modelling: calibration of the coupled ecological library AED2 - - PowerPoint PPT Presentation

Telemac3D for aquatic ecological modelling: calibration of the coupled ecological library AED2

Autors:

Francesco Piccioni1, Brigitte Vinçon-Leite1, Minh-Hoang Le2, Bruno J. Lemaire1, Céline Casenave4, Yi Hoang1, Magali Jodeau2,3, Chi-Tuan Pham3, Javier Vidal3, Nicole Goutal2,3 1 LEESU, Ecole des Ponts ParisTech

2 LHSV, Ecole des Ponts ParisTech, CEREMA, EDF R&D, Chatou, France 3 LNHE, EDF R&D, Chatou, France 4 UMR MISTEA, Univ Montpellier, INRA, Montpellier SupAgro, Montpellier

17/10/2019

Context

2

Introduction Materials and methods Results Further steps

C/N/P HABs Urbanization Climate change

• Urbanization
• Heavy precipitation

Washes additional nutrients

Global warming

Favors cyanobacteria

………...….

(Figure from: H. Paerl, 2014)

A complex ecological library AED2 (Aquatic EcoDynamics) has been recently coupled with TELEMAC3D

3

• First calibration of the coupled T3D /

AED2 in a natural ecosystem

• Through high-frequency biological

data (10 mins) on a short 2-weeks period

• In terms of:

– thermal stratification (temperature) – Biology: [Chl-a], [O2], [Cyanobacteria]

Introduction Materials and methods Results Further steps

Objectives

Calibration: July 2018 Validation tests:

• July 2015
• July 2016

4

Study site

Lake Champs-sur-Marne

• Surface: 0.12 km2
• Mean depth: 2.5 m
• Max depth: 4 m

Recreational area Severe cyanobacteria blooms High-frequency data since 2015:

• Chl-a fluorescence
• Phycocyanin fluorescence
• DO
• Water temperature

Temp. O2 Chl-a Cyano.

Introduction Materials and methods Results Further steps

TELEMAC 3D: set up and domain

• Grid & bathymetry : via BlueKenue
• 620 nodes
• Interpolated bathymetry
• 10 layers, σ-transformation
• Hydrodynamics
• Turbulence closure model: mixing length

+ Damping function

• Activated air-water heat exchange
• Meteorological forcing:

Orly airport, (Météo France)

5 Main heat exchange parameters Wind reduction coefficient 0.00135 [-] Vertical diffusion 10-6 m2/s Light extinction coefficient Computed by AED

Introduction Materials and methods Results Further steps

100 m

AED2: main characteristics

• Activated modules :

– Oxygen – Carbon – Inorganic nutrients – Organic matter (PON, DON) – Phytoplankton

• 2 phytoplankton

groups:

– Green algae – Cyanobacteria 6

Introduction Materials and methods Results Further steps

AED2: main parameters

7

Green Algae Cyanobacteria Max growth rate 1.4 1.1

• Resp. rate

0.17 0.14 C / Chl-a 30 40

• Temp. coefficient

1.08 1.09 Optimal temp. (Min- Max) 26 °C (20-33°C) 28 °C (19-36°C) Half-Sat. const. for light limitation (IK) 10 μE m-2 s-1 15 μE m-2 s-1 Half-Sat. conc. of nitrogen (KN) 0.9 (mmol N m-3) 0.4 (mmol N m-3)

• I. C.
• Wat. Temp.

25.5°C O2 225 mmol O2 / m3 DIC 800 mmol C / m3 pH 7.5 Ammonium 2 mmol N / m3 Nitrate 0.5 mmol N / m3 Si 12.5 mmol Si / m3 P 0.3 mmol P / m3 Green agae Cyanobacteria 7 mmol C / m3 9 mmol C / m3 Nutrient and mineralization Nitrification rate 1.4 d-1 DON (PON) mineral. Rate 0.12 d-1 (0.4 d-1) Half sat. const. (for nitrif. And mineral.) 78.1 mmol N m-3 Sediment Oxygen demand 10 mmol O2 m-2 d-1

Source:

• AED2 manual
• Gal et al., 2009
• Trial and error calibration

Introduction Materials and methods Results Further steps

Results

Calibration period, July-August 2018

• Temperature: thermal stratification
• [Chl-a], [O2], [Cyanobacteria]

Validation tests: July 2015, 2016

• Temperature
• [Chl-a], [O2]

8 Introduction Materials and methods Results Further steps

Calibration period: august 2018

9

Introduction Materials and methods Results Further steps

Calibration period: august 2018

10

Introduction Materials and methods Results Further steps

Validation test: 2015

11

Introduction Materials and methods Results Further steps

Validation test: 2016

12

Introduction Materials and methods Results Further steps

Discussion and further steps

• Model results are satifying in terms of surface water

temperature; bottom water temperature is overestimated

• The biogeochemical library is able to correctly reproduce

– Dynamics of growth and mortality over short periods – The timing and the concentration of the peak of growth

• Complete validation:

– vertical profiles – remaining measuring sites – longer simulation periods

• Meteorological data need to be carefully assessed
• Nutrient input dynamics into the lake need to be better

assessed

13 Introduction Materials and methods Results Discussion

Introduction Materials and methods Results Further steps

14

THANK YOU !

Calibration period: inorganic nutrients

15

Introduction Materials and methods Results Further steps

NO3 NH4 PO4

Mixing length and damping function

Mixing length model:

• expresses the turbulent viscosity 𝜉 (or diffusion coefficient) as a function of the mean

velocity gradient and the mixing length (Prandtl’s theory): 𝜉 = 𝑔(𝑆𝑗)𝑀𝑛

2

𝐸𝑗𝑘𝐸𝑗𝑘, with: 𝐸𝑗𝑘

1 2 𝜖𝑉 𝑗 𝜖𝑦𝑘 + 𝜖𝑉 𝑘 𝜖𝑦𝑗

Damping function:

• Mixing reduction into a stable stratified: a damping function can be introduced into the

turbulence mixing length model according to the Richardson number. 𝑔(𝑆𝑗) = (1 + 14𝑆𝑗)−0.75, 𝑆𝑗 =

𝑕𝛼𝜍 𝜍 𝛼𝑣 2

• Ri is the Richardson number:

ratio between the buoyancy terms and the flow shear stress. Used to quantify the effects of the gravity terms in the turbulent power balance

16 Introduction Materials and methods Results Further steps

Calibration period: august 2018

17

Introduction Materials and methods Results Further steps