Page 258 - Freshwater-Biology-and-Ecology-Handbook

CHAPTER 5 3.2.2 – 3.2.3 3.2.2 RIVPACS Model 44 Currently, RIVPACS Model 44 exists only as an experimental

The width, depth and nature of the substrate measured at a
site may not represent the natural state of the river because
system for Great Britain, so that its performance can be
tested. GIS data has been compiled by CEH UK (Centre for
many have been re-sectioned, particularly over-deepened
Ecology and Hydrology) for all the environmental parameters
for land drainage, flood conveyance or navigation. The flow
used by Model 44 (except for alkalinity) but based around
in many rivers may also be very different to the natural state
because of abstraction or discharges. As a result, when
measured values are used in RIVPACS, the fauna that it
for which do not allow it to be released to the public domain
other than one site at a time.
predicts may not be the natural fauna. To overcome this, a
new predictive model, RIVPACS Model 44 (Clark & Davy- a proprietary GIS river network, the licencing requirements
Bowker 2017, 2018) (119) (120) has been developed in which Until the new model has been tested and its performance is
these environmental parameters have been replaced by better that the original RIVPACS model, it will not be released
parameters from Geographic Information Systems (GIS- for operational use. An advantage of obtaining environmental
systems) that are not affected by human interventions. Model parameters from GIS is that users will not have to compile
44 is potentially very useful for water resource assessment; the environmental input data themselves, other than the
but before it can be used operationally, it needs to be tested, Ordnance Survey grid reference and alkalinity. Obtaining the
because substrate, width and depth are strong predictors. correct environmental input data relies on the grid reference
The loss of predictive power caused by their omission could being correct.
balance any improvements that Model 44 provides.

RIVPACS Model 44

An experimental, predictive model utilising GIS to substitute environmental information from pristine environments.
This enables the natural state of the river to be simulated into the predictions.

Comparison of environmental predictors for RIVPACS Model 1 and Model 44

RIVPACS IV general model 1 RIVPACS IV model 44
Sample data Sample data

Width Key
Depth Geo-chemistry
Substrate % clay / silt One of: Red parameters entered
into RICT
% sand alkalinity total hardness * Calculated by RICT
% gravel / pebbles calcium conductivity
% cobbles / boulders ** Obtained from database
Map data hosted with RICT using
mean particle size* OS grid reference OS grid reference
mean air temperature*
Geo-chemistry
air temperature range*
One of: latitude *
alkalinity total hardness Replaced longitude*
calcium conductivity by Altitude **
Map data Distance from source **
Slope **
OS grid reference Discharge category **
mean air temperature*
air temperature range* % drift geology class in upstream catchment
latitude Class 1 – Peat**
longitude* Class 3 – Clay**
Class 6 – Chalk**
Altitude Class 7 – Limestone**
Distance from source Class 8 – Hard Rocks**
Slope Upstream catchment area**
Discharge category or velocity Mean altitude of upstream catchment**

Figure 5.12
Anything in red font can be entered into RICT. You can either enter alkalinity, calcium concentration, total hardness or electrical conductivity.
RIVPACS converts these to alkalinity which is the predictor variable. Likewise, you can either enter discharge category or flow velocity. If you enter
the latter, RIVPACS will estimate discharge category, which it uses as the predictor variable. Model 44 is a flow and sediment independent model.

258 | Freshwater Biology and Ecology Handbook
–