MODELING THE CONTRIBUTION OF WILDFIRE EMISSIONSTO AIR POLLUTION IN KATHMANDU VALLEY

KUNDAN LAL SHRESTHA KATHMANDU UNIVERSITY KUNDAN@KU.EDU.NP

PROBLEMThe mid-western part of Nepal is at

high risk from forest fires. The fire counts(see figure below) show, on average, 2000occurrence of fire events every year in Nepal[1].

WRF-Chem modeling system can be usedto assess the impact of such wildfires on theair quality and health of people [2]. The fireemissions are available due to the use ofremote sensing and GIS [3]. But this impactassessment is difficult due to three aspectsin Nepal:

1. Complex terrain of Nepal2. Availability of emission data3. Ground-based monitoring data

Fire

co

un

ts in

Ne

pal

0

1000

2000

3000

4000

Year1998 2000 2002 2004 2006 2008 2010 2012 2014

RESULTS

Contribution of wildfire to:AOD (550 nm) CO PM10 PM2.5

27°N

28°N

84°E 85°E 86°E 0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

17.5

Cont

ribut

ion

of w

ildfir

e to

AOD

(%)

27°N

28°N

84°E 85°E 86°E 0

3

6

9

12

15

18

Cont

ribut

ion

of w

ildfir

e to

CO

(%)

27°N

28°N

84°E 85°E 86°E 0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

17.5

Cont

ribut

ion

of w

ildfir

e to

PM

10 (%

)

27°N

28°N

84°E 85°E 86°E 0

3

6

9

12

15

18

21

24

Cont

ribut

ion

of w

ildfir

e to

PM

2.5 (

%)

AOD (550 nm) from WRF-Chem Fire Simulation (KTM Domain)

27°N

28°N

84°E 85°E 86°E

WRF-Chem: 2015 - 03 - 12

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

27°N

28°N

84°E 85°E 86°E

WRF-Chem: 2015 - 03 - 21

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

27°N

28°N

84°E 85°E 86°E

WRF-Chem: 2015 - 03 - 22

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

27°N

28°N

84°E 85°E 86°E

WRF-Chem: 2015 - 03 - 23

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

AOD (550 nm) from Daily Level 2 of MODIS

27°N

28°N

84°E 85°E 86°E

12/03/2015

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

27°N

28°N

84°E 85°E 86°E

21/03/2015

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

27°N

28°N

84°E 85°E 86°E

22/03/2015

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

27°N

28°N

84°E 85°E 86°E

23/03/2015

0.0

0.2

0.4

0.6

0.8

1.0

1.2

AOD

March 12, 2015 March 21, 2015 March 22, 2015 March 23, 2015The three polygons in the center are the districts of the Kathmandu Valley (Kathmandu,Lalitpur and Bhaktapur).

METHOD

WRF simulations comprised of twoseparate domains with resolution 48 km(ASIA Domain) and 6 km (KTM Domain).

Boundary conditions: NCEP Final analysis(FNL); Microphysics: Purdue Lin scheme;Cumulus: New Kain-Fritsch; Longwaveradiation: RRTM; Shortwave radiation:Dudhia shortwave scheme; Surface: MYJ;Planetary Boundary Layer: YSU.

Anthropogenic emissions: EDGAR HTAPv2; Chemical initial and boundary condi-tions: MOZART-4; The biogenic emissions:MEGAN; Fire emissions: FINN v 1.5. Inchemistry parametrization, MOZART chem-istry with GOCART aerosol was used.

The MODIS Terra and MERRA-2 AOD:from NASA LAADS Web.

VALIDATION

AO

D (

550

nm)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Day of March, 201508 10 12 14 16 18 20 22 24

MERRAWRF-Chem

AO

D (

550

nm

)

0

0.2

0.4

0.6

0.8

1

1.2

Day of March, 20158 10 12 14 16 18 20 22 24

MODIS

ASIA-FIREASIA-NOFIRE

KTM-FIREKTM-NOFIRE

CONTRIBUTIONSTill now, the distribution of wildfire in

Nepal has been studied but the effect of thewildfire emissions on the air quality and hu-man health has not been quantified. This re-search has tried to quantify the contributionof wildfire emissions to the air quality in theKathmandu Valley. A high-resolution mod-eling system has been used to simulate thedifferent scenarios of wildfire emissions. Themain contributions are:

1. WRF-Chem modeling system at high-resolution that can be used for overallassessment of air quality.

2. Quantification of the effect of wildfireemissions on air quality.

3. Validation of WRF-Chem with datafrom remote sensing.

4. Importance of local wildfire emissions.

REFERENCES[1] Parajuli, A. et al. Spatial and temporal distribu-

tion of forest fires in Nepal. In XIV World ForestryCongress (Durban, South Africa, 2015).

[2] Crippa, P. et al. Population exposure to hazardousair quality due to the 2015 fires in equatorial Asia.Scientific Reports 6, 37074 (2016).

[3] Matin, M. A. et al. Understanding forest fire patternsand risk in Nepal using remote sensing, geographicinformation system and historical fire data. Interna-tional Journal of Wildland Fire 26, 276–286 (2017).

FUTURE WORK1. Improving the emission data.2. Increasing the model resolution.3. Validation of WRF-Chem with pollu-

tion data of Kathmandu Valley.4. Long-term simulations with nested do-

mains.