improved assessment of personal exposure to air … · equivalent pm10 average daily concentration...

2nd European Exposure Science Strategy Workshop 1 / 15 July 4-5, 2019 Bilthoven, The Netherlands

Environmental

Engineering

Laboratory

Dep. of Chemical Engineering

School of Engineering

Aristotle Univ. of Thessaloniki

IMPROVED ASSESSMENT OF PERSONAL

EXPOSURE TO AIR POLLUTANTS USING AGENT

BASED MODELLING (ABM) COUPLED WITH MULTI-

SENSORS NETWORKS

D. Chapizanis1, D.A. Sarigiannis1,2,3, S. Karakitsios1,2

1Aristotle University of Thessaloniki, Department of Chemical Engineering, Environmental Engineering Laboratory,

University Campus, Thessaloniki 54124, Greece

2HERACLES Research Center on the Exposome and Health, Center for Interdisciplinary Research and Innovation,

Balkan Center, Bldg. B, 10th km Thessaloniki-Thermi Road, 57001, Greece

3School for Advanced Study (IUSS), Science, Technology and Society Department, Environmental Health

Engineering, Piazza della Vittoria 15, Pavia 27100, Italy

http://www.enve-lab.eu


Environmental

Engineering

Laboratory




• Use sensors for tracking personal location and activities in order to assess

exposure at a personal level.

• Hurdles involved in collecting real data for large populations.

• Development of an individual exposure model based on Agent Based

Modelling (ABM).

Introduction


Environmental

Engineering

Laboratory




Time

Data Analysis & Interpretation Multi-sensor Campaign

Time – Activity Diary

& Questionnaire

PM,

Location,

Temp,

Humidity

Steps,

Activities,

Heart Rate,

Organic

chemicals

Exposure

modelling

Location

Activities

Pollutants’

Concentration

Statistical

Models

CO, CO2, VOCs, PM2.5, O3, NO2,

Temp, Humidity, Air pressure.

Methodology

2019 wearable sensors campaign

Personal Exposure

9 EU cities

100 participants/city

Duration: 7 days

Summer & Winter period

Online Platform

Smart Data Collection


Environmental

Engineering

Laboratory




Methodology

PM and PAHs sampling


Environmental

Engineering

Laboratory




Methodology

PAHs analysis

Extraction of filters with organic solvents

GC-MS analysis

Determination of concentrations using

internal standards


Environmental

Engineering

Laboratory




Place Settings Duration

(h)

LR NLR MLR MLR_RV ANN

PM1 PM2.5 PM10 PM1 PM2.5 PM10 PM1 PM2.5 PM10 PM1 PM2.5 PM10 PM1 PM2.5 PM10

Lab Laboratory Conditions 24 0.79 0.75 0.71 0.84 0.82 0.81 0.83 0.81 0.80 0.86 0.85 0.84 0.87 0.85 0.85

Closed Windows / AC Off 17 0.84 0.81 0.70 0.89 0.86 0.86 0.85 0.83 0.80 0.95 0.94 0.95 0.96 0.96 0.96

Closed Windows / AC On 24 0.85 0.83 0.79 0.92 0.92 0.90 0.89 0.89 0.88 0.96 0.96 0.95 0.97 0.97 0.97

Open Windows / AC Off 24 0.86 0.84 0.67 0.90 0.90 0.86 0.88 0.83 0.67 0.96 0.95 0.91 0.97 0.96 0.94

Closed Windows - Battery Low 6 0.52 0.53 0.45 0.58 0.60 0.56 0.63 0.75 0.77 0.78 0.87 0.90 0.84 0.91 0.91

Mixed - Fri/Sat/Sun 72 0.85 0.86 0.77 0.91 0.89 0.83 0.88 0.86 0.78 0.95 0.93 0.87 0.97 0.96 0.90

Office Closed Windows / AC On 24 0.85 0.84 0.47 0.91 0.88 0.64 0.92 0.88 0.62 0.97 0.96 0.80 0.97 0.97 0.84

Open Windows / AC On 10 0.70 0.56 0.19 0.74 0.62 0.20 0.67 0.60 0.37 0.84 0.77 0.68 0.89 0.79 0.65

Open Windows / AC Off 11 0.81 0.78 0.72 0.85 0.83 0.81 0.84 0.83 0.80 0.86 0.84 0.82 0.86 0.86 0.85

Living Open Windows / Fan Off 18 0.86 0.84 0.71 0.90 0.89 0.80 0.90 0.89 0.78 0.97 0.95 0.87 0.97 0.96 0.89

Room Open Windows / Fan On 12 0.88 0.86 0.70 0.92 0.90 0.80 0.92 0.91 0.88 0.97 0.96 0.96 0.96 0.96 0.97

Closed Windows / Fan On 12 0.87 0.83 0.77 0.91 0.88 0.85 0.91 0.89 0.82 0.96 0.96 0.92 0.96 0.97 0.94

Closed Windows / Fan Off 12 0.81 0.68 0.37 0.85 0.77 0.49 0.85 0.73 0.46 0.93 0.82 0.56 0.94 0.86 0.68

Methodology

AQ sensors data validation

• Test device under diverse environmental conditions prior deployment

• Include additional environmental parameters that influence concentration

• Properly interpret the physicochemical phenomena and use the right statistical tools

*work in progress – testing period: 15 days (Aug – Sept 2018), results from 5 sensors


Environmental

Engineering

Laboratory




Methodology

PM and PAHs deposition

• Μultiple Path Particle Deposition Model: Tidal Volume, Functional

Residual Capacity, URT volume and Breathing Frequency as inputs

Major:

• Diffusion

• Sedimentation

• Inertial Impaction

Minor:

• Interception

• Electrostatic

Aerosol property

• Size distribution

• Concentration

• Particle hygroscopicity

• Gas-particle interaction

• Chemical reaction

• Particle surface

Air Flow property

• Lung capacity

• Breathing frequency

Respiratory tract

• Lung structure and morphology


Environmental

Engineering

Laboratory




• A system is modelled as a collection of agents and the relationships

between them.

• The agents, programmed as autonomous decision-makers, react and act in

their environment.

• Take into account the heterogeneity of the entities

composing the system.

• Behaviours emerge through the agents interactions. building agents

human agents

Methodology

Agent Based Modelling (ABM)

road agents

spatially explicit ABM platform


Environmental

Engineering

Laboratory




Buildings Network

Roads Network

Methodology

Buildings & Road Network

road agents

building agents

human agents


Environmental

Engineering

Laboratory




Gender

Age

movement

Time-Use

Surveys

Campaigns

findings

Literature

associations

----- sources ----- -------------- attributes -------------- ------------- behavioural rules ------------

IF

THEN

Pactivity = f(age, gender, income, education, Po )

activities

Interaction

between

human agents

Socio-Economic Status

o Level of education

o Occupational status

o Marital status

o Income

Methodology

Human agents spatiotemporal behaviour


Environmental

Engineering

Laboratory




ABM simulation preview

Sleeping

Indoor activities

In transit

Working

Relaxing

Sports

Activities

Road Network

Buildings,

Land Use

Time Activity Place Vehicle

0:00 sleep home

8:10 commute road car

8:20 paidwork work

11:20 commute road car

12:00 eatdrink home

13:00 tvradio home

18:30 selfcare home

19:10 clean home

20:10 selfcare home

20:50 tvradio home

Human Agent

Trajectories

Air Quality

Data

Methodology

Personal exposure modelling Human agent X – Exposure to PAHs (ng/m3)

Human agent X –

Daily intake of PAHs (ng/kg-day)


Environmental

Engineering

Laboratory




Results

Ambient air PAHs

0

20

40

60

80

100

120

140

160

PM1 - ΣPAHs

PM1 -TEQ

PM2.5 - ΣPAHs

PM2.5 -TEQ

PM10 - ΣPAHs

PM10 -TEQ

PM1 - ΣPAHs

PM1 -TEQ

PM2.5 - ΣPAHs

PM2.5 -TEQ

PM10 - ΣPAHs

PM10 -TEQ

Urban background Traffic

ΣP

AH

s -

TE

Q (

ng

/m3 )

ΣPAHs - TEQ

5%-95%

Median

Mean


Environmental

Engineering

Laboratory




31,5 28,2 28,8

25,4

0,0

10,0

20,0

30,0

40,0

Equivalent PM10 average daily concentration (ug/m3)

Outdoor Concentration

Indoor Concentration

Exposure

Inhalation Adj. Exposure

Civstatus

1

Education

2

Income

2

Employment

1

Age

23

Gender

M

Region

centre

M_House

D_Thessalonikis

M_Work/School

D_PavlouMela

0

10

20

30

40

50

60

70

80

90

PA

Hs (

ng

/m3)

Indoors (home/store/bar) Indoors (work/school) Outdoors (park/forest) In transit

00:00 | leisure

08:30 | paidwork

15:20 | commute (car)

15:30 | selfcare


18:20 | foodprep

21:40 | shopserv


00:20 | sleep

16:30 | forest walk


19:10 | tvradio

20:20 | maintain


23:40 | eatdrink Gender: Female

Age: 27 years old

Civil status: in a relationship

Employment: full time job

Income: medium

Education: secondary

Residential place: Centre

Work place: North

14,9

13,3 13,6

12,0

10,0

15,0

20,0

PAHs Daily intake dose per body weight (ng/kg-day)



Exposure


Female, 27 years, medium income, secondary education, full-time job, in a relationship

Results

Human agent exposure profile – PAHs



Environmental

Engineering

Laboratory




0

10

20

30

40

50

60

70

80

90

PA

Hs (

ug

/m3)

Indoors (home/store/bar) Indoors (work/school) Outdoors (park/forest) In transit

Civstatus

2

Education

3

Income

2

Employment

1

Age

35

Gender

M

Region

centre

M_House

D_Thessalonikis

M_Work/School

D_Kordeliou_Evosmou

00:00 | eat-drink

09:30 | paidwork

18:20 | selfcare

19:20 | leisure

20:20 | tv-radio

22:20 | compint

08:40 | commute (motorbike)

01:40 | sleep

17:30 | commute (motorbike)

22:50 | eatdrink

23:20 | read

Gender: Male

Age: 37 years old

Civil status: single

Employment: full time job

Income: medium

Education: higher

Residential place: Centre

Work place: West

14,9

13,6 13,9

17,1

10,0

15,0

20,0

PAHs Daily intake dose per body weight (ng/kg-day)



Exposure


31,5 28,8 29,4

36,1

0,0

10,0

20,0

30,0

40,0

Equivalent PM10 average daily concentration (ug/m3)



Exposure


Results

Human agent exposure profile – PAHs

Male, 37 years, medium income, higher education, full-time job, single


Environmental

Engineering

Laboratory




5,0

10,0

15,0

20,0

25,0

To

tal

Lo

w E

du

c. le

ve

l

Hig

h E

du

c. le

ve

l

Lo

w In

co

me

Hig

h In

co

me

Ma

le

Fe

ma

le

Ce

ntr

e

Ea

st

West

Outdoors

Da

ily i

nta

ke

(n

g/k

g-d

ay)

Intake Dose Variation (ABM)

Results

Human agent population subgroups

• Exposure can differ by 32% between housemates and up to 77% for neighbours

• Exposure was found to vary based on SES conditions


Environmental

Engineering

Laboratory




Results

Human agent population subgroups


Environmental

Engineering

Laboratory




Results

campaign exposure vs. ABM exposure data

sensors

campaign

ABM


Environmental

Engineering

Laboratory




• cost-effective construction of refined time-activity diaries & exposure profiles

• estimations based onto emerging properties of the agent system behaviour

• ΑΒΜ model especially useful for vulnerable groups of population

• evaluation of the probable impacts of different public health policies, prior to

implementation

• Combined data of sensors, chemical analysis and advanced exposure

modelling are able to further elucidate differences among individuals and

population groups

Conclusions


Environmental

Engineering

Laboratory



Aristotle Univ. of Thessaloniki Acknowledgements

Funding

HBM4EU project

ICARUS project

The HBM4EU project has received funding from the Horizon 2020

Research and Innovation Programme under grant agreement No 733032

The ICARUS project has received funding from the European Union’s

H2020 Framework Programme under grant agreement No - 690105


Environmental

Engineering

Laboratory




Agent Based

improved exposure assessment

Demographic & socioeconomic determinants

Personal exposure Artificial societies

Heterogeneity

vulnerable population

integration of Modelling

from the bottom up

emergent behaviour

Thank you!

Thessaloniki, Greece

improved assessment of personal exposure to air … · equivalent pm10 average daily concentration...

Documents