Skip to main content

Predicting aquatic ecosystem quality using artificial neural networks: impact of environmental characteristics on the structure of aquatic communities (algae, benthic and fish fauna).

Objective

Problems to be solved
Setting aside historical factors, the structure and diversity of aquatic communities in running waters is primarily dependent on a complex of physical, chemical and biotic factors. Physical and chemical variables taking place inside the hydro system are heavily dependent on climatic and catchments natural properties (hydrology, geology, topography, etc.), but are also strongly influenced by anthropogenic impacts (hydraulic management, land use and agricultural practices, waste water discharge, etc.). In the purpose to really assess the ecosystem health, with the object of better managing it or of restoring its integrity in the future, the problem is to make the part between the natural variability of the ecosystem conditions and biogenesis variation in one hand, and the anthropogenic alterations and their effects on the other hand. Now for several decades, hydro biological studies have identified the main factors determining freshwater communities, but very few have been able to really establish deterministic links between ecological factors and the structure of key aquatic communities. Despite these uncertainties, key aquatic communities have been utilised to evaluate environmental and biological quality of streams and rivers. Practical methods for calculating "biotic indices" have been designed, with the requirement that they be simple enough to be applied in routine surveys. As a result, a variety of standard methods have been proposed and used for assessing water quality in different river orders, often without regard to a reference to the natural state of the community. By contrast, recent approaches based on the concept of ecosystem integrity and biodiversity are more promising, especially for integrated water management. The goal of the project is to develop general methodologies, based on advanced modelling techniques which will be utilised both for assessing the ecosystem status, and predicting structure and diversity of key aquatic communities (benthic diatoms, benthic micro-invertebrates and fish), under natural (i.e. undisturbed by human activities) and under man-made disturbance (i.e. submitted to various pollutions, discharge regulation, etc.).
Scientific objectives and approach
The main scientific objectives will be
i) setting up a standardised methodological approach (we have defined a set of technical procedures which will be used in a common framework to analyse or to predict community structure of studied ecosystems; each reference site is sampled in a standardised way and this will allow to compare different sites for regional conservation priorities);
ii) linking the environmental characteristics and the community structure at each reference site by using a defined set of parameters and a combination of target groups representing the main functional levels of the ecosystems (rapid assessment procedures will be implemented on these hypotheses that regulative and functional factors, the resources describe ecosystem functioning in a unifying way);
iii) evaluating at a functional level the sensitivity of the studied ecosystems and their response to disturbance through implementation of sensitivity indices and modelling (the main threats on living communities and on local endangered species will be identified as we shall build predictive models of community structure for a set of critical habitats); and
iv) investigating the effects of human impacts on the functioning of the ecosystem, i.e. on the composition and change in structural and functional organism groups in comparison to nearby natural reference conditions. Special attention is directed to summarising ecosystem functioning by exploring the chance of community restoration in selected sites submitted to the most common types of disturbance. In this project, the assessment of river ecosystem integrity will be based on the relationships between environmental impacts and organism groups, i.e. community structure as depending upon the environmental variables. However, ecosystem analysis and prediction with empirical statistical and analytical methods are often limited by the spatial complexity, the non-linear relationships between variables and the temporal dynamics of complex ecological processes. The artificial neural network (ANN) models can be used as predictive tools, but also for explaining and understanding the complex relationships between variables. These tools could be applied in other river networks throughout Europe. They will be simple, easy to handle and applicable to stream management and stream policy making.
Expected impacts
The main applied objective is to propose a set of tools for water management and water policies which allow to easily assess ecological quality and perturbations of stream ecosystems. These tools will provide information about running water quality as well as community structure. The assessment tools will allow identifying measures that should be taken to restore biological integrity in running waters. Hopefully, the study can be considered as a first step toward linking the improvement of water quality through specific management measures (e.g. waste water treatment, habitat restoration, etc.) to the expected improvement in ecological and biological value of running water systems.
This project could have impacts on two end-users, namely the scientific community and freshwater managers.
· Scientific results for the scientific community: outputs for the scientific community will consist in comprehensive insights into the driving mechanisms of community dynamics, as well as explicative and predictive artificial neural network models.
Concentrating and coordinating the investigations of different teams on different sites in Europe, with diverse ecological problems (diatoms, benthic invertebrate and fish fauna) will give considerable impetus to the elucidation of the major questions related to community structure in freshwater research. This progress will act as an original comparison point for different ecosystems in Europe.
· Models for preservation, policy implementation, and for freshwater managers: There is a growing interest in Europe for databases and model implementations. A very strong demand exists from river preservation managers who want to know which habitats and which species have to be considered as conservation priorities, and how to insure efficient management and monitoring of freshwater streams they have in charge. In the PAEQANN project, we shall be able to provide access to a very large database in Europe, on which it will be possible to implement real tools for freshwater management. These tools, clearly written and easily accessible by non-specialists, will be available for all the environmental managers (e.g. Water Agencies) who would like to use them to improve the quality of aquatic ecosystems. These tools will improve the harmonisation of freshwater quality assessment in Europe and increase the comparability of river quality data from different European regions.

Setting aside historical factors, the structure and diversity of aquatic communities in running waters is primarily dependent on a complex of physical, chemical and biotic factors. Physical and chemical variables taking place inside the hydrosystem are heavily dependent on climatic and catchment natural properties (hydrology, geology, topography, etc.), but are also strongly influenced by anthropogenic impacts (hydraulic management, land use and agricultural practices, waste water discharge, etc.). In the purpose to really assess the ecosystem health, with the object of better managing it or of restoring its integrity in the future, the problem is to make the part between the natural variability of the ecosystem conditions and biocenosis variation in one hand, and the anthropogenic alterations and their effects on the other hand. Now for several decades, hydrobiological studies have identified the main factors determining freshwater communities, but very few have been able to really establish deterministic links between ecological factors and the structure of key aquatic communities. Despite these uncertainties, key aquatic communities have been utilised to evaluate environmental and biological quality of streams and rivers. Practical methods for calculating "biotic indices" have been designed, with the requirement that they be simple enough to be applied in routine surveys. As a result, a variety of standard methods have been proposed and used for assessing water quality in different river orders, often without regard to a reference to the natural state of the community. By contrast, recent approaches based on the concept of ecosystem integrity and biodiversity are more promising, especially for integrated water management.

The goal of the project is to develop general methodologies, based on advanced modelling techniques which will be utilised both for assessing the ecosystem status, and predicting structure and diversity of key aquatic communities (benthic diatoms, benthic micro-invertebrates and fish), under natural (i.e. undisturbed by human activities) and under man-made disturbance (i.e. submitted to various pollutions, discharge regulation, etc.).

Expected impacts
The main applied objective is to propose a set of tools for water management and water policies which allow to easily assess ecological quality and perturbations of stream ecosystems. These tools will provide information about running water quality as well as community structure. The assessment tools will allow to identify measures which should be taken to restore biological integrity in running waters. Hopefully, the study can be considered as a first step toward linking the improvement of water quality through specific management measures (e.g. waste water treatment, habitat restoration, etc.) to the expected improvement in ecological and biological value of running water systems. This project could have impacts on two end-users, namely the scientific community and freshwater managers.

Scientific results for the scientific community: outputs for the scientific community will consist in comprehensive insights into the driving mechanisms of community dynamics, as well as explicative and predictive artificial neural network models. Concentrating and coordinating the investigations of different teams on different sites in Europe, with diverse ecological problems (diatoms, benthic invertebrate and fish fauna) will give considerable impetus to the elucidation of the major questions related to community structure in freshwater research. This progress will act as an original comparison point for different ecosystems in Europe.

Models for preservation, policy implementation, and for freshwater managers: There is a growing interest in Europe for databases and model implementations. A very strong demand exists from river preservation managers who want to know which habitats and which species have to be considered as conservation priorities, and how to insure efficient management and monitoring of freshwater streams they have in charge. In the PAEQANN project, we shall be able to provide access to a very large database in Europe, on which it will be possible to implement real tools for freshwater management. These tools, clearly written and easily accessible by non-specialists, will be available for all the environmental managers (e.g. Water Agencies) who would like to use them to improve the quality of aquatic ecosystems. These tools will improve the harmonisation of freshwater quality assessment in Europe and increase the comparability of river quality data from different European regions.

Funding Scheme

CSC - Cost-sharing contracts

Coordinator

UNIVERSITE PAUL SABATIER DE TOULOUSE III
Address
118,Route De Narbonne 118 Bftiment Ivr3
31062 Toulouse
France

Participants (5)

CENTRE DE RECHERCHE PUBLIC - GABRIEL LIPPMAN
Luxembourg
Address
Avenue De La Faïencerie 162A
1511 Luxembourg
CENTRE NATIONAL DU MACHINISME AGRICOLE, DU GENIE RURAL, DES EAUX ET DES FORETS
France
Address
Avenue De Verdun 50 , Gazinet
33612 Canejan
ECON S.R.L.
Italy
Address
Vico S. Domenico Maggiore, 9
80134 Napoli
FACULTES UNIVERSITAIRES NOTRE-DAME DE LA PAIX DE NAMUR
Belgium
Address
Rue De Bruxelles 61
5000 Namur / Namen
THE UNIVERSITY OF PHARMACEUTICAL SCIENCES
Denmark
Address
Universitetsparken 2
2100 Koepenhagen