Estimating relative abundance from catch and effort data, using neural networks

We develop and test a method to estimate relative abundance from catch and effort data using neural networks. Most stock assessment models use time series of relative abundance as their major source of information on abundance levels. These time series of relative abundance are frequently derived from catch-per-unit-of-effort (CPUE) data, using general linearized models (GLMs). GLMs are used to attempt to remove variation in CPUE that is not related to the abundance of the population. However, GLMs are restricted in the types of relationships between the CPUE and the explanatory variables. An alternative approach is to use structural models based on scientific understanding to develop complex non-linear relationships between CPUE and the explanatory variables. Unfortunately, the scientific understanding required to develop these models may not be available. In contrast to structural models, neural networks uses the data to estimate the structure of the non-linear relationship between CPUE and the explanatory variables. Therefore neural networks may provide a better alternative when the structure of the relationship is uncertain. We use simulated data based on a habitat based-method to test the neural network approach and to compare it to the GLM approach. Cross validation and simulation tests show that the neural network performed better than nominal effort and the GLM approach. However, the improvement over GLMs is not substantial. We applied the neural network model to CPUE data for bigeye tuna (Thunnus obesus) in the Pacific Ocean.

A GIS-based model of soil erosion and transport

Soil erosion is a natural process that occurs when the force of wind, raindrops or running water on the soil surface exceeds the cohesive forces that bind the soil together. In general, vegetation cover protects the soil from the effects of these erosive forces. However, land management activities such as ploughing, burning or heavy grazing may disturb this protective layer, exposing the underlying soil. The decision making process in rural catchment management is often supported by the predictive modelling of soil erosion and sediment transport processes within the catchment, using established techniques such as the Universal Soil Loss Equation [USLE] and the Agricultural Nonpoint Source pollution model [AGNPS]. In this article, the authors examine the range of erosion models currently available and describe the application of one of these to the Burrishoole catchment on the north-west coast of Ireland, which has suffered heavy erosion of blanket peat in recent years.

Progress report on fish counting on the Rivers Itchen and Test

This progress report summarises work on NSHEB Mark 10 fish counters which are installed at Woodmill on the River Itchen and Nurseling Mill and Connegar Bridge on the River Test. Counters are evaluated and salmon behaviour regarding the counters examined. The report includes a a list of equipment needed for the efficient running of the project in the future.

Use of macroinvertebrate communities to detect environmental stress in running waters

This paper presents an account of some current uses of RIVPACS (River Invertebrate Prediction and Classification System), a software package developed by the Institute of Freshwater Ecology (UK). Background information is also given on the unique data-set on which the system is based. Before discussing RIVPACS, we consider the range of environmental stresses encountered in flowing-water systems and some of the ways in which stresses may affect macroinvertebrate communities. The wide application and relevance of the RIVPACS approach was recognised when it was chosen as the biological method for use throughout the UK in the 1990 River Quality Survey (RQS). In the concluding section we list some lessons learnt both from the 1990 survey and from our own testing exercise, and we outline current developments which will lead to a new version of RIVPACS for use in the 1995 RQS.

A Test of Computer-assisted Matching Using the North Pacific Humpback Whale, Megaptera novaeangliae, Tail Flukes Photograph Collection

Testing was conducted of a computer-assisted system for matching humpback whale tail flukes photographs. Trials with a 12,000-photographs database found no differences in match success between matching by computer and matching by comparing smaller catalogs ranging in size from 200 to 400 photographs. Tests with a 24,000-photographs database showed that, on average, the first match was found after examining about 130 photographs whether the photograph quality was excellent, good, or poor. Match success did not appear to be strongly related to whether the tail flukes had especially distinctive markings or pigment patterns (recognition quality). An advantage of computer-assisted matching is the ability to compare new photographs to the entire North Pacific collection, where no bias is introduced based on expectation of resightings within or between specific areas, or based on expectation of behavioral role (e.g. matching “known” females to “known” females).

PCR-based assay for detection of four coral pathogens

Several microorganisms have been identified as pathogenic agents responsible for various outbreaks of coral disease. Little has been learned about the exclusivity of a pathogen to given disease signs. Most pathogens have only been implicated within a subset of corals, leaving gaps in our knowledge of the host range and geographic extent of a given pathogen. PCR-based assays provide a rapid and inexpensive route for detection of pathogens. Pathogen-specific 16S rDNA primer sets were designed to target four identified coral pathogens: Aurantimonas coralicida, Serratia marcescens, Vibrio shilonii, and Vibrio coralliilyticus. Assays detected the presence of targets at concentrations of less than one cell per microliter. The assay was applied to 142 coral samples from the Florida Keys, Puerto Rico, and U.S. Virgin Islands as an in situ specificity test. Assays displayed a high-level of specificity, seemingly limited only by the resolution of the 16S rDNA.

Species identification of family Lutjanidae and separation of populations of John’s snapper (Lutjanus johnii) in Persian Gulf and Oman Sea based on otolith shape analysis

The anatomical and morphometric (shape indices, contour descriptors and otolith weight) characterizations of sagittal otoliths were investigated in 13 species of Lutjanus spp. inhabiting the Persian Gulf. This is the first study that compares the efficiency of three different image analysis techniques for discriminating species based on the shape of the outer otolith contour, including elliptical Fourier descriptors (EFD), fast Fourier transform (FFT) and wavelet transform (WT). Sagittal otoliths of snappers are morphologically similar with some small specific variations. The use of otolith contour based on wavelets (WT) provided the best results in comparison with the two other methods based on Fourier descriptors, but only the combination of the all three methods (EFD, FFT and WT) was useful to obtain a robust classification of species. The species prediction improved when otolith weight was included. In relation to the shape indices, only the aspect ratio provided a clear grouping of species. Also, another study was carried on to test the possibility of application of shape analysis and comparing otolith contour of otoliths of Lutjanus johnii from Persian Gulf and Oman Sea to identify potential stocks. The results showed the otoliths have differences in contour shape and can be contribute to two different stocks.