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.     

Dominant processes concept,model simplification and classification framework in catchment hydrology

Technological and methodological advances have facilitated tremendous growth in hydrology during the last century; however, there are also concerns that these advances indirectly contribute to additional problems in our research. An insight into hydrologic literature reveals our tendency to develop more complex models than perhaps needed, and our increasing emphasis on individual mathematical techniques rather than general hydrologic issues. Some recent studies of diverse forms have suggested that simplification in modeling and development of a common framework may help alleviate these problems. The present study is intended to bring such studies together towards a more coherent approach to research in catchment hydrology. This is done by highlighting the need for model simplification and generalization and proposing some potential directions for achieving such. Through a discussion of difficulties in data measurements, the need for moving beyond the notion of “modeling everything” to the notion of “capturing the essential features” is explained; the concept of dominant processes in model simplification and the utility of integration of concepts for modeling improvement are discussed. Formulation of a catchment classification framework is advocated as a possible means for a common framework in hydrology, and the role of dominant processes in this formulation is presented; the problems due to adoption of different modeling terminologies are highlighted and potential ways to overcome such are also discussed.     

OpenDA-NEMO framework for ocean data assimilation

Data assimilation methods provide a means to handle the modeling errors and uncertainties in sophisticated ocean models. In this study, we have created an OpenDA-NEMO framework unlocking the data assimilation tools available in OpenDA for use with NEMO models. This includes data assimilation methods, automatic parallelization, and a recently implemented automatic localization algorithm that removes spurious correlations in the model based on uncertainties in the computed Kalman gain matrix. We have set up a twin experiment where we assimilate sea surface height (SSH) satellite measurements. From the experiments, we can conclude that the OpenDA-NEMO framework performs as expected and that the automatic localization significantly improves the performance of the data assimilation algorithm by successfully removing spurious correlations. Based on these results, it looks promising to extend the framework with new kinds of observations and work on improving the computational speed of the automatic localization technique such that it becomes feasible to include large number of observations.     

Decision Support Modeling: Data Assimilation,Uncertainty Quantification,and Strategic Abstraction

We present a framework for design and deployment of decision support modeling based on metrics which have their roots in the scientific method. Application of these metrics to decision support modeling requires recognition of the importance of data assimilation and predictive uncertainty quantification in this type of modeling. The difficulties of implementing these procedures depend on the relationship between data that is available for assimilation and the nature of the prediction(s) that a decision support model is required to make. Three different data/prediction contexts are identified. Unfortunately, groundwater modeling is generally aligned with the most difficult of these. It is suggested that these difficulties can generally be ameliorated through appropriate model design. This design requires strategic abstraction of parameters and processes in a way that is optimal for the making of one particular prediction but is not necessarily optimal for the making of another. It is further suggested that the focus of decision support modeling should be on the ability of a model to provide receptacles for decision-pertinent information rather than on its purported ability to simulate environmental processes. While models are compromised in both of these roles, this view makes it clear that simulation should serve data assimilation and not the other way around. Data assimilation enables the uncertainties of decision-critical model predictions to be quantified and maybe reduced. Decision support modeling requires this.     

Assimilation of hydrologic and hydrometeorological data into distributed hydrologic model: Effect of adjusting mean field bias in radar-based precipitation estimates

This paper investigates the effect of adjusting the mean field bias (MFB) in radar-based precipitation data on analysis and prediction of streamflow and soil moisture in assimilating streamflow or streamflow and in situ soil moisture data into distributed hydrologic models. To evaluate the effect of adjusting the MFB under realistic as well as idealized conditions, both real-world and synthetic experiments are carried out for the Eldon Catchment on the border of Oklahoma and Arkansas in the US. In the synthetic experiment, the MFB is modeled as a stationary Markov chain process. The synthetic experiment showed that adjusting the MFB in the assimilation process significantly improves streamflow analysis when the initial conditions are known with reasonable certainty, and that assimilating soil moisture in addition to streamflow improves analysis of streamflow as well as soil moisture if the initial conditions are largely uncertain. Adjusting the MFB during the assimilation process noticeably improved streamflow analysis over ranges of the MFB and random noise in the precipitation data. On the other hand, increasing the MFB and random noise in the precipitation data tended to degrade soil moisture analysis due possibly to over-adjusting soil moisture to mitigate the precipitation error. The real-world experiment with one-year dataset showed that adjusting the MFB during the assimilation process helped capture the peak as well as volume of outlet flow analysis as well as prediction, and that additionally assimilating interior flow observations was necessary to improve analysis and prediction of peak flows at interior locations.     

A synthetic hydrologic‐response dataset

Synthetic data have long been employed in hydrology for model development and testing. The objective of this study was to generate a synthetic dataset of hydrologic response with higher spatial and temporal resolution than could presently be obtained in the field, spanning a longer period than the typical duration of monitoring campaigns in experimental catchments. The synthetic dataset was generated for a rangeland catchment with the Integrated Hydrology Model (InHM), and is presented for future use by the community. The InHM boundary‐value problem is based upon the previously reported hypothetical reality of Tarrawarra‐like hydrologic response. Whereas the emphasis in developing the hypothetical reality was on parameterising InHM to reproduce observations from the Tarrawarra catchment, the emphasis in generating the synthetic dataset is on developing an internally valid hydrologic‐response dataset that extends well beyond the period of observations at Tarrawarra. The synthetic dataset spans 11 years of continuous forcing and response data (e.g. integrated response, distributed fluxes, state variable dynamics). The dataset should be useful for a wide range of problems including evaluation of simple rainfall runoff modelling techniques, design of measurement networks, development of data‐assimilation algorithms, and studies on information theory. The dataset is available at: ftp://pangea.stanford.edu/pub/loague/ . Copyright © 2011 John Wiley & Sons, Ltd.     

A new method for establishing hydrologic fidelity of snow depth measurements based on snowmelt–runoff hydrographs

For snowmelt-driven flood studies, snow water equivalent (SWE) is frequently estimated using snow depth data. Accurate measurements of snow depth are important in providing data for continuous hydrologic simulations of such watersheds. A new hydrologic fidelity metric is proposed in this study to evaluate the potential contribution of particular snow depth datasets to flow characteristics using observed data and hydrologic modeling using the Variable Infiltration Capacity (VIC) model. Data-based hydrologic fidelity of snow depth measurements is defined as a categorical skill score between the snow depth in the watershed and the hydrograph peak or volume at the watershed outlet. Similarly, model-based hydrologic fidelity is defined as a categorical skill score between the model-simulated snow depth and the model-simulated hydrograph peak or volume. The proposed framework is illustrated using the Pecatonica River watershed in the USA, indicating which sites have a higher hydrologic fidelity, which is preferred in hydrologic studies.     

Nonlinear dynamics and chaos in hydrologic systems: latest developments and a look forward

During the last two decades or so, studies on the applications of the concepts of nonlinear dynamics and chaos to hydrologic systems and processes have been on the rise. Earlier studies on this topic focused mainly on the investigation and prediction of chaos in rainfall and river flow, and further advances were made during the subsequent years through applications of the concepts to other problems (e.g. data disaggregation, missing data estimation, and reconstruction of system equations) and other processes (e.g. rainfall-runoff and sediment transport). The outcomes of these studies are certainly encouraging, especially considering the exploratory stage of the concepts in hydrologic sciences. This paper discusses some of the latest developments on the applications of these concepts to hydrologic systems and the challenges that lie ahead on the way to further progress. As for their applications, studies in the important areas of scaling, groundwater contamination, parameter estimation and optimization, and catchment classification are reviewed and the inroads made thus far are reported. In regards to the challenges that lie ahead, particular focus is given to improving our understanding of these largely less-understood concepts and also finding ways to integrate these concepts with the others. With the recognition that none of the existing one-sided ‘extreme-view’ modeling approaches is capable of solving the hydrologic problems that we are faced with, the need for finding a balanced ‘middle-ground’ approach that can integrate different methods is stressed. To this end, the viability of bringing together the stochastic concepts and the deterministic concepts as a starting point is also highlighted.     

Data assimilation methods in the Earth sciences

Although remote sensing data are often plentiful, they do not usually satisfy the users’ needs directly. Data assimilation is required to extract information about geophysical fields of interest from the remote sensing observations and to make the data more accessible to users. Remote sensing may provide, for example, measurements of surface soil moisture, snow water equivalent, snow cover, or land surface (skin) temperature. Data assimilation can then be used to estimate variables that are not directly observed from space but are needed for applications, for instance root zone soil moisture or land surface fluxes. The paper provides a brief introduction to modern data assimilation methods in the Earth sciences, their applications, and pertinent research questions. Our general overview is readily accessible to hydrologic remote sensing scientists. Within the general context of Earth science data assimilation, we point to examples of the assimilation of remotely sensed observations in land surface hydrology.     

Optimal estimation of irrigation schedule – An example of quantifying human interferences to hydrologic processes

Reliable records of water use for irrigation are often lacking. This presents a difficulty for a qualified water use and water availability assessment. Quantification of the hydrologic cycle processes in regions of intensive agricultural practice requires irrigation as an input to hydrologic models. This paper presents a coupled forward-inverse framework to estimate irrigation schedule using remote-sensed data and data assimilation and optimization techniques. Irrigation schedule is treated as an unknown input to a hydro-agronomic simulation model. Remote-sensed data is used to assess actual crop evapotranspiration, which is used as the “observation” of the computed crop evapotranspiration from the simulation model. To handle the impact of model and observation error and the unknown biased error with irrigation inputs, a coupled forward-inverse approach is proposed, implemented and tested. The coupled approach is realized by an integrated ensemble Kalman filter (EnKF) and genetic algorithm (GA). The result from a case study demonstrates that the forward and inverse procedures in the coupled framework are complementary to each other. Further analysis is provided on the impact of model and observation errors on the non-uniqueness problem with inverse modeling and on the exactness of irrigation estimates.     

Spatio-temporal evolution and time-stable characteristics of soil moisture within remote sensing footprints with varying soil, slope, and vegetation

Air-borne passive microwave remote sensors measure soil moisture at the footprint scale, a scale of several hundred square meters or kilometers that encompasses different characteristic combinations of soil, topography, vegetation, and climate. Studies of within-footprint variability of soil moisture are needed to determine the factors governing hydrologic processes and their relative importance, as well as to test the efficacy of remote sensors. Gridded ground-based impedance probe water content data and aircraft-mounted Electronically Scanned Thinned Array Radiometer (ESTAR) pixel-average soil moisture data were used to investigate the spatio-temporal evolution and time-stable characteristics of soil moisture in three selected (LW03, LW13, LW21) footprints from the Southern Great Plains 1997 (SGP97) Hydrology Experiment. Better time-stable features were observed within a footprint containing sandy loam soil than within two pixels containing silty loam soil. Additionally, flat topography with split wheat/grass land cover produced the largest spatio-temporal variability and the least time stability in soil moisture patterns. A comparison of ground-based and remote sensing data showed that ESTAR footprint-average soil moisture was well calibrated for the LW03 pixel with sandy loam soil, rolling topography, and pasture land cover, but improved calibration is warranted for the LW13 (silty loam soil, rolling topography, pasture land) and LW21 (silty loam soil, flat topography, split vegetation of wheat and grass land with tillage practice) pixels. Footprint-scale variability and associated nonlinear soil moisture dynamics may prove to be critical in the regional-scale hydroclimatic models.     

Rapid simulated hydrologic response within the variably saturated near surface

Column and field experiments have shown that the hydrologic response to increases in rainfall rates can be more rapid than expected from simple estimates. Physics‐based hydrologic response simulation, with the Integrated Hydrology Model (InHM), is used here to investigate rapid hydrologic response, within the variably saturated near surface, to temporal variations in applied flux at the surface boundary. The factors controlling the speed of wetting front propagation are discussed within the Darcy–Buckingham conceptual framework, including kinematic wave approximations. The Coos Bay boundary‐value problem is employed to examine simulated discharge, pressure head, and saturation responses to a large increase in applied surface flux. The results presented here suggest that physics‐based simulations are capable of representing rapid hydrologic response within the variably saturated near surface. The new InHM simulations indicate that the temporal discretization and measurement precision needed to capture the rapid subsurface response to a spike increase in surface flux, necessary for both data‐based analyses and evaluation of physics‐based models, are smaller than the capabilities of the instrumentation deployed at the Coos Bay experimental catchment. Copyright © 2007 John Wiley & Sons, Ltd.     

Analysis on observing optimization for the wind-driven circulation by an adjoint approach

The adjoint approach is a variational method which is often applied to data assimilation widely in meteorology and oceanography. It is used for analyses on observing optimization for the wind-driven Sverdrup circulation. The adjoint system developed by Thacker and Long (1992), which is based on the GFDL Byran-Cox model, includes three components, i. e. the forward model, the adjoint model and the optimal algorithm. The GFDL Byran-Cox model was integrated for a long time driven by a batch of ideal wind stresses whose meridional component is set to null and zonal component is a sine function of latitudes in a rectangle box with six vertical levels and 2 by 2 degree horizontal resolution. The results are regarded as a "real" representative of the wind-driven Sverdrup circulation, from which the four dimensional fields are allowed to be sampled in several ways, such as sampling at the different levels or along the different vertical sections. To set the different samples, the fields of temperature, salinity and velocities function as the observational limit in the adjoint system respectively where the same initial condition is chosen for 4D VAR data assimilation. By examining the distance functions which measure the misfit between the circulation field from the control experiment of the adjoint system with a complete observation and those from data assimilation of adjoint approach in these sensitivity experiments respectively, observing optimizations for the wind-driven Sverdrup circulation will be suggested under a fixed observational cost.     

Calibration framework for a Kalman filter applied to a groundwater model

The paper presents a novel approach to the setup of a Kalman filter by using an automatic calibration framework for estimation of the covariance matrices. The calibration consists of two sequential steps: (1) Automatic calibration of a set of covariance parameters to optimize the performance of the system and (2) adjustment of the model and observation variance to provide an uncertainty analysis relying on the data instead of ad-hoc covariance values. The method is applied to a twin-test experiment with a groundwater model and a colored noise Kalman filter. The filter is implemented in an ensemble framework. It is demonstrated that lattice sampling is preferable to the usual Monte Carlo simulation because its ability to preserve the theoretical mean reduces the size of the ensemble needed. The resulting Kalman filter proves to be efficient in correcting dynamic error and bias over the whole domain studied. The uncertainty analysis provides a reliable estimate of the error in the neighborhood of assimilation points but the simplicity of the covariance models leads to underestimation of the errors far from assimilation points.     

Dual state-parameter estimation of root zone soil moisture by optimal parameter estimation and extended Kalman filter data assimilation

With well-determined hydraulic parameters in a hydrologic model, a traditional data assimilation method (such as the Kalman filter and its extensions) can be used to retrieve root zone soil moisture under uncertain initial state variables (e.g., initial soil moisture content) and good simulated results can be achieved. However, when the key soil hydraulic parameters are incorrect, the error is non-Gaussian, as the Kalman filter will produce a persistent bias in its predictions. In this paper, we propose a method coupling optimal parameters and extended Kalman filter data assimilation (OP-EKF) by combining optimal parameter estimation, the extended Kalman filter (EKF) assimilation method, a particle swarm optimization (PSO) algorithm, and Richards’ equation. We examine the accuracy of estimating root zone soil moisture through the optimal parameters and extended Kalman filter data assimilation method by using observed in situ data at the Meiling experimental station, China. Results indicate that merely using EKF for assimilating surface soil moisture content to obtain soil moisture content in the root zone will produce a persistent bias between simulated and observed values. Using the OP-EKF assimilation method, estimates were clearly improved. If the soil profile is heterogeneous, soil moisture retrieval is accurate in the 0-50 cm soil profile and is inaccurate at 100 cm depth. Results indicate that the method is useful for retrieving root zone soil moisture over large areas and long timescales even when available soil moisture data are limited to the surface layer, and soil moisture content are uncertain and soil hydraulic parameters are incorrect.     

Assimilation of snow covered area information into hydrologic and land-surface models

This paper describes a data assimilation method that uses observations of snow covered area (SCA) to update hydrologic model states in a mountainous catchment in Colorado. The assimilation method uses SCA information as part of an ensemble Kalman filter to alter the sub-basin distribution of snow as well as the basin water balance. This method permits an optimal combination of model simulations and observations, as well as propagation of information across model states. Sensitivity experiments are conducted with a fairly simple snowpack/water-balance model to evaluate effects of the data assimilation scheme on simulations of streamflow. The assimilation of SCA information results in minor improvements in the accuracy of streamflow simulations near the end of the snowmelt season. The small effect from SCA assimilation is initially surprising. It can be explained both because a substantial portion of snowmelts before any bare ground is exposed, and because the transition from 100% to 0% snow coverage occurs fairly quickly. Both of these factors are basin-dependent. Satellite SCA information is expected to be most useful in basins where snow cover is ephemeral. The data assimilation strategy presented in this study improved the accuracy of the streamflow simulation, indicating that SCA is a useful source of independent information that can be used as part of an integrated data assimilation strategy.     

复杂岩性储层参数评价中神经网络技术的应用

在复杂岩性储层中，储层四性关系比较复杂，表现为非线性，用传统的方法已经难以解决这类问题，为此引入了目前比较流行的人工神经网络技术、在前人基础上，以SN油田9井区为例进行储层研究工作，该区非均质性比较强，经对该区364口井常规到井资料进行储层参数重新解释，并做平面展布，与实际资料吻合较好。由此表明，神经网络技术在解决非线性问题上表现出了较大的优越性，值得我们做进一步的研究工作。     

On the use of multi‐algorithm,genetically adaptive multi‐objective method for multi‐site calibration of the SWAT model

With the availability of spatially distributed data, distributed hydrologic models are increasingly used for simulation of spatially varied hydrologic processes to understand and manage natural and human activities that affect watershed systems. Multi‐objective optimization methods have been applied to calibrate distributed hydrologic models using observed data from multiple sites. As the time consumed by running these complex models is increasing substantially, selecting efficient and effective multi‐objective optimization algorithms is becoming a nontrivial issue. In this study, we evaluated a multi‐algorithm, genetically adaptive multi‐objective method (AMALGAM) for multi‐site calibration of a distributed hydrologic model—Soil and Water Assessment Tool (SWAT), and compared its performance with two widely used evolutionary multi‐objective optimization (EMO) algorithms (i.e. Strength Pareto Evolutionary Algorithm 2 (SPEA2) and Non‐dominated Sorted Genetic Algorithm II (NSGA‐II)). In order to provide insights into each method's overall performance, these three methods were tested in four watersheds with various characteristics. The test results indicate that the AMALGAM can consistently provide competitive or superior results compared with the other two methods. The multi‐method search framework of AMALGAM, which can flexibly and adaptively utilize multiple optimization algorithms, makes it a promising tool for multi‐site calibration of the distributed SWAT. For practical use of AMALGAM, it is suggested to implement this method in multiple trials with relatively small number of model runs rather than run it once with long iterations. In addition, incorporating different multi‐objective optimization algorithms and multi‐mode search operators into AMALGAM deserves further research. Copyright © 2009 John Wiley & Sons, Ltd.     

Characterizing hydrological processes in a semiarid rangeland watershed: A hydrogeophysical approach

The complex ecohydrological processes of rangelands can be studied through the framework of ecological sites (ESs) or hillslope‐scale soil–vegetation complexes. High‐quality hydrologic field investigations are needed to quantitatively link ES characteristics to hydrologic function. Geophysical tools are useful in this context because they provide valuable information about the subsurface at appropriate spatial scales. We conducted 20 field experiments in which we deployed time‐lapse electrical resistivity tomography (ERT), variable intensity rainfall simulation, ground‐penetrating radar (GPR), and seismic refraction, on hillslope plots at five different ESs within the Upper Crow Creek Watershed in south‐east Wyoming. Surface runoff was measured using a precalibrated flume. Infiltration data from the rainfall simulations, coupled with site‐specific resistivity–water content relationships and ERT datasets, were used to spatially and temporally track the progression of the wetting front. First‐order constraints on subsurface structure were made at each ES using the geophysical methods. Sites ranged from infiltrating 100% of applied rainfall to infiltrating less than 60%. Analysis of covariance results indicated significant differences in the rate of wetting front progression, ranging from 0.346 m min^?1/2 for sites with a subsurface dominated by saprolitic material to 0.156 m min^?1/2 for sites with a well‐developed soil profile. There was broad agreement in subsurface structure between the geophysical methods with GPR typically providing the most detail. Joint interpretation of the geophysics showed that subsurface features such as soil layer thickness and the location of subsurface obstructions such as granite corestones and material boundaries had a large effect on the rate of infiltration and subsurface flow processes. These features identified through the geophysics varied significantly by ES. By linking surface hydrologic information from the rainfall simulations with subsurface information provided by the geophysics, we can characterize the ES‐specific hydrologic response. Both surface and subsurface flow processes differed among sites and are directly linked to measured characteristics.