Hydrogeochemistry and strontium isotopes in the Arno River Basin (Tuscany, Italy): Constraints on natural controls by statistical modeling

In this paper the chemistry of major and trace elements and, for the first time, strontium isotopic ratios measured in running waters from the Arno River Basin (Tuscany, central-northern Italy) and thermal springs discharging in the same hydrographic system are presented and discussed. Classical graphical methods (e.g. mixing diagrams) have here been improved to identify, in a correct statistical sample space, extreme chemical compositions attributable to the action of geochemical processes and/or inherited from specific lithologies (namely contributions or components 1, 2 and 3) to be used in inverse modeling procedures, due to the absence of clear end-members. A linear least squares problem, with non-negativity constraints and distances, as required for compositional data (convex linear mixing) was solved by considering the contribution of the most discriminant variables given by the ⁸⁷Sr/⁸⁶Sr isotopic ratios and the concentration of Ca²⁺, Mg²⁺, Sr and Rb. Following such approach, it can be assumed (p < 0.05) that component 1, characterized by a Ca²⁺(Mg²⁺)– radiogenic-rich (⁸⁷Sr/⁸⁶Sr = 0.71274; Rb/Sr = 0.039), represents a dominant feature at basin scale, with a weight ranging from 69% to 100%. Much lower percentages are related to component 2, represented by a Ca²⁺(Mg²⁺)– facies with intermediate ⁸⁷Sr/⁸⁶Sr (0.70874) and low Rb/Sr (2.8 × 10⁻⁴) ratios and component 3, identified by Ca²⁺– facies with less radiogenic ⁸⁷Sr/⁸⁶Sr (0.70827) and low Mg²⁺/Ca²⁺ (0.011) ratios. These components are mainly dominated by the dissolution of evaporitic rocks and/or mixing with thermal waters in the southern part of the Arno River Basin and by dissolution of the carbonatic fraction, kinetically favored with respect to that of the silicatic minerals, in the upper reaches of the main course and its tributaries, respectively.     

An empirical model of event scale cohesive bank profile evolution

The highly stochastic nature of riverbank erosion has driven the need for spatially explicit empirical models. Detailed bank profile surveys along a meander bend of the Brandywine Creek in Pennsylvania, USA, before and after 28 high flow events over a 2·5 year period are used to develop an empirical model of cohesive bank profile erosion. Two hundred and thirty‐six bank erosion observations are classified as hydraulic erosion or subaerial erosion. Threshold conditions required to initiate bank erosion cannot be defined based on field measurements. Using the near‐bank velocity and the number of freeze–thaw cycles as predictors, regression equations are derived for hydraulic erosion that specify the length, thickness, and location on the bank face of eroded blocks. An empirical discriminant function defines the critical geometry of overhang failures, and the volumes removed by overhang failures are computed using another regression equation. All the regression equations are significant, but have low correlation coefficients, suggesting that cohesive bank erosion has a strong stochastic component. Individual events typically remove small masses of soil (average volume 0·084 m³/m) a few centimeters thick (median = 0·057 m) and a few decimeters in length (median = 0·50 m) from the lower third of the bank. Hydraulic erosion is responsible for 87% of all erosion. When applied to three survey sites not used in its development, the profile model predicts the total volume of erosion with errors of 23%, 5% and 1%. Twenty‐four percent of computed erosion volumes for single events are within 50% of observed volumes at these three sites. Extending the approach to decadal timescales and to entire bends will require three‐dimensional observations of bank failure, and spatially and temporally explicit methods to account for the influence of individual large trees on bank failures and near‐bank hydraulic processes. Copyright © 2009 John Wiley & Sons, Ltd.     

The effects of interdune vegetation changes on eolian dune field evolution: a numerical‐modeling case study at Jockey's Ridge,North Carolina,USA

Changes in vegetation cover within dune fields can play a major role in how dune fields evolve. To better understand the linkage between dune field evolution and interdune vegetation changes, we modified Werner's (Geology, 23, 1995: 1107–1110) dune field evolution model to account for the stabilizing effects of vegetation. Model results indicate that changes in the density of interdune vegetation strongly influence subsequent trends in the height and area of eolian dunes. We applied the model to interpreting the recent evolution of Jockey's Ridge, North Carolina, where repeat LiDAR surveys and historical aerial photographs and maps provide an unusually detailed record of recent dune field evolution. In the absence of interdune vegetation, the model predicts that dunes at Jockey's Ridge evolve towards taller, more closely‐spaced, barchanoid dunes, with smaller dunes generally migrating faster than larger dunes. Conversely, the establishment of interdune vegetation causes dunes to evolve towards shorter, more widely‐spaced, parabolic forms. These results provide a basis for understanding the increase in dune height at Jockey's Ridge during the early part of the twentieth century, when interdune vegetation was sparse, followed by the decrease in dune height and establishment of parabolic forms from 1953‐present when interdune vegetation density increased. These results provide a conceptual model that may be applicable at other sites with increasing interdune vegetation cover, and they illustrate the power of using numerical modeling to model decadal variations in eolian dune field evolution. We also describe model results designed to test the relative efficacy of alternative strategies for mitigating dune migration and deflation. Installing sand‐trapping fences and/or promoting vegetation growth on the stoss sides of dunes are found to be the most effective strategies for limiting dune advance, but these strategies must be weighed against the desire of many park visitors to maintain the natural state of the dunes. Copyright © 2009 John Wiley & Sons, Ltd.     

Enhanced application of root‐reinforcement algorithms for bank‐stability modeling

Riparian vegetation is known to exert a number of mechanical and hydrologic controls on bank stability. In particular, plant roots provide mechanical reinforcement to a soil matrix due to the different responses of soils and roots to stress. Root reinforcement is largely a function of the strength of the roots crossing potential shear planes, and the number and diameter of such roots. However, previous bank stability models have been constrained by limited field data pertaining to the spatial and temporal variability of root networks within stream banks. In this paper, a method is developed to use root‐architecture data to derive parameters required for modeling temporal and spatial changes in root reinforcement. Changes in root numbers over time were assumed to follow a sigmoidal curve, which commonly represents the growth rates of organisms. Regressions for numbers of roots crossing potential shear planes over time showed small variations between species during the juvenile growth phase, but extrapolation led to large variations in root numbers by the time the senescent phase of the sigmoidal growth curve had been reached. In light of potential variability in the field data, the mean number of roots crossing a potential shear plane at each year of tree growth was also calculated using data from all species and an additional sigmoidal regression was run. After 30 years the mean number of roots predicted to cross a 1 m shear plane was 484, compared with species‐specific curves whose values ranged from 240 roots for black willow trees to 890 roots for western cottonwood trees. In addition, the effect of spatial variations in rooting density with depth on stream‐bank stability was modeled using the bank stability and toe erosion model (BSTEM). Three root distributions, all approximating the same average root reinforcement (5 kPa) over the top 1 m of the bank profile, were modeled, but with differing vertical distributions (concentrated near surface, non‐linear decline with depth, uniform over top meter). It was found that stream‐bank F_S varied the most when the proportion of the failure plane length to the depth of the rooting zone was greatest. Copyright © 2008 John Wiley & Sons, Ltd.     

基于三维模拟的海洋CSEM资料处理

海洋可控源电磁法已经成为海洋油气勘探一个重要工具,但是其资料处理和解释还处于定性和一维模拟阶段.在积分方程三维模拟的基础上对Troll油田实测数据进行了处理,采用人机交互三维模拟寻找背景模型和异常体初始模型,最后对异常体电阻率采用准线性近似快速反演,取得了定量的结果.同时,说明对于二维测线和二维模型依然可以用三维来模拟,其结果优于二维反演.在电子计算机技术快速发展的今天,可以预计三维反演将成为资料处理解释的主流.     

An easily implemented agro-hydrological procedure with dynamic root simulation for water transfer in the crop–soil system: Validation and application

Models for water transfer in the crop–soil system are key components of agro-hydrological models for irrigation, fertilizer and pesticide practices. Many of the hydrological models for water transfer in the crop–soil system are either too approximate due to oversimplified algorithms or employ complex numerical schemes. In this paper we developed a simple and sufficiently accurate algorithm which can be easily adopted in agro-hydrological models for the simulation of water dynamics. We used a dual crop coefficient approach proposed by the FAO for estimating potential evaporation and transpiration, and a dynamic model for calculating relative root length distribution on a daily basis. In a small time step of 0.001 d, we implemented algorithms separately for actual evaporation, root water uptake and soil water content redistribution by decoupling these processes. The Richards equation describing soil water movement was solved using an integration strategy over the soil layers instead of complex numerical schemes. This drastically simplified the procedures of modeling soil water and led to much shorter computer codes. The validity of the proposed model was tested against data from field experiments on two contrasting soils cropped with wheat. Good agreement was achieved between measurement and simulation of soil water content in various depths collected at intervals during crop growth. This indicates that the model is satisfactory in simulating water transfer in the crop–soil system, and therefore can reliably be adopted in agro-hydrological models. Finally we demonstrated how the developed model could be used to study the effect of changes in the environment such as lowering the groundwater table caused by the construction of a motorway on crop transpiration.     

Modeling the dynamic process of tsunami earthquake by liquid-solid coupling model

Tsunami induced by earthquake is an interaction problem between liquid and solid.Shallow-water wave equation is often used to modeling the tsunami,and the boundary or initial condition of the problem is determined by the displacement or velocity field from the earthquake under sea floor,usually no interaction between them is consid-ered in pure liquid model.In this study,the potential flow theory and the finite element method with the interaction between liquid and solid are employed to model the dynamic processes of the earthquake and tsunami.For model-ing the earthquake,firstly the initial stress field to generate the earthquake is set up,and then the occurrence of the earthquake is simulated by suddenly reducing the elastic material parameters inside the earthquake fault.It is dif-ferent from seismic dislocation theory in which the relative slip on the fault is specified in advance.The modeling results reveal that P,SP and the surface wave can be found at the sea surface besides the tsunami wave.The surface wave arrives at the distance of 600 km from the epicenter earlier than the tsunami 48 minutes,and its maximum amplitude is 0.55 m,which is 2 times as large as that of the sea floor.Tsunami warning information can be taken from the surface wave on the sea surface,which is much earlier than that obtained from the seismograph stations on land.The tsunami speed on the open sea with 3 km depth is 175.8 m/s,which is a little greater than that pre-dicted by long wave theory,(gh)1/2=171.5 m,and its wavelength and amplitude in average are 32 km and 2 m,respectively.After the tsunami propagates to the continental shelf,its speed and wavelength is reduced,but its amplitude become greater,especially,it can elevate up to 10 m and run 55 m forward in vertical and horizontal directions at sea shore,respectively.The maximum vertical accelerations at the epicenter on the sea surface and on the earthquake fault are 5.9 m/s2 and 16.5 m/s2,respectively,the later is 2.8 times the former,and therefore,sea water is a good shock     

利用流固耦合模型模拟地震和海啸全过程

海底地震引起的海啸过程在力学上是一个流固耦合问题。地震引起的海底变形会影响流体的运动,流体运动会影响地震引起的海底变形。海啸的数值模拟,通常采用浅水波控制方程,把地震引起的海底变形作为海啸波动的边界条件或初始条件,不考虑它们之间的相互作用。本文采用势流体的流固耦合有限元方法模拟了地震和海啸的全过程。地震过程的模拟与地震位错模型不同,在位错模型中,断层的位错是事先指定的;而在本文中,首先形成自重作用下的初始应力场,然后通过断层材料的突然软化引起的错动,模拟地震震源的动力学过程。模拟结果显示,在海面除了可以看到大振幅的海啸波外,还可以发现体波震相和面波震相。在600 km的海面震中距上,它们要比海啸波早到48分钟,在此处面波的最大平均振幅可达0.55 m,是相同震中距海底面波最大平均振幅的2倍。因此,海啸预警信息在海面可以比在地表更早地得到。海啸波的传播速度在水深3 km的开阔海面是175.8 m/s,它要比理想长波理论预测的大,其平均振幅为2 m,波长可达32 km. 到达大陆架后其速度、波长都减小,在岸边可以激起10 m高的巨浪,水平方向深入陆地达53 m。震中附近海面和地震断层上的最大垂直加速度分别为5.9 m/s2和16.3 m/s2,后者是前者的2.8倍。由此看来,海水是很好的减震器。海啸波的加速度到达岸边会衰减10倍。与加速度不同,海面震中处的振动速度为3.2 m/s, 是海底震源处的1.4倍。震源处的最大位移小于震中海面的最大位移, 其差就是海啸波源的振幅。值得注意的是,海底地震的最大位错在震后23 s达到,不是发生在断层滑动的开始。      

Generalized likelihood uncertainty estimation (GLUE) using adaptive Markov Chain Monte Carlo sampling

In the last few decades hydrologists have made tremendous progress in using dynamic simulation models for the analysis and understanding of hydrologic systems. However, predictions with these models are often deterministic and as such they focus on the most probable forecast, without an explicit estimate of the associated uncertainty. This uncertainty arises from incomplete process representation, uncertainty in initial conditions, input, output and parameter error. The generalized likelihood uncertainty estimation (GLUE) framework was one of the first attempts to represent prediction uncertainty within the context of Monte Carlo (MC) analysis coupled with Bayesian estimation and propagation of uncertainty. Because of its flexibility, ease of implementation and its suitability for parallel implementation on distributed computer systems, the GLUE method has been used in a wide variety of applications. However, the MC based sampling strategy of the prior parameter space typically utilized in GLUE is not particularly efficient in finding behavioral simulations. This becomes especially problematic for high-dimensional parameter estimation problems, and in the case of complex simulation models that require significant computational time to run and produce the desired output. In this paper we improve the computational efficiency of GLUE by sampling the prior parameter space using an adaptive Markov Chain Monte Carlo scheme (the Shuffled Complex Evolution Metropolis (SCEM-UA) algorithm). Moreover, we propose an alternative strategy to determine the value of the cutoff threshold based on the appropriate coverage of the resulting uncertainty bounds. We demonstrate the superiority of this revised GLUE method with three different conceptual watershed models of increasing complexity, using both synthetic and real-world streamflow data from two catchments with different hydrologic regimes.     

A land surface data assimilation framework using the land information system: Description and applications

The Land Information System (LIS) is an established land surface modeling framework that integrates various community land surface models, ground measurements, satellite-based observations, high performance computing and data management tools. The use of advanced software engineering principles in LIS allows interoperability of individual system components and thus enables assessment and prediction of hydrologic conditions at various spatial and temporal scales. In this work, we describe a sequential data assimilation extension of LIS that incorporates multiple observational sources, land surface models and assimilation algorithms. These capabilities are demonstrated here in a suite of experiments that use the ensemble Kalman filter (EnKF) and assimilation through direct insertion. In a soil moisture experiment, we discuss the impact of differences in modeling approaches on assimilation performance. Provided careful choice of model error parameters, we find that two entirely different hydrological modeling approaches offer comparable assimilation results. In a snow assimilation experiment, we investigate the relative merits of assimilating different types of observations (snow cover area and snow water equivalent). The experiments show that data assimilation enhancements in LIS are uniquely suited to compare the assimilation of various data types into different land surface models within a single framework. The high performance infrastructure provides adequate support for efficient data assimilation integrations of high computational granularity.