SSA 13 : Thau Lagoon

1
Science and Policy Integration for COastal System
Assessment

• SSA 13 Thau Lagoon
• Managing the microbiological contamination of
  the Thau Lagoon

José A. Pérez Agúndez, Annie Fiandrino, Johanna
Béganton, Rémi Mongruel , Thierry Laugier,
Valérie Derolez, Ludovic Cesmat, Ophélie Serais
Hélène Rey-Valette, Sébastien Roussel,François
Valette
Gilles Brocard
(NB GEYSER, in charge of mediation with
stakeholders, has retired from SPICOSA)
SAF cluster meeting, 20-21 October 2009,
Thessaloniki
2
Content

• Introduction  policy issue, scenarios,
  socio-economic dimensions and structure of the
  system to be modelled
• Building the model with Extend (modular use,
  connections)
• Watersheds and the Water treatment system
• Microbiological contamination monitoring and
  law enforcement
• Economic assessment
• Conclusion  state of work interactions with
  stakeholders

3
A coastal zone close to thecities of Montpellier
and Sète
4
The detailed structure of the Thau watershed
5
An overview of the policy issue

• The general problem because of rapid local
  demographic growth and also the importance of
  tourism, the watersheds of the Thau lagoon are
  subject to pollution. In particular, events of
  microbiological contamination may affect the
  lagoon, with direct consequences on human
  activities in the Lagoon and also feedbacks to
  the overall development pattern of the region.
• The microbiological contamination raises four
  basic questions1) What are the impacts of
  microbiological contamination on shellfish
  farming and other human activities in the lagoon
  ?
• 2) What are the main contamination sources ?
• 3) What is the efficiency of the current water
  treatment system as regards microbiological
  contamination ?
• 4) Which management options would reduce the
  impacts of MC in a way which would be coherent
  with the more global local policy objectives ?
• Another research project, OMEGATHAU (to be
  completed by the end of 2010), which has been
  conducted jointly with the local public
  organisation in charge of Thau Lagoon water
  management (SMBT), has addressed the 2 first
  questions. Our approach for the SPICOSA SSA work
  in the Thau Lagoon is to rely on OMEGATHAU
  results in order to explore questions 3 and 4.

6
Policy context two corresponding management
plans
SAGE The local water management plan,which
defines the water policyfor the Lagoon and its
watersheds
Coherence of - perimeters- diagnosis- objectives
A shared diagnosis
SCOT The territorialmanagement plan,whose
perimeter iscoherent with theone defined
forthe water policy
7
The current policy debate and theexpectations
from the SPICOSA experiment

• The local policy framework, which has been set
  up during the preparation of the Territorial
  Management Plan (SCoT) gives the general
  objectives
• - economic development, including tourism and
  thermal industry- environment preservation,
  including water quality and ecosystems
  protection- maintaining of the cultural
  patrimony, including traditional activities
  (shellfish farming, professional fisheries in the
  lagoon, recreational activities, etc.)
• NB of course, these objectives generate internal
  conflicts
• The local water management plan (SAGE) defines
  several specific objectives related to the
  environment policy. One specific objective is the
  reduction of the water microbiological
  contamination. The current political debate
  focuses the ways of translating these objectives
  into operational management options.
• In this context, the main purpose of our SPICOSA
  experiment is twofold- to assess the efficiency
  and the costs of new water treatment
  systems Cost-effectiveness analysis (1
  restricted)- to estimate the impacts of various
  operational management options on the traditional
  activities and the local economy Multi-criteri
  a analysis (2 extended)

8
Scenarios definition and system assessment

• The definition of scenarios and the assessment of
  the resulting states of the system will
  contribute to the local political debate in three
  different ways.
• Designing the Water Policy the objective of the
  MC reduction has to be translated into
  operational objectives, in particular as regards
  the sanitary classification of the Lagoon (A or
  B) and the occurrence of commercial bans for the
  shellfish farming industry, according to the
  trends of the system (forcings) definition of
  scenarios trends, operational objectives,
  technical options
• Searching for the more efficient water treatment
  systems
• the assessment of possible new water treatment
  systems (technical options), as regards their
  costs and their ability to reach one particular
  operational objectives costs/effectiveness
  analysis of simple management options
• Assessing the impacts of one operational policy
  objectives- on the regional economy, through the
  financial analysis of impacted sectors and the
  induced effects on the other industries and
  activities- on the broader development concerns
  (as set up by the SCOT), considering eventually
  additional management options for traditional
  activities
• selection of indicators multi-criteria analysis
  of more complex management options

9
The economic dimensions
Economic dimension 1 cost-effectiveness analysis
of the water treatment systems, according to
different technical options. Approach
investment and running costs of water treatment
settlements, considering local public budget
constraints Economic dimension 2 economic
dynamics in the shellfish farming
industry. Approach financial analysis and the
vulnerability of businesses to the closures of
the shellfish farming area (period during which
sales are forbiden) Additional management options
to be tested in complex scenarios re-allocation
of the production capacity of the farms which
stop their activity Economic dimension 3 a
regional economy matrix will be used so as to
provide macro-economic drivers and indicators at
the site scale. Approach incorporating direct
(pressure and impacts) and indirect (induced
effects) relationships between the environment
and the macro-economic dynamics of the Thau
Lagoon territory (economic feedback loops)
10
The system and its components
11
The structure of the Extend Model
12
The watershed ecological modules
13
The watershed ecological modules

1. Every day permanent (and/or seasonal) populations
   activities produce E.coli pollutions
2. These E.coli loads can be reduced (or not) by
   specific treatments
3. These reduced (or not) E.coli loads discharge
   into the lagoon via local streams

What we need to achieve this goal in extend model
? To identify Main types of E.coli source on
watershed To formulate (in an analytic way)
E.coli daily load for each type of source
14
The watershed ecological modules
In Extend model (in order to be as well generic
as possible) we distinguish these 3 steps
3 different modules
E.coli pollutions production blocks
E.coli pollutions treatment blocks
E.coli pollutions transfer blocks
15
The watershed modules
E.coli pollutions production blocks
Parameters read in Database
For each town on watershed
Nb. of Perm. Inh. Nb. of tourists
living in houses connected to WWTP SRC1
of pop. connected to WWTP
living in houses with their own septic tanks
SRC2
Nb. of house with ind. WWT System
Seasonal population living in campings with their
own WWT systems SRC3
Nb. of tourists in campings during summer
Farming (poultry, horse,) SRC4
Nb. of farms Nb. of animals in each farm
1 Eq. Inh 5,7x10 10 E.coli/d
.
For each type of SRC
State variable saved in Database
16
The watershed modules
E.coli pollutions treatment blocks
Parameters and state variables read in Database
For each SRC on watershed
Daily Produced E.coli load
Daily Produced E.coli load
E.coli Abatement techniques (or not)
Parameters for treatment method
State variable saved in Database
Daily Treated E.coli load lt/ Daily Produced
E.coli load
Empirical values for abatement in each WWTP
depends on
For instance
SRC1 Daily E.coli load produced by pop.
connected to WWTP

• Season (treatment is more efficient in summer)
• Meteorological condition (by-pass during rainfall
  events)

.
17
The watershed modules
E.coli pollutions transfer blocks
Parameters and state variables read in Database
For each SRC on watershed
Daily Treated E.coli load
Daily Treated E.coli load
Abatement in local stream depends on
E.coli Abatement due to natural processes during
transfer in the local stream (from point source
to the lagoon)

• Distance from point source to the lagoon
• Season (T90 in stream is higher in summer)
• Meteorological condition (speed of stream higher
  during rainfall event)

Daily Transferred E.coli load lt/ Daily Treated
E.coli load
State variable saved in Database
.
18
The watershed modules
E.coli pollutions transfer blocks
In Watershed Database Number of tables
(containing fields associated to Daily
Produced,Treated and Transferred E.coli load)
numbers of SRC on the watershed

Custom block
Re-arrange daily E.coli load create one
Dynamic_Database per outlet that contains
Total Daily E.coli load (sum of all E.coli load
of all SRC on this watershed) and each component
of this total load
19
The water treatment module

• There are infinite technical ways for reaching
  the objectives fixed by stakeholders
• creating new water treatment structures
• modifying existing structures (size, technology)
• improving the collection nets
• management of punctual treatment under-capacity
  (storms)
• But the aim of the model is to explore
  prospective scenarios, not to optimize a
  public policy

20
The cost efficient analysis
v
21
Operational use of the water treatment module by
Extend

• The water treatment module, associated with the
  hydrologic component, is integrated in the model
  following two complementary steps
• 1st ? Restrictive model comparative assessment
  of water treatment options (several runs)
• Selection of best possibilities considering
  general constraints
• total budget (operational and investment
  costs), land availability, opportunity costs,
  etc.
• Comparative assessment of water treatment
  options
• Hierarchical classification of each option by
  considering costs and abatement efficiency
• 2nd ? Global model integration of the selected
  water treatment options in the prospective model
• Strategic planning at long run in terms of
  investments
• Estimation on global impacts in the model
  depending of selected scenarios

22
Lagoon contamination module
INTO THE LAGOON
National monitoring network (REMI) controls
microbiological quality of oysters and mussels to
determine category of shellfish production areas.
9 monitoring stations are sampling every first
Monday of the month
If E.coli concentration in shellfish passes a
legal level, administration can decide to
temporarily forbid shellfish harvesting
Second step Simulate surveillance procedure
and regulation rules
23
Lagoon contamination simulation using 3D model
First step Simulate impact of E.coli load
discharge into the lagoon by each local
streams using results of 3D-hydrodynamics model
coupled to a biological module that simulate
transport, dispersion, mortality of E.coli cells
in water lagoon
100 m x 100 m grid 10 s level along vertical
direction
Calibration and validation of this 3D model
achieve in the framework of OMEGA-THAU project
24
Lagoon contamination simulation using 3D model
Comparisons in situ measurements / results of
simulations for salinity
Salinity sensor
High frequency measurements
Precise in situ measurements
25
Lagoon contamination simulation using 3D model
Comparisons in situ measurements / results of
simulations for E.coli in water
Precise in situ measurements
Best results obtained with T90 48h
26
Lagoon contamination simulation using 3D model
Sensitivity studies with 3D- model shown
1) E.coli in water can be summed
2) High impact of wind conditions on E.coli in
water
No wind
Daily E.coli load 13 Log
Wind conditions O-NO, 36 km/h
5 characteristics wind conditions are retained
315N, 330N, 90N, 150N / 10m/s, No wind)
27
Lagoon contamination simulation using 3D model
Sensitivity studies with 3D- model shown
3) High impact of temporal distribution of stream
flow on E.coli in water
Daily E.coli load Stream flow x E.colistream
E.coli x Stream Flow E.coli x Stream
High x Low
Low x High
E.coli discharge during dry weather
E.coli discharge during rainfall event
Daily E.coli load 13 Log
Wind conditions N-NO, 36 km/h
2 characteristics meteorological conditions are
retained dry period, rainfall event 70 mm in
one day
28
Lagoon contamination module
First step Simulate impact of E.coli discharge
into the lagoon at each time step, at each REMI
station (i 1,9)
E.coli REMI Station
Daily E.coli load (DEcL) from a local stream
For a fixed wind condition and a fixed
meteorological condition Transfer function
depends on 4 parameters
Cmax Linear function (DEcL) Tdeb 2nd order
Poly. function (DEcL) Tfin 2nd order Poly.
function (DEcL) Tmax independent of DEcL
At current time, a DEcL can induce non zero
E.coli REMI Station d-day and days after
E.coli REMI Station (state variable) is saved
in Database at current time and following time
step if necessary
29
Lagoon contamination module
30
The governance module regulation of sanitary
risks due to the microbiologic water charge
31
The governance module monitoring and regulation
Risk of contamination (rainfall, incident, etc..)
?
MONITORING
ALERT Level 0
J 1er Monday of the month
If all values lt Threshold (9/9)
9 sampling (REMI)
Alert
MONITORING J1
If 1 value gt Threshold (1/9)
J J détection du risque
9 sampling (REMI)
ALERT Level 1
MONITORING J2
If all values lt Threshold (9/9)
J J 2
9 sampling (REMI)
If 1 value gt Threshold (1/9)
Alert Level 2
MONITORING
J every monday
All values (9/9), Must have 2 values lt
threshold (9/9)
9 sampling (REMI)
If 1 value gt Threshold (1/9)
32
The Shellfish farming module (1/3)

• - Figuring the shellfish farming sector
• Thau lagoon is the principal shellfish farming
  production area of the French Mediterranean
• There are globally 540 companies that represent
  locally a source of direct employment for 2,000
  workers (fulltime equivalent estimates). The
  global production can be estimated slightly
  higher than 20,000 tons by year and distributed
  into 60 of oysters and 40 of mussels
• Companies are segmented into three categories
  depending on their production capacity. This
  capacity is directly linked to the number of
  concessions they own
• - Small companies (owning until 4 tables)
• - Medium companies (owning 4-8 tables)
• - Big companies (owning more than 8 tables)

33
The Shellfish farming module (2/3)
- Rationale of the approach . No bioeconomic
model (such as in SSA10) . Each commercial
ban potential economic losses . Linking the
production capacity by type of company, their
representative market distribution structure and
the selling prices practiced, it is possible to
estimate the global revenue obtained by class of
company in a normal year. . A normal year can
be considered as an annual economic period with
no sanitary closure events. Consequently, it
translates the maximum revenue which can be
obtained by shellfish companies in the absence of
sanitary events.
34
The Shellfish farming module (3/3)

• Expected output of the shellfish farming module
• . The shellfish farming box of the model
  evaluates the economic impacts directly supported
  by companies due to microbiological water
  overcharge
• . Estimation of the economic losses associated to
  commercial bans due to microbiologic peaks .
  Effects at short-run determined by intra-annual
  economic dynamics and sanitary rules in the
  governance box
• . Effects at mid and long-un image impacts,
  economic sensibility of companies

35
The macro-economic module
      SOL et ROUTES SOL et ROUTES SOL et ROUTES SOL et ROUTES SOL et ROUTES LOCAUX et CONSTRUCTIONS LOCAUX et CONSTRUCTIONS LOCAUX et CONSTRUCTIONS etc
Noms   PS. NON CONST. PS.CONSTR AMNGT.SOL RES.ROUTES AMNGT.RTES HABIT CONC HABIT LOT. P.HAB.ISOL
  Unités 100HA 100HA 100HA 100KM 100KM 100LGT 100LGT 100LGT
1 Us.SOL.CONSTR. 100HA   1.00 1.00     -0.010 -0.020 -0.060
2 Us.SOL NON CONSTR. 100HA 1.00   -1.00 -0.46 -0.46 0.00   0.00
3 HAB.CONC. LOGT           100.00   0.00
4 HAB.LOTISST. LOGT             100.00  
5 HAB.DISPERS. LOGT           0.00   100.00
6 RESID.SEC LOGT           0.00   0.00
7 SEJ.TOUR. KJTRS           0.00   0.00
8 LOC.PROF. 1000M2           0.00   0.00
9 LOC.TECH. 1000M2           0.00   0.00
10 Us.MACH.AGR. KHUSM -1.20   -2.00 -0.55 -0.55 0.00   0.00
11 Us.RES.ROUTES 100KM 0.00     1.00 1.00 -0.01 -0.01 -0.01
12 RES.EAU.POT. 100KM                
13 RES.EAU.NON POT. 100KM                
14 RES.COL.EAU.US 100KM                
etc                  
36
The tourism industry
There are two ways to integrate the tourism
industry in the model (this question is actually
under discusion) - Endogenously (feed back) the
site attractiveness generates a tourism demand .
The objective is to calibrate a function
comparing data series of tourism frequentation
and an indicator of water quality (sanitary
closures as a proxy) . Problem in water quality
other environmental components are included.
Separability question - Exogenously if
problems to assess an attractiveness function,
the dynamics of the sector can be forced in the
scenarios box
37
Conclusion state of work

• The modeling activities are going on
• The model is not complete at the whole but some
  associated blocks are starting to run
• Some modules of the model need to be quickly
  developed
• High efforts have been made in restructuring the
  model, programming and integrating
• We have some incertitude about additional
  information required for some processes but
  globally the structure of the model seems enough
  robust
• Interaction with local stakeholders will be
  assured by our Spicosa colleagues participating
  to the expertise requirements of Scot and Sage

38
Thank you for your attention