Developing an empirical model of phytoplankton primary production: a neural network case study

We describe the development of a neural network model for estimating primary production of phytoplankton. Data from an enriched estuary in the eastern United States, Chesapeake Bay, were used to train, validate and test the model. Two error backpropagation multilayer perceptrons were trained: a simpler one (3-5-1) and a more complex one (12-5-1). Both neural networks outperformed conventional empirical models, even though only the latter, which exploits a larger suite of predictive variables, provided truly accurate outputs. The application of this neural network model is thoroughly discussed and the results of a sensitivity analysis are also presented.     

Case study: Life‐cycle design strategies at the saturn corporation

This case study is the third in a four‐part series profiling environmental decision making at leading companies. © 2000 John Wiley & Sons, Inc.     

Geophysical and hydrological data assimilation to monitor water content dynamics in the rocky unsaturated zone

In recent years, geophysics is increasingly used to study the flow and transport processes in the vadose zone. Particularly, when the vadose zone is made up of rocks, it is difficult to install sensors in the subsurface to measure hydrological state variables directly. In these cases, the electrical resistivity tomography (ERT) represents a useful tool to monitor the hydrodynamics of the infiltration and to estimate hydraulic parameters and state variables, such as hydraulic conductivity and water content. We propose an integrated approach aimed at predicting water content dynamics in calcarenite, a sedimentary carbonatic porous rock. The uncoupled hydrogeophysical approach proposed consists in 4D ERT monitoring conducted during an infiltrometer test under falling head conditions. Capacitance probes were installed to measure water content at different depths to validate the estimations derived from ERT. A numerical procedure, based on a data assimilation technique, was accomplished by combining the model (i.e., Richards’ equation) with the observations in order to provide reliable water content estimations. We have used a new data assimilation method that is easy to implement, based on the ensemble Kalman filter coupled with Brownian bridges. This approach is particularly suitable for strongly non-linear models, such as Richards’ equation, in order to take into account both the model uncertainty and the observation errors. The proposed data assimilation approach was tested for the first time on field data. A reasonable agreement was found between observations and predictions confirming the ability of the integrated approach to predict water content dynamics in the rocky subsoil.     

Population-level effects of augmented herbivory on Lespedeza cuneata: implications for biological control.

Invasive species pose significant ecological costs, and therefore successful management techniques are important. One commonly employed method is biological control. The success of biological control depends largely on whether additional inflicted damage can successfully reduce the fitness and population growth rate of a target species. Here, we simulate herbivory on the invasive Lespedeza cuneata and create stage-structured projection models to determine if augmented herbivory by a leaf-chewing biological control agent would regulate the population growth rate of this species. We found that augmented herbivory influenced stage transitions of plants in the smallest stage class, causing higher mortality and reduced growth. No other effect was found on stage transitions or fecundities, despite manipulation of herbivory at exceptionally high levels (up to 80% leaf loss). None of the clipping treatments significantly reduced the population growth rate of L. cuneata. We conclude that biological control by a leaf chewing herbivore would not likely be successful, even if an exceptionally large amount of each plant were consumed. We suggest that this approach, a combination of simulated herbivory and demographic modeling, will provide essential information for understanding the utility of biological control to curb the population growth of invasive plant species.     

Being trusted: How team generational age diversity promotes and undermines trust in cross‐boundary relationships

We examine how demographic context influences the trust that boundary spanners experience in their dyadic relationships with clients. Because of the salience of age as a demographic characteristic as well as the increasing prevalence of age diversity and intergenerational conflict in the workplace, we focus on team age diversity as a demographic social context that affects trust between boundary spanners and their clients. Using social categorization theory and theories of social capital, we develop and test our contextual argument that a boundary spanner's experience of being trusted is influenced by the social categorization processes that occur in dyadic interactions with a specific client and, simultaneously, by similar social categorization processes that influence the degree to which the client team as a whole serves as a cooperative resource for demographically similar versus dissimilar boundary spanner–client dyads. Using a sample of 168 senior boundary spanners from the consulting industry, we find that generational diversity among client team members from a client organization undermines the perception of being trusted within homogeneous boundary spanner–client dyads while it enhances the perception of being trusted within heterogeneous dyads. The perception of being trusted is an important aspect of cross‐boundary relationships because it influences coordination and the costs associated with coordination. © 2015 The Author Journal of Organizational Behavior Published by John Wiley & Sons Ltd     

Incidence of thyroid disease following exposure to polybrominated biphenyls and polychlorinated biphenyls, Michigan, 1974-2006

Thyroid hormones, which influence body metabolism and development, could be affected by persistent organic pollutants. We sought to examine the relationship between polybrominated biphenyls (PBBs) and polychlorinated biphenyls (PCBs) and thyroid disease. We employed incidence density sampling to perform a nested case control analysis of the Michigan Long-Term PBB Cohort. Cohort members (n = 3333) were exposed to PBBs through contaminated cattle feed in 1973-1974 and to PCBs through daily life. Those with detectable serum PBB and PCB concentrations at enrollment were categorized into tertiles of PBB and PCB exposure. Case-patients were cohort members answering “Yes” to “Has a healthcare provider ever told you that you had a thyroid problem?” during follow-up interviews; control-patients were cohort members answering “No”. We used odds ratios (OR) with 95% confidence intervals (CI) to compare odds of thyroid disease by PBB and PCB exposure and by various risk factors. Total cumulative thyroid disease incidence after 33 years was 13.9% among women and 2.6% among men. After adjusting for body mass index, we found no statistically significant differences in odds of any type of thyroid disease among women or men with elevated PBB or PCB exposure. Compared to control-patients, women with thyroid disease had increased odds of being overweight/obese (OR = 2.82, 95% CI: 1.94-4.11) and developing infertility (OR = 1.71, 95% CI: 1.08-2.69), diabetes (OR = 1.61, 95% CI: 1.04-2.51), or arthritis (OR = 1.71, 95% CI: 1.18-2.50) during follow-up. Additional research should explore potential associations between PBBs/PCBs and thyroid disease among children exposed in utero.     

Experimental evaluation of PCDD/Fs and PCBs release and mass balance of a WTE plant

The behaviour of waste incineration plants with respect to organic toxic trace contaminants such as PCDDs, PCDFs and, to a minor extent, PCBs, is still a matter of concern for the public opinion and the decision makers. It is therefore very important, first, to evaluate the release of these organic toxic trace contaminants in the environment during waste incineration, not only through the stack gas emission but also with the solid and liquid residues, and then to compare the total release with the input through the treated waste in order to assess the plant behaviour as a “sink” rather than a “source” of organic toxic trace contaminants. The experimental investigation carried out on an Italian full scale incineration plant has shown a total 17 PCDD/Fs and 12 dioxin-like PCBs release of 5.5-27 μg WHO-TEQ per tonne of treated waste and an input flux of 1.6-44 μg WHO-TEQ per tonne of waste, with the difference between the input and the output fluxes rather small and the plant behaviour toward organic trace toxic contaminants in average neutral. Results are compared with similar evaluations conducted in the last decade on a number of waste-to-energy (WTE) plants operating in Italy.     

Predicting dioxin-like PCBs soil contamination levels using milk of grazing animal as indicator

Dioxin-like PCBs (dl-PCBs) are ubiquitous persistent organic pollutants of recognized negative effects on human health. Assessing highly polluted areas should be an important public health issue. This study proposes to use the milk of grazing animals as a bioindicator of dl-PCB contamination in the environment. The hypothesis is that milk concentration of dl-PCBs are related to soil concentrations of these compounds, and that soils are generally reflective of a larger environmental issue of dl-PCB contamination. In this study, we evaluate the possibility of predicting soil concentrations using milk of sheep, cows and buffalos in a spatial model. For this purpose, samples of soil and milk collected in Campania (Italy) were introduced in a GIS platform to perform geostatistical analysis for building a simple predictive model. The ordinary least squares regressions (OLS) showed a statistically significant correlation (p < 0.05) between soil and milk contamination. However, this relationship was spatially variable. Thus, a geographically weighted regression (GWR) was performed, obtaining R² values of 0.91, 0.77 and 0.66 for sheep’s, buffalo’s and cow’s milk respectively. Assessed the mathematical relationships between the variables, new data was introduced to evaluate the performance of the model. Predictions of soil contamination with dl-PCBs using sheep’s, cow’s and buffalo’s milk showed a mean error of 23%, 25% and 36% respectively. According to these results the sheep’s milk can be considered the best bioindicator of dl-PCBs contamination among the three species. The results of this project evidence the potentialities of the proposed approach to assess bioindicator performance in a spatial predictive model.     

Implications of a changing climate on food security and smallholders’ livelihoods in Bogotá, Colombia

Small farmers who supply the city of Bogotá with food are facing many challenges that are jeopardizing their livelihoods and by extension, the food security of Colombia’s capital. We expect future changes in climatic conditions to exacerbate the plight of the small farmers and this is expected to compromise Bogota’s food security even further. This paper specifically seeks to assess the impact of climate change (CC) on the livelihoods of smallholders who supply Bogota with most of its food. In our multidisciplinary methodology, we translated the exposure to CC into direct impact on crops and assessed sensitivity and adaptive capacity using the sustainable rural livelihoods framework. The results show that rainfall (by average of 100 mm) and temperature (by average of 2.1 °C) will increase over the study area, while the future climate suitability of the most important crops such as mango (Mangifera indica), papaya (Carica papaya), corn (Zea mays) and plantain (Musa balbisiana) shows a decrease of 19 % to 47 % climate suitability by the year 2050. The assessment of sensitivity and adaptive capacity demonstrates that farmers participating in a farmers’ market, initiated by several local and international non-governmental organizations (NGOs), are less vulnerable to CC than farmers who sell through intermediaries. Those farmers selling directly to consumers in the farmers’ market have a higher adaptive capacity (3 on a scale of 3) in social and financial capital than those selling to intermediaries with less adaptive capacity (1 on a scale of 3). In light of the reduction in overall climatic suitability of some of the major crops and the change of geographic location of suitability for others, there are likely to be serious threats for Bogotá’s food security, the ecological landscape around the city, and farmers’ livelihoods. We further conclude that unless proper adaptation measures are implemented, the geographical shift in climate suitability may also force farmers to shift their crops to higher elevations including remaining forests and páramos (the Colombian alpine tundra ecosystems), which may be threatened in the near future.