No abstract

We describe continental-scale increases in lake and stream total phosphorus (TP) concentrations, identified through periodic probability surveys of thousands of water bodies in the conterminous U.S. The increases, observed over the period 2000-2014 were most notable in sites in relatively undisturbed catchments and where TP was initially low (e.g., less than 10 μg L(-1)). Nationally, the percentage of stream length in the U.S. with TP ≤ 10 μg L(-1) decreased from 24.5 to 10.4 to 1.6% from 2004 to 2009 to 2014; the percentage of lakes with TP ≤ 10 μg L(-1) decreased from 24.9 to 6.7% between 2007 and 2012. Increasing TP concentrations appear to be ubiquitous, but their presence in undeveloped catchments suggests that they cannot be entirely attributed to either point or common non-point sources of TP.

Voice, the animus of storytelling, the author's presence in written works, concerns both sides of the modernist fence. Authors are urged to restrain and regulate their voices in deference to disciplinary expectations. We advocate developing an audibfe writer's voice that reflects our empirical experiences. Voice ranges from the evocative to the analytic and varies according to: (a) the freshness of the inquiry, (b) relationships with respondents, and (c) the place of the studied phenomena in larger systems of meaning and practice. Two different examples of our writing illustrate how fieldwork influenced the emergence of evocative and analytic voices.

No abstract

Regression analysis with a bounded outcome is a common problem in applied statistics. Typical examples include regression models for percentage outcomes and the analysis of ratings that are measured on a bounded scale. In this paper, we consider beta regression, which is a generalization of logit models to situations where the response is continuous on the interval (0,1). Consequently, beta regression is a convenient tool for analyzing percentage responses. The classical approach to fit a beta regression model is to use maximum likelihood estimation with subsequent AIC-based variable selection. As an alternative to this established - yet unstable - approach, we propose a new estimation technique called boosted beta regression. With boosted beta regression estimation and variable selection can be carried out simultaneously in a highly efficient way. Additionally, both the mean and the variance of a percentage response can be modeled using flexible nonlinear covariate effects. As a consequence, the new method accounts for common problems such as overdispersion and non-binomial variance structures.

No abstract

No abstract

No abstract