SepNet: A neural network for directionally correlated data.

Multi-dimensional tensor data appear in diverse settings, including multichannel signals, spectrograms, and hyperspectral data from remote sensing. In many cases, these data are directionally correlated, i.e.

Application of stereolithographic custom models for studying the impact of biofilms and mineral precipitation on fluid flow.

Here we introduce the use of transparent experimental models fabricated by stereolithography for studying the impacts of biomass accumulation, minerals precipitation, and physical configuration of flow paths on liquid flow in fracture apertures. The internal configuration of the models ranged in complexity from simple geometric shapes to those that incorporate replicated surfaces of natural fractures and computationally derived fracture surfaces. High-resolution digital time-lapse imaging was employed to qualitatively observe the migration of colloidal and soluble dyes through the flow models.

Toxicity of metals and metal mixtures: analysis of concentration and time dependence for zinc and copper.

The toxic effects of heavy metals, zinc and copper, in unary and binary solutions were studied using the Microtox acute toxicity test which relies upon the attenuation of light intensity emitted by Vibrio fischeri. The toxic effect Gamma (ratio of the light intensity lost at time t to the light intensity remaining at time t) of zinc could be related to its concentration [X] by a two-parameter equation Gamma=a(1-exp(-b[X])), where parameter a was a function of time and b equal to 0.88L/mg.