2 A Modelling Approach for a Comprehensive
Understanding of the Ecological Functioning of the Seine
River
The Riverstrahler model, first developed in the 1990s [23, 24], is still one of the few
tools available to model nutrient cycling and ecological functioning at the scale of an
entire drainage network based on the network’s morphological characteristics
(e.g. length and depth of the rivers), the hydro-meteorological conditions and the
human activities in the watershed (point and diffuse pollution sources related to
domestic and agricultural water usages). It combines a generic model that mimics
ecological and biogeochemical processes (RIVE) with a water flow routing scheme
through the drainage network partly based on its Strahler ordination [8] (Fig. 1a).
Starting from a conceptualisation with one community of phytoplankton taking
up only N and P, and one bacterial community degrading organic matter, the RIVE
model (www.fire.upmc.fr/rive) has gradually evolved over the last 30 years
(Fig. 1b). Among the significant developments, the introduction of diatoms and
silica uptake allowed us to specifically examine river primary production and oxygen
production since diatoms are the dominant species in rivers, specifically in the Seine
River [24]. Besides building up an organic stock mineralised by bacteria, phytoplankton is grazed by microzooplankton, which is also represented in the model
[9]. Nitrifying bacteria and nitrification were additionally introduced for a realistic
simulation of ammonium dynamics, a typical element of point sources [25–27],
taking into account N 2 O as an intermediate of nitrification [28]. Phosphorus, coming
from both point and diffuse sources, was specifically studied for a parametrisation of
its adsorption–desorption dynamics [29–31]. The model also includes a calculation
of nutrient exchanges across the sediment–water interface as a result of a given
sedimentation flux of organic material, taking into account organic matter degradation, oxygen consumption and N, P and Si processes, mixing in the interstitial and
solid phases and accretion of the sedimentary column by inorganic matter sedimentation [15, 32, 33] (Fig. 1c).
Recent developments concern the introduction of inorganic forms of carbon and
greenhouse gas emissions (carbon dioxide, CO 2 [34]; methane, CH 4 ; nitrous oxide,
N 2 O). The current version of RIVE comprises 30 state variables.
The RIVE model applied to the whole upstream–downstream network with the
same parameters implies the recognition of the unicity of the kinetics formalisation
of ecological processes, whatever the location in the network, a strong assumption
that has proved realistic, not only along a land-to-sea continuum in a gradient of
watershed sizes (the Seine River [23]; the Danube River [35]) but also in stagnant
systems [9, 10] and in a variety of regions (sub-tropical [36]; subarctic [37]). The
RIVE model also performs fairly well in the case of transitory events such as
combined sewer overflows [38, 39]. For the study of urban systems, it is
implemented in the ProSe model [32, 38, 40], which was used to study the plurennial
impact of transient events on the carbon and oxygen cycles [41, 42], as well as
nutrients [27, 31, 43]. The implementation of RIVE in the ProSe model simulates
192
J. Garnier et al.
Understanding of the Ecological Functioning of the Seine
River
The Riverstrahler model, first developed in the 1990s [23, 24], is still one of the few
tools available to model nutrient cycling and ecological functioning at the scale of an
entire drainage network based on the network’s morphological characteristics
(e.g. length and depth of the rivers), the hydro-meteorological conditions and the
human activities in the watershed (point and diffuse pollution sources related to
domestic and agricultural water usages). It combines a generic model that mimics
ecological and biogeochemical processes (RIVE) with a water flow routing scheme
through the drainage network partly based on its Strahler ordination [8] (Fig. 1a).
Starting from a conceptualisation with one community of phytoplankton taking
up only N and P, and one bacterial community degrading organic matter, the RIVE
model (www.fire.upmc.fr/rive) has gradually evolved over the last 30 years
(Fig. 1b). Among the significant developments, the introduction of diatoms and
silica uptake allowed us to specifically examine river primary production and oxygen
production since diatoms are the dominant species in rivers, specifically in the Seine
River [24]. Besides building up an organic stock mineralised by bacteria, phytoplankton is grazed by microzooplankton, which is also represented in the model
[9]. Nitrifying bacteria and nitrification were additionally introduced for a realistic
simulation of ammonium dynamics, a typical element of point sources [25–27],
taking into account N 2 O as an intermediate of nitrification [28]. Phosphorus, coming
from both point and diffuse sources, was specifically studied for a parametrisation of
its adsorption–desorption dynamics [29–31]. The model also includes a calculation
of nutrient exchanges across the sediment–water interface as a result of a given
sedimentation flux of organic material, taking into account organic matter degradation, oxygen consumption and N, P and Si processes, mixing in the interstitial and
solid phases and accretion of the sedimentary column by inorganic matter sedimentation [15, 32, 33] (Fig. 1c).
Recent developments concern the introduction of inorganic forms of carbon and
greenhouse gas emissions (carbon dioxide, CO 2 [34]; methane, CH 4 ; nitrous oxide,
N 2 O). The current version of RIVE comprises 30 state variables.
The RIVE model applied to the whole upstream–downstream network with the
same parameters implies the recognition of the unicity of the kinetics formalisation
of ecological processes, whatever the location in the network, a strong assumption
that has proved realistic, not only along a land-to-sea continuum in a gradient of
watershed sizes (the Seine River [23]; the Danube River [35]) but also in stagnant
systems [9, 10] and in a variety of regions (sub-tropical [36]; subarctic [37]). The
RIVE model also performs fairly well in the case of transitory events such as
combined sewer overflows [38, 39]. For the study of urban systems, it is
implemented in the ProSe model [32, 38, 40], which was used to study the plurennial
impact of transient events on the carbon and oxygen cycles [41, 42], as well as
nutrients [27, 31, 43]. The implementation of RIVE in the ProSe model simulates
192
J. Garnier et al.