Neural Network Input Selection for Hydrological Forecasting Affected by Snowmelt1

Abstract: Snowmelt largely affects runoff in watersheds in Nordic countries. Neural networks (NN) are particularly attractive for streamflow forecasting whereas they rely at least on daily streamflow and precipitation observations. The selection of pertinent model inputs is a major concern in NNs implementation. This study investigates performance of auxiliary NN inputs that allow short‐term streamflow forecasting without resorting to a deterministic snowmelt routine. A case study is presented for the Rivière des Anglais watershed (700 km²) located in Southern Québec, Canada. Streamflow (Q), precipitations (rain R and snow S, or total P), temperature (T) and snow lying (A) observations, combined with climatic and snowmelt proxy data, including snowmelt flow (Q_SM) obtained from a deterministic model, were tested. NN implemented with antecedent Q and R produced the largest gains in performance. Introducing increments of A and T to the NNs further improved the performance. Long‐term averages, seasonal data, and Q_SM failed to improve the networks.     

Task knowledge overlap and knowledge variety: the role of advice network structures and impact on group effectiveness

Drawing from the structural perspective of social capital theory, this research investigates how internal and external advice network structures influence knowledge overlap and variety and, how these knowledge dimensions in turn influence group effectiveness. Findings from two studies on knowledge‐intensive groups indicate that different advice network structures are associated with knowledge overlap and knowledge variety, and only knowledge variety was significantly associated with group effectiveness. In addition, despite implicit understanding that advice networks aid performance through enhancing knowledge outcomes, only knowledge variety was found to mediate the relationship between external network and group effectiveness. Implications for theory and practice are discussed. Copyright © 2007 John Wiley & Sons, Ltd.     

Gastrointestinal Bacterial Transmission among Humans,Mountain Gorillas,and Livestock in Bwindi Impenetrable National Park,Uganda

Abstract: Habitat overlap can increase the risks of anthroponotic and zoonotic pathogen transmission between humans, livestock, and wild apes. We collected Escherichia coli bacteria from humans, livestock, and mountain gorillas (Gorilla gorilla beringei) in Bwindi Impenetrable National Park, Uganda, from May to August 2005 to examine whether habitat overlap influences rates and patterns of pathogen transmission between humans and apes and whether livestock might facilitate transmission. We genotyped 496 E. coli isolates with repetitive extragenic palindromic polymerase chain reaction fingerprinting and measured susceptibility to 11 antibiotics with the disc‐diffusion method. We conducted population genetic analyses to examine genetic differences among populations of bacteria from different hosts and locations. Gorilla populations that overlapped in their use of habitat at high rates with people and livestock harbored E. coli that were genetically similar to E. coli from those people and livestock, whereas E. coli from gorillas that did not overlap in their use of habitats with people and livestock were more distantly related to human or livestock bacteria. Thirty‐five percent of isolates from humans, 27% of isolates from livestock, and 17% of isolates from gorillas were clinically resistant to at least one antibiotic used by local people, and the proportion of individual gorillas harboring resistant isolates declined across populations in proportion to decreasing degrees of habitat overlap with humans. These patterns of genetic similarity and antibiotic resistance among E. coli from populations of apes, humans, and livestock indicate that habitat overlap between species affects the dynamics of gastrointestinal bacterial transmission, perhaps through domestic animal intermediates and the physical environment. Limiting such transmission would benefit human and domestic animal health and ape conservation.     

A Behaviorally Explicit Demographic Model Integrating Habitat Selection and Population Dynamics in California Sea Lions

Abstract: Although there has been a call for the integration of behavioral ecology and conservation biology, there are few tools currently available to achieve this integration. Explicitly including information about behavioral strategies in population viability analyses may enhance the ability of conservation biologists to understand and estimate patterns of extinction risk. Nevertheless, most behavioral‐based PVA approaches require detailed individual‐based data that are rarely available for imperiled species. We present a mechanistic approach that incorporates spatial and demographic consequences of behavioral strategies into population models used for conservation. We developed a stage‐structured matrix model that includes the costs and benefits of movement associated with 2 habitat‐selection strategies (philopatry and direct assessment). Using a life table for California sea lions (Zalophus californianus), we explored the sensitivity of model predictions to the inclusion of these behavioral parameters. Including behavioral information dramatically changed predicted population sizes, model dynamics, and the expected distribution of individuals among sites. Estimated population sizes projected in 100 years diverged up to 1 order of magnitude among scenarios that assumed different movement behavior. Scenarios also exhibited different model dynamics that ranged from stable equilibria to cycles or extinction. These results suggest that inclusion of behavioral data in viability models may improve estimates of extinction risk for imperiled species. Our approach provides a simple method for incorporating spatial and demographic consequences of behavioral strategies into population models and may be easily extended to other species and behaviors to understand the mechanisms of population dynamics for imperiled populations.