English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The prediction of the dynamics of algal blooms—the nonlinear oscillations in the concentration of algal species, with attendant implications on water quality, is a notoriously difficult problem in ecological sciences and environmental engineering. This problem is important because of the practical need to develop early warning systems for harmful algal blooms. There is an enormous literature on eutrophication modeling, in which the prediction of algal dynamics often plays a central role. In the past two decades, most of the major advances were made using deterministic water quality models that mimic the phytoplankton growth and nutrient cycles (e.g. Thomann and Mueller 1987; Di Toro 2001). When properly calibrated against extensive field data, these models can predict trends reasonably well and are often adequate for water quality management purposes (e.g. Lee and Lee 1995). On the other hand, the use of data-driven methods to model algal dynamics appeared only very recently. Recknagel et al. (2002) compare the performance of an artificial neural network model (ANNA) with the SALMO model with parameters estimated by a genetic algorithm and a linguistic model built by a genetic algorithm. The data used in the paper are from Lake Kasumigaura. The main conclusions from this paper are: (i) that multivariate modelling by machine learning techniques is superior to deterministic modelling techniques; (ii) artificial neural networks can be powerful short-term predictors of the timing of algal bloom events; and (iii) causal knowledge can be discovered from the evolutionary algorithms presented. Although the paper outlines an interesting approach to algal bloom modelling, we believe that certain essential details of the data analysis are missing. In the absence of the needed details, the conclusions of the paper are perhaps pre-mature and over-stated, and indeed can be misleading if the results are taken at face value. The following points are discussed in relation to a well-established body of knowledge of algal dynamics—derived from many field and modeling studies by many investigators.

journal of hydroinformatics

Comment on ‘Comparative application of artificial nueral networks and genetic algorithms for multivariate time-series modelling of algal blooms in freshwater lakes’