Field‐Scale Application of Three Types of Neural Networks to Predict Ground‐Water Levels1

Abstract: In this paper, a field‐scale applicability of three forms of artificial neural network algorithms in forecasting short‐term ground‐water levels at specific control points is presented. These algorithms are the feed‐forward back propagation (FFBP), radial basis networks (RBN), and generalized regression networks (GRN). Ground‐water level predictions from these algorithms are in turn to be used in an Optimized Regional Operations Plan that prescribes scheduled wellfield production for the coming four weeks. These models are up against each other for their accuracy of ground‐water level predictions on lead times ranging from a week to four weeks, ease of implementation, and execution times (mainly training time). In total, 208 networks of each of the three algorithms were developed for the study. It is shown that although learning algorithms have emerged as a viable solution at field scale much larger than previously studied, no single algorithm performs consistently better than others on all the criteria. On average, FFBP networks are 20 and 26%, respectively, more accurate than RBN and GRN in forecasting one week ahead water levels and this advantage drops to 5 and 9% accuracy in forecasting four weeks ahead water levels, whereas GRN posted a training time that is only 5% of the training time taken by that of FFBP networks. This may suggest that in field‐scale applications one may have to trade between the type of algorithm to be used and the degree to which a given objective is honored.     

Analysis of two-state multivariate phenotypic change in ecological studies

Analyses of two-state phenotypic change are common in ecological research. Some examples include phenotypic changes due to phenotypic plasticity between two environments, changes due to predator/non-predator character shifts, character displacement via competitive interactions, and patterns of sexual dimorphism. However, methods for analyzing phenotypic change for multivariate data have not been rigorously developed. Here we outline a method for testing vectors of phenotypic change in terms of two important attributes: the magnitude of change (vector length) and the direction of change described by trait covariation (angular difference between vectors). We describe a permutation procedure for testing these attributes, which allows non-targeted sources of variation to be held constant. We provide examples that illustrate the importance of considering vector attributes of phenotypic change in biological studies, and we demonstrate how greater inference can be made than by evaluating variance components with MANOVA alone. Finally, we consider how our method may be extended to more complex data.     

Efficient and equitable design of marine protected areas in Fiji through inclusion of stakeholder‐specific objectives in conservation planning

The efficacy of protected areas varies, partly because socioeconomic factors are not sufficiently considered in planning and management. Although integrating socioeconomic factors into systematic conservation planning is increasingly advocated, research is needed to progress from recognition of these factors to incorporating them effectively in spatial prioritization of protected areas. We evaluated 2 key aspects of incorporating socioeconomic factors into spatial prioritization: treatment of socioeconomic factors as costs or objectives and treatment of stakeholders as a single group or multiple groups. Using as a case study the design of a system of no‐take marine protected areas (MPAs) in Kubulau, Fiji, we assessed how these aspects affected the configuration of no‐take MPAs in terms of trade‐offs between biodiversity objectives, fisheries objectives, and equity in catch losses among fisher stakeholder groups. The achievement of fisheries objectives and equity tended to trade‐off concavely with increasing biodiversity objectives, indicating that it is possible to achieve low to mid‐range biodiversity objectives with relatively small losses to fisheries and equity. Importantly, the extent of trade‐offs depended on the method used to incorporate socioeconomic data and was least severe when objectives were set for each fisher stakeholder group explicitly. We found that using different methods to incorporate socioeconomic factors that require similar data and expertise can result in plans with very different impacts on local stakeholders.     

Facilitating disaster preparedness through local radio broadcasting

The 2008 Disaster Mitigation Preparedness (DMP) study took place in Aceh province, Indonesia. It sought to help develop radio programmes and messages to increase resilience to disasters. The role of radio was evaluated during and after the 2004 Asian tsunami disaster. The study team interviewed 984 tsunami survivors from nine sub-districts of Banda Aceh, and local nongovernmental organisations convened eight focus groups around the area of Aceh Besar. Six key informant interviews were held with government disaster management agencies. The DMP survey is the first of its kind to interview a representative random sample of Banda Aceh residents. It reveals the importance of community and social networks, during disaster situations, when essential communications are down. A disaster warning information system based on a multi-media approach needs to be developed. The wider community should be involved in the planning, education and training of Banda Aceh and Aceh Besar residents to facilitate appropriate personal and community survival strategies.     

Mercury and selenium levels in lemon sharks (Negaprion brevirostris) in relation to a harmful red tide event

Tissue levels of mercury (Hg; total, organic) and selenium (Se) were assessed in juvenile lemon sharks (Negaprion brevirostris) from Florida nearshore waters collected during a harmful algal bloom (HAB, brevetoxin) event and compared with sharks not exposed to HABs. In all sharks studied, total Hg levels in the muscle were generally present in a molar excess over Se (which may protect against Hg toxicity) and mean muscle Hg levels (0.34  $\upmu $ g/g) exceed safe human consumption guidelines. While there was generally no difference in tissue Hg and Se levels following exposure of sharks to HABs, hepatic Hg levels were significantly lower (56% reduction) in the HAB-exposed sharks compared to controls. As Hg and HABs are globally increasing in scope and magnitude, further work is warranted to assess their interactions and biotic impacts within aquatic ecosystems, especially for a species such as the lemon shark that is classified as a near-threatened species by the International Union for the Conservation of Nature.     

Changes in polycyclic aromatic hydrocarbon content of atmospheric particulate monitored using fuzzified chromatograms

A simple rapid chromatographic technique for polycyclic aromatic hydrocarbons which is poor at speciation has been used to obtain data on the presence of the class in air particulate. Matching of such chromatograms to a fuzzified reference from these data is illustrated as a practical way to obtain overviews of changes in species distributions summarized in few values. Such information is lost in more commonly used surrogate approaches to “class analyses”.