吸附的污染物对环境的影响。实验通过柱实验研究了微米级生物炭颗粒在饱和多孔介质中的迁移和滞留特性,主要选取了四种影响生物炭迁移的潜在因素:原材料、裂解温度、背景溶液pH值和离子强度。实验结果显示微米级生物炭在饱和多孔介质中具有一定的迁移能力,但是大部分的微米级生物炭会滞留在饱和多孔介质的表面和孔隙间。生物炭的制备原材料对于生物炭的表面电势特性有较大影响,从而影响了微米级生物炭在饱和多孔介质中的迁移能力。热解温度越高,生物炭的表面电势越大,迁移能力越弱。随着背景溶液离子强度增加和pH值减小,微米级生物炭颗粒在多孔介质中滞留增加,迁移能力减弱。     

不同土水比土壤浸提液与饱和泥浆电导率的比较研究

针对饱和泥浆调制时饱和点的不确定性及其电导率测定的不稳定性等问题,为了统一和准确地表征暗管改碱的盐碱土盐渍化程度,采用室内化验与统计分析的方法,系统比较研究了饱和泥浆所需水量与饱和土壤溶液所需水量及其电导率和全盐与土水比1︰5、1︰2.5、1︰1、1︰0.5系列的相关和换算关系。研究结果表明:在黄河三角洲盐碱土(氯化物类型)区,可采用通过计算土壤孔隙度得到的饱和土壤溶液含水量来定量确定原来定性判定饱和点的饱和泥浆含水量,以饱和土壤溶液电导率可完全代替饱和泥浆电导率;饱和泥浆电导率数值和全盐含量均不是土水比1︰5、1︰2.5、1︰1、1︰0.5系列中的最高值,但均与其存在极显著的相关关系,饱和泥浆电导率与土水比1︰5土壤浸提液的电导率、饱和泥浆全盐含量与土水比1︰0.5全盐含量相关性最高,可分别用最佳拟合回归函数模型进行换算:一元线性函数模型Ece=2.042 8×EC1︰5+0.089 5、一元二次函数模型Tse=0.064 1×(TS1︰0.5)2?0.059×TS1︰0.5+0.397 9。     

SWAT模型土壤空间离散化对径流和输沙模拟结果的影响

以东江源为研究区,采用SWAT （Soil and Water Assessment Tool）模型,通过设置11种子流域划分方案,并保持土地利用和地形2个要素不变,以定量分析子流域划分引起的土壤离散化对产流产沙过程的影响。结果表明：（1）随着子流域数量的增加,土壤类型空间离散化趋势越明显,其中面积最大的泥红壤面积显著减少（P<0.05）,面积第2和第3的渗育质水稻土和黄壤显著增加（P<0.05）,面积最小的石质土在子流域数量少（11~267个）时该地类被概化完全消失,子流域数量增加到524个后,该地类面积呈增加趋势（P<0.05）。（2）土壤空间离散化导致年均径流量（1.32%）和输沙量减少（P<0.05）,且对输沙量影响更显著（18.07%,P<0.01）。（3）随着土壤离散程度增加,年平均最大1天、连续最大5天、连续最大7天输沙量均显著减少（16.64%,17.54%,17.34%,P<0.01）,而径流量变化不显著（0.59%,0.89%,0.83%,P>0.05）。（4）土壤空间离散化导致洪水过程输沙量峰值显著减少,而对洪峰径流量影响并不显著。子流域划分引起的土壤离散化主要通过土壤侵蚀因子K值变化而引起输沙量变化。研究结果将为分布式水文模型不确定性研究和提高模拟精度提供参考。     

Transport of Explosive Residue Surrogates in Saturated Porous Media

Department of Defense operational ranges may become contaminated by particles of explosives residues (ER) as a result of low-order detonations of munitions. The goal of this study was to determine the extent to which particles of ER could migrate through columns of sandy sediment, representing model aquifer materials. Transport experiments were conducted in saturated columns (2?×?20?cm) packed with different grain sizes of clean sand or glass beads. Fine particles (approximately 2 to 50?μm) of 2,6-dinitrotoluene (DNT) were used as a surrogate for ER. DNT particles were applied to the top 1?cm of sand or beads in the columns, and the columns were subsequently leached with artificial groundwater solutions. DNT migration occurred as both dissolved and particulate phases. Concentration differences between unfiltered and filtered samples indicate that particulate DNT accounted for up to 41% of the mass recovered in effluent samples. Proportionally, more particulate than dissolved DNT was recovered in effluent solutions from columns with larger grain sizes, while total concentrations of DNT in effluent were inversely related to grain size. Of the total DNT mass applied to the uppermost layer of the column, <3% was recovered in the effluent with the bulk remaining in the top 2?cm of the column. Our results suggest there is some potential for subsurface migration of ER particles and that most of the particles will be retained over relatively short transport distances.     

Processes in Pathogenic Biocolloidal Contaminants Transport in Saturated and Unsaturated Porous Media: A Review

There are several classes of subsurface colloids, abiotic and biotic. Basically, small particles of inorganic, organic and pathogenic biocolloids variety exist in natural subsurface system. Transport of these pathogenic biocolloidal contaminants (Viruses, bacteria and protozoa) pose a great risk in water resources and have caused large outbreaks of waterborne diseases. Biocolloid transport processes through saturated and unsaturated porous media is of significant interest, from the perspective of protection of groundwater supplies from contamination, assessment of risk from pathogens in groundwater and for the design of better water treatment systems to remove biocolloids from drinking water supplies This paper has reviewed the large volume of work that has already been done and the progress that has been made towards understanding the various basic multi-processes to predicting the biocolloid transport in saturated and unsaturated porous media. There are several basic processes such as physical, chemical and biological processes which are important in biocolloid transport. The physical processes such as advection, dispersion, diffusion, straining and physical filtration, adsorption and biological processes such as growth/decay processes and include active adhesion/detachment, survival and chemotaxis are strongly affected on biocolloid transport in saturated and unsaturated porous media. The unsaturated zone may play an important role in protecting aquifers from biocolloidal contamination by retaining them in the solid phase during their transport through the zone. Finally, author here highlighted the future research direction based on his critical review on biocolloid transport in saturated and unsaturated porous media.     

土壤厚度演化模型理论方法研究进展

全球平均土壤厚度仅约为1 m，但土壤厚度的空间分布信息在地貌、生态及水文科学等领域的研究和实践中具有重要价值。由于其具有显著的空间异质性，基于现有土壤制图产品、地球物理勘测及经验统计模型难以获取流域尺度土壤厚度分布信息，亟待发展土壤厚度预测的过程机理模型。本文回顾了土壤厚度演化模型理论方法的研究进展，评价了不同土壤生成及输移模型的适用性。研究指出土壤化学风化成土等机理仍不清晰是制约模型发展的理论瓶颈。此外，模型的方法体系有待完善，亟待进一步发展土壤生成率和输移率的函数形式及其参数的估计方法等，指出物理与随机结合的模拟方法以及基于数学物理途径的参数确定方法等有望解决模型应用中遇到的难题。最后，在土壤厚度演化模型基础上，提出发展基于流域协同演化理论的土壤发生学模型是定量预测土壤理化全要素发生所亟需突破的难点之一。     

Soil degradation: I. Basic processes

This paper examines soil and land degradation. It describes basic processes and factors responsible for degradation, illustrates the cause-effect relationships and differentiates between natural and man-induced regressive effects. The ‘critical limit’ concept is described in terms of properties beyond which the soil will not support an economically-viable agriculture. This paper is not an exhaustive literature review but emphasizes the scientific principles involved and highlights natural against man-induced processes. Important natural processes are: laterization, hard-setting, fragipan and clay pan formation, and geologic erosion. In comparison, man-induced processes consist of: soil compaction, accelerated erosion, desertification, salt accumulation and leaching and acidification. One of the principal constraints is the problem of data reliability. A reliable database and precise criteria are definitely lacking and hinder the assessment of the extent, type and degree of soil degradation and establishing the cause-effect scenario. Improving our database is, therefore, of a high priority if we are to adopt land use policy for sustainable soil management and long-range resource management. Also outlined, are vital research and development strategies. Judicious resource management policy should emphasize managing prime agricultural land to produce to its maximum potential so that there is no need to cultivate marginal and easily-degraded fragile ecosystems. A strict code of conduct is needed for utilizing marginal/fragile lands. Methods of restoring the productivity of degraded lands must be researched so as to minimize the need to clear and develop new lands.     

参数优化方法对微生物模型预测土壤有机碳时空演变的影响

准确把握土壤有机碳（SOC）的时空演变规律对于土壤资源的高效持续利用、发挥土壤生态系统服务功能，以及应对气候变化等均具有重要意义。以江苏省南部为研究区，以明确表达微生物分解作用的微生物模型MIMICS为对象，以模型参数敏感性分析为切入点，分析了不同参数优化方法对MIMICS模型预测苏南农田表层（0 ~ 20 cm）SOC时空演变动态的影响。结果表明，批处理和点对点两种参数优化方法下，MIMICS模型均能较好地模拟1980─2015年苏南农田表层SOC密度先增加后减少的总体趋势；采用考虑模型参数空间异质性的点对点参数优化方法时，MIMICS模型预测精度最高，其预测误差（RMSE）较采用默认参数值时分别降低22.2%（2000年独立验证）和14.7%（2015年独立验证），但2015年SOC密度预测精度依然偏低（R2 = 0.13，RMSE = 1.22 kg?m–2）。上述结果表明进一步改进微生物模型的结构、提高模型输入数据的精度及分辨率，将是微生物模型建模区域尺度SOC时空动态所面临的重要挑战。     

Modelling of Atmospheric Transport and Deposition of Pesticides

Modelling of atmospheric transport and deposition of pesticides is presented and discussed. Modelling on regional scale builds on the existing knowledge gained in other air pollution fields. An overview of current modelling studies on transport and deposition on a regional scale (typically 30-3000 km) is given. From these studies it is concluded that the models are capable in simulating the spatial distribution of the concentrations and depositions. However, large uncertainties are present in this type of modelling and are for the greater part induced by the uncertainty in the emissions and subsequently in the exchange process parameterisations and the physicochemical properties needed in the parameterisations. Many more measurement data are needed to validate the models.     

农田土壤压实过程及模型研究进展

农田土壤受到农业机械田间作业的影响发生压实板结,造成土壤孔隙率降低,容重和紧实度增大,限制水分入渗和根系生长,影响作物产量。随着我国农业机械化水平不断提高,土壤压实对农业可持续发展的影响引起了广泛的关注。本文通过文献调研,总结了土壤压实过程的国内外研究进展,对土壤压缩行为、压缩曲线与预固结压力的计算方法进行了梳理,综述了土壤压实机理和压实模型的发展历程和未来动向,可为推进农田土壤压实研究提供参考。     

基于非线性参数优化的水稻花前生育期参数估计

水稻生育期模型为复杂的非线性模型,其参数的合理标定是模型应用的重要环节。本文采用两种不同温度响应函数的花前生育期模型（MBETA和MBILN）,利用基于GML（Gauss-Marquardt-Levenberg）算法的模型独立参数优化程序PEST（model-independent parameter estimation）对模型参数进行优化,并在优化中引入参数先验信息和参数初始值扰动方法,以提高参数优化结果的可靠性。结果显示,参数先验信息有效降低了待优化参数的不确定性。最优参数值的95%置信区间较初始值域显著缩小。在优化得到的参数相关系数矩阵中也未显现出高度相关的参数。从目标函数值（?）序列看,MBETA和MBILN的?值最终收敛至相当接近的最小值,分别为11.71和11.82。但该最小值下两个模型的温度、光周期效应等参数值存在一定差异。这种差异平衡了不同温度响应方程与模型其它方程对水稻生育期模拟误差的贡献。在最优参数值组合下,两个模型验证结果表现一致。其中,抽穗开花期模拟值与实测值的相关性均通过了0.01水平的显著性检验。模拟误差主要来自幼穗分化期,与缺少对水稻光周期敏感始期的观测有关。本文优化方法降低了待优化参数收敛于局部小值的几率,对稳定参数优化和提高优化结果的可靠性具有重要作用。     

腐殖酸-高岭石胶体对铀(VI)在饱和多孔介质中迁移过程的影响

为探讨可溶性有机质存在条件下黏土矿物胶体对铀（VI）在饱和多孔介质中迁移的影响,选取2种石英砂（细砂和粗砂）分别构建均质构型的石英砂柱,研究了饱和水流条件下腐殖酸存在时高岭石胶体对铀在2种不同粒径石英砂中迁移的影响。结果表明,示踪溶质在2种均质石英砂柱中的穿透曲线相似且分布对称。当在石英砂柱中只通入铀（VI）溶液时,铀（VI）在2种粒径石英砂柱中的穿透量均较小（<25%）,此时大量的铀（VI）通过络合作用吸附在石英砂表面上。当通入溶液中存在腐殖酸和高岭石胶体时,铀（VI）在2种石英砂柱中的穿透量增加,且在600~850 μm石英砂中（58.36%）比250~425 μm石英砂中（42.68%）穿透量更多。这充分表明腐殖酸—高岭石胶体和石英砂粒径是影响铀（VI）在地下环境中迁移的重要因素。