Michael N. Fienen

CrowdHydrology: Crowdsourcing Hydrologic Data and Engaging Citizen Scientists

Spatially and temporally distributed measurements of processes, such as baseflow at the watershed scale, come at substantial equipment and personnel cost. Research presented here focuses on building a crowdsourced database of inexpensive distributed stream stage measurements. Signs on staff gauges encourage citizen scientists to voluntarily send hydrologic measurements (e.g., stream stage) via text message to a server that stores and displays the data on the web. Based on the crowdsourced stream stage, we evaluate the accuracy of citizen scientist measurements and measurement approach. The results show that crowdsourced data collection is a supplemental method for collecting hydrologic data and a promising method of public engagement.

Development of a joint hydrogeophysical inversion approach and application to a contaminated fractured aquifer

This paper presents a joint inversion approach for combining crosshole seismic travel time and borehole flowmeter test data to estimate hydrogeological zonation. The approach is applied to a complex, fractured Department of Energy field site located at the Oak Ridge National Laboratory in Tennessee, United States. We consider seismic slowness (the inverse of seismic velocity) and hydrogeological zonation indicators as unknown variables and use a physically based model with unknown parameters to relate the seismic slowness to the zonation indicators. We jointly estimate all the unknown parameters in the model by conditioning them to the crosshole seismic travel times as well as the borehole flowmeter data using a Bayesian model and a Markov chain Monte Carlo sampling method. The fracture zonation estimates are qualitatively compared to bromide tracer breakthrough data and to uranium biostimulation experiment results. The comparison suggests that the joint inversion approach adequately estimated the fractured zonation and that the fracture zonation influenced biostimulation efficacy. Our study suggests that the new joint hydrogeophysical inversion approach is flexible and effective for integrating various types of data sets within complex subsurface environments and that seismic travel time data have the potential to provide valuable information about fracture zonation.

On constraining pilot point calibration with regularization in PEST

Ground water model calibration has made great advances in recent years with practical tools such as PEST being instrumental for making the latest techniques available to practitioners. As models and calibration tools get more sophisticated, however, the power of these tools can be misapplied, resulting in poor parameter estimates and/or nonoptimally calibrated models that do not suit their intended purpose. Here, we focus on an increasingly common technique for calibrating highly parameterized numerical models-pilot point parameterization with Tikhonov regularization. Pilot points are a popular method for spatially parameterizing complex hydrogeologic systems; however, additional flexibility offered by pilot points can become problematic if not constrained by Tikhonov regularization. The objective of this work is to explain and illustrate the specific roles played by control variables in the PEST software for Tikhonov regularization applied to pilot points. A recent study encountered difficulties implementing this approach, but through examination of that analysis, insight into underlying sources of potential misapplication can be gained and some guidelines for overcoming them developed.

Scripting MODFLOW model development using Python and FloPy

Graphical user interfaces (GUIs) are commonly used to construct and postprocess numerical groundwater flow and transport models. Scripting model development with the programming language Python is presented here as an alternative approach. One advantage of Python is that there are many packages available to facilitate the model development process, including packages for plotting, array manipulation, optimization, and data analysis. For MODFLOW-based models, the FloPy package was developed by the authors to construct model input files, run the model, and read and plot simulation results. Use of Python with the available scientific packages and FloPy facilitates data exploration, alternative model evaluations, and model analyses that can be difficult to perform with GUIs. Furthermore, Python scripts are a complete, transparent, and repeatable record of the modeling process. The approach is introduced with a simple FloPy example to create and postprocess a MODFLOW model. A more complicated capture-fraction analysis with a real-world model is presented to demonstrate the types of analyses that can be performed using Python and FloPy.

Approaches to highly parameterized inversion: Pilot-point theory, guidelines, and research directions

.........................................................................................................................................................

Using prediction uncertainty analysis to design hydrologic monitoring networks: Example applications from the Great Lakes water availability pilot project

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

A cross-validation package driving Netica with python

Bayesian networks (BNs) are powerful tools for probabilistically simulating natural systems and emulating process models. Cross validation is a technique to avoid overfitting resulting from overly complex BNs. Overfitting reduces predictive skill. Cross-validation for BNs is known but rarely implemented due partly to a lack of software tools designed to work with available BN packages. CVNetica is open-source, written in Python, and extends the Netica software package to perform cross-validation and read, rebuild, and learn BNs from data. Insights gained from cross-validation and implications on prediction versus description are illustrated with: a data-driven oceanographic application; and a model-emulation application. These examples show that overfitting occurs when BNs become more complex than allowed by supporting data and overfitting incurs computational costs as well as causing a reduction in prediction skill. CVNetica evaluates overfitting using several complexity metrics (we used level of discretization) and its impact on performance metrics (we used skill). Cross-validation avoids overfitting, improving predictive power of Bayesian Networks.CVNetica is a Python tool for cross-validation of Bayesian Networks.Cross-validation illustrated on a data-driven and a model emulation Bayesian Network.

Understanding the DayCent model

The ability of biogeochemical ecosystem models to represent agro-ecosystems depends on their correct integration with field observations. We report simultaneous calibration of 67 DayCent model parameters using multiple observation types through inverse modeling using the PEST parameter estimation software. Parameter estimation reduced the total sum of weighted squared residuals by 56% and improved model fit to crop productivity, soil carbon, volumetric soil water content, soil temperature, N2O, and soil NO 3 - compared to the default simulation. Inverse modeling substantially reduced predictive model error relative to the default model for all model predictions, except for soil NO 3 - and NH 4 + . Post-processing analyses provided insights into parameter-observation relationships based on parameter correlations, sensitivity and identifiability. Inverse modeling tools are shown to be a powerful way to systematize and accelerate the process of biogeochemical model interrogation, improving our understanding of model function and the underlying ecosystem biogeochemical processes that they represent. Several DayCent submodels were calibrated simultaneously using inverse modeling.Parameter estimation reduced DayCent total sum of weighted squared residuals by 56%.Soil temperature and water content are highly informative in DayCent calibration.Parameter estimation is an efficient way to calibrate soil biogeochemical models.Post-estimation analyses provide unique insights into model structure and function.

A python framework for environmental model uncertainty analysis

We have developed pyEMU, a python framework for Environmental Modeling Uncertainty analyses, open-source tool that is non-intrusive, easy-to-use, computationally efficient, and scalable to highly-parameterized inverse problems. The framework implements several types of linear (first-order, second-moment (FOSM)) and non-linear uncertainty analyses. The FOSM-based analyses can also be completed prior to parameter estimation to help inform important modeling decisions, such as parameterization and objective function formulation. Complete workflows for several types of FOSM-based and non-linear analyses are documented in example notebooks implemented using Jupyter that are available in the online pyEMU repository. Example workflows include basic parameter and forecast analyses, data worth analyses, and error-variance analyses, as well as usage of parameter ensemble generation and management capabilities. These workflows document the necessary steps and provides insights into the results, with the goal of educating users not only in how to apply pyEMU, but also in the underlying theory of applied uncertainty quantification. pyEMU is a python framework for model-independent uncertainty analysis and supports highly-parameterized inversion.pyEMU exposes several methods for data-worth analysis for designing observation networks and data collection activities.pyEMU can be used to estimate parameter and forecast uncertainty before inversion.pyEMU can be used to design objective functions and parameterizations.

Partial least squares for efficient models of fecal indicator bacteria on Great Lakes beaches.

At public beaches, it is now common to mitigate the impact of water-borne pathogens by posting a swimmers advisory when the concentration of fecal indicator bacteria (FIB) exceeds an action threshold. Since culturing the bacteria delays public notification when dangerous conditions exist, regression models are sometimes used to predict the FIB concentration based on readily-available environmental measurements. It is hard to know which environmental parameters are relevant to predicting FIB concentration, and the parameters are usually correlated, which can hurt the predictive power of a regression model. Here the method of partial least squares (PLS) is introduced to automate the regression modeling process. Model selection is reduced to the process of setting a tuning parameter to control the decision threshold that separates predicted exceedances of the standard from predicted non-exceedances. The method is validated by application to four Great Lakes beaches during the summer of 2010. Performance of the PLS models compares favorably to that of the existing state-of-the-art regression models at these four sites.