Analyses Of Aerosol Optical Properties And Development Of Its Algorithm During Seasonal Monsoon Circulation
Loading...
Date
2015-11
Authors
Tan, Fuyi
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
The impact of biomass burning and pollution on Southeast Asia (SEA) is of
considerable concern to global climate change researchers. Geographical location,
topography and meteorological factors affecting SEA all contribute to the complexity
of the aerosols system worldwide. However, frequent cloud cover in the region
results in missing data during observations by remote sensing techniques. Therefore
obtaining continuous aerosol optical depth (AOD) measurements is a difficult task.
As a way to address such issues, this study first investigates the inter-relationship
among fire activity, ground smoke emission, rainfall events, typhoon activity and
outgoing longwave radiation in both the northern and southern regions of SEA
during seasonal monsoon between 2011 and 2013. The analysis of the correlation
among these parameters in northern and southern SEA shows that southern SEA is
frequently covered by cloud. In this study an empirical algorithm is developed to
compensate for the eliminated data due to frequent cloud coverage in a selected
study area in southern SEA, namely Penang. Ground-based measurements such as
visibility and air pollutant index are used in the model as predictor data to retrieve
the missing AOD data from Aerosol Robotic Network (AERONET). The empirical
model coefficients are determined through multiple non-linear regression analysis;
the calibrated model coefficients have a coefficient of determination of R2 = 0.72.
The predicted AOD of the model is generated on the basis of these calibrated
coefficients and is compared with data measured through standard statistical tests,
yielding R2 = 0.68 as validation accuracy. The error in weighted mean absolute
percentage error is determined to be less than 0.50% than that of the actual data. The
results reveal that the proposed model efficiently predicts the AOD data.
Additionally, the results of the model are compared with other independent source,
i.e., light detection and ranging system (LIDAR) data to yield good correspondence,
with R2 = 0.86. The optical properties of aerosols in Penang, Malaysia, are analyzed
for four monsoonal seasons including northeast monsoon, pre-monsoon, southwest
monsoon, and post-monsoon based on data from the AERONET recorded from
February 2012 to November 2013. The aerosol types in Penang for each monsoonal
period are quantitatively identified according to the scattering plots of the Ångström
exponent against the AOD. Both aerosol optical properties and aerosol types are used
to examine the proposed AOD-predicting model. The established seasonal AOD
prediction models are observed to have a clear relationship in the dominant aerosol
type with different optical properties and back-trajectories patterns. The proposed
model is also applicable for predicting the AOD at each studied wavelength within
340 nm and 1,020 nm. As an illustration to apply the model, the aerosol size
determined using measured AOD data for Penang was compared with that from the
model. This was done by examining the curvature in the ln [AOD] versus ln
[wavelength] plots using both measured and predicted AOD data in 2012 and 2013.
Based on the consistency in the curvature of the log-log plots, it was concluded that
Penang was dominated by fine mode aerosol in 2012 and 2013. These results
indicate that the proposed AOD prediction model using routine measurements as
input is a promising tool for regular monitoring of aerosol variation during nonretrieval
times. The AOD prediction model can serve as an alternative tool for
measuring short- and long-term AOD and can provide supplementary information for
climatological studies and monitoring of aerosol variation.
Description
Keywords
Aerosol