Causes of systematic over‐ or underestimation of low streamflows by use of index‐streamgage approaches in the United States

Low‐flow characteristics can be estimated by multiple linear regressions or the index‐streamgage approach. The latter transfers streamflow information from a hydrologically similar, continuously gaged basin (‘index streamgage’) to one with a very limited streamflow record, but often results in biased estimates. The application of the index‐streamgage approach can be generalized into three steps: (1) selection of streamflow information of interest, (2) definition of hydrologic similarity and selection of index streamgage, and (3) application of an information‐transfer approach. Here, we explore the effects of (1) the range of streamflow values, (2) the areal density of streamgages, and (3) index‐streamgage selection criteria on the bias of three information‐transfer approaches on estimates of the 7‐day, 10‐year minimum streamflow (Q_{7, 10}). The three information‐transfer approaches considered are maintenance of variance extension, base‐flow correlation, and ratio of measured to concurrent gaged streamflow (Q‐ratio invariance). Our results for 1120 streamgages throughout the United States suggest that only a small portion of the total bias in estimated streamflow values is explained by the areal density of the streamgages and the hydrologic similarity between the two basins. However, restricting the range of streamflow values used in the index‐streamgage approach reduces the bias of estimated Q_{7, 10} values substantially. Importantly, estimated Q_{7, 10} values are heavily biased when the observed Q_{7, 10} values are near zero. Results of the analysis also showed that Q_{7, 10} estimates from two of the three index‐streamgage approaches have lower root‐mean‐square error values than estimates derived from multiple regressions for the large regions considered in this study. Published in 2011 by John Wiley & Sons, Ltd.     

Understanding and modelling spatial drain–aquifer interactions in a low‐lying coastal aquifer—the Thurne catchment,Norfolk, UK

Seawater intrusion into fresh groundwater formations generally results inadvertently from human activities, such as over‐abstraction from coastal aquifers. This article describes the data analysis to quantify drain–aquifer interactions in a low‐lying pump‐drained coastal aquifer, which is subject to saline intrusion due to widespread land drainage, and the resulting development and application of a numerical groundwater model to understand the spatial groundwater system behaviour (including groundwater salinity fluxes). Without measured flow data in this pump‐drained catchment, a novel groundwater head‐dependent approach to hydrograph separation is described. Time‐variant and time‐invariant MODFLOW analyses are utilised to examine the flow processes. A new approach to calculate drain coefficients, which represent the extensive network of drainage ditches in the regional model, using field information, is described; the sum of the drainage coefficients are close to the values independently estimated from the head‐dependent hydrograph separation. Results show that (1) the groundwater flows into the drainage systems are well reproduced using the new drain coefficients, (2) particle tracking of fresh and saline water can explain observed spatial salinity distribution within drainage networks and (3) the modelled flow of seawater across the coast is approximately 25% greater than that discharged by the pumps, demonstrating the need for drainage management to be aware of the slow response of groundwater systems to past drainage system changes. The article demonstrates that numerical groundwater modelling can produce the improved understanding needed to inform management decisions in such complex environments. Copyright © 2010 John Wiley & Sons, Ltd.     

Geophysical investigation of the Sandalp rock avalanche deposits

In the study of rock avalanche phenomena, numerical modelling makes use of back analyses of the rock avalanche propagation for calibration of the modelling assumptions and parameters. The back analyses require knowledge of the run-out area boundaries and the thickness distribution of the deposit. Geophysical methods can be applied to retrieve the thickness distribution, but, due to strong heterogeneities and logistic problems they are seldom applied. The aim of this work is to assess the potential of integrated geophysical methods to recognise and characterise a deposit created by two rock avalanches which occurred in the Sandalp valley (Switzerland) in 1996. The topography of the site before and after the rock avalanche is known and can be used as a benchmark. Resistivity tomography, seismic P-wave tomography, and active and passive surface wave analysis have been applied on several profiles deployed both on the rock avalanche deposit and in the surrounding area. Innovative approaches for surface wave analysis based on laterally constrained inversion and multimodal inversion have been applied to the data. A comparison of the results of the geophysical investigations with the topographic benchmark has shown the capability of the geophysical methods to locate the bottom of the deposit in the areas where the contrast with the host sediments properties is significant. In these areas, the deposit has higher resistivities and lower seismic velocities than the underlying materials. In the areas where the deposit is thicker and richer in fine-grained materials the geophysical parameters are not able to discriminate between the rock avalanche deposit and the underlying sediments. As a secondary task, the geophysical methods also allowed the bedrock pattern to be outlined.     

Lagrangian observation of phytoplankton dynamics at an artificially enriched subsurface water in Sagami Bay,Japan

Phytoplankton dynamics in the lower euphotic zone were observed by tracking a subsurface water released at 20-m depth from Takumi, an artificial upwelling device. Takumi continually discharged seawater pumped up from a depth of 205 m: this water was mixed with 5-m depth water to adjust the density to that of 20-m depth water of Sagami Bay, Japan. The discharged water was pulse-labeled at Takumi with uranine and tracked for 63.9 h with a drifting buoy equipped with a drogue at 20-m depth. We present a simple model to estimate in situ phytoplankton net growth rates from temporal changes in phytoplankton abundance in the discharged water with correction for the influence of water exchange between the discharged water and neighboring layers. Lagrangian observation showed active growth of pico- and nanophytoplankton, especially cryptophytes and Synechococcus (Cyanobacteria), in the subsurface layer. In contrast, diatoms grew little in spite of micromolar concentrations of nutrients. The active growth of pico- and nanophytoplankton was in good agreement with shipboard serial dilution culture experiments. The low growth activity of diatoms was suggested to be related to low light availability in the subsurface layer.     

Improved numerical analysis for ultimate behavior of elastomeric seismic isolation bearings

Elastomeric isolation bearings consist of multiple rubber layers with their top and bottom surfaces bonded to steel plates to restrict compressive deformation. Deformation constraints result in a variation of elastic modulus over the cross section of the rubber layers. In this paper, we describe a normalized compression modulus distribution on a circular rubber pad. The compressive and bending moduli of the rubber pad can be reproduced by applying the distribution to a series of axial springs. We also present a mechanical model for predicting the behavior of elastomeric seismic isolation bearings subject to large shear deformation and high compressive load. The mechanical model consists of a series of multiple shear springs at midheight and a series of axial springs at the top and bottom interfaces of the bearing. Simulation analyses of bearing tests were conducted to validate the proposed model. The analyses demonstrated that a model for circular lead-rubber bearings can successfully capture the influence of the axial load magnitude on the bearing shear behavior. The new model can simulate much more realistic behavior than prior models based on a uniform modulus assumption.     

Correlation between Expansion Rate of the Coronal Magnetic Field and Solar Wind Speed in a Solar Activity Cycle

Thirteen synoptic maps of expansion rate of the coronal magnetic field (CMF; RBR) calculated by the so-called ‘potential model’ are constructed for 13 Carrington rotations from the maximum phase of solar activity cycle 22 through the maximum phase of cycle 23. Similar 13 synoptic maps of solar wind speed (SWS) estimated by interplanetary scintillation observations are constructed for the same 13 Carrington rotations as the ones for the RBR. The correlation diagrams between the RBR and the SWS are plotted with the data of these 13 synoptic maps. It is found that the correlation is negative and high in this time period. It is further found that the linear correlation is improved if the data are classified into two groups by the magnitude of radial component of photospheric magnetic field, |B^pho_r|; group 1, 0.0 G ≦ |B_r^pho| < 17.8 G and group 2, 17.8 G ≦ |B_r^pho|. There exists a strong negative correlation between the RBR and the SWS for the group 1 in contrast with a weak negative correlation for the group 2. Group 1 has a double peak in the density distribution of data points in the correlation diagram; a sharp peak for high-speed solar wind and a low peak for low-speed solar wind. These two peaks are located just on the axis of maximum variance of data points in the correlation diagram. This result suggests that the solar wind consists of two major components and both the high-speed and the low-speed winds emanating from weak photospheric magnetic regions are accelerated by the same mechanism in the course of solar activity cycle. It is also pointed out that the SWS can be estimated by the RBR of group 1 with an empirical formula obtained in this paper during the entire solar activity cycle.     

High-resolution X-ray fluorescence profiling of hybrid event beds: Implications for sediment gravity flow behaviour and deposit structure

Hybrid event beds form when turbidity currents that transport or locally acquire significant quantities of mud decelerate. The mud dampens turbulence driving flow transformations, allowing both mud and sand to settle into dense, near-bed fluid layers and debris flows. Quantifying details of the mud distribution vertically in what are often complex tiered deposits is critical to reconstructing flow processes and explaining the diverse bed types left by mud-bearing gravity flows. High-resolution X-ray fluorescence core scanning provides continuous vertical compositional profiles that can help to constrain mud distribution at sub-millimetre scale, offering a significant improvement over discrete sampling. The approach is applied here to cores acquired from the Pennsylvanian Ross Sandstone Formation, western Ireland, where a range of hybrid event beds have been identified. Raw X-ray fluorescence counts are calibrated against element concentrations and mineral abundances determined on coincident core plugs, with element and element log-ratios used as proxies to track vertical changes in abundances of quartz, illite (including mica), chlorite and calcite cement. New insights include ‘stepped’ (to higher values) as opposed to ‘saw-tooth’ vertical changes in mud content and the presence of compositional banding that would otherwise be overlooked. Hybrid event beds in basin floor sheets that arrived ahead of the prograding fan system have significantly cleaner sandy components than those in mid-fan lobes. The latter may imply that the heads of the currents emerging from mid-fan channels entrained significant mud immediately before they collapsed. Many of the H3 debrites are bipartite with a sandier H3a division attributed to re-entrainment and mixing of a trailing debris or fluid mud flow (H3b) with sand left by the forward part of the flow. Hybrid event bed structure may thus partly reflect substrate interaction and mixing during deposition, and the texture of the bed divisions may not simply mirror those in the suspensions from which they formed.     

Tropospheric delay correction with dense GPS network in L1-SAIF augmentation

The L1-SAIF (L1 Submeter-class Augmentation with Integrity Function) signal is one of the Quasi-Zenith Satellite System (QZSS) navigation signals, which provides an augmentation function for mobile users in Japan. The tropospheric delay correction in the L1-SAIF augmentation is discussed in detail. Because the topographical features in Japan are complicated, the correction information is generated from GPS observation data collected at 200 GPS stations which are densely distributed over Japan. A total of 210 Tropospheric Grid Points (TGPs) are arranged to fully cover Japan. The TGPs that provide the correction information are selected adaptively to achieve the expected correction accuracy. This selection of TGPs is provided by the TGP mask message. Mobile users acquire the zenith tropospheric delay (ZTD) value at neighboring TGPs from the correction messages, and can estimate the local ZTD value accurately by using a suitable ZTD model. Only up to seven L1-SAIF messages are sufficient to provide the full correction information. Accuracy evaluations have proven that it is possible to achieve a correction accuracy of 13.4 mm RMS. The strategy presented here has been implemented into the augmentation system using the L1-SAIF signal, and its application guidance is presented in the QZSS interface specification.     

Simple Resistivity Probe System for Real-Time Monitoring of Injected Reagents

In situ treatment usually requires contact between an injected reagent and target contaminant to realize mass removal from source zones and plumes. Despite significant site characterization efforts, unknown heterogeneities that exist at all spatial scales often hinder prediction of the distribution of injected reagents. To provide remediation practitioners with additional information on the distribution of injected reagents, a prototype real-time monitoring probe was designed and tested. This new probe comprises a two-wire resistivity circuit, hence the designation dipole resistivity probe (DRP). The low-cost DRPs were built to be attached as arrays to a central stalk, and installed by direct-push techniques. An installed network of multilevel DRPs can be used to detect the arrival, persistence, and relative concentration of a high-conductivity reagent solution in real-time across a zone of interest. Static cell and sandbox experiments were conducted to test and refine the DRP design before field testing. Static cell experimental results indicated that the DRP was sensitive to solution electrical conductivity (EC), and that the probe response and EC relationship is nonlinear and dependent on the fixed resistor used. The choice of fixed resistor can be adjusted to optimize the DRP response over a critical EC range of interest. Under dynamic conditions in the sandbox, the DRP was able to reproduce breakthrough profiles collected by a commercial EC sensor. The results from two field studies demonstrated the utility of the DRPs to generate information regarding the arrival and persistence of reagents in an efficient and cost-effective manner. The first field study used a network of DRPs to monitor the land application of a sodium sulfate solution, and the second field study employed DRPs to monitor the distribution and longevity of a persulfate solution. While additional field testing is warranted, the results presented are encouraging and suggest that this low-cost system can be used to improve our understanding of the detailed migration of injected reagents in treatment zones.