Effects of the agricultural sector on the physico-chemistry and the ecosystem quality of inland waters

The aim of this research was to identify the links between the agricultural sector and the physicochemical condition and ecosystem health of inland waters. In order to create a comprehensive model, two analytical tools have been used, on the one hand the "Driving forces – Pressure - State – Impact - Response" (DPSIR) modeling framework and, on the other, the decision trees data mining technique with the use of an improved version of the C4.5 algorithm called See5. The datamining applications were performed in three consecutive phases. In the first phase “State” data were tested against “Impact” data in order to identify possible links. In the second phase the same was done with “Pressure” data against “State” data and in the third phase “Driving Forces” data where tested against “Pressure” data. The results of the first phase indicate, as expected, that there is a linkage between several physicochemical characteristics of the lakes examined and the amount of fish caught from those lakes. These characteristics are the electrical conductivity, the Cl- concentration, the pH and the oxygen saturation percentage of the water. When these parameters were examined against the agricultural “Pressures” data in the second phase, the results showed that the agricultural production area and the agricultural production amounts, in general as well as for specific products, were linked to the physicochemical characteristics. The results of the third phase showed that although three non-environmentally friendly financial assistance programmes were found to be directly linked to agricultural production, three environmentally friendly programmes showed no links to the agricultural production which can be attributed to the larger size and geographical scope of the non-environmentally friendly programmes. Apart of the detailed results obtained, it was also found that data mining techniques are able to assess the effects of the agricultural sector on the physicochemical of inland waters and it is expected that they could also perform well with other types of environmental data. Thus, data mining is a powerful tool in the hands of environmental policy maker and should be used wherever the available data permit it.

Author dc:creator

• Vorloou, A

Contributors dc:contributor

• Prof. Dr. Ir.Vansteenskiste, G
• Prof. Dr. Janssen, C

Statement dc:rights

• info:eu-repo/semantics/openAccess

Language dc:language

und

Identifier

https://biblio.ugent.be/publication/467947
http://doi.org/1854/6094
https://biblio.ugent.be/publication/467947/file/1878691

OAI identifier oai:identifier

oai:archive.ugent.be:467947