RICT article by RP chair Steve Brooks

The below article was written by Steve Brooks — founder member and co-chair of the Riverfly Partnership. Many thanks to Steve for sharing it with us.

Have you ever wondered what your riverfly scores would be like if your river was in pristine condition?

The River Invertebrate Classification Tool (RICT) is a mathematical model developed by scientists at the Freshwater Biological Association and the four UK environmental agencies to predict just this.

RICT is based on a dataset of the macroinvertebrates (riverflies in the broadest sense) found in 200 top quality UK rivers. Twelve physical variables have been found that best predict the macroinvertebrates that are most likely to occur on those rivers. These variables are longitude and latitude of the sample site, altitude, distance of the sample site from the river’s source, width and depth of the channel, substrate composition, discharge, alkalinity, slope, mean air temperature and mean air temperature range. These variables are the most important in influencing the macroinvertebrate composition at a given sampling site. Therefore, using this tool, it is possible to predict the macroinvertebrate assemblage of any site on any river in the UK in a relatively undisturbed state by inputting into the model the values of those 12 key variables at the site in question. The macroinvertebrate data predicted for the site can then be converted into an RMI score.

How does an example of this work in practice? Charlotte Hawkins, my local EA Ecology Officer contact, has predicted RMI scores, on an annual and seasonal basis, for all of the EA sampling sites on Hertfordshire’s rivers. These scores are derived from the values of the 12 key variables measured at each of the EA sampling sites. Charlotte has kindly passed on to me the predicted scores for the River Ash in east Hertfordshire, where I do my RMI monitoring, at the five EA sampling sites on that river. Given the relatively short length of the river and the fairly uniform topography of the river basin, it is not surprising that the predicted RMI scores are quite similar, ranging between 10 and 13 throughout the length of the river, with the higher scores occurring in spring and summer and lower scores in the autumn. Winter scores are not predicted by RICT.

How do these predicted scores compare with the actual RMI scores that I typically find? At Watersplace Farm, the most downstream of the EA sampling sites on the River Ash, the observed RMI scores are similar to the predicted scores in spring and often exceed the predicted scores in the summer. This suggests that the river in this part of its course is in better condition than expected for a river of its type. However, further upstream at Hadham Mill, observed RMI scores are at around 8-9, so 2-4 points lower than the predicted scores. At Maltings Lane, the most upstream EA sampling site, observed RMI scores are even lower, at around 5-7, about half the predicted score.

Why do we see these discrepancies between the observed and predicted RMI scores? It is difficult to be certain, but one factor for the higher-than-expected RMI scores at Watersplace Farm (and the other RMI sampling sites on the Easneye Estate) may be thanks to the river restoration work that has been carried out on the Ash as it flows through the Easneye Estate. On the other hand, low flows and siltation in the upper reaches of the Ash at Hadham Mill and Maltings Lane may be contributing to the poor performance of the river there.

This example from the River Ash illustrates how the predicted RMI scores, derived from RICT, can be useful. They help us see which parts of the river are in better-than-expected condition and which are worse and suggest which drivers may be responsible for these results. The predicted RMI scores also provide targets, and a measure of success, for river restorations.

In a wider context, this approach allows us to generate appropriate RMI targets across many different types of watercourses throughout the UK. Furthermore, these targets are based on a robust methodology that is already widely used by the UK environmental agencies.

If you would like to know the predictive RICT-based RMI scores for your own sampling site, contact your river coordinator.

With thanks to John Davy-Bowker for helpful comments on a draft of this article.