Developing a macrophyte index of ecological status for Northern Ireland
Dodkins, Ian R 2003
Unviersity of Ulster, 313 pp.
greybar.jpg - 2645 Bytes

The recent European Water Framework Directive requires rivers to be assessed on their ecological status. Macrophytes are one of the biological elements to be assessed. This study was conducted to develop a method of measuring the ecological quality of macrophytes in Northern Ireland’s rivers.

A prior data set of 281 monitoring sites was utilised to define different macrophyte communities. The environmental variables which were best at defining the communities were used to generate a preliminary typology. From the environmental typology and biological classification reference sites were chosen as representatives of reference conditions. After resurvey of the reference sites to ensure high hydromorphological quality they were reclassified, and a final typology developed.

Different methods for assessing ecological quality were examined:

- A ‘System A’ approach, utilising a fixed typology.
- A RIVPACS approach producing ‘expected species’ lists from interpolated reference conditions.
- Metrics indicative of different impacts, produced from CCA and utilising species optima and niche widths.
- Two novel AI techniques.

Within System A, altitude (obligatory in the WFD typology) was not a major gradient, and the arbitrary boundaries between ecotypes did not enable natural communities to be accurately defined within the typology. This is despite the fact that macrophytes tend to exist as looser associations rather than distinct communities

Natural variation (especially disturbance) may be producing functionally equivalent, but structurally different (different species) sites. This makes direct species comparisons difficult, reducing the effectiveness of RIVPACS type predictions. Metrics acted as intermediaries between disparate species lists enabling comparisons between monitoring and reference sites. The AI techniques are only suitable for examining simple gradients and lack statistical support, though they are user friendly and have future potential.

Metrics were found to be the most suitable for assessing ecological status.