Channel representation in physically based models coupling groundwater and surface water: pitfalls and how to avoid them.

Recent models that couple three-dimensional subsurface flow with two-dimensional overland flow are valuable tools for quantifying complex groundwater/stream interactions and for evaluating their influence on watershed processes. For the modeler who is used to defining streams as a boundary condition, the representation of channels in integrated models raises a number of conceptual and technical issues. These models are far more sensitive to channel topography than conventional groundwater models. On all spatial scales, both the topography of a channel and its connection with the floodplain are important. For example, the geometry of river banks influences bank storage and overbank flooding; the slope of the river is a primary control on the behavior of a catchment; and at the finer scale bedform characteristics affect hyporheic exchange. Accurate data on streambed topography, however, are seldom available, and the spatial resolution of digital elevation models is typically too coarse in river environments, resulting in unrealistic or undulating streambeds. Modelers therefore perform some kind of manual yet often cumbersome correction to the available topography. In this context, the paper identifies some common pitfalls, and provides guidance to overcome these. Both aspects of topographic representation and mesh discretization are addressed. Additionally, two tutorials are provided to illustrate: (1) the interpolation of channel cross-sectional data and (2) the refinement of a mesh along a stream in areas of high topographic variability.

Development and validation of a nephelometric immunoassay for IgG1 in milk.

A nephelometric immunoassay, with a detection range of 0.3 to 5 g IgG1/l, was (leveloped for the determination of immunoglobulin in bovine milk. The assay exhibited no significant cross-reactivity with alphas1-casein, alphas2-casein, beta-casein, K-casein or beta-lactoglobulin and 39% cross-reactivity with IgG2. The nephelometric assay was compared with ELISA and RID (24 h and 48 h incubations) assays using 105 duplicate milk samples covering IgG1 values ranging from 0.45 to 1.8 g. The results obtained from all assays showed good agreement with the exception of those obtained by the RID assay (24 h incubation) which gave lower results in samples containing more than 1.2 g IgG1/l. It was concluded that the nephelometric assay is a reliable, rapid and convenient method suitable for the quantification of IgG1 in milk. The assay can be configured for routine high-throughput milk quality assurance for IgG1 in dairy laboratories.