Search

Streamflow is one of the most important variables in hydrology, but it is difficult to measure continuously. As a result, nearly all streamflow time series are estimated from rating curves that define a mathematical relationship between streamflow and some easy-to-measure proxy like water surface elevation (stage). Despite the existence of automated methods, most rating curves are still fit manually, which can be time-consuming and subjective. Although several automated methods exist, they vary greatly in performance because of the non-convex nature of the problem. In this work, we develop a parameterization of the segmented power law that works reliably with minimal data, which could serve operationally or as a benchmark for evaluating other methods. The model, along with test data and tutorials, is available as an open-source Python package called ratingcurve. The implementation uses a modern probabilistic machine-learning framework, which is relatively easy to modify so that others can improve upon it.

The root-mean-squared error (RMSE) and mean absolute error (MAE) are widely used metrics for evaluating models. Yet, there remains enduring confusion over their use, such that a standard practice is to present both, leaving it to the reader to decide which is more relevant. In a recent reprise to the 200-year debate over their use, Willmott and Matsuura (2005) and Chai and Draxler (2014) give arguments for favoring one metric or the other. However, this comparison can present a false dichotomy. Neither metric is inherently better: RMSE is optimal for normal (Gaussian) errors, and MAE is optimal for Laplacian errors. When errors deviate from these distributions, other metrics are superior.

Much of modern science takes place in a computational environment, and, increasingly, that environment is programmed using R, Python, or Julia. Furthermore, most scientific data now live on the cloud, so the first step in many workflows is to query a cloud database and load the response into a computational environment for further analysis. Thus, tools that facilitate programmatic data retrieval represent a critical component in reproducible scientific workflows. Earth science is no different in this regard. To fulfill that basic need, we developed R, Python, and Julia packages providing programmatic access to the U.S. Geological Survey’s National Water Information System database and the multi-agency Water Quality Portal. Together, these packages create a common interface for retrieving hydrologic data in the Jupyter ecosystem, which is widely used in water research, operations, and teaching. Source code, documentation, and tutorials for the packages are available on GitHub. Users can go there to learn, raise issues, or contribute improvements within a single platform, which helps foster better engagement and collaboration between data providers and their users.

Cell adhesion molecules (CAMs) are important in prokaryotes and eukaryotes for cell-cell and cell-substratum interactions. The characteristics of adhesive proteins in the model diatom Phaeodactylum tricornutum were investigated by bioinformatic analysis and in vivo characterization. Bioinformatic analysis of the protein coding potential of the P. tricornutum genome used an amino-acid profile that we developed as a new system to identify uncharacterized or novel CAMs. Putative diatom CAMs were identified and seven were characterized in vivo, by generation of transgenic diatom lines overexpressing genes encoding C-terminal yellow fluorescent protein (YFP) fusion proteins. Three of these selected genes encode proteins with weak similarity to characterized proteins, a c-type lectin and two fasciclins, whereas the others are novel. The resultant cell lines were investigated for alterations in their adhesive ability. Whole cell-substratum adhesion strength was measured in a fully turbulent flow chamber, while atomic force microscopy was used to quantify the relative frequency of adhesion, as well as the length and strength of single molecules in the secreted mucilage. Finally, quartz crystal microbalance analysis characterized the visco-elastic properties and interaction of the mucilage-substratum interface. These combined studies revealed a range of phenotypes affecting adhesion, and led to the identification of candidate proteins involved in diatom adhesion. In summary, our study has for the first time combined bioinformatics and molecular physiological studies to provide new insights into diatom adhesive molecules., (© 2014 Phycological Society of America.)