, Volume 415, Issue 0, pp 131-138

First online:

Macrophyte functional variables versus species assemblages as predictors of trophic status in flowing waters

  • M. M. AliAffiliated withDepartment of Botany, Faculty of Science at Aswan, South Valley University
  • , K. J. MurphyAffiliated withInstitute of Biomedical & Life Sciences, Division of Environmental & Evolutionary Biology, University of Glasgow Email author 
  • , V. J. AbernethyAffiliated withInstitute of Biomedical & Life Sciences, Division of Environmental & Evolutionary Biology, University of Glasgow

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


A series of models was developed using functionally-derived variables (mainly based on morphological attributes of freshwater macrophytes) to predict the trophic status of river and associated channel systems. The models were compared with an existing species-assemblage based procedure for predicting British river trophic conditions (the Macrophyte Trophic Ranking scheme, MTR). We compared sites in cooler temperate conditions (in Scotland) and warmer, sub-tropical conditions (in Egypt). In total, we made measurements of 13 traits from >600 individual plant specimens of 33 species growing at 42 sites (divided into independent input and test site datasets). N status (as annual mean concentration in water of total oxidised nitrogen, TON) was only very poorly predicted by this approach. However, P (as annual mean concentration in water of soluble reactive phosphate, SRP) was better predicted: both by a model based on MTR (r = −0.585, p<0.001), and by models using functional attributes of the macrophyte vegetation. River Trophic Status Indicator (RTSI) models based on ranked plant functional group relationship to river water P concentrations (RTSIFG), or field-measured trait sets of the plants (RTSITR) could also individually explain up to about 34% of the variation in P, both for the total dataset and for subsets from Egypt or Scotland alone or for high v. low-flow sites. Combining both types of RTSI measure produced the most powerful predictive model (r = 0.72, p<0.001), explaining just over half the variability in P.

phosphate modelling eutrophication aquatic plants rivers irrigation channels