- Email: [email protected]
Comparison of MODIS aerosol optical depth retrievals with ground-based measurements in the tropics
Available online at www.sciencedirect.com Available online at www.sciencedirect.com
Procedia Engineering
ProcediaProcedia Engineering 00 (2012) 000–000 Engineering 32 (2012) 392 – 398 www.elsevier.com/locate/procedia
I-SEEC2011
Comparison of MODIS aerosol optical depth retrievals with ground-based measurements in the tropics T. Jantarach∗, I. Masiri, S. Janjai Laboratory of Tropical Atmospheric Physics, Department of Physics, Faculty of Science, Silpakorn University, Nakhon Pathom, 73000, Thailand Elsevier use only: Received 30 September 2011; Revised 10 November 2011; Accepted 25 November 2011.
Abstract Aerosols are small particles suspended in the atmosphere. They may have profound effects on human health. Their effects on the physical environment can also be of importance as they have the ability to both scatter and absorb incident solar radiation. The ability of aerosols to deplete solar radiation can be quantified in terms of aerosol optical depth (AOD). AOD can be retrieved from the MODIS satellite. However, AOD from this satellite has a wide range of uncertainty, depending on environments and climatic zones. In this work, AOD retrieved from MODIS was compared to that obtained from ground-based measurements at four sites in the tropical environment of Thailand. These are Chiang Mai (18.78 ºN, 98.98 ºE), Ubon Ratchathani (15.25 ºN, 104.87 ºE), Nakhon Pathom (13.82 ºN, 100.04 ºE) and Songkhla (7.2 ºN, 100.60 ºE). AOD at these sites was measured by using Cimel sunphotometers. The AOD data from these sites over a period of 2-5 years was used in the comparison. It was found that the discrepancy in terms of root mean squared difference between the daily AOD retrieved from MODIS and that of the ground-based measurements was in the range of 33.8%-53.7%.
© 2010 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of I-SEEC2011 Keywords: aerosol optical depth; MODIS; ground-base measurement; tropical envirinment; Thailand
1. Introduction Aerosols are small particles suspended in the atmosphere that may have profound effects on health [14]. Added to this their effects on the physical environment may also be of importance, as they have the ability to both scatter and absorb incident solar radiation [5] and also modify cloud properties [6-10]. The aerosol optical depth (AOD), which is an indicator of aerosol loading in the vertical column of the
* Corresponding author. Tel.: +66-34-270761; fax: +66-34-271189. E-mail address: [email protected].
1877-7058 © 2012 Published by Elsevier Ltd. doi:10.1016/j.proeng.2012.01.1284
2
T.T.Jantarach Engineering 3200 (2012) – 398 Jantarachetetal.al./ Procedia / Procedia Engineering (2012)392 000–000
atmosphere, is the main parameter by which the ability of aerosols to deplete solar radiation is measured. One of the more critical problems is the lack of information on aerosol optical properties in these regions, which is still very limited and not sufficient to draw conclusions about climatic effects. The aerosol optical properties can be measured directly using ground-based measurements or derived from remote sensing observations. An estimation of aerosol optical properties from ground-based observations often involves measurements of direct and diffuse solar radiation in distinct narrow spectral bands [11]. In terms of satellite data, the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the NASA Earth Observation System’s Terra satellite is the major instrument designed to provide aerosol optical properties. Comparisons between the AOD from MODIS and ground-based measurements were carried out by a number of studies [12-13]. It can be concluded that the discrepancies of MODIS-derived AOD are within an acceptable range of ±0.2AOD [14]. The purpose of this paper is to compare AOD data retrieved from MODIS to that obtained from ground-based measurements at four sites in the tropical environment of Thailand. The paper starts with a brief description of the data including the ground-based measurements and the satellite data in the second section. The third section presents the results of the comparison between the MODIS derived aerosol optical depth values with the ones obtained from the ground-based measurements. The conclusions are presented in the last section. 2. The data 2.1 Ground-based measurements For this paper we procured sunphotometers and then installed them at our existing solar radiation monitoring stations in four main regions of Thailand. These stations are situated at Chiang Mai (18.78 ºN, 98.98 ºE) in the Northern region, Ubon Ratchathani (15.25 ºN, 104.87 ºE) in the Northeastern region, Nakhon Pathom (13.82 ºN, 100.04 ºE) in the Central region and Songkhla (7.2 ºN, 100.60 ºE) in the Southern region (Fig. 1). The sunphotometers at Chiang Mai, Nakhon Pathom and Songkhla were fabricated by Cimel (model CE-318). In Ubon Ratchathani two sunphotometers were employed, the first sunphotometer was produced by Pread Co., Ltd. and the second was a Cimel sunphotometer. The Cimel sunphotometers take solar radiation measurements with the almucantar scan once per hour and at optical airmass of 4,3,2 and 1.7 in both the morning and afternoon. The instrument measures direct sun measurements at nominal wavelengths of 340, 380, 440, 500, 675, 870, 940 and 1020 nm and diffuse sky radiances at 440, 675, 870 and 1020 nm, with a bandwidth of 10 nm except 2 nm in the UV [15]. All Cimel sunphotometers were incorporated into the Aerosol Robotic Network (AERONET) of NASA. These sunphotometers were calibrated by AERONET every 1-1.5 years, resulting in AOD accuracy of ~0.01-0.02, with the higher error in the UV [16]. AERONET AOD data screened for clouds by the algorithm of Smirnov et al. [17] that relies on the greater temporal variance of cloud versus aerosol optical depths. It should also be mentioned that only Version 2, level 2 AOD data were utilized in this study. Table 1 and Fig. 1 provide more details, including locations and periods of measurements. The Pread sunphotometer was equipped with a sun tracker (Kipp&Zonen, model 2AP). This sunphotometer measured only the spectral direct solar radiation at the wavelengths 350-1050 nm with a bandwidth of 3.6 nm. It takes solar radiation measurements every 5 minutes. This sunphotometer was calibrated by the manufacturer before and after the measurement period.
393
394
T. Jantarach al. / Procedia Engineering 32 (2012) 392 – 398 T. Jantarach et al.et/ Procedia Engineering 00 (2012) 000–000
3
Northeast monsoon (November-February)
A C
B
D
Southwest monsoon (May-October) Fig. 1. Instruments and locations of the measurement sites. A, B, C and D indicate the main region of Thailand, namely the North, The Northeast, the Central and the South, respectively Table 1. Station details Station name
Latitude
Longitude
Instrument and period of data
Chiang Mai
18.78 ºN
98.98 ºN
Cimel sunphotometer
Ubon Ratchathani
15.25 ºN
104.87 ºN
Pread sunphotometer
(October 2006 – September 2009) (June 2008 – July 2009) and Cimel sunphotometer (October 2009 – December 2009) Nakhon Pathom
13.82 ºN
100.04 ºN
Cimel sunphotometer (September 2006 – December 2009)
Songkhla
7.2 ºN
100.60 ºN
Cimel sunphotometer (January 2007 – December 2009)
4
T.T.Jantarach Engineering 3200 (2012) – 398 Jantarachetetal.al./ Procedia / Procedia Engineering (2012)392 000–000
2.2 Satellite data Moderate Resolution Imaging Spectroradiometer (MODIS) is an instrument on the Terra and Aqua satellites. The Terra and Aqua are polar-orbiting satellites which orbit the Earth in morning descending and afternoon ascending directions, respectively. The MODIS instrument collects spectral irradiance data in 36 wavelength bands ranging from 0.4 to 14.4 μm (from visible to thermal infrared). Two spectral bands are imaged with the resolution of 250 m at nadir pixel dimensions, five spectral bands with the resolution of 500 m and 29 spectral bands with the resolution of 1000 m (http://modis.gsfc.nasa.gov/). In this study, we have used data from MODIS Terra AOD (MOD08_D3, in HDF format) for the period of 2006 to 2009. Which corresponds to the ground-based measurements. The data has been obtained in a daily format (spatial resolution 1ox1o) from the Giovanni website (http://giovanni.gsfc.nasa.gov/). The AOD data from MODIS were sectorized over the locations of Chiang Mai (18.78 ºN, 98.98 ºE), Ubon Ratchathani (15.25 ºN, 104.87 ºE), Nakhon Pathom (13.82 ºN, 100.04 ºE) and Songkhla (7.2 ºN, 100.60 ºE). 3. Comparison of MODIS data using ground-based measurements The AOD data from MODIS at four stations were compared with the measurements over the four main regions of Thailand, situated at Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla. This validation was accomplished by a comparison of the aforementioned data at a wavelength of 550 nm with the ground-based data, measured from the sunphotometers in the same regions, at a wavelength of 500 nm. Fig. 2 (a)-(d) presents the comparison of the AOD measured at Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla with the AOD retrieved from MODIS over these regions based on the daily values. Overall, the root mean squared difference (RMSD) for Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla are in the range of 33.8% to 53.7%. The comparison of the monthly AOD values was shown in Fig. 3. The RMSD for Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla are in the range of 26.4% to 43.6%. The highest RMSD values (53.7% for daily and 43.6% for monthly) were observed for Chiang Mai. This suggests that the aerosol model used for the inversion of the MODIS radiance measurements might be rather well adapted to the locally produced aerosols but not to the case of the biomass burning aerosols in Chiang Mai. In extreme conditions, the AOD values retrieved from MODIS are overestimated for clean atmosphere and underestimated for the turbid conditions from ground-based measurements [12]. This could be a result from a cease to agricultural clean-up activities which lead to a large decrease in biomass burning aerosols in Chiang Mai.
395
396
T. Jantarach al. / Procedia Engineering 32 (2012) 392 – 398 T. Jantarach et al.et/ Procedia Engineering 00 (2012) 000–000 1.5
Chiang Mai : 2006-2009
1.2
1.2
0.9
0.9
0.6
0.6
0.3
0.3
N= 266 RMSD= 53.7%
0.0
1.5
0.3
0.6 0.9 AODMEASUREMENT
1.2
1.5
0.0
(d)
Nakhon Pathom : 2006-2009
0.3
1.5
1.2
1.2
0.9
0.9
0.6 0.9 AODMEASUREMENT
1.2
1.5
Songkhla : 2007-2009
AODMODIS
AODMODIS
N= 100 RMSD= 40.3%
0.0
0.0
(c)
Ubon Ratchathani : 2008-2009
AODMODIS
AODMODIS
1.5
5
0.6
0.6
0.3
0.3
N= 488 RMSD= 33.8%
0.0 0.0
0.3
0.6
0.9
AODMEASUREMENT
1.2
1.5
N= 158 RMSD= 41.3%
0.0 0.0
0.3
0.6 0.9 AODMEASUREMENT
1.2
1.5
Fig. 2. Comparison of the scatter plots daily AOD between MODIS (AODMODIS) with that measurements (AODMEASUREMENT) by ground-based at four sites namely: (a) Chiang Mai; (b) Ubon Ratchathani; (c) Nakhon Pathom and (d) Songkhla.
397
T.T.Jantarach Engineering 3200 (2012) – 398 Jantarachetetal.al./ Procedia / Procedia Engineering (2012)392 000–000
6
(a)
(b)
Chiang Mai
1.0
0.8
0.8
0.6
0.6
AODMODIS
AODMODIS
1.0
0.4 0.2
N= 13 RMSD= 33.1%
0.0 0.0
0.2
1.0
0.4 0.6 AODMEASUREMENT
0.8
1.0
0.0
0.8
0.8
0.6
0.6
0.4 0.2
0.2
1.0
(d)
Nakhon Pathom
AODMODIS
AODMODIS
0.4 0.2
N= 19 RMSD= 43.6%
0.0
(c)
Ubon Ratchathani
0.8
1.0
Songkhla
0.4 0.2
N= 30 RMSD= 26.4%
0.4 0.6 AODMEASUREMENT
N= 20 RMSD= 27.4%
0.0
0.0 0.0
0.2
0.4 0.6 AODMEASUREMENT
0.8
1.0
0.0
0.2
0.4 0.6 AODMEASUREMENT
0.8
1.0
Fig. 3. Comparison of the monthly average AOD between MODIS MODIS (AODMODIS) with that measurements (AODMEASUREMENT) by ground-based at four sites namely: (a) Chiang Mai; (b) Ubon Ratchathani; (c) Nakhon Pathom and (d) Songkhla.
4. Conclusions This paper has compared the AOD data from MODIS with those from ground-based measurements at four different regions in Thailand. They are located in the north of the country (Chiang Mai), in the northeast (Ubon Ratchathani), the central part (Nakhon Pathom) and the south (Songkhla). MODIS AOD compares better with those from ground-based measurements at the Nakhon Pathom, Ubon Ratchathani and Songkhla, but poorer correlations are found at Chiang Mai station where they have large scale biomass burning. The root mean square difference (RMSD) between daily AOD retrieved from MODIS and that obtained from the ground-based measurements is in the range 33.8%-53.7%.
398
T. Jantarach / Procedia Engineering 32 (2012) 392 – 398 T. Jantarach et al.et/ al. Procedia Engineering 00 (2012) 000–000
7
Acknowledgements The authors would like to thank the Thailand Research Fund for providing financial support to this research work. The authors would also like to thank Dr. Brent Holben for incorporating our four sunphotometers into AERONET. Dr. David Giles is gratefully acknowledged for the arrangement of the data processing. References [1] Salma I, Balashazy I, Hofmann W, Zaray G. Effect of physical exertion on the deposition of urban aerosols in the human respiratory system. J Aero Sci 2002;33:983-97. [2] Hauck H, Berner A, Frischer T, Gomiscek B, Kundi M, Neuberger M et al. AUPHEP-Austrian project on health effects of particulates-general overview. Atmos Environ 2004; 38:3905-15. [3] Andersson KG, Roed J, Byrne MA, Hession H. Deposition of contaminant aerosol on human skin. J Environ Radio 2006;85:182-95. [4] Kennedy IM. The health effects of combustion-generated aerosols. Proceeding of the Combustion Institute 2007;31:2757-70. [5] Ramannathan V, Crutzen PJ, Lelieveld J, Mitra AP, Althausen D, Anderson J et al. Indian ocean experiment: An intergrated analysis of the climate forcing and effects of the great Ondo-Asian haze. J Geophys Res 2001;106:28371-98. [6] Flossmann AI. Interaction of aerosol particles and clouds. J Atmos Sci 1998;55:879-87. [7] O’Dowd CD, Lowe JA, Smith MH. The effect of clouds on aeroaol growth in the rural atmosphere. Atmos Res 2000;54:20121. [8] Cattani E, Costa MJ, Torricella F, Levizzani V, Silva AM. Influence of aerosol particles from biomass burning on cloud microphysical properties and radiative forcing. Atmos Res 2006;82:310-27. [9] Kelly JT, Chuang CC, Wexler AS. Influence of dust composition on cloud droplet formation. Atmos Environ 2007;41:290416. [10] Myhre G, Stordal F, Johnsrud M, Kaufman YJ, Rosenfeld D, Storelvmo T. Aerosol-cloud interaction inferred from MODIS satellite data and global aerosol models. Atmos Chem Phys 2007;7:3081-101. [11] Iqbal M. An introduction to solar radiation. Academic. New York; 1983. [12] He Q, Li C, Tang X, Li H, Geng F, Wu Y. Validation of MODIS derived aerosol optical depth over the Yangtze River Delta in China. Remote Sens Environ 2010;114:1649-61. [13] Bennouna YS, Cachorro VE, Toledano C, Berjon A, Prats N, Fuertes D et al. Comparison of atmospheric climatologies over southwestern Spain derived from AERONET and MODIS. Remote Sens Environ 2011;115:1272-84. [14] Chu DA, Kaufman YJ, Zibordi G, Chern JD, Mao JT, Li CC et al. Global monitoring of air pollution over land from the Earth Observing System-Terra Moderate Resolution Imaging Spectroradiometer (MODIS). J Geophys Res 2003;108:4661, doi:10.1029/2002JD003179. [15] Holben BN, Eck TF, Slutsker I, Tanre D, Buis JP, Setzer A et al. AERONET-a federated instrument network and data archive for aerosol characterization. Remote Sens Environ 1998;66:1-16. [16] Eck TF, Holben BN, Reid JS, Dubovik O, Smirnov A, O,Neill NT et al. Wavelength dependence of the optical depth of biomass burning, urban, and desert dust aerosols. J Geophys Res 1999;104(D24):31333-350. [17] Smirnov A, Holben BN, Eck TF, Dubovik O, Slusker I. Cloud-screening and quality control algorithms for AERONET database. Remote Sens Environ 2000;73:337-49.
Procedia Engineering
ProcediaProcedia Engineering 00 (2012) 000–000 Engineering 32 (2012) 392 – 398 www.elsevier.com/locate/procedia
I-SEEC2011
Comparison of MODIS aerosol optical depth retrievals with ground-based measurements in the tropics T. Jantarach∗, I. Masiri, S. Janjai Laboratory of Tropical Atmospheric Physics, Department of Physics, Faculty of Science, Silpakorn University, Nakhon Pathom, 73000, Thailand Elsevier use only: Received 30 September 2011; Revised 10 November 2011; Accepted 25 November 2011.
Abstract Aerosols are small particles suspended in the atmosphere. They may have profound effects on human health. Their effects on the physical environment can also be of importance as they have the ability to both scatter and absorb incident solar radiation. The ability of aerosols to deplete solar radiation can be quantified in terms of aerosol optical depth (AOD). AOD can be retrieved from the MODIS satellite. However, AOD from this satellite has a wide range of uncertainty, depending on environments and climatic zones. In this work, AOD retrieved from MODIS was compared to that obtained from ground-based measurements at four sites in the tropical environment of Thailand. These are Chiang Mai (18.78 ºN, 98.98 ºE), Ubon Ratchathani (15.25 ºN, 104.87 ºE), Nakhon Pathom (13.82 ºN, 100.04 ºE) and Songkhla (7.2 ºN, 100.60 ºE). AOD at these sites was measured by using Cimel sunphotometers. The AOD data from these sites over a period of 2-5 years was used in the comparison. It was found that the discrepancy in terms of root mean squared difference between the daily AOD retrieved from MODIS and that of the ground-based measurements was in the range of 33.8%-53.7%.
© 2010 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of I-SEEC2011 Keywords: aerosol optical depth; MODIS; ground-base measurement; tropical envirinment; Thailand
1. Introduction Aerosols are small particles suspended in the atmosphere that may have profound effects on health [14]. Added to this their effects on the physical environment may also be of importance, as they have the ability to both scatter and absorb incident solar radiation [5] and also modify cloud properties [6-10]. The aerosol optical depth (AOD), which is an indicator of aerosol loading in the vertical column of the
* Corresponding author. Tel.: +66-34-270761; fax: +66-34-271189. E-mail address: [email protected].
1877-7058 © 2012 Published by Elsevier Ltd. doi:10.1016/j.proeng.2012.01.1284
2
T.T.Jantarach Engineering 3200 (2012) – 398 Jantarachetetal.al./ Procedia / Procedia Engineering (2012)392 000–000
atmosphere, is the main parameter by which the ability of aerosols to deplete solar radiation is measured. One of the more critical problems is the lack of information on aerosol optical properties in these regions, which is still very limited and not sufficient to draw conclusions about climatic effects. The aerosol optical properties can be measured directly using ground-based measurements or derived from remote sensing observations. An estimation of aerosol optical properties from ground-based observations often involves measurements of direct and diffuse solar radiation in distinct narrow spectral bands [11]. In terms of satellite data, the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the NASA Earth Observation System’s Terra satellite is the major instrument designed to provide aerosol optical properties. Comparisons between the AOD from MODIS and ground-based measurements were carried out by a number of studies [12-13]. It can be concluded that the discrepancies of MODIS-derived AOD are within an acceptable range of ±0.2AOD [14]. The purpose of this paper is to compare AOD data retrieved from MODIS to that obtained from ground-based measurements at four sites in the tropical environment of Thailand. The paper starts with a brief description of the data including the ground-based measurements and the satellite data in the second section. The third section presents the results of the comparison between the MODIS derived aerosol optical depth values with the ones obtained from the ground-based measurements. The conclusions are presented in the last section. 2. The data 2.1 Ground-based measurements For this paper we procured sunphotometers and then installed them at our existing solar radiation monitoring stations in four main regions of Thailand. These stations are situated at Chiang Mai (18.78 ºN, 98.98 ºE) in the Northern region, Ubon Ratchathani (15.25 ºN, 104.87 ºE) in the Northeastern region, Nakhon Pathom (13.82 ºN, 100.04 ºE) in the Central region and Songkhla (7.2 ºN, 100.60 ºE) in the Southern region (Fig. 1). The sunphotometers at Chiang Mai, Nakhon Pathom and Songkhla were fabricated by Cimel (model CE-318). In Ubon Ratchathani two sunphotometers were employed, the first sunphotometer was produced by Pread Co., Ltd. and the second was a Cimel sunphotometer. The Cimel sunphotometers take solar radiation measurements with the almucantar scan once per hour and at optical airmass of 4,3,2 and 1.7 in both the morning and afternoon. The instrument measures direct sun measurements at nominal wavelengths of 340, 380, 440, 500, 675, 870, 940 and 1020 nm and diffuse sky radiances at 440, 675, 870 and 1020 nm, with a bandwidth of 10 nm except 2 nm in the UV [15]. All Cimel sunphotometers were incorporated into the Aerosol Robotic Network (AERONET) of NASA. These sunphotometers were calibrated by AERONET every 1-1.5 years, resulting in AOD accuracy of ~0.01-0.02, with the higher error in the UV [16]. AERONET AOD data screened for clouds by the algorithm of Smirnov et al. [17] that relies on the greater temporal variance of cloud versus aerosol optical depths. It should also be mentioned that only Version 2, level 2 AOD data were utilized in this study. Table 1 and Fig. 1 provide more details, including locations and periods of measurements. The Pread sunphotometer was equipped with a sun tracker (Kipp&Zonen, model 2AP). This sunphotometer measured only the spectral direct solar radiation at the wavelengths 350-1050 nm with a bandwidth of 3.6 nm. It takes solar radiation measurements every 5 minutes. This sunphotometer was calibrated by the manufacturer before and after the measurement period.
393
394
T. Jantarach al. / Procedia Engineering 32 (2012) 392 – 398 T. Jantarach et al.et/ Procedia Engineering 00 (2012) 000–000
3
Northeast monsoon (November-February)
A C
B
D
Southwest monsoon (May-October) Fig. 1. Instruments and locations of the measurement sites. A, B, C and D indicate the main region of Thailand, namely the North, The Northeast, the Central and the South, respectively Table 1. Station details Station name
Latitude
Longitude
Instrument and period of data
Chiang Mai
18.78 ºN
98.98 ºN
Cimel sunphotometer
Ubon Ratchathani
15.25 ºN
104.87 ºN
Pread sunphotometer
(October 2006 – September 2009) (June 2008 – July 2009) and Cimel sunphotometer (October 2009 – December 2009) Nakhon Pathom
13.82 ºN
100.04 ºN
Cimel sunphotometer (September 2006 – December 2009)
Songkhla
7.2 ºN
100.60 ºN
Cimel sunphotometer (January 2007 – December 2009)
4
T.T.Jantarach Engineering 3200 (2012) – 398 Jantarachetetal.al./ Procedia / Procedia Engineering (2012)392 000–000
2.2 Satellite data Moderate Resolution Imaging Spectroradiometer (MODIS) is an instrument on the Terra and Aqua satellites. The Terra and Aqua are polar-orbiting satellites which orbit the Earth in morning descending and afternoon ascending directions, respectively. The MODIS instrument collects spectral irradiance data in 36 wavelength bands ranging from 0.4 to 14.4 μm (from visible to thermal infrared). Two spectral bands are imaged with the resolution of 250 m at nadir pixel dimensions, five spectral bands with the resolution of 500 m and 29 spectral bands with the resolution of 1000 m (http://modis.gsfc.nasa.gov/). In this study, we have used data from MODIS Terra AOD (MOD08_D3, in HDF format) for the period of 2006 to 2009. Which corresponds to the ground-based measurements. The data has been obtained in a daily format (spatial resolution 1ox1o) from the Giovanni website (http://giovanni.gsfc.nasa.gov/). The AOD data from MODIS were sectorized over the locations of Chiang Mai (18.78 ºN, 98.98 ºE), Ubon Ratchathani (15.25 ºN, 104.87 ºE), Nakhon Pathom (13.82 ºN, 100.04 ºE) and Songkhla (7.2 ºN, 100.60 ºE). 3. Comparison of MODIS data using ground-based measurements The AOD data from MODIS at four stations were compared with the measurements over the four main regions of Thailand, situated at Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla. This validation was accomplished by a comparison of the aforementioned data at a wavelength of 550 nm with the ground-based data, measured from the sunphotometers in the same regions, at a wavelength of 500 nm. Fig. 2 (a)-(d) presents the comparison of the AOD measured at Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla with the AOD retrieved from MODIS over these regions based on the daily values. Overall, the root mean squared difference (RMSD) for Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla are in the range of 33.8% to 53.7%. The comparison of the monthly AOD values was shown in Fig. 3. The RMSD for Chiang Mai, Ubon Ratchathani, Nakhon Pathom and Songkhla are in the range of 26.4% to 43.6%. The highest RMSD values (53.7% for daily and 43.6% for monthly) were observed for Chiang Mai. This suggests that the aerosol model used for the inversion of the MODIS radiance measurements might be rather well adapted to the locally produced aerosols but not to the case of the biomass burning aerosols in Chiang Mai. In extreme conditions, the AOD values retrieved from MODIS are overestimated for clean atmosphere and underestimated for the turbid conditions from ground-based measurements [12]. This could be a result from a cease to agricultural clean-up activities which lead to a large decrease in biomass burning aerosols in Chiang Mai.
395
396
T. Jantarach al. / Procedia Engineering 32 (2012) 392 – 398 T. Jantarach et al.et/ Procedia Engineering 00 (2012) 000–000 1.5
Chiang Mai : 2006-2009
1.2
1.2
0.9
0.9
0.6
0.6
0.3
0.3
N= 266 RMSD= 53.7%
0.0
1.5
0.3
0.6 0.9 AODMEASUREMENT
1.2
1.5
0.0
(d)
Nakhon Pathom : 2006-2009
0.3
1.5
1.2
1.2
0.9
0.9
0.6 0.9 AODMEASUREMENT
1.2
1.5
Songkhla : 2007-2009
AODMODIS
AODMODIS
N= 100 RMSD= 40.3%
0.0
0.0
(c)
Ubon Ratchathani : 2008-2009
AODMODIS
AODMODIS
1.5
5
0.6
0.6
0.3
0.3
N= 488 RMSD= 33.8%
0.0 0.0
0.3
0.6
0.9
AODMEASUREMENT
1.2
1.5
N= 158 RMSD= 41.3%
0.0 0.0
0.3
0.6 0.9 AODMEASUREMENT
1.2
1.5
Fig. 2. Comparison of the scatter plots daily AOD between MODIS (AODMODIS) with that measurements (AODMEASUREMENT) by ground-based at four sites namely: (a) Chiang Mai; (b) Ubon Ratchathani; (c) Nakhon Pathom and (d) Songkhla.
397
T.T.Jantarach Engineering 3200 (2012) – 398 Jantarachetetal.al./ Procedia / Procedia Engineering (2012)392 000–000
6
(a)
(b)
Chiang Mai
1.0
0.8
0.8
0.6
0.6
AODMODIS
AODMODIS
1.0
0.4 0.2
N= 13 RMSD= 33.1%
0.0 0.0
0.2
1.0
0.4 0.6 AODMEASUREMENT
0.8
1.0
0.0
0.8
0.8
0.6
0.6
0.4 0.2
0.2
1.0
(d)
Nakhon Pathom
AODMODIS
AODMODIS
0.4 0.2
N= 19 RMSD= 43.6%
0.0
(c)
Ubon Ratchathani
0.8
1.0
Songkhla
0.4 0.2
N= 30 RMSD= 26.4%
0.4 0.6 AODMEASUREMENT
N= 20 RMSD= 27.4%
0.0
0.0 0.0
0.2
0.4 0.6 AODMEASUREMENT
0.8
1.0
0.0
0.2
0.4 0.6 AODMEASUREMENT
0.8
1.0
Fig. 3. Comparison of the monthly average AOD between MODIS MODIS (AODMODIS) with that measurements (AODMEASUREMENT) by ground-based at four sites namely: (a) Chiang Mai; (b) Ubon Ratchathani; (c) Nakhon Pathom and (d) Songkhla.
4. Conclusions This paper has compared the AOD data from MODIS with those from ground-based measurements at four different regions in Thailand. They are located in the north of the country (Chiang Mai), in the northeast (Ubon Ratchathani), the central part (Nakhon Pathom) and the south (Songkhla). MODIS AOD compares better with those from ground-based measurements at the Nakhon Pathom, Ubon Ratchathani and Songkhla, but poorer correlations are found at Chiang Mai station where they have large scale biomass burning. The root mean square difference (RMSD) between daily AOD retrieved from MODIS and that obtained from the ground-based measurements is in the range 33.8%-53.7%.
398
T. Jantarach / Procedia Engineering 32 (2012) 392 – 398 T. Jantarach et al.et/ al. Procedia Engineering 00 (2012) 000–000
7
Acknowledgements The authors would like to thank the Thailand Research Fund for providing financial support to this research work. The authors would also like to thank Dr. Brent Holben for incorporating our four sunphotometers into AERONET. Dr. David Giles is gratefully acknowledged for the arrangement of the data processing. References [1] Salma I, Balashazy I, Hofmann W, Zaray G. Effect of physical exertion on the deposition of urban aerosols in the human respiratory system. J Aero Sci 2002;33:983-97. [2] Hauck H, Berner A, Frischer T, Gomiscek B, Kundi M, Neuberger M et al. AUPHEP-Austrian project on health effects of particulates-general overview. Atmos Environ 2004; 38:3905-15. [3] Andersson KG, Roed J, Byrne MA, Hession H. Deposition of contaminant aerosol on human skin. J Environ Radio 2006;85:182-95. [4] Kennedy IM. The health effects of combustion-generated aerosols. Proceeding of the Combustion Institute 2007;31:2757-70. [5] Ramannathan V, Crutzen PJ, Lelieveld J, Mitra AP, Althausen D, Anderson J et al. Indian ocean experiment: An intergrated analysis of the climate forcing and effects of the great Ondo-Asian haze. J Geophys Res 2001;106:28371-98. [6] Flossmann AI. Interaction of aerosol particles and clouds. J Atmos Sci 1998;55:879-87. [7] O’Dowd CD, Lowe JA, Smith MH. The effect of clouds on aeroaol growth in the rural atmosphere. Atmos Res 2000;54:20121. [8] Cattani E, Costa MJ, Torricella F, Levizzani V, Silva AM. Influence of aerosol particles from biomass burning on cloud microphysical properties and radiative forcing. Atmos Res 2006;82:310-27. [9] Kelly JT, Chuang CC, Wexler AS. Influence of dust composition on cloud droplet formation. Atmos Environ 2007;41:290416. [10] Myhre G, Stordal F, Johnsrud M, Kaufman YJ, Rosenfeld D, Storelvmo T. Aerosol-cloud interaction inferred from MODIS satellite data and global aerosol models. Atmos Chem Phys 2007;7:3081-101. [11] Iqbal M. An introduction to solar radiation. Academic. New York; 1983. [12] He Q, Li C, Tang X, Li H, Geng F, Wu Y. Validation of MODIS derived aerosol optical depth over the Yangtze River Delta in China. Remote Sens Environ 2010;114:1649-61. [13] Bennouna YS, Cachorro VE, Toledano C, Berjon A, Prats N, Fuertes D et al. Comparison of atmospheric climatologies over southwestern Spain derived from AERONET and MODIS. Remote Sens Environ 2011;115:1272-84. [14] Chu DA, Kaufman YJ, Zibordi G, Chern JD, Mao JT, Li CC et al. Global monitoring of air pollution over land from the Earth Observing System-Terra Moderate Resolution Imaging Spectroradiometer (MODIS). J Geophys Res 2003;108:4661, doi:10.1029/2002JD003179. [15] Holben BN, Eck TF, Slutsker I, Tanre D, Buis JP, Setzer A et al. AERONET-a federated instrument network and data archive for aerosol characterization. Remote Sens Environ 1998;66:1-16. [16] Eck TF, Holben BN, Reid JS, Dubovik O, Smirnov A, O,Neill NT et al. Wavelength dependence of the optical depth of biomass burning, urban, and desert dust aerosols. J Geophys Res 1999;104(D24):31333-350. [17] Smirnov A, Holben BN, Eck TF, Dubovik O, Slusker I. Cloud-screening and quality control algorithms for AERONET database. Remote Sens Environ 2000;73:337-49.