Top Banner
L. Boithias, R. Marcé, V. Acuña, J. Aldekoa, V. Osorio, M. Petrović, A. Ginebreda, F. Francés, S. Pérez, S. Sabater 4th SCARCE International Conference 2526 November 2013, Cádiz, Spain Assessment of the water purification ecosystem service regarding instream pharmaceutical residues: Exploring the GREATER model parameters based on data uncertainty
16

of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Apr 21, 2018

Download

Documents

vanbao
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

L. Boithias, R. Marcé, V. Acuña, J. Aldekoa, V. Osorio, M. Petrović, A. Ginebreda, F. Francés, S. Pérez, S. Sabater

4th SCARCE International Conference25‐26 November 2013, Cádiz, Spain

Assessment of the water purification ecosystem service 

regarding in‐stream pharmaceutical residues: 

Exploring the GREAT‐ER model parameters 

based on data uncertainty

Page 2: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Barcelona

Igualada

Manresa

Anoia river

Llobregat river

Cardener river

N

Drinking water treatment facilityWaste water treatment plants

Pharmaceuticals in the Llobregat basin

Boithias et al. SCARCE 20132

• 5000 km2

• High industrial, agricultural and urban activity: 60 WWTP

• PCP = 700 mm . At outlet, discharge is 19 m3 s‐1

• The basin supplies drinking water for the 3 million inhabitants Barcelona area

• Throughout the basin, discharge may be provided by the WWTP

• High concentrations of pollutants, including pharmaceuticals

• Water purification service from WWTP and ecosystems is a major issue

Page 3: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Modelling the fate of pharmaceuticals

• Pharmaceuticals are ubiquitous in densely urbanized areas

• Attenuation of pharmaceutical contamination in the aquatic environment depends on:

• Few studies about the in‐stream attenuation of pharmaceuticals

• Objective: assess the ability of the spatially explicit GREAT‐ER model to simulate the concentration of 13 pharmaceuticals in two contrasted hydrological conditions

• Uncertainty analysis approach

Boithias et al. SCARCE 20133

‐ Physico‐chemical properties‐ Hydrology (sediments)

Page 4: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

The GREAT‐ER model

• Steady‐state spatially explicit model (Boeije and Koorman, 2003) for personal care and pharmaceutical compounds

• Previously applied over several catchments

– UK: triclosan (Sabaliunas et al., 2003), LAS (Price et al., 2009), diclofenac and propranolol (Johnson et al., 2007)

– Austria: LAS, EDTA, triclosan (Wind et al., 2004)

– Germany: carbamazepine and diclofenac (Heberer et al., 2005)

– Swiss:  Estrogens (Vermeirssen et al., 2006)

– Spain: Diclofenac in the Llobregat (Aldekoa et al., 2013)

Boithias et al. SCARCE 20134

Page 5: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

The GREAT‐ER model

• 3levels of complexity ‐> simplest one

• Inputs: 

Boithias et al. SCARCE 20135

Attenuation = Degradation / Sorption

Catchment descriptionPharmaceuticals 

properties

‐ River stretches location (confluences, dams, WWTP, gauging stations)‐ River stretches annual discharges, velocity, depth

‐WWTP location ‐WWTP annual discharge

‐ Annual pharmaceutical emissions to sewage (kg cap‐1 yr‐1)

‐WWTP removal rates (percentage ‐ %)

‐ River removal rate (decay ‐ h‐1)

Outputs: annual average concentration in each river stretch 

Page 6: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

LLO1

LLO4

LLO3

LLO2

LLO5LLO6

LLO7

CAR3

CAR2

CAR1

CAR4

ANO1ANO2

ANO3

Data ‐ Pharmaceuticals of interest

Boithias et al. SCARCE 20136

• Discharge : ACA• Pharmaceuticals in‐stream 

concentrations: 14 sampling points of SCARCE– 2 campaigns : high flow (2010) and low 

flow (2011) conditions• Selected pharmaceuticals based on:

– Medical use (point source through WWTP)– At least 8 out of 14 samples with 

concentration > LOQ for both campaigns– Availability of WWTP removal efficiencies 

values (Gros et al., 2010; Jelic et al., 2011)– Availability of river removal efficiencies 

values (literature review – 30 references)– 13 selected pharmaceuticals

• Calculated percentiles to provide a 95% confidence interval

Page 7: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Input data uncertainty

Boithias et al. SCARCE 20137

• Uncertainty is high, depends on:– Molecules– Number of available data

• Model minimal, median and maximal scenarios

• Simulation of the statistical distribution of observed data ‐> avoiding the calibration step

• Is the simplest model of GREAT‐ER able to model pharmaceutical fate in the Llobregat?

1<n<27

6<n<38

7<n<40

Page 8: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Results

Boithias et al. SCARCE 20138

Nonsteroidal anti‐inflammatory‐ 9 < RMSEmedian < 65‐Min & max encompass the 1:1 line‐Well simulated for both low flow and high flow 

Page 9: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Results

Boithias et al. SCARCE 20139

Antiepileptic‐ 7 < RMSEmedian < 29‐Min & max encompass the 1:1 line‐Well simulated for both low flow and high flow 

Antibiotic‐ 2 < RMSEmedian < 3‐Min & max encompass the 1:1 line‐Well simulated for both low flow and high flow 

Sim. C

onc. (n

g L‐1)

Sim. C

onc. (n

g L‐1)

Page 10: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Results

Boithias et al. SCARCE 201310

Lipid regulator‐ 126 < RMSEmedian < 507‐Min & max encompass the 1:1 line‐Well simulated for both low flow and high flow 

Sim. C

onc. (n

g L‐1)

Page 11: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Results

Boithias et al. SCARCE 201311

Sim. C

onc. (n

g L‐1)

Sim. C

onc. (n

g L‐1)

Antihistaminic‐ 4 < RMSEmedian < 5‐Min & max encompass the 1:1 line‐ Badly simulated for low flow

Beta‐blocker‐ 49 < RMSEmedian < 84‐Min & max encompass the 1:1 line‐ Badly simulated for low flow

Page 12: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Results

Boithias et al. SCARCE 201312

Beta‐blocker‐ 1 < RMSEmedian < 217‐ Overestimated during high flow

Lipid regulator‐ 5 < RMSEmedian < 18‐ Overestimated during low flow

Page 13: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Results

Boithias et al. SCARCE 201313

Sim. C

onc. (n

g L‐1)

Analgesic‐ 2 < RMSEmedian < 6‐ Underestimated during both low flow and high flow

Analgesic‐ 43 < RMSEmedian < 62‐ Underestimated during high flow and ovestimated during low flow

Page 14: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Discussion

• For some molecules, the simplest model is enough to describe the fate, when using the median of the available data– Non‐steroidal anti‐inflammatory drugs were well simulated– A simple statistical analysis showed that the RMSE was lower for

lower Henry low constants ‐> more volatilizable molecules

• Some molecules are badly simulated during low flow: REWWTPand RERivers may increase during low flow depending on the pharmaceutical concentration ‐> this was not simulated:

– future work: simulate 27 Med/Min/max combinations to check thisassumption

• Other uncertainty source : emissions may change spatially, but not depending on the hydrological conditions

Boithias et al. SCARCE 201314

Page 15: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Conclusions and future work

• Well simulated molecules could already be used to assess the water purification service, i.e. the contaminants removal, at basin scale

• Next step: run the 27 scenarios of Min/Max/Median combinations, to get more details about the effect of the hydrological conditions on the removal efficiencies

• Ongoing work: automatic sensitivity analysis and an automatic calibration (GREAT‐ER more complex tiers)

Boithias et al. SCARCE 201315

Page 16: of the water purification ecosystem service regarding in ... SCARCE International... · Waste water treatment plants ... Catchment description Pharmaceuticals ... – Medical use

Contact: [email protected]

Thank you !

16Boithias et al. SCARCE 2013