The policies and procedures adopted by the Biological Monitoring Working Party in developing its score system are summarised. The conceptual basis of the reappraisal is described and then formulated mathematically. The analysis is based on some 17,000 biological samples and covers 85 families. The results are expressed in terms of derived scores which mirror the scores allocated to the families by the BMWP. Some of the differences between the derived and original scores are quite substantial, and are shown to relate to policies and practices adopted by the Working Party. It is argued that the derived scores offer a more realistic representation of the families' sensitivities to pollution than do the BMWP scores. The most significant differences are discussed in relation to their likely causes and are shown to be consistent with expectations. It is concluded that the system's poor performance in some recent tests may be explained by these findings, and that corrective action could improve performance.
[1]
Niels De Pauw,et al.
Biological monitoring of river water quality
,
1994
.
[2]
J. R. Chandler.
A biological Approach to water quality management
,
1970
.
[3]
D. Balloch,et al.
Biological assessment of water quality in three British rivers: the North Esk (Scotland), the Ivel (England) and the Taf (Wales)
,
1976
.
[4]
Mike T. Furse,et al.
The prediction of the macro‐invertebrate fauna of unpolluted running‐water sites in Great Britain using environmental data
,
1987
.
[5]
Sašo Džeroski,et al.
Biological Monitoring: a Comparison between Bayesian, Neural and Machine Learning Methods of Water Quality Classification.
,
1996
.