Search

The shallow and deep hypothesis suggests that stream concentration-discharge (CQ) relationships are shaped by distinct source waters from different depths. Under this hypothesis, baseflows are typically dominated by groundwater and mostly reflect groundwater chemistry, whereas high flows are typically dominated by shallow soil water and mostly reflect soil water chemistry. Aspects of this hypothesis draw on applications like end member mixing analyses and hydrograph separation, yet direct data support for the hypothesis remains scarce. This work tests the shallow and deep hypothesis using co-located measurements of soil water, groundwater, and streamwater chemistry at two intensively monitored sites, the W-9 catchment at Sleepers River (Vermont, United States) and the Hafren catchment at Plynlimon (Wales). At both sites, depth profiles of subsurface water chemistry and stream CQ relationships for the 10 solutes analyzed are broadly consistent with the hypothesis. Solutes that are more abundant at depth (e.g., calcium) exhibit dilution patterns (concentration decreases with increasing discharge). Conversely, solutes enriched in shallow soils (e.g., nitrate) generally exhibit flushing patterns (concentration increases with increasing discharge). The hypothesis may hold broadly true for catchments that share such biogeochemical stratifications in the subsurface. Soil water and groundwater chemistries were estimated from high- and low-flow stream chemistries with average relative errors ranging from 24 to 82%. This indicates that streams mirror subsurface waters: stream chemistry can be used to infer scarcely measured subsurface water chemistry, especially where there are distinct shallow and deep end members.

Given the variable biogeochemical, physical, and hydrological processes driving fluvial sediment and nutrient export, the water science and management communities need data-driven methods to identify regions prone to production and transport under variable hydrometeorological conditions. We use Bayesian analysis to segment concentration-discharge linear regression models for total suspended solids (TSS) and particulate and dissolved phosphorus (PP, DP) using 22 years of monitoring data from 18 Lake Champlain watersheds. Bayesian inference was leveraged to estimate segmented regression model parameters and identify threshold position. The identified threshold positions demonstrated a considerable range below and above the median discharge—which has been used previously as the default breakpoint in segmented regression models to discern differences between pre and post-threshold export regimes. We then applied a Self-Organizing Map (SOM), which partitioned the watersheds into clusters of TSS, PP, and DP export regimes using watershed characteristics, as well as Bayesian regression intercepts and slopes. A SOM defined two clusters of high-flux basins, one where PP flux was predominantly episodic and hydrologically driven; and another in which the sediment and nutrient sourcing and mobilization were more bimodal, resulting from both hydrologic processes at post-threshold discharges and reactive processes (e.g., nutrient cycling or lateral/vertical exchanges of fine sediment) at prethreshold discharges. A separate DP SOM defined two high-flux clusters exhibiting a bimodal concentration-discharge response, but driven by differing land use. Our novel framework shows promise as a tool with broad management application that provides insights into landscape drivers of riverine solute and sediment export.

We present evidence of increasing persistence in daily precipitation in the northeastern United States that suggests that global circulation changes are affecting regional precipitation patterns. Meteorological data from 222 stations in 10 northeastern states are analyzed using Markov chain parameter estimates to demonstrate that a significant mode of precipitation variability is the persistence of precipitation events. We find that the largest region-wide trend in wet persistence (i.e., the probability of precipitation in 1 day and given precipitation in the preceding day) occurs in June (+0.9% probability per decade over all stations). We also find that the study region is experiencing an increase in the magnitude of high-intensity precipitation events. The largest increases in the 95th percentile of daily precipitation occurred in April with a trend of +0.7 mm/d/decade. We discuss the implications of the observed precipitation signals for watershed hydrology and flood risk.

[1] Organizing or clustering data into natural groups is one of the most fundamental aspects of understanding and mining information. The recent explosion in sensor networks and data storage associated with hydrological monitoring has created a huge potential for automating data analysis and classification of large, high-dimensional data sets. In this work, we develop a new classification tool that couples a Naive Bayesian classifier with a neural network clustering algorithm (i.e., Kohonen Self-Organizing Map (SOM)). The combined Bayesian-SOM algorithm reduces classification error by leveraging the Bayesian's ability to accommodate parameter uncertainty with the SOM's ability to reduce high-dimensional data to lower dimensions. The resulting algorithm is data-driven, nonparametric and is as computationally efficient as a Naive Bayesian classifier due to its parallel architecture. We apply, evaluate and test the Bayesian-SOM network using two real-world hydrological data sets. The first uses genetic data to classify the state of disease in native fish populations in the upper Madison River, MT, USA. The second uses stream geomorphic and water quality data measured at ∼2500 Vermont stream reaches to predict habitat conditions. The new classification tool has substantial benefits over traditional classification methods due to its ability to dynamically update prior information, assess the uncertainty/confidence of the posterior probability values, and visualize both the input data and resulting probabilistic clusters onto two-dimensional maps to better assess nonlinear mappings between the two.

The abundance of cosmogenic 36 Cl, measured in 41 limestone samples from a 9 m high bedrock fault scarp, allows us to construct the 14 kyr fault displacement history of the Nahef East normal fault, northern Israel (300 m above sea level, N33° latitude). The Nahef East fault is one of a series of fault scarps located along the 700 m high Zurim Escarpment, a major geomorphic feature. Samples at the top of the scarp have the highest nuclide concentrations (79 x 10 4 atoms (g rock) -1 ); samples at the base have the lowest (11 x 10 4 atoms (g rock) -1 ), Using chemical data from the samples, Nahef East fault scarp geometry, and surface and subsurface production rates for the 36 Cl-producing reactions, we have constructed a numerical model that calculates 36 Cl accumulation on a scarp through time, given a series of unique displacement scenarios. The resulting model 36 Cl concentrations are compared to those measured in the scarp samples. Faulting histories that result in a good match between measured and modeled 36 Cl abundances show three distinct periods of fault activity during the past 14 kyr with over 6 vertical meters of motion occurring during a 3 kyr time period in the middle Holocene. Smaller amounts of displacement occurred before and after the period of most rapid motion. The episodic behavior of the Nahef East fault indicates that the average displacement rate of this fault system has varied through time.

A technique for obtaining a (nearly) optimal scheme using multiple management periods has been developed. The method has been developed for very large scale combinatorial optimization problems. Simulated annealing has been extended to this problem. An importance function is developed to accelerate the search for good solutions. These tools have been applied to groundwater remediation problems at Lawrence Livermore National Laboratory (LLNL). A deterministic site-specific engineering-type flow and transport model (based on the public domain code SUTRA) is combined with the heuristic optimization technique. The objective is to obtain the time-varying optimal locations of the remediation wells that will reduce concentration levels of volatile organic chemicals in groundwater below a given threshold at specified areas on the LLNL site within a certain time frame and subject to a variety of realistic complicating factors. The cost function incorporates construction costs, operation and maintenance costs for injection and extraction wells, costs associated with piping and treatment facilities, and a performance penalty for well configurations that generate flow and transport simulations that exceed maximum concentration levels at specified locations. The resulting application reported here comprises a huge optimization problem. The importance function detailed in this paper has led to rapid convergence to solutions. The performance penalty allows different goals to be imposed on different geographical regions of the site; in this example, short-term off-site plume containment and long-term on-site cleanup are imposed. The performance of the optimization scheme and the effects of various trade-offs in management objectives are explored through examples using the LLNL site.

(1) Microbial biodiversity in groundwater and soil presents a unique opportunity for improving characterization and monitoring at sites with multiple contaminants, yet few computational methods use or incorporate these data because of their high dimensionality and variability. We present a systematic, nonparametric decision-making methodology to help characterize a water quality gradient in leachate-contaminated groundwater using only microbiological data for input. The data-driven methodology is based on clustering a set of molecular genetic-based microbial community proles. Microbes were sampled from groundwater monitoring wells located within and around an aquifer contaminated with landll leachate. We modied a self-organizing map (SOM) to weight the input variables by their relative importance and provide statistical guidance for classifying sample similarities. The methodology includes the following steps: (1) preprocessing the microbial data into a smaller number of independent variables using principal component analysis, (2) clustering the resulting principal component (PC) scores using a modied SOM capable of weighting the input PC scores by the percent variance explained by each score, and (3) using a nonparametric statistic to guide selection of appropriate groupings for management purposes. In this landll leachate application, the weighted SOM assembles the microbial community data from monitoring wells into groupings believed to represent a gradient of site contamination that could aid in characterization and long-term monitoring decisions. Groupings based solely on microbial classications are consistent with classications of water quality from hydrochemical information. These microbial community prole data and improved decision-making strategy compliment traditional chemical groundwater analyses for delineating spatial zones of groundwater contamination.

[1] Groundwater biogeochemistry is adversely impacted when municipal solid waste leachate, rich in nutrients and anthropogenic compounds, percolates into the subsurface from leaking landfills. Detecting leachate contamination using statistical techniques is challenging because well strategies or analytical techniques may be insufficient for detecting low levels of groundwater contamination. We sampled profiles of the microbial community from monitoring wells surrounding a leaking landfill using terminal restriction fragment length polymorphism (T-RFLP) targeting the 16S rRNA gene. Results show in situ monitoring of bacteria, archaea, and the family Geobacteraceae improves characterization of groundwater quality. Bacterial T-RFLP profiles showed shifts correlated to known gradients of leachate and effectively detected changes along plume fringes that were not detected using hydrochemical data. Experimental sediment microcosms exposed to leachate-contaminated groundwater revealed a shift from a β-Proteobacteria and Actinobacteria dominated community to one dominated by Firmicutes and δ-Proteobacteria. This shift is consistent with the transition from oxic conditions to an anoxic, iron-reducing environment as a result of landfill leachate-derived contaminants and associated redox conditions. We suggest microbial communities are more sensitive than hydrochemistry data for characterizing low levels of groundwater contamination and thus provide a novel source of information for optimizing detection and long-term monitoring strategies at landfill sites.

[1] This work contributes a combination of laboratory-based aquifer tracer experimentation and bias-aware Ensemble Kalman Filtering (EnKF) to demonstrate that systematic modeling errors (or bias) in source loading dynamics and the spatial distribution of hydraulic conductivity pose severe challenges for groundwater transport forecasting under uncertainty. The impacts of model bias were evaluated using an ammonium chloride tracer experiment conducted in a three dimensional laboratory tank aquifer with 105 near real-time sampling locations. This study contributes a bias-aware EnKF framework that (i) dramatically enhances the accuracy of concentration breakthrough forecasts in the presence of systematic, spatio-temporally correlated modeling errors, (ii) clarifies in space and time where transport gradients are maximally impacted by model bias, and (iii) expands the size and scope of flow-and-transport problems that can be considered in the future.

[1] A novel data-driven artificial neural network (ANN) that quantitatively combines large numbers of multiple types of soft data is presented for performing stochastic simulation and/or spatial estimation. A counterpropagation ANN is extended with a radial basis function to estimate parameter fields that reproduce the spatial structure exhibited in autocorrelated parameters. Applications involve using three geophysical properties measured on a slab of Berea sandstone and the delineation of landfill leachate at a site in the Netherlands using electrical formation conductivity as our primary variable and six types of secondary data (e.g., hydrochemistry, archaea, and bacteria). The ANN estimation fields are statistically similar to geostatistical methods (indicator simulation and cokriging) and reference fields (when available). The method is a nonparametric clustering/classification algorithm that can assimilate significant amounts of disparate data types with both continuous and categorical responses without the computational burden associated with the construction of positive definite covariance and cross-covariance matrices. The combination of simplicity and computational speed makes the method ideally suited for environmental subsurface characterization and other Earth science applications with spatially autocorrelated variables.