Procedia Environmental Science (2016)
Atmospheric impacts of Indonesian fire emissions:
Assessing remote sensing data & air quality during 2013 Malaysian haze
De Lin Show-, Su-Chin Chang-
Department of Earth Sciences, The University of Hong Kong, Pokfulam Road, Hong Kong.
Results (cont’d)
Introduction
2013 Malaysian haze severely diminished nationwide horizontal visibility as thick smoke blanketed Peninsular
Malaysia in June 2013. On 23 June 2013, a state of emergency was declared in two southern districts on Peninsular
Malaysia as the air pollution worsened to extremely hazardous levels unprecedented in Malaysian history (Nature
2013).
Methodology
The time when PM10 concentration peaked at Petaling Jaya matches well with the time when southwesterlies prevailed
at Malaysia. Southwesterlies brought relatively less rainfall from the Straits of Melaka to the west coast of Peninsular
Malaysia. In addition, precipitation processes were inhibited by a high pressure system over the Central Sumatra
basin. Less rainfall in Peninsular Malaysia resulted in surging PM10 concentration at Petaling Jaya, as observed in
June 2013 (Fig 2).
Air quality data
Satellite remote sensing
Wind vector maps
Assessment of air quality at Petaling
Jaya was performed based on
concentration of PM10, particulate
matter that is 10 micrometers or less
in diameter. Air quality data were
collected daily over a sampling
period of 1 April to 31 August 2013.
Daily PM10 concentration were
recorded using β-ray attenuation
mass monitor (BAM-1020). Since
southern districts on Peninsular
Malaysia were most affected by the
2013 Malaysian haze (Juneng et al.
2009), our study area has focused on
Petaling Jaya (3° 06' N, 101° 39' E)
on Fig 1a.
In order to locate and detect active
fires occurring in Sumatra, satellite
remote sensing was employed to
retrieve active fire data over the
Central Sumatra basin. Active fire
data were retrieved using Moderate
Resolution Imaging
Spectroradiometer (MODIS) from
Terra and Aqua satellite. A polygon
area was delineated to include Riau
(0° 32′ N, 101° 27′ E) and Jambi
(1° 35′ S, 103° 36′ E) in Sumatra.
Frequency of hotspots occurring in
the designated polygon area (Fig
1b, 1c) was plotted into hotspot
counts.
For regional wind vector analysis
over Malaysia and Indonesia,
daily mean composites were
computed at an upper air analysis
level of 925mb over the Indian
Pacific region. Daily composites
from 1 April 2013 to 31 August
2013 were averaged and
constructed into monthly wind
vector maps. A Cylindrical
Equidistant projection was used
to construct a mapping domain of
latitude 20 °N to 20 °S and
longitude 60 °E to 180 °W using
NCEP/NCAR Reanalysis Product
(Kalnay et al. 1996).
1a
Riau
b
Petaling Jaya
Fig 1a. Geographical
locations of Petaling
Jaya, Riau, Jambi.
Fig 1b. MODIS
polygon area over the
Central Sumatra
basin.
Fig 1c. Distribution
of hotspots occurring
at Riau and Jambi.
Jambi
4
Monthly wind vector maps in the Indian Pacific region
1-30 April 2013
1-31 May 2013
c
Jambi
Riau
On a regional scale, wind vector maps (Fig 4) show a weakening of southwesterlies over the Indian Pacific region
from April to May 2013. This marks the arrival of inter-monsoon period at Peninsular Malaysia in late April and early
May (Juneng et al. 2011; Azmi et al. 2010). In June 2013, southwesterlies from the Indian Ocean began to strengthen
and gain intensity. This observation corresponded to the intensification of Intertropical Convergence Zone (ITCZ)
over South China Sea, a low pressure belt between latitude 10 ° N and 15 ° N (Oozeer et al. 2016).
1-30 June- June 2013
1-31 July 2013
Results
This study presents new evidence for attributing PM10 variations at Petaling Jaya to foreign emission sources in
Sumatra. On 23 June 2013, PM10 concentration at Petaling Jaya reached a record high of 290 µg/m3 (Fig 2). The
prominent PM10 peak in June was likely due to stronger transport energy in the atmosphere during summer monsoon
season. PM10 variations at Petaling Jaya were influenced by the direction and strength of wind transport, in addition to
the proximity of emission sources to the target area (Juneng et al. 2009).
2
1-31 August 2013
PM10 concentration at Petaling Jaya
Max PM10 concentration
on 23 June 2013
3
Previous research suggested that El Nino modulations could weaken aerosol dispersion and exacerbate haze pollution
during summer monsoon (Hyer et al. 2012; Parameswaran et al. 2004). However, our study finds that El Nino
influence is negligible for the Petaling Jaya case in 2013 Malaysian haze. Here we suggest two controls that are more
influential on Petaling Jaya PM10 concentration:
1. Foreign emission sources in Sumatra
2. A change of monsoonal wind direction over the Indian Pacific region
Hotspot counts in Sumatra
Max hotspot counts
on 19 June 2013
Conclusions
Air quality assessment has enabled a detailed analysis of PM10 temporal variations at Petaling Jaya, whereas remote
sensing data have been useful for plotting hotspot counts in Sumatra. Hotspot counts were highest on 19 June 2013,
diminishing subsequently on 21 June 2013 and 23 June 2013. PM10 concentration reached a record high of 290 µg/m3
on 23 June 2013. Wind vector maps showed how airborne pollutants were transported from Sumatra, Indonesia to
Peninsular Malaysia during the summer monsoon season in June 2013. Both the strong monsoonal flow of
southwesterlies and high pressure system over Sumatra led to reduced rainfall at Petaling Jaya. Consequently, PM10
concentration at Petaling Jaya soared markedly in response to dryer weather conditions in June 2013.
Selected References
Azmi, S., et al. 2010. Trend and status of air quality at three different monitoring stations in the Klang Valley, Malaysia. Air Qual. Atmos. Health. 3(1): 53-64.
Hyer, E. et al. 2012. Patterns of fire activity over Indonesia and Malaysia from polar and geostationary satellite observations. Atmos. Res. 122, 504–519.
‘Indonesian fires choke nearby countries’, Nature 498:- June 2013. doi:10.1038/498412a
4-day time lag
Juneng et al. 2011. Factors influencing the variations of PM10 aerosol dust in Klang Valley, Malaysia during the summer. Atmos. Environ. 45,-.
Juneng, L., et al. 2009. Spatio-temporal characteristics of PM10 concentrations across Malaysia. Atmos. Environ. 43,-.
PM10 emission sources were attributed to Riau and Jambi in Sumatra using hotspot counts. Hotspot counts peaked
abruptly on 19 June 2013 (Fig 3). As many as 676 fire activities were recorded in Sumatra on 19 June. Second and
third highest hotspot counts were recorded on 21 June and 23 June 2013 respectively. A 4-day time lag was noted from
the day with maximum hotspot counts (19 June) to the day with maximum PM10 peak (23 June). A possible
explanation for the 4-day time lag: airborne particulate matter was lifted from Sumatra on 19 June, suspended in the
troposphere and transported for 4 days before reaching Petaling Jaya on 23 June (Kanniah et al. 2014).
Kalnay et al. 1996. The NCEP/NCAR 40-year reanalysis project. Bull. Amer. Meteor. Soc. 77, 437-470.
Kanniah, K., et al. 2014. Investigating aerosol properties in Peninsular Malaysia via the synergy of satellite remote sensing and ground-based measurements.
Atmos. Res. 138, 223–239.
Oozeer, M., et al. 2016 (In press). Numerical study of the transport and convective mechanisms of biomass burning haze in South-Southeast Asia. Aerosol
Air Qual. Res. doi: 10.4209/aaqr-
Parameswaran, K., et al. 2004. Impact of Indonesia forest fires during the 1997 El Nino on the aerosol distribution over the Indian Ocean. Adv. Space Res.
33,-.
Acknowledgement
Air quality data provided by the Malaysian Meteorological Department
MODIS Active Fire Data provided by NASA/LANCE – FIRMS
NCEP/NCAR Reanalysis Product developed by NOAA – ESRL