A distributed modelling approach to assess the use of Blue and Green Infrastructures to fulfil stormwater management requirements

Abstract

Blue and Green Infrastructures (B&GI) are nature-based solutions considered as particularly efficient to reduce the potential impact of new and existing developments with respect to stormwater issues. In order to assess their performance at some large scales compatible with urban projects, adapted distributed rainfall-runoff models are required. The latest advancements of the Multi-Hydro platform have made possible the representation of such B&GI. Applied in a virtual new urban development project located in the Paris region, Multi-Hydro has been used to simulate the impact of B&GI implementation, and their ability to fulfil regulation rules authorizing the connexion to the sewer network. The results show that a combination of several B&GI, if they are widely implemented, could represent an efficient tool to meet regulations at the parcel scale, as they can reduce runoff volume about 90%.

Full text

HAL Id: hal-02060156
https://hal-enpc.archives-ouvertes.fr/hal-02060156
Submitted on 15 Mar 2019
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
A distributed modelling approach to assess the use of
Blue and Green Infrastructures to full stormwater
management requirements
P.-A. Versini, N. Kotelnikova, A. Poulhes, I. Tchiguirinskaia, D. Schertzer, F.
Leurent
To cite this version:
P.-A. Versini, N. Kotelnikova, A. Poulhes, I. Tchiguirinskaia, D. Schertzer, et al.. A dis-
tributed modelling approach to assess the use of Blue and Green Infrastructures to full stormwa-
ter management requirements. Landscape and Urban Planning, Elsevier, 2018, 173, pp.60-63.
�10.1016/j.landurbplan.2018.02.001�. �hal-02060156�

Elsevier Editorial System(tm) for Landscape
and Urban Planning
Manuscript Draft
Manuscript Number: LAND-D-18-00056R1
Title: A distributed modelling approach to assess the use of Blue and
Green Infrastructures to fulfil stormwater management requirements
Article Type: Research Note
Keywords: blue green infrastructures, stormwater management, distributed
modelling
Corresponding Author: Dr. Pierre-Antoine Versini, Dr
Corresponding Author's Institution: Ecole des Ponts ParisTech
First Author: Pierre-Antoine Versini, Dr
Order of Authors: Pierre-Antoine Versini, Dr; Natalia Kotelnikova, Dr.;
Alexis Poulhes, Dr.; Ioulia Tchiguirinskaia, Dr.; Daniel Schertzer,
Prof.; Fabien Leurent, Prof.
Abstract: Blue and green infrastructures (B&GI) are nature-based
solutions considered as particularly efficient to reduce the potential
impact of new and existing developments with respect to stormwater
issues. In order to assess their performances at some large scales
compatible with urban projects, adapted distributed rainfall-runoff
models are required. The latest advancements of the Multi-Hydro platform
have made possible the representation of such B&GI. Applied in a virtual
new urban development project located in the Paris region, Multi-Hydro
has been used to simulate the impact of B&GI implementation, and their
ability to fulfil regulation rules authorizing the connexion to the sewer
network. The results show that a combination of several B&GI, if they are
widely implemented, could represent an efficient tool to meet regulations
at the parcel scale, as they can reduce runoff volume about 90%.

A distributed modelling approach to assess the use of Blue and Green
Infrastructures to fulfil stormwater management requirements
P.-A. Versini1,*, N. Kotelnikova2, A. Poulhes2, I. Tchiguirinskaia1, D. Schertzer1, F.
Leurent2
1 HMCo, Ecole des Ponts ParisTech, Champs-sur-Marne, France
2 LVMT, Ecole des Ponts ParisTech, Champs-sur-Marne, France
* Corresponding author: pierre-antoine.versini@enpc.fr
*Title Page with Author Identifiers

1
A distributed modelling approach to assess the use of Blue and Green 1
Infrastructures to fulfil stormwater management requirements 2
3
4
Blue and green infrastructures (B&GI) are nature-based solutions considered as 5
particularly efficient to reduce the potential impact of new and existing developments 6
with respect to stormwater issues. In order to assess their performance at some large 7
scales compatible with urban projects, adapted distributed rainfall-runoff models are 8
required. The latest advancements of the Multi-Hydro platform have made possible 9
the representation of such B&GI. Applied in a virtual new urban development project 10
located in the Paris region, Multi-Hydro has been used to simulate the impact of 11
B&GI implementation, and their ability to fulfil regulation rules authorizing the 12
connexion to the sewer network. The results show that a combination of several 13
B&GI, if they are widely implemented, could represent an efficient tool to meet 14
regulations at the parcel scale, as they can reduce runoff volume about 90%. 15
16
Key words: blue green infrastructures, stormwater management, distributed modelling 17
18
19
1 Introduction 20
21
Blue and Green Infrastructures (B&GI), including green roof, bio-retention swale, 22
porous pavement, harvesting tank, soakaway or pond for instance, can provide 23
multiple benefits to urban areas affected by both climate change and urbanization 24
effects: urban heat island reduction, biodiversity conservation, reduced buildings 25
energy requirements,... Last but not least, they appear to be particularly efficient in 26
stormwater management (Liao et al, 2017). By detention, infiltration and 27
evapotranspiration processes, they can be used to control urban runoff at the local 28
scale. 29
30
The hydrological performance and benefit of B&GI have been shown in numerous 31
studies conducted at small scales: Kamali et al. (2017) for porous pavement, 32
Chapman and Horner (2010) for bioretention system, or Stovin et al. (2012) for green 33
roofs. Nevertheless, their performance and interaction at higher scales (urban project) 34
*Blinded Manuscript with No Author Identifiers
Click here to view linked References

2
are still uncertain and insufficiently quantified. Modelling tools are required to 35
consider B&GI configuration and optimize their performance, as most of the existing 36
models are focused on one or very few assets such as green roofs (Versini et al., 37
2015). Few of them are technically able to combine dynamically several 38
infrastructures, but usually in a semi-distributed approach that mixes several types of 39
landcover (road, house, grass, park…), and implies some huge difficulties to adjust a 40
priori the parameters without observed data. It is the case of the Storm Water 41
Management Model, as shown in Lucas et al. (2015) or Palla and Gnecco (2015) 42
among others. To properly assess B&GI performance on a large set of spatial scales, a 43
hydrologic model characterized by a high spatial resolution is also required. Such a 44
structure is necessary to consider heterogeneous surfaces, and the associated 45
dynamics due to the layout of impervious and pervious areas. 46
47
Based on these considerations, the main objective of this research note is to assess the 48
performance of B&GI in stormwater management at the urban project scale. A 49
distributed modelling approach has been chosen to especially study the respective 50
performance of a B&GI set, and their evolution regarding storm event return periods. 51
52
2 Presentation of the case study: the “Echangeur” project 53
54
The virtual urban project called “Echangeur” has been designed by a group of 55
students during a specialized master training devoted to the “Ecodesign of Sustainable 56
Cities”. Supported by the Academic Chair on the Eco-design of building sets and 57
infrastructure established by ParisTech and the Vinci group (see Kotelnikova et al., 58
2016 for a detailed presentation), the main activity of this course is to design a 59
sustainable neighborhood materialized by a layout plan. Located in the eastern 60
suburbs of Greater Paris (Champs-sur-Marne, France) and covering an area of 10.66 61
ha, the plan proposed by the students for the Echangeur project (Figure 1) hosts 62
accommodation for 5900 inhabitants and activities with the creation of 1150 jobs. 63
This plan must also fulfill stormwater management requirements concerning the 64
connection to the stormwater network. Here the discharge at the parcel outlet has to 65
be lower than a reference threshold of 10 l/s/ha for a rainy event characterized by a 66
20-years return period. In the Paris region, this corresponds to a 30-minute rainfall 67
event characterized by a 50 mm/h intensity. 68

3
69
Figure 1. Layout of the Echangeur catchment differentiating the different land use 70
classes 71
72
Due to a lack of space, the construction of a large storage unit has not been 73
considered. Several blue and green infrastructures have been planned to fulfil this 74
stormwater regulation rule: (i) Green spaces (grass, forest and vegetable gardens), (ii) 75
green roofs, (iii) green swales (swaled drainage course with sloped sides and filled 76
with vegetation and riprap), (iv) small retention basins, (v) porous pavement. 77
78
3 Materials and method 79
80
3-1 The Multi-Hydro model 81
82
The Multi-Hydro distributed rainfall-runoff model represents a well-adapted tool to 83
assess hydrological impacts at the urban scale (Giangola-Murzyn, 2014, Ichiba et al, 84
2017). For each time step, Multi-Hydro provides overland water depth (flooding) and 85
infiltration maps, but also discharge values for each pipe and junction of the 86
stormwater network. Multi-Hydro is currently being developed at the XXX to take 87
into account the wide complexity of urban environments. The latest advancements 88
have made possible the representation of several “resilience infrastructures” such as 89
basins, barriers, and green roof (see Versini et al., 2016 for details). Based on these 90
previous works, Multi-Hydro has been adapted to reproduce the hydrological 91
behaviour of the mentioned B&GI planned in the Echangeur project. 92
93
Multi-Hydro has been implemented on this case study to simulate its hydrological 94
response with a resolution of 5 m in space and 5 minutes in time. Based on the layout 95
plan, the input data required by the model (map of topography, landuse and 96
stormwater network) were produced by using adapted GIS tools. 97
98
3-2 Land use scenarios 99
100
In order to study the relative contribution of each implemented B&GI, different land 101
use scenarios have been established: (0) there is no blue or green infrastructure, but 102

4
only impervious surfaces such as roads, buildings and pavements, (1) Green spaces 103
are implemented, (2) Every building roof is covered with an extensive green roof, (3) 104
Green swales are implemented, (4) Impervious pavements are replaced with porous 105
ones on the pedestrian area, (5) Most of the stormwater network outfalls are 106
connected to two small retention basins, (6) All of the B&GI mentioned above are 107
implemented. 108
109
3-3 Rainfall scenarios 110
111
To quantify the relative performance of B&GI regarding stormwater management 112
issue, several rainfall scenarios have been provided. These are synthetic hyetographs 113
characterized by a homogenous precipitation and based on the specific Intensity-114
Duration-Frequency relationship (established in a station located 20 km away from 115
the studied area by Météo-France). They were computed for a 30-minute duration 116
(close to the watershed concentration time) and several return periods (see Table 1). 117
118
Table 1. Rainfall intensity (expressed in mm/h) for the 8 considered return periods 119
120
3-4 Work plan 121
122
Multi-Hydro was applied on every land use and rainfall scenario (7x8 simulations). 123
Some of the resulting hydrographs are illustrated in Figure 2 and analysed in the 124
following. Note that two indicators were used to assess B&GI performance: runoff 125
volume (

V) and peak discharge (

Qp) reductions: 126
ΔQp(%)=
(Qp0- Qpi)
Qp0
100
(Eq. 2) 127
128
ΔV(%)=
(V0- Vi)
V0
100
(Eq. 3) 129
Where Qp0 and V0 refer to peak discharge and runoff volume computed for the 130
impervious situation, whereas Qpi and Vi correspond to those computed for the 131
different B&GI scenarios. 132
133
4 Presentation of the results 134
For the impervious situation, most of the rainfall volume is transferred to the basin 135
outfalls. Only initial losses and water stored in local depression can be deduced. 136

5
Regarding 30-minute duration events, peak discharge reaches 200 l/s to 1200 l/s. It is 137
worth noting that regulation threshold is exceeded whatever the return period of the 138
considered storm event. 139
140
When 11.6% of the total area is covered by green spaces (Scenario 1), runoff volume 141
and peak discharge decrease by approximately 10-15% for the more frequent events, 142
and less than 10% for the strongest ones. In these cases, infiltration capacity of green 143
spaces is reduced, and some water is finally drained to the stormwater network 144
145
The green roof implantation proposed in Scenario 2 -representing 42.3% of the 146
watershed area- induces both runoff volume and peak discharge reduction starting 147
from 15% to 25% for the more frequent storm events, and dropping to about 5% for 148
the heaviest ones. Green roofs appear to be particularly efficient at the beginning of 149
the storm, when they can temporarily store water in the substrate. 150
151
In Scenario 3, Green swales represent a small part of the studied basin (5.5%), but 152
they drain water from surrounding elements (almost 30% of the total area). It is 153
illustrated by some runoff volume and peak discharge reductions that vary from 30% 154
for the 1-month event to 17% for the 20-year one. 155
156
As porous pavements represent 31.5% of the whole area, their implementation in 157
Scenario 4, characterized by a high storage capacity, significantly influences the 158
hydrological response of the catchment. Both runoff volume and peak discharge 159
decrease about 30-40% depending on the considered rainfall event. 160
161
Retention basins drawn up in Scenario 5 represent the most effective infrastructure in 162
terms of runoff reduction as they drain two thirds of the catchment area. Both runoff 163
volume and peak discharge decrease of about 70% for the more common storm 164
events. For the highest events, the total storage capacity (1300 m3) is reached. From 165
that time, the exceeded water is routed to the stormwater network and produces a 166
“step” in the catchment response. 167
168
The implementation of all of the B&GIs on the Echnageur project (Scenario 6) is 169
obviously the most effective configuration. Both peak discharge and runoff volumes 170

6
are reduced by about 90% on the wide range of return periods, and the regulation rule 171
of 10 l/s/ha is almost always met (except for the two main events). 172
173
Figure 2. Presentation of the simulated hydrographs for different rainfall events and 174
B&GI scenarios. Orange horizontal solid line corresponds to the 10 l/s/ha regulation. 175
176
5 Conclusions and perspectives 177
178
A combination of B&GI appears to be the best solution to significantly reduce the 179
quantity of water flowing into the sewage network during storm events, and to fulfil 180
regulation rules established by local stormwater managers. The distributed structure 181
of Multi-Hydro and the possibility to reproduce a large set of B&GI allow the 182
realization of such detailed and dynamic impact studies. As Multi-Hydro is still in 183
development, additional B&GI could be added in the future, and among these, 184
different configurations could be tested (ie. several green roofs differentiated by their 185
substrate porosity or thickness). 186
187
The presented results must be taken with caution, as they depend on the catchment 188
configuration, especially on the combination of impervious and pervious surfaces, but 189
also on its geometry and on the sewage network arrangement. Moreover, it should 190
also be noticed that initial conditions have not been considered in this study. Every 191
B&GI was assumed to be empty / unsaturated at the beginning of every rainfall event. 192
Future versions of Multi-Hydro should take into account evapotranspiration processes 193
and detention basins draining during dry periods to better estimate the initial state of 194
the system. The succession of several rainfall events should also be possible to study 195
B&GI performance in more realistic conditions, as it is usually the case for rainwater 196
harvesting tank or detention basin sizing. 197
198
Acknowledgments 199
200
References 201
Chapman, C., & Horner, R.R. (2010). Performance assessment of a street- drainage 202
bioretention system. Water Environ. Res., 82 (2), 109–119 203

7
Giangola-Murzyn, A. (2014). Modélisation et paramétrisation hydrologique de la 204
ville, résilience aux inondations, PhD thesis, Université Paris-Est, 260 pp. 205
Ichiba, A., Gires, A., Tchiguirinskaia, I., Schertzer, D., Bompard, P., & Ten Veldhuis, 206
M.-C. (2017). Scale effect challenges in urban hydrology highlighted with a 207
distributed hydrological model. Hydrol. Earth Syst. Sci. Discuss., 208
https://doi.org/10.5194/hess-2017-286 209
Kamali, M., Delkash, M., & Tajrishy, M. (2017). Evaluation of permeable pavement 210
responses to urban surface runoff, Journal of Environmental Management, 211
187, 43-53 212
Kotelnikova, N., De Bartoli, A., Féraille,A., & Leurent, F. (2016). Integrating Urban 213
Ecodesign in French engineering curricula: an example at École des Ponts 214
ParisTech. International Conference on Sustainable Built Environment, 215
Hamburg, Germany 216
Liao, KH., Deng, S., Tan, P.Y. (2017). Blue-Green Infrastructure: New Frontier for 217
Sustainable Urban Stormwater Management. In: Tan P., Jim C. (eds) Greening 218
Cities. Advances in 21st Century Human Settlements. Springer, Singapore 219
Lucas, W.C., & Sample, D.J. (2015). Reducing combined sewer overflows by using 220
outlet controls for Green Stormwater Infrastructure: case study in Richmond, 221
Virginia. Journal of Hydrology, 520, 473–488 222
Palla, A., & Gnecco, I. (2015). Hydrologic modeling of Low Impact Development 223
systems at the urban catchment scale, Journal of Hydrology, 528, 361-368 224
Stovin, V., Vesuviano, G. & Kasmin, H. (2012). The hydrological performance of a 225
green roof test bed under UK climatic conditions. Journal of Hydrology, 414-226
415, 148-161. 227
Versini, P.A., Gires, A., Tchiguirinskaia, I., & Schertzer, D. (2016). Toward an 228
operational tool to simulate green roof hydrological impact at the basin scale: 229
a new version of the distributed rainfall-runoff model Multi-Hydro. Water 230
Science and Technology, 74(8), 1845-1854. 231
Versini, P.A., Jouve, P., Ramier, D., Berthier, E. & de Gouvello, B. (2015). Use of 232
green roofs to solve storm water issues at the basin scale - Study in the Hauts-233
de-Seine county (France). Urban Water Journal. 234
http://dx.doi.org/10.1080/1573062X.2014.993993235

8
List of Tables 236
237
Table 1. Rainfall intensity (expressed in mm/h) for the 8 considered return periods 238
239
240
241
242
243

9
244
1 month
3 months
6 months
1 year
2 years
5 years
10 years
20 years
D=30min
8.0
14.4
19.7
25.9
31.4
42.0
50.1
58.2
245
246
247

Figure 1
Click here to download high resolution image

Figure 2
Click here to download high resolution image