SEARCH

SEARCH BY CITATION

Keywords:

  • Dust storms;
  • Indo-Gangetic plains;
  • AERONET;
  • MODIS;
  • Aerosols

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[1] The Indo-Gangetic (IG) plains is one of the largest and most densely populated regions in the world. Recent studies over the IG plains using multi-year (2000–2004) satellite (including Moderate Resolution Imaging Spectroradiometer: MODIS) and ground Aerosol Robotic Network (AERONET) data show strong seasonal variability of aerosol optical depth (AOD) with maximum aerosol loading (>0.6–0.7 at 500 nm) during the pre-monsoon (summer) season. A number of major dust storms, originating from western arid and desert regions of Africa, Arabia and western part of India (Thar Desert), affect the whole IG plains during the pre-monsoon season (April–June). The mean AOD increases from 0.4–0.5 to more than 0.6–0.7 throughout the plains (>0.8–0.9 on the western side) as a result of the dust storm events. Pronounced changes in the aerosol optical parameters, derived from AERONET, have been observed over Kanpur (26.45°N, 80.35°E) during dust storm events (2001–2005). The maximum AOD (at 500 nm) during dust event days show increase from ∼1 to ∼2.4 with advance of the pre-monsoon season (April–June). The aerosol size distribution (ASD) shows increase in radius from 1.71 to 2.24 μm (in coarse fraction) and decrease in the distribution width from 3.76 to 2.56 μm showing changes in the aerosol characteristics during dust events. The aerosol parameters [ASD, single scattering albedo (SSA, total and coarse mode) and real and imaginary parts of the refractive index] change significantly during dust events. The National Oceanic and Atmospheric Administration (NOAA) Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model (5-day back-trajectory) and MODIS level-3 daily data (AOD and Ångström exponent) have been used to trace the source, path and spatial extent of dust storm events. During major dust events, enhancement of the total column water vapor is observed from MODIS level-3 daily water vapor data (near-infrared clear column) showing a strong association (72% correlation) with the AOD along the track of dust storms over the IG plains. A significant rise of 50–100% is observed in the ground level respirable suspended particulate matter (RSPM) concentration showing alarming health risks to the people living in the IG plains during dust storm events.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[2] Natural mineral dust aerosols and dust storms have attracted attention of scientists in the last few decades as they constitute one of the largest fractions of total aerosols [Goudie, 1983; Duce, 1995; Ginoux et al., 2001; Kohfeld and Harrison, 2001]. Carlson and Prospero [1972], Carlson and Caverly [1977], Duce et al. [1980], Tegen and Fung [1994] were one of the first to undertake studies on the mineral dust and their properties, transport and dynamics. Recently, dust aerosols have been found to influence the hydrologic cycle, monsoon system and climate as they are capable of changing radiative characteristics of the atmosphere [Duce, 1995; Sokolin and Toon, 1996; Tegen et al., 1996; Miller and Tegen, 1998; Ramanathan et al., 2001; Tegen, 2003;Miller et al., 2004; Tegen et al., 2004]. Dust storms originating from the arid and desert areas are found to be prime contributor of mineral dust throughout the world. The arid regions of Africa (Saharan Desert), Middle East, Arabian Desert, Afghanistan and Thar Desert (Rajasthan, India), located several thousand kilometers upwind of the western side of the Indo-Gangetic (IG) plains, are some of the hot spots of the world [Gillette, 1999] and contribute a large fraction of mineral dust with respect to total annual loading. The characteristics of the dust storms in Asia and southwest Asia have been discussed in detail by Middleton [1986] and Husar et al. [2001].

[3] Recently, ground based Aerosol Robotic Network (AERONET), [Holben et al., 1998; Smirnov et al., 2000] and satellite data (Polarization and Directionality of the Earth's Reflectance's (POLDER), Total Ozone Mapping Spectrometer (TOMS), Moderate Resolution Imaging Spectroradiometer (MODIS), and Multiangle Imaging SpectroRadiometer (MISR)) have been used extensively to study aerosol optical properties over the IG plains [Golitsyn and Gillette, 1993; Goloub et al., 2001; Di Girolamo et al., 2004; Massie et al., 2004; Singh et al., 2004; Prasad et al., 2004, 2005a, 2005b; Ramanathan and Ramana, 2005; Deepshikha et al., 2005; Jethva et al., 2005]. In the IG plains, aerosol loading is found to be highest during the pre-monsoon season (summer season, April–June) because of large influx from long range transport of mineral dust aerosols from the western arid regions [Singh et al., 2004; Prasad et al., 2004, 2005a, 2005b; El-Askary et al., 2004, 2006; Dey et al., 2004]. Recently, El- Askary et al. [2004, 2006] and Dey et al. [2004] have studied dust storms in the IG plains, limited to few events, but does not cover characteristics, sources and implications for air quality during the whole dust storm season (pre-monsoon season). El-Askary et al. have shown the importance of multisensor satellites data in detection, and monitoring of major dust storm events. Dey et al. have covered optical characteristics and source regions of two dust events during May 2002 and 2003. Long term data set obtained from ground based stations such as Kanpur AERONET (2001–2005) (26.45°N, 80.35°E) as well as daily MODIS satellite aerosol parameters and water vapor data till date provide unique opportunity to characterize dust storms that affect the IG plains every year during the pre-monsoon season.

[4] The air quality of the urban and rural areas lying along the track of dust storms changes significantly (quality and particulate matter concentrations) affecting health of people exposed during dust events [El-Askary et al., 2004, 2006; Dey et al., 2004]. The real time monitoring and forecast of the dust storm events and their tracks will be useful in reducing exposure of these dust events to a large population (∼600 million) living in the plains. Significant depletion of total ozone column (TOC) has also been observed over the IG plains [Sahoo et al., 2005] compared to the southern parts of India from the analysis of TOMS data. The cause of this depletion is difficult to ascertain, but could be attributed to the complex atmospheric chemistry from the higher aerosol loading [Bonasoni et al., 2004]. Dust storms contribute largest fraction of aerosol optical depth (AOD) over the IG plains during the pre-monsoon season. Daily ground data (AERONET) provide optical characterization of the dust for the whole season. MODIS satellite data having a daily overpass provide spatial information about dust storms along its track [Kaufman et al., 1997, 2002] will be of great use in studying the dust characteristics and the areas affected by the dust events through dust circulation models and also in the calculations of radiative forcing to quantify impact on the climate.

[5] In the present paper, we present physical properties and optical characterization of all major dust storm events occurred during the pre-monsoon season 2001–2005 (April–June) over Kanpur using ground based AERONET data where we focus in detail on the radical changes observed in aerosol optical properties (AOD, Ångström exponent: α, aerosol size distribution: ASD, single scattering albedo: SSA and real and imaginary parts of refractive index: RI) during dusty and non-dust days (or clean days) for the year 2005. Air mass trajectories (5 day back-trajectory) have been used to trace the source, path and spatial extent of these dust storm events using National Oceanic and Atmospheric Administration (NOAA) Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model and MODIS level-3 daily data. We have also investigated into the spatial coverage of major dust storms over the IG plains to demarcate total area affected during the event days using satellite AOD data. Here we also discuss the impact of these dust storms on the air quality of major cities lying in the IG plains using Central Pollution Control Board (CPCB) data. The dust storms are dominant during the summer season that lead to enhanced aerosol loading and adversely affect the air quality over the IG plains [Singh et al., 2004; Prasad et al., 2004, 2005a]. Detailed analysis of daily total column MODIS level-3 near-infrared water vapor under cloud free conditions and corresponding MODIS level-3 AOD data have been carried out to show enhancement of water vapor with dust storms along its track over the IG plains.

2. Data Used

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[6] The AERONET (http://aeronet.gsfc.nasa.gov/) program is an extensive ground-based remote sensing aerosol network to measure aerosol optical properties and validate satellite retrievals of aerosol optical properties [Holben et al., 1998]. We have used the AERONET version 1, level 1.5 cloud screened data throughout the analysis of optical properties of dust storms for the year 2005 as the level 2 quality assured was not available at the time of analysis. A new version of AERONET data – version 2, level 2 quality assured – is now available that is more adequate and recommended for further studies. We have used the total column aerosol optical properties such as AOD (τa500 nm) and Ångström exponent (α440–870 nm), ASD, peak and geometrical width of volume particle size distribution [dV(r)/dlnr (μm3/μm2)], real [n(λ)] and imaginary parts [k(λ)] of the RI, and SSA [ωo(λ)] from Kanpur AERONET station lying in the center of the IG plains, to identify and characterize major dust storms. The aerosol products are routinely retrieved from AERONET data following the approach discussed in detail by Dubovik and King [2000] and Dubovik et al. [2000]. The error in the retrieved aerosol parameters is described by Dubovik et al. for urban-industrial, biomass burning and desert dust aerosols. The inversion algorithm [Dubovik et al., 2000] shows the sensitivity to non-sphericity of the dust particle especially for particles in coarse size range. Expected accuracy as described by Dubovik et al. for urban-industrial, biomass burning or other aerosol not dominated by coarse particle, is 15–25% for dV(r)/lnr (0.1 μm ≤ r ≤ 7 μm); 25–100% (or <10% of dV(r)/dlnr in maximum) for r < 0.1 μm and r >7μm. Accuracy is expected to be 0.03 for ωo(λ) and n(λ). For k(λ), expected accuracy is 30% to 50% for strongly and weakly absorbing aerosols, respectively. For aerosols dominated by coarse particles or desert dust, expected accuracy of dV(r)/lnr measurement is 15–25% for r ≥ 0.5 μm; 25–100% [or < 10% of dV(r)/dlnr in maximum] for r < 0.5 μm. Accuracy is expected to be 0.03 for ωo(λ) and 50% for both n(λ) and k(λ). Expected accuracy for n(λ) at 870 nm and 1020 nm is 0.05 and 0.04 respectively.

[7] MODIS is a major instrument on the EOS (Earth Observing System) polar orbiting Terra satellite [Salomonson et al., 1989; King et al., 1992, 2003; Gao and Kaufman, 2003] which is designed to measure atmospheric, biological and physical processes on a global scale every 1 to 2 days. MODIS has higher spatial resolution (250 m at nadir), wide swath (2330 km), and large spectral range (36 channels between 0.412 to 14.2 μm). The channels between 0.4 and 2.1μm are useful for retrieval of aerosol properties while five channels in near-infrared are useful for remote sensing of water vapor. We have analyzed Level-3 MODIS daily global product (MOD08_D3, Collection 4, http://modis-atmos.gsfc.nasa.gov/) containing data collected from the Terra platform to show spatial extent and association of water vapor with these dust storms [King et al., 1992]. A new collection—Collection 5 (C5) – of MODIS product is now available that is recommended for further studies. The daily MODIS Level -3 product (MOD08_D3) gives combination of various aerosol, water vapor and ozone parameters in a single HDF (Hierarchical Data Format) file and are particularly useful to study the interaction between aerosols, energy budget and hydrological cycle. We have used daily Level 3 (MOD08_D3) gridded product (AOD, Ångström exponent, and near-infrared water vapor with clear column) having spatial resolution of 1 degree. The Level-2 product is generated at 10 km resolution for aerosol products (AOD and Ångström exponent) and at 1-km spatial resolution for water vapor. The daily maximum of the aerosol optical thickness at 0.55 μm for both ocean (best) and land (corrected) and mean of the Ångström exponent (land) for 0.47 and 0.66 microns have been used to study daily spatial extent and movement of dust storms from source to sink. The analysis of the association and enhancement of water vapor with dust storms along its path from source (west) to sink (east) have been carried out using mean water vapor (near infrared-clear column). The retrieval algorithm and accuracy of the aerosol products (AOD and Ångström exponent) are discussed by Tanré et al. [1996], Chu et al. [2003], and Remer et al. [2005]. A small uncertainty of ±0.05 ± 0.15 τ in MODIS AOD measurement over land [Remer et al., 2005] has been observed. MODIS water vapor is retrieved at near-IR wavelength that is based on the attenuation of solar radiation at the near-IR. The band ratio technique for water vapor retrieval in MODIS Terra uses near-infrared channels at 0.935, 0.940, and 0.905 μm (water vapor absorption channels with decreasing absorption coefficients) with atmospheric window channels at 1.24 and 0.865 μm (non-absorption channels). The channel at 0.935 μm wavelength is relatively strongly absorbing in nature and is most useful for dry conditions. While the channel at 0.905 μm wavelength is weakly absorbing and is most useful for very humid conditions, or low solar elevation. The theory of radiative transfer of solar radiation (DISORT: Discrete Ordinate Radiative Transfer model) is used to derive water vapor in the atmosphere [Stamnes et al., 1988]. The sensitivity analysis of channel ratio techniques show that a 0.01 error in derived transmittance gives roughly 2.5% error in the retrieved column water vapor. However, errors can be 10% or slightly greater if the aerosol effects are not corrected under hazy conditions (visibilities <10 km) or when the surface reflectance near 1 μm is small (less than about 0.1) [Gao and Kaufman, 1998].

[8] The HYSPLIT_4 (http://www.arl.noaa.gov) model is a system with simple graphical user interface for computing trajectories and air concentrations [Draxler and Hess, 1998; Draxler, 1999]. Gridded meteorological data, at regular time intervals, are used in calculation of air mass trajectories. For back-trajectories, data are obtained from existing archives. A complete description of input data, methodology, equations involved, and sources of error for calculation of air mass trajectory can be found in Draxler and Hess [1997]. The model is run directly on the web (http://www.arl.noaa.gov/ready/hysp_info.html) by giving necessary inputs or on local PC after installing the software and input data set. The executables and meteorological data are provided by the NOAA ARL (Air Resources Laboratory) for free for back-trajectory analysis and registration is required for forecast analysis. The model gives output in the form of post-script image and well as ASCII form that can be imported in other programs for plotting.

[9] The ground concentrations of respirable suspended particulate matter (RSPM) during dust storm events (year 2005) over Kanpur and Delhi have been obtained from Central Control Pollution Board (CPCB) (http://www.cpcb.nic.in/) network.

3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[10] The dust events occur mainly during the pre-monsoon season [Middleton, 1986]. The frequency of occurrence of dust storms is larger in the western region (6–8 in Delhi–Kanpur region) and lesser in the eastern part of the IG plains. During the pre-monsoon season of the year 2005 (April–June), five major dust storm events were recorded at Kanpur. The aerosol characteristics during dust storm event days April 9 (Figure 1), 8 May, 3, 10 and 14 June (year 2005) clearly show the presence of heavy loading of mineral dust in the atmosphere compared to aerosol data few days prior to the event. Detailed analysis of changes observed in the aerosol optical properties and comparison between dusty and non-dusty days is presented here.

image

Figure 1. MODIS TERRA images (band combination 1-4-3) showing long range transport of dust storms. The white color represents clouds. The dust appears in pale beige color, obscuring the land features and seawater (Arabian Sea) below. Dust (plume) can be easily identified against background of black seawater (Arabian Sea) as pale beige color on MODIS image of April 7, 2005. Source: http://rapidfire.sci.gsfc.nasa.gov.

Download figure to PowerPoint

3.1. Diurnal Variation of AOD and Ångström Exponent

[11] Figure 2 shows variations in AOD (τa) and α observed by AERONET using level-2 quality assured data during dust events for period 2001–2004. The AOD is found to vary in the range 1–2.6 with corresponding low Ångström value (closer to zero) during major dust events (2001–2004, Figure 2). Figure 3 shows diurnal variations of AOD (τa) and Ångström exponent (α) for dust storms during the year 2005. The vertical lines in Figures 2 and 3 show the diurnal structure of τa (local time) and each dot represents corresponding α. Average τa during non-dusty days is usually in the range 0.4–0.6 that rapidly increases to >1 on arrival of major dust storms. On the event days during the year 2005 (9 April, 9 May, 3 June, 10 June and 14 June), τa shows average value up to 1.25 ± 0.09, 0.8 ± 0.14, 0.9 ± 0.13, 1.5 ± 0.3, 0.9 ± 0.2 and α value −0.007 ± 0.004, 0.15 ± 0.23, 0.3 ± 0.09, 0.1 ± 0.06, 0.16 ± 0.03, respectively (Figure 3). Table 1 gives τa (500 nm) and α (440–870 nm) for non-dusty and dusty days for each major dust storm event during year 2005. Diurnal variations of ∼0.1–0.2 in τa are found during non-dusty days because of dust originating from vast dry fallow agricultural lands in the IG plains. The τa is found to show large increase from 0.4–0.6 to >1 and corresponding sharp decline in α (close to range 0–0.4) with arrival of dust storms compared to non-dusty days. A sharp decrease in α to low values (even negative values) is observed over Delhi during dust storm events (Apr–June 2003) [Singh et al., 2005] and also in case of the Saharan dust [Hamonou et al., 1999]. A clear distinction can be seen in τa and α value because of major dust storms compared to local mineral dust from the agricultural fields. An increase is found in τa from 0.4–0.5 (April) to 0.6–0.7 (June) (Figure 4). The intensity of dust storm events is found to increase with the increase of temperature during April–June. The relatively low value of τa (maximum) 1.6 during April is found to increase gradually up to 2.4 by the end of the season (June) (Figure 3).The Northeastern Mediterranean region suffers from dust storms originating from the central Sahara, the eastern Sahara, and the Middle East/Arabian peninsula. The AOD retrieved from AERONET shows increase up to 1.8 with very low angstrom values (near to zero) characterized by almost neutral spectral dependency [Kubilay et al., 2003]. In comparison, the dust storm affecting IG plains shows some spectral dependence that varies with each dust storm (Figures 4a and 4b).

image

Figure 2. AERONET record of aerosol optical depth (AOD) and Ångström exponent (α) during major dust storm event days since installation at Kanpur (2001–2004). The vertical lines show the diurnal structure of τa (local time) and each dot represents corresponding α.

Download figure to PowerPoint

image

Figure 3. AERONET aerosol optical depth (AOD) and Ångström exponent (α) during major dust storm event days in the year 2005. The vertical lines show the diurnal structure of τa (local time) and each dot represents corresponding α.

Download figure to PowerPoint

image

Figure 4. Comparison of wavelength dependence of aerosol optical depth (AOD) during (a) non-dust and (b) dusty days (during major dust storms). (c) Two distinct class of dusty and non-dust days are visible on AERONET AOD (500 nm) versus Ångström exponent (440–870 nm) plot.

Download figure to PowerPoint

Table 1. AERONET AOD (500 nm) and Ångström Exponent (440–870 nm) for Non-Dusty and Dusty Days for Each Major Dust Storm Event During Year 2005
Dust Storm EventDusty/Non-dustDaysAOD (500 nm)Ångström Exponent (440–870 nm)
MeanStd. dev.MeanStd. dev.
1Non-dust5 April0.3400.0430.9230.047
Dusty9 April1.2510.091−0.0070.004
2Non-dust4 May0.4070.0541.0950.082
Dusty8 May1.2630.5840.1370.099
3Non-dust29 May0.4570.0490.9120.042
Dusty3 June0.8530.1320.2580.090
4Non-dust5 June0.6480.0630.7180.077
Dusty10 June1.4460.2320.1050.062
5Non-dust9 June0.7380.1060.5440.088
Dusty11 June1.6250.5070.0530.060
3.1.1. Wavelength Dependence of Aerosol Optical Depth

[12] During non-dusty days, aerosol particles show strong wavelength dependence of aerosol optical depth especially in lower wavelength regions (340, 380 and 440 nm) compared with higher wavelengths (870, 1020 nm) (Figure 4a). Figure 4a clearly shows increase in aerosol loading from April to June that shows effect of major dust events on the aerosol loading over the IG plains during the pre-monsoon season.

[13] The AOD shows less sensitivity to wavelength dependence on dusty days compared with non-dusty days. Aerosol loading, during dust storms, is found to increase from April to June that mark increase in intensity of dust storms (Figure 4b). Exception is that of 3 June 2006 event where observed aerosol loading is lesser compared with other events. Dust event of 3 June shows relatively higher wavelength dependence (higher slope) compared to other dusty days. The wavelength dependence of AOD is found to be lowest for dust event (9 April 2006) (Figure 4). The event of April 9 marks the start and initial phase of the dust storms. Wavelength dependence is found to be lowest at higher wavelengths (870, 1020 nm) as compared to non-dusty days (Figure 4). The wavelength dependence of AOD during dust storms observed over the IG plains is found to be in conformity with the recent results of Singh et al. [2005].

3.1.2. Statistical Test on Aerosol Optical Depth Between Non-Dusty and Dusty Days

[14] The paired t-test [Box et al., 1978] is a parametric statistical method that is used to check if the effect of a single treatment or condition (here effect of dust storm on the aerosol optical depth) is found to be significant based on the assumption that the treatment or changed condition effects (i.e., the changes before and after the treatment or event) are normally distributed. The paired t-test examines the changes which occur before (non-dusty days) and during the dust storm (dusty day) on the AOD (at all available wavelengths, 340, 380, 440, 500, 670, 870 and 1020 nm paired together for non-dusty and dusty days). The statistical test provides confidence about the significant impact of dust events on the AOD. The difference in the mean values of the two groups (non-dusty and dusty) is found to be higher than the expected by chance (at alpha = 0.05) showing statistically significant difference between the input groups (non-dusty and dusty days) (P = <0.001). The result of the statistical test is given in Table 2.

Table 2. Paired t-test on AERONET AOD Observed During Non-Dusty and Dusty Daysa,b
Pair NameNMissingMeanStd. Dev.SEM
  • a

    Output: t = −21.155 with 34 degrees of freedom. (P = <0.001). The 95% confidence interval for difference of means: −0.845 to −0.696.

  • b

    Result: The change that occurred with the treatment (i.e. effect of dust storms) is greater than would be expected by chance; there is a statistically significant change (P = <0.001) at alpha = 0.05.

AOD (Non-dust days)3500.5160.1990.0337
AOD (Dusty days)3501.2860.2630.0445
Difference350−0.7700.2150.0364

3.2. Aerosol Size Distribution (ASD)

[15] The ASD clearly shows distinction between dusty and non-dusty days. The ASD shows bimodal nature, with peaks in the range 0.5–10 and 0.05–0.5μm (Figure 5). The moderate to coarse aerosol particles (0.5–10 μm) show a large increase in the volume compared to non-dusty days that indicate presence of moderate to coarse mineral dust particles because of the dust storms. The ASD peak (maxima), corresponding radius (in μm) as well as its geometrical width clearly show distinction between dusty and non-dusty days and also show significant change in the characteristics of dust storm events over the IG plains with advance of the pre-monsoon season. Non-dusty days (4 April, 4 May, 29 May, 5 June, 9 June; Figure 5) show bimodal size distribution with relatively large fraction of fine mode particles. Fraction of medium to coarse mode particles (∼1–10 μm) are found to gradually increase with increase in AOD due to dust events (8–9 April, 7–8–9 May, 3 June, Figure 5) showing dominance of mineral dust particles. The low Ångström values (0–0.4) during dust events show the presence of coarse particles. The peak of volume (size) distribution is found to be at radius (μm) 1.71, 1.71, 2.24, 2.24, 1.71 with corresponding height (dV/dlnR, μm3/μm2) of peak is 0.81, 0.41 (0.55 on 8 May), 0.40, 0.99, 0.51 on the event days 9 April, 9 May, 3 June, 10 June and 14 June, respectively (Figure 5). The ASD peak shows decline in the range of width (μm) 3.76, 3.76, 3.76, 2.56, 2.56 on the event days with advance of the pre-monsoon season. The ASD maxima peak in coarse mode (1–10 μm) is found around radius 1.71 μm during event days in the months of April and May. However, during last phase of the pre-monsoon season (dust event days, 10 and 14 June), ASD maxima peak is found to be higher (radius 1.71–2.24 μm) as compared to start of the season (r = 1.71 μm, April–May). Similarly width of maxima peak is found to be broad during April–May (3.76 μm) as compared to dust storms at the end of the season (2.56 μm, 10 and 14 June). Table 3 gives AERONET size distribution for non-dusty and dusty days for each major dust event during the year 2005.

image

Figure 5. AERONET size (volume) distribution during major dust storm event days in the year 2005.

Download figure to PowerPoint

Table 3. AERONET Size Distribution For Non-Dust and Dusty Days for Each Major Dust Storm Event During Year 2005a
Size Distribution
Radius (μm)5-Apr9-Apr4-May8-May29-May3-Jun5-Jun10-Jun9-Jun11-Jun
  • a

    Classification of days into dusty and non-dusty days for each major dust dtorm event is shown in Table 1.

0.0500.0040.0930.0030.0770.0580.0230.0290.0470.0270.085
0.0660.0090.0510.0050.0330.0270.0300.0420.0260.0380.048
0.0860.0160.0280.0100.0260.0150.0370.0460.0160.0410.025
0.1130.0230.0110.0190.0200.0110.0280.0390.0110.0330.013
0.1480.0220.0050.0290.0150.0120.0140.0270.0090.0210.009
0.1940.0150.0050.0240.0130.0150.0090.0160.0100.0130.009
0.2550.0080.0100.0110.0170.0150.0100.0090.0150.0090.018
0.3350.0060.0270.0060.0290.0090.0150.0080.0270.0090.046
0.4390.0050.0570.0060.0500.0060.0260.0110.0480.0140.094
0.5760.0070.0710.0120.0820.0080.0510.0220.0790.0310.118
0.7560.0130.0860.0340.1350.0230.0970.0530.1390.0730.152
0.9920.0250.1790.0700.2580.0740.1670.0950.2720.1320.278
1.3020.0350.5130.0640.4730.1150.2700.1230.5430.1940.640
1.7080.0400.8120.0520.5490.1230.3560.1510.9080.2551.045
2.2410.0480.6660.0510.5080.1380.3970.2010.9930.3221.178
2.9400.0610.5920.0570.4960.1400.3960.2790.6310.3811.252
3.8570.0770.4370.0540.3880.0990.2660.3390.3110.3090.740
5.0610.1040.1950.0330.1740.0460.1030.2790.1760.1800.233
6.6410.1540.0650.0150.0630.0180.0330.1620.1290.1160.061
8.7130.1910.0280.0070.0280.0080.0130.0970.1010.0790.026
11.4320.1930.0190.0040.0190.0050.0070.0750.0740.0490.022
15.0000.2830.0220.0040.0220.0060.0080.0700.0480.0290.032

3.3. Single Scattering Albedo (SSA)

[16] SSA (ωo(λ)) provides important information regarding scattering and absorption properties of aerosols. SSA is estimated from extinction optical thickness τext(λ) and scattering optical thickness τscatt(λ). The detailed procedure of estimating τext(λ) and τscatt(λ) and retrieval of SSA is discussed by Dubovik et al. [2000]. Figure 6 shows the variability of SSA for event days.

image

Figure 6. AERONET total and coarse mode single scattering albedo (SSA) during major dust storm event days in the year 2005.

Download figure to PowerPoint

[17] The SSA (total and coarse mode) clearly shows increase on dust storm days compared with non-dusty days (Figure 6) for all wavelengths (441, 673, 873 and 1022 nm). SSA (coarse mode) is found to increase with arrival of dust and found to be a common feature during dust events in the year 2005 with similar behavior in the total SSA. During dust storm events, SSA is found to be ∼0.95 that shows the presence of dust [Shettle and Fenn, 1979]. The decrease in α and corresponding increase in τa together with the increase in SSA support the presence of coarser scattering particles (mineral dust) compared to the nature of absorbing aerosols from biomass or forest fire. For λ = 673, SSA is found to be > 0.9 during dust storm days and systematically shows lower values during non-dusty days, showing the presence of scattering and larger size particles which is also supported by earlier observations of Shettle and Fenn [1979]. SSA shows large difference between dusty and non-dusty days at higher wavelengths (λ > 673 nm) compared to λ = 441 nm and SSA values are found to be higher in higher wavelengths. Table 4 gives AERONET SSA (total and coarse mode) for non-dusty and dusty days for each major dust storm event during the year 2005.

Table 4. AERONET SSA (Total and Coarse) For Non-Dusty and Dusty Days for Each Major Dust Storm Event During Year 2005a
DaysSSA (Total)SSA(Coarse)
44167387310224416738731022
  • a

    Classification of days into dusty and non-dust days for each major dust storm event is shown in Table 1.

5-Apr0.8640.8440.8360.8370.6700.7340.7700.791
9-Apr0.9270.9630.9730.9780.9070.9540.9680.976
4-May0.9410.9340.9290.9300.8530.8870.9010.913
8-May0.9080.9490.9610.9680.8720.9350.9530.963
29-May0.9490.9540.9590.9630.9010.9340.9490.958
3-Jun0.8870.9250.9400.9480.8460.9060.9290.941
5-Jun0.8510.8690.8780.8880.7510.8250.8530.871
10-Jun0.8730.9180.9340.9430.8380.9000.9230.936
9-Jun0.8610.8900.9010.9120.7930.8610.8850.902
11-Jun0.9060.9610.9740.9790.8770.9510.9690.976

[18] The spectral variations of SSA are found to be greater than 0.05 for mineral dust by Dubovik et al. [2002]. For AERONET stations located in desert regions, Dubovik et al. [2002] have shown the spectral difference (ΔSSA) (SSA1022 nm − SSA441 nm) greater than 0.05 for mineral dust aerosols. Our observations also match very well with the observations made by Dubovik et al. [2002] during dusty days. The ΔSSA (total mode) is found to be 0.051, 0.60, 0.061, 0.070 and 0.073 for dusty days 9 April, 8 May, 3 June, 10 June and 11 June, respectively (Table 4). Like mineral dust affecting the IG plains, desert dust of Sahara and Middle East also shows a higher SSA (0.95 ± 0.03) with further increase in SSA values slightly at longer wavelengths. Prior to April 9 event in the IG plains, SSA (total) (Figure 6a) shows lower values at higher wavelengths which is characteristic of both biomass burning and anthropogenically derived aerosols. This feature is also observed for Mediterranean region before and after the dust storms implying presence of anthropogenic aerosols [Dubovik et al., 2002; Kubilay et al., 2003]. The SSA of dust particles over East Asia (China) are found to be around 0.9 at 500 nm. The optical properties of Asian dust around Korea and Japan are found to be similar to those found in dust source regions (Dunhuang and Mandalgovi) [Kim et al., 2004a, 2004b]. Kim et al. [2004a, 2004b] have found small values of SSA during the dust event days around Korea and Japan whereas SSA values are found to increase over the IG plains, Sahara and Middle East regions. The decrease in the SSA values with higher wavelength during dust events can be attributed to mixing of the large pool of anthropogenic aerosol pollution over East Asia (China-Japan) due to absorbing soot and organic aerosols from fossil fuel (coal-petroleum) combustion and biomass burning [Chameides et al., 1999]. The ΔSSA values show increasing dominance of mineral dust aerosols over the IG plains during pre-monsoon season and mask the effect of anthropogenic aerosols as volume of coarse particles increase during dust storms. The effect of anthropogenic aerosols on optical characteristics of aerosols in the IG plains due to fossil fuel combustion in industries, thermal power plants, automobiles and bio-mass burning is more visible during the winter season as dust storms are absent [Prasad et al., 2006].

3.4. Refractive Index (RI)

[19] Refractive index [real: n(λ) and imaginary: k(λ)] is one of the important optical properties that provide information about the nature of aerosols. k(λ) and n(λ) values are dependent on the chemical composition of aerosols and provide information on the scattering or absorbing nature of aerosols. In the visible region, mineral dust typically shows n(λ) values 1.53 ± 0.05 and k(λ) ∼0.006 and lesser [Shettle and Fenn, 1979; Levin et al., 1980; WMO, 1983; Sokolik et al., 1993; Koepke et al., 1997]. Higher values of k(λ) represent absorbing type of aerosols and higher values of n(λ) represent scattering types of aerosols [Bohren and Huffman, 1983; Sinyuk et al., 2003]. The n(λ) shows high values in the range 1.5–1.6 and k(λ) is found to decrease for dusty days during dust events compared with non-dusty days (Figure 7). k(λ) value shows a general decrease with advance of the summer season, k(λ) value is mostly found to be > 0.0045 for non-dusty days during start of the pre-monsoon season that gradually decrease to less than 0.0045. Decrease in k(λ) is found compared to non-dusty days prior to the dust storm event. The decrease in k(λ) shows dominance of mineral dust aerosols during the dust storms and is likely to be attributed to the decrease in fraction of anthropogenic (smoke) aerosols. Prasad et al. [2006] have studied effect of anthropogenic aerosols (emitted from fossil fuel burning in the thermal power plants and vehicles including bio-mass burning) on the aerosol optical properties during the winter season (December–January) when the dust storms are absent in the IG plains. During the summer season, mineral dust aerosols dominate over the IG plains and effects of dust are clearly visible from the aerosol optical properties. Variations of RI [n(λ) and k(λ)] (Figure 7) support the presence and dominance of mineral dust during dust storm events that is also observed from SSA values (total and coarse) (Figure 6). Table 5 gives AERONET RI (real and imaginary parts) for non-dusty and dusty days for each major dust storm event during the year 2005. The n(λ) shows slightly higher values at longer wavelengths and varies in the range 1.5 ± 0.05 for Saharan and Middle East Desert dust [Kubilay et al., 2003]. The n(λ) under moderate-to strong dust event is 1.51 ± 0.07 at 440 nm at Mediterranean region near Turkey, 1.55 ± 0.03 at Bahrain (Persian Gulf), 1.56 ± 0.03 at Solar-Vil (Saudi Arabia), and 1.48 ± 0.05 at Cape Verde [Dubovik et al., 2002]. Though the dust storms in the IG plains show a similar high value greater than 1.5 in the wavelength range 441–673 nm, showing a drop in the higher wavelength range. However, the k(λ) shows lower absorption (<0.002) for Saharan and Middle East Desert dust. The k(λ) for Saharan and Middle East Desert is found to be around 0.0015 and 0.0005, respectively; k(λ) at 440 nm varies 0.0025 ± 0.001 over Bahrain, 0.0029 ± 0.001 over Solar Vil, and 0.0025 ± 0.001 over Cape Verde [Kubilay et al., 2003]. The k(λ) values are found to decrease with the arrival of dust storms over the IG plains showing low absorption value (less than 0.0045) during dusty days with small decrease at higher wavelengths which is a characteristic of mineral dust compared to absorbing aerosols. The difference in spectral dependence at higher wavelength [n(λ)] and absolute value of k(λ) could be due to complex mineralogy during long range transportation and mixing with anthropogenic pollutants emitted from different cities located along the dust storms track. Kubilay et al. [2003] have also found the difference in k(λ) values for Saharan and Middle East Desert dust due to changes in mineralogy of dust, k(λ) provides information of the absorption capacity of the medium [Sokolik and Toon, 1999]. The difference in the mineralogy of Saharan and Middle East Desert dusts can also be inferred from k(λ) values [Kubilay et al., 2003] it may also be concluded that the small changes in k(λ) values of the observed dusts over the IG plains (Figure 7) may be due to the difference of the source region of dusts.

image

Figure 7. AERONET real and imaginary parts of refractive index (RI) during major dust storm event days in the year 2005.

Download figure to PowerPoint

Table 5. AERONET Refractive Index (RI (Real) and RI (Imaginary)) for Non-Dusty and Dusty Days for Each Major Dust Storm Event During Year 2005a
DaysRI (real) at Wavelength (nm)RI (imaginary) at Wavelength (nm)
44167387310224416738731022
  • a

    Classification of days into dusty and non-dusty days for each major dust storm event is shown in Table 1.

5-Apr1.60001.60001.60001.60000.01020.00870.00840.0083
9-Apr1.48251.52621.50671.45830.00210.00140.00130.0012
4-May1.56231.56391.57541.57990.00450.00470.00530.0055
8-May1.55871.56481.55181.53030.00320.00220.00200.0020
29-May1.60001.60001.60001.60000.00210.00200.00190.0019
3-Jun1.52951.55621.54161.52590.00410.00340.00320.0032
5-Jun1.52451.57571.59461.59640.00690.00580.00590.0058
10-Jun1.58601.56471.52871.50090.00410.00330.00330.0033
9-Jun1.52611.56931.56921.55010.00560.00470.00480.0048
11-Jun1.56841.57791.54801.51810.00270.00130.00110.0010

4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[20] The source and path followed by dust storms before arrival over Kanpur at the central part of the IG plains, have been studied using a 5 days back-trajectory obtained from NOAA HYSPLIT (http://www.arl.noaa.gov/) air mass transport model (Figure 8). We have shown air mass back-trajectory at three heights above surface 500, 1000 and 1500 m to locate source of aerosols during dust storm events over Kanpur. HYSPLIT back trajectory was initialized with starting date as day of dust storm event recorded over Kanpur and its position at every hour interval up to previous 5 days (Figure 8). These back-trajectories extend by more than 4000 km, up to the neighboring continent (Africa) and total length of path traveled by dust storm varies with each dust event. Results from the air mass transport model show multiple sources and extent of dust transport that affect the IG plains. Dust storm event of 9 April 2005 shows origin from the Oman and from the United Arab Emirates region (for all three levels) while 9 May 2005 event extend farthest up to Bahrain (Arabian arid region) and Egypt in Africa. The dust storm event of 3 June 2005 extends NW side toward Afghanistan, Turkmenistan region. The dust storm events of 10 and 14 June 2005 extend toward Thar Desert and the Arabian Ocean indicating possible source to be Saharan Desert as the Arabian Ocean is affected by dust transport from these regions. Most of the dust storms are observed to be passing over the Thar Desert before entering into the IG plains while event of 3 June 2005 enters over the IG plains through the western provinces (Punjab and Delhi). Dust storm event of 3 June 2005 shows different behavior in peak AOD, higher wavelength dependence of AOD (Figure 4a), Ångström and ASD. Event of 3 June 2006 shows unique characteristic compared with other events and it is attributed to its unique source region (Afghanistan Desert) and path of travel (less influence of Thar Desert as it crosses Punjab to enters into the IG plains). Singh et al. [2005] have discussed change in wavelength dependence of AOD due to dust storms using handheld Sun photometer observations over Delhi lying about 400 km north-west of Kanpur. Our study shows that change in wavelength dependence between dust storms over the IG plains at different days between April and June is governed by source region as well as path of travel. The arid regions of the Thar Desert, Afghanistan, Middle East and the Africa (Saharan Desert) have been found to be major contributors of the mineral dust over the IG plains during the pre-monsoon season. All the major dust events have been found to originate from the arid regions west of the IG plains which travel a distance of the order of 1000's km over a period of 5–7 days showing transcontinental nature of dust storms affecting the IG plains (Figures 1, 8) every year. Tables 1, 3, 4, and 5 show change in aerosol optical properties prior and during dust storms that can be of great use in quantitative estimation of change in radiative forcing and the effect of dust storms on the climate. Data from AERONET and MODIS during dust storms can be assimilated into regional and global climate models to study its effect in the Indo-Gangetic plains.

image

Figure 8. HYSPLIT 5 day back-trajectory starting from AERONET station at Kanpur showing major sources of dust events affecting IG plains.

Download figure to PowerPoint

5. Spatial Extent of Major Dust Storms and Health Risks

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[21] The air mass transport model showing very long range transport of mineral dust passing through arid and desert regions up to the IG plains, show the importance of aerosol characteristics over the region lying along its path. The IG plains is characterized by low and unique valley shape topography extending east-west from Kolkata (located in the eastern part of the IG plains) up to alluvial plains of Pakistan in the west (Figure 8) where about 600 million people live and is one of the most agriculturally productive plains of the world. Daily overpass visible image from MODIS shows area affected by these dust storms (dust event 7–9 April 2006, Figure 1). An increase in the ground based concentration of RSPM by ∼100% over Kanpur and Delhi region have been observed during this period (Figure 9). The increase in RSPM generally has an adverse impact on the health affecting respiratory system of people living in the cities and rural areas. Monitoring of these dust storms from source region and along track of dust is possible using multisensor satellite data [El-Askary et al., 2004, 2006]. The total area affected by these storms can be mapped by near real-time MODIS data (http://rapidfire.sci.gsfc.nasa.gov/) and health related warnings may be issued in advance. Wang and Christopher [2003] have recently reported relationship between satellites derived aerosol loading and ground particulate matter (PM2.5) concentration and have discussed potential application in the air quality studies over cities. Engel-Cox et al. [2004] has presented quantitative application of MODIS satellite data in urban and regional air quality. Dust storm events can be effectively tracked using ground and satellite data and holds potential for air quality studies.

image

Figure 9. Spatial extent and area affected by dust storms in the IG plains as seen on daily MODIS AOD (maximum), Ångström exponent (α) and water vapor (NIR-clear column) map of May 4 and 8, 2005.

Download figure to PowerPoint

6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[22] Figure 10 shows daily MODIS derived AOD (maximum at 0.55 microns), Ångström exponent (mean for 0.47 and 0.66 microns) and mean water vapor (near-infrared, clear column) variations. Enhanced level of water vapor is found over the dust storms track from west to east (4 and 8 May 2005, Figure 10). MODIS AOD (max. >1.2) and Ångström exponent (<0.25) of 4 May 2005 show concentration of dust storm over Punjab, Delhi and the western Uttar Pradesh (UP) region that gradually moves to the eastern UP, Bihar region by 8 May 2005 (Figures 8, 10). The location and aerial extent of dust storm is clearly visible from MODIS water vapor (near-infrared clear column) image where values are found to be greater than 5 cm (Figure 10).

image

Figure 10. Power correlation between mean MODIS AOD (τmodis-mean) and mean MODIS water vapor (near infra-red, clear column) (ψmodis-mean) during dust storm events (May 4 and May 8) over different regions.

Download figure to PowerPoint

[23] We have studied total column aerosol loading and associated water vapor content during presence of dust storms over Pakistan, the western part of the IG plains on 4 May and the eastern part of the IG plains on 8 May 2005. The eastern part of the IG plains clearly shows increase in power correlation between mean MODIS AOD (τmodis-mean) and mean MODIS water vapor (near-infrared clear column) (ψmodis-mean) from 32% (4 May 2005) to 72% (8 May 2005) due to the influence of dust storms (Figure 11). Similarly, over the eastern part of the IG plains, increase is found in the power correlation between maximum MODIS AOD (τmodis-max) and ψmodis-mean from 41% (4 May 2005) to 61% (8 May 2005) with absolute rise in total water vapor and AOD. The presence of dust storm over the western part of the IG plains during 4–8 May 2005 shows increase in power correlation between τmodis-mean and ψmodis-mean from 48% to 58% (4 to 8 May 2005). Dust storm gradually covers whole of the IG plains (>4000 km wide from west to east) during 4 to 8 May 2005 and also shows large increase in power correlation between τmodis-mean and ψmodis-mean from 44% to 68% (R2 between τmodis-max. and ψmodis-mean increases from 0.41 to 0.54) (Figure 11). Similarly, Pakistan region also shows drop in power correlation from 41% to 15% between τmodis-mean and ψmodis-mean (R2 between τmodis-max. and ψmodis-mean decreases from 0.47 to 0.20) during dust storms (4 May 2005) and few days after the dust storm (8 May 2005) (Figures 10 and 11). The power correlation of τmodis-mean (or τmodis-max) with ψmodis-mean shows increase in correlation coefficient due to the influence of dust storms. However, a power correlation study of ψmodis-mean with τmodis-mean is found to be more sensitive for such studies than use of τmodis-max. The association of enhanced level of water vapor with dust storms and its effect on radiative forcing have been recently discussed by Kim et al. [2004a, 2004b] and Yoon et al. [2006] for the Asian dust storms. Strong association of water vapor as observed from MODIS data with dust storms affecting the IG plains during the pre-monsoon season is likely to affect the radiative heating or cooling rates over the IG plains and it may affect the monsoon system. Chinnam et al. [2006] have calculated the radiative forcing due to dust storm during 12–22 May 2004 due to dust over Kanpur based on modeled data as AERONET data is largely missing for this period. Earlier studies on the heating rate of mineral dust in short and long wavelength usually do not consider the role of atmospheric water vapor that is found to be an essential part of dust storms and it can modify radiative properties significantly [Kim et al., 2004a, 2004b].

image

Figure 11. Ground observations (CPCB) of RSPM over Delhi and Kanpur cities during major dust storm events (April–June, 2005).

Download figure to PowerPoint

7. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[24] Analysis of the aerosol parameters retrieved over Kanpur, situated in the central part of the IG plains, from ground-based AERONET (2001–2005) and satellite (MODIS) measurements show significant changes compared to non-dusty days prior to the dust storm events. Kanpur AERONET data for all major dust storm events (year 2005) show large increase in AOD (τa500 nm) during event days (>1) and corresponding decrease in Ångström exponent (α440–870 nm) showing presence of larger fraction of coarser particles which is also supported from ASD values. There is also increase in the intensity of dust storms (increase in maximum observed τa from >1 to >2.4 with advance of the pre-monsoon season) depending upon the source region and the travel path of dust storm. Significant changes have been found in aerosol optical properties (ASD, SSA, RI) during dust storm event days compared with non-dust days. The paired t-test shows that the dust storms change the aerosol optical properties (AOD) significantly. ASD also shows increase in radius of maxima value from 1.71 to 2.24 μm implying presence of more coarse particles with advance of the pre-monsoon season. The SSA (total and coarse mode) shows high value (>0.9) during event days supported by high value of real part (1.5–1.6) and decrease in the imaginary part (<0.0045) of the refractive index which are typical of desert dust and scattering in nature. Various aerosol parameters (τa, α, ASD, SSA and RI) during event days support the presence of desert mineral dust which are found to be transported from arid and desert regions of Thar Desert (Rajasthan), Afghanistan, Arabia and Africa (Saharan Desert). The HYSPLIT 5 days back-trajectory supports multiple source and travel path of these dust storm events occurring in a season. Daily MODIS visible image (at 250m resolution) and aerosol optical properties data (τa and α) show spatial extent (area affected by dust storms) along the track of dust storms. The difference in the observed optical properties of mineral dust aerosols over East Asia (China, Korea and Japan) are more likely due to mixing of the mineral dust and the anthropogenic aerosols from industries and bio-mass burning [Kim et al., 2004a, 2004b]. This shows higher absorbing nature of the mineral dust storms over East-Asia compared with the Saharan Desert, Middle East and Thar Desert (west of the IG plains). Saharan dust has a higher absorbing and lower scattering nature (as visible in SSA and RI) as compared to that from the Middle East [Kubilay et al., 2003]. This could be explained by contrasting mineralogy of dust emitted from land surface and presence of more bio-mass burning aerosols over Africa as compared to that of the Middle East. The IG plains, which is invariably affected by dust storms originating from western regions (Saharan, Middle East and Afghanistan), partially show spectral signatures of these regions. Though RI (k(λ)) shows low absorption values as expected for mineral dust of western origin, the absolute values are found to be relatively higher. Similarly, n(λ) shows high values over the IG plains which are similar to values reported for Saharan and Middle East dust. Values and spectral variations of SSA also support the mineral dust nature of dust. The weak spectral dependence of AOD for some of the dust events show the difference in the dust mineralogy of the dusts observed over the IG plains and that of Saharan, Middle East.

[25] The dissimilarities in the optical characteristics of the dust between Saharan-Middle East and that of the IG plains can be explained by change in characteristics during long range transport caused by mixing of mineral dust with dust from other regions (Thar Desert dust, soil dust), increasing association with water vapor over the IG plains, and mixing with anthropogenic aerosols over the IG plains. One of the unique features of dust storm season over the IG plains is gradual rise of the aerosol loading with gradual increase in the intensity of dust storms with passage of the season (April–June). Increasing dominance of mineral dust suppresses signature of anthropogenic aerosols and optical properties show more scattering signature. Absorbing aerosols are found to be dominant during the winter season when mineral dust aerosols is washed away by the monsoon rainfall and dust storm activity is absent leading to more absorbing signature in optical characteristics [Prasad et al., 2006]. The increase in level of RSPM associated with dust storm events based on CPCB data shows serious health risks for people living in the IG plains, although no statistics are available but during the pre-monsoon season children and old people are found to suffer with respiratory problems. An enhanced level of water vapor (>3 cm, in near-infra-red, clear column) have been found to be associated with higher aerosol loading during these dust storms. Increase in power correlation between mean MODIS AOD (τmodis-mean) and mean MODIS water vapor (ψmodis-mean) is found to be pronounced over the IG plains due to the influence of dust storms. Power correlation, up to 72%, has been found between τmodis-mean and ψmodis-mean during major dust storm events over the IG plains. An extensive database of aerosol optical properties of all these dust storm events from ground (AERONET) and satellite (MODIS) record will be useful in quantitative evaluation of its effect on the regional weather and climate conditions. Besides, the quantitative estimates of increase of water vapor during dust storms over the IG plains (including alluvial plains of Pakistan) may have direct impact in the global climate modeling and also on the rainfall pattern of the region.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information

[26] We thankfully acknowledge National Aeronautics and Space Administration (NASA) MODIS Team for providing MODIS TERRA data and National Oceanic and Atmospheric Administration Air Resources Laboratory (NOAA ARL) for HYSPLIT transport and dispersion model. We thank NASA and IIT Kanpur for supporting operation of Kanpur AERONET site. Efforts of S. S. Chauhan in preparation of the manuscript are thankfully acknowledged. The part of the work is supported through a research grant to one of the authors (RPS) through a research project supported by the Indian Space Research Organisation-Geosphere Biosphere Program.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information
  • Bohren C. F., and D. R. Huffman (1983), Absorption and Scattering of Light by Small Particles, 550 pp., John Wiley, Hoboken, N. J.
  • Bonasoni, P., P. Cristofanelli, F. Calzolari, U. Bonafè, F. Evangelisti, A. Stohl, S. Z. Sajani, R. van Dingenen, T. Colombo, and Y. Balkanski (2004), Aerosol-ozone correlations during dust transport episodes, Atmos. Chem. Phys., 4(5), 12011215.
  • Box, G. E. P., W. G. Hunter, and J. S. Hunter (1978), Statistics for experimenters: an introduction to design, analysis and model building,, John Wiley & Sons, New York.
  • Carlson, T. N., and R. S. Caverly (1977), Radiative characteristics of Saharan dust at solar wavelengths, J. Geophys. Res., 82, 31413152.
  • Carlson, T. N., and J. M. Prospero (1972), The large-scale movement of Saharan air outbreaks over the northern equatorial Atlantic, J. Appl. Meteorol., 11, 283297.
  • Chameides, W. L., et al. (1999), Case study of the effects of atmospheric aerosols and regional haze on agriculture: An opportunity to enhance crop yields in China through emission controls, Proc. Natl. Acad. Sci., 96, 13,62613,633.
  • Chinnam, N., S. Dey, S. N. Tripathi, and M. Sharma (2006), Dust events in Kanpur, northern India: Chemical evidence for source and implications to radiative forcing, Geophys. Res. Lett., 33, L08803, doi:10.1029/2005GL025278.
  • Chu, D. A., Y. J. Kaufman, G. Zibordi, J. D. Chern, J. Mao, C. Li, and B. N. Holben (2003), Global monitoring of air pollution over land from EOS-Terra MODIS, J. Geophys. Res., 108(D21), 4661, doi:10.1029/2002JD003179.
  • Deepshikha, S., S. K. Satheesh, and J. Srinivasan (2005), Regional distribution of absorbing efficiency of dust aerosols over India and adjacent continents inferred using satellite remote sensing, Geophys. Res. Lett., 32, L03811, doi:10.1029/2004GL022091.
  • Dey, S., S. N. Tripathi, R. P. Singh, and B. N. Holben (2004), Influence of dust storms on aerosol optical properties over the Indo-Gangetic basin, J. Geophys. Res., 109, D20211, doi:10.1029/2004JD004924.
  • Di Girolamo, L., T. C. Bond, D. Bramer, D. J. Diner, F. Fettinger, R. A. Kahn, et al. (2004), Analysis of Multi-angle Imaging SpectroRadiometer (MISR) aerosol optical depths over greater India during winter 2001–2004, Geophys. Res. Lett., 31, L23115, doi:10.1029/2004GL021273.
  • Draxler, R. R. (1999), HYSPLIT_4 User's Guide,, NOAA Technical Memorandum ERL ARL-230,, June, pp. 35.
  • Draxler, R. R., and G. D. Hess (1997), Description of the HYSPLIT_4 modelling system, NOAA Technical Memorandum,, ERL ARL-224,, 24 pp.
  • Draxler, R. R., and G. D. Hess (1998), An overview of the HYSPLIT_4 modelling system for trajectories, dispersion and deposition, Aust. Meteorol. Mag., 47(4), 295308.
  • Dubovik, O., and M. D. King (2000), A flexible inversion algorithm for retrieval of aerosol optical properties from sun and sky radiance measurements, J. Geophys. Res., 105, 20,67320,696.
  • Dubovik, O., A. Smirnov, B. N. Holben, M. D. King, Y. J. Kaufman, T. F. Eck, and I. Slutsker (2000), Accuracy assessments of aerosol optical properties retrieved from Aerosol Robotic Network (AERONET) Sun and sky radiance measurements, J. Geophys. Res., 105, 97919806.
  • Dubovik, O., B. N. Holben, T. F. Eck, A. Smirnov, Y. J. Kaufman, M. D. King, D. Tanre, and I. Slutsker (2002), Variability of absorption and optical properties of key aerosol types observed in worldwide locations, J. Atmos. Sci., 59, 590608.
  • Duce, R. A. (1995), Sources, distributions, and fluxes of mineral aerosols and their relationship to climate, Aerosol Forcing of Climate,, pp. 4372., R. Charlson, J. Heintzenberg (Eds.), Wiley, New York.
  • Duce, R., C. K. Unni, B. Ray, J. M. Prospero, and J. Merrill (1980), Long range transport of soil dust from Asia to the tropical North Pacific: Temporal variability, Science, 209, 15221524.
  • El-Askary, H., R. Gautam, and M. Kafatos (2004), Monitoring of dust storms over Indo-Gangetic Basin, Indian J. Rem. Sens., 32(2), 121124.
  • El-Askary, H., R. Gautam, R. Singh, and M. Kafatos (2006), Dust storms detection over the Indo-Gangetic Basin using multi sensor data, Adv. Space Res., (in press).
  • Engel-Cox, J. A., C. H. Holloman, B. W. Coutant, and R. M. Hoff (2004), Qualitative and quantitative evaluation of MODIS satellite sensor data for regional and urban scale air quality, Atmos. Environ., 38(16), 24952509.
  • Gao, B.-C., and Y. J. Kaufman (1998), The MODIS Near-IR Water Vapor Algorithm, Products: MOD05, MOD08, ATBD Reference Number: ATBD-MOD-03.
  • Gao, B.-C., and Y. J. Kaufman (2003), Water vapor retrievals using Moderate Resolution Imaging Spectroradiometer (MODIS) near-infrared channels, J. Geophys. Res., 108(D13), 4389, doi:10.1029/2002JD003023.
  • Gillette, D. A. (1999), A qualitative geophysical explanation for ‘hot spot’ dust emitting source regions, Contrib. Atmos. Phys., 72(1), 6777.
  • Ginoux, P., M. Chin, I. Tegen, J. M. Prospero, B. Holben, and O. Dubovik (2001), Sources and distributions of dust aerosols simulated with the GOCART model, J. Geophys. Res., 106(D17), 20,25520,273.
  • Golitsyn, G., and D. A. Gillette (1993), Introduction: a joint Soviet-American experiment for the study of Asian Desert dust and its impact on local meteorological conditions and climate, Atmos. Environ., Part A, 27A(16), 24672470.
  • Goloub, P., J. L. Deuze, M. Herman, D. Tanre, I. Chiapello, B. Roger, and R. P. Singh (2001), Aerosol remote sensing over land using the spaceborne polarimeter POLDER, in Current Problems in Atmospheric Radiation,, edited by W. L. Smith, and Yu. M. Timofeyev, 113116, A. Deepak, Hampton, Va.
  • Goudie, A. S. (1983), Dust storms in space and time, Prog. Phys. Geogr., 7(4), 502530.
  • Hamonou, E., P. Chazette, D. Balis, F. Dulac, X. Schneider, E. Galani, E. Ancellet, and A. Papayannis (1999), Characterization of the vertical structure of Saharan dust export to the Mediterranean basin, J. Geophys. Res., 104, 22,25722,270.
  • Holben, B. N., et al. (1998), AERONET a federated instrument network and data archive for aerosol characterization, Rem. Sens. Environ., 66(1), 116.
  • Husar, R. B., et al. (2001), Asian dust event 1998, J. Geophys. Res., 06, 18,31718,330.
  • Jethva, H., S. K. Satheesh, and J. Srinivasan (2005), Seasonal variability of aerosols over the Indo-Gangetic basin, J. Geophys. Res., 112, D21204, doi:10.1029/2005JD005938.
  • Kaufman, Y. J., D. Tanré, L. A. Remer, E. F. Vermote, A. Chu, and B. N. Holben (1997), Operational remote sensing of tropospheric aerosol over land from EOS moderate resolution imaging spectroradiometer, J. Geophys. Res., 102, 17,05117,067.
  • Kaufman, Y., D. Tanré, and O. Boucher (2002), A satellite view of aerosols in the climate system, Nature, 419, 215223.
  • Kim, D. H., B. J. Sohn, T. Nakajima, T. Takamura, T. Takemura, B. C. Choi, and S. C. Yoon (2004a), Aerosol optical properties over east Asia determined from ground-based sky radiation measurements, J. Geophys. Res., 109, D02209, doi:10.1029/2003JD003387.
  • Kim, S.-W., S.-C. Yoon, A. Jefferson, J.-G. Won, E. G. Dutton, J. A. Ogren, and T. L. Anderson (2004b), Observation of enhanced water vapor in Asian dust layer and its effect on atmospheric radiative heating rates, Geophys. Res. Lett., 31, L18113, doi:10.1029/2004GL020024.
  • King, M. D., Y. J. Kaufman, W. P. Menzel, and D. Tanre (1992), Remote sensing of cloud, aerosol, and water vapor properties from the Moderate Resolution Imaging Spectrometer (MODIS), IEEE Trans. Geosci. Remote Sensing, 30, 127.
  • King, M. D., W. P. Menzel, Y. J. Kaufman, D. Tanre, B.-C. Gao, S. Platnick, S. A. Ackerman, L. A. Remer, R. Pincus, and P. A. Hubanks (2003), Cloud and aerosol and water vapor properties, precipitable water, and profiles of temperature and humidity from MODIS, IEEE Trans. Geosci. Remote Sens., 41, 442458.
  • Koepke, P.M. Hess, I. Schult, and E. P. Shettle (1997), Global aerosol data set Rep.,, 243,, p. 44, Max Planck Inst. for Meteorol.,, Hamburg, Germany.
  • Kohfeld, K. E., and S. P. Harrison (2001), DIRTMAP: The geological record of dust, Earth Sci. Rev., 54(1–3), 81114.
  • Kubilay, N., T. Cokacar, and T. Oguz (2003), Optical properties of mineral dust outbreaks over the northeastern Mediterranean, J. Geophys. Res., 108(D21), 4666, doi:10.1029/2003JD003798.
  • Levin, Z., J. H. Joseph, and Y. Mekler (1980), Properties of Sharav (Khamsin) dust—Comparison of optical and direct sampling data, J. Atmos. Sci., 37, 182191.
  • Massie, S. T., O. Torres, and S. J. Smith (2004), Total Ozone Mapping Spectrometer (TOMS) observations of increases in Asian aerosol in winter from 1979 to 2000, J. Geophys. Res., 109, D18211, doi:10.1029/2004JD004620.
  • Middleton, N. J. (1986), A geography of dust storms in southwest Asia, J. Clim., 6, 183196.
  • Miller, R. L., and I. Tegen (1998), Climate response to soil dust aerosols, J. Clim., 11(12), 32473267.
  • Miller, R. L., I. Tegen, and J. Perlwitz (2004), Surface radiative forcing by soil dust aerosols and the hydrologic cycle, J. Geophys. Res., 109, D04203, doi:10.1029/2003JD004085.
  • Prasad, A. K., R. P. Singh, and A. Singh (2004), Variability of aerosol optical depth over Indian subcontinent using MODIS data, J. Indian Soc. Rem. Sens., 32(4), 313316.
  • Prasad, A. K.,R. P. Singh, M. Kafatos, and A. Singh (2005a). Effect of the growing population on the air pollution, climatic variability and on the hydrological regime of the Ganga basin, India; in Regional Hydrological Impacts of Climatic Change—Impact Assessment and Decision Making (Proceedings of symposium S6 held during the Seventh IAHS Scientific Assembly at Foz do Iguaçu, Brazil, April 2005),, International Association of Hydrological Sciences, Publ., 295.
  • Prasad, A. K., R. P. Singh, A. Singh, and M. Kafatos (2005b), Seasonal variability of aerosol optical depth over Indian subcontinent,, Analysis of Multi-Temporal Remote Sensing Images, 2005 International Workshop,, IEEE, 3538.
  • Prasad, A. K., R. P. Singh, and M. Kafatos (2006), Influence of coal based thermal power plants on aerosol optical properties in the Indo-Gangetic basin, Geophys. Res. Lett., 33, L05805, doi:10.1029/2005GL023801.
  • Ramanathan, V., and M. V. Ramana (2005), Persistent, widespread and strongly absorbing haze over the Himalayan Foothills and the Indo-Ganges Plains, Pure Appl. Geophys., doi:10.1007/s00024-005-2685-8.
  • Ramanathan, V., P. J. Crutzen, J. T. Kiehl, and D. Rosenfeld (2001), Aerosols, climate, and the hydrologic cycle, Science, 294, 21192124.
  • Remer, L. A., and Y. J. Kaufmanet al. (2005), The MODIS aerosol algorithm, products, and validation, J. Atmos. Sci., 62(4), 947973.
  • Sahoo, A., S. Sarkar, R. P. Singh, M. Kafatos, and M. E. Summers (2005), Declining trend of total ozone column over the northern parts of India, Int. J. Remote Sens., 26(16), 34333440.
  • Salomonson, V. V., and W. L. Barneset al. (1989), MODIS: Advanced facility instrument for studies of the Earth as a system, IEEE Trans. Geosci. Remote Sens., 27, 145153.
  • Shettle, E. P., and R. W. Fenn (1979), Models of aerosols lower troposphere and the effect of humidity variations on their optical properties, AFCRL Tech. Rep. 79 0214,, 100 pp., Air Force Cambridge Res. Lab.,, Hanscom Air Force Base, Mass.
  • Singh, R. P., S. Dey, S. N. Tripathi, V. Tare, and B. Holben (2004), Variability of aerosol parameters over Kanpur, northern India, J. Geophys. Res., 109, D23206, doi:10.1029/2004JD004966.
  • Singh, S., S. Nath, R. Kohli, and R. Singh (2005), Aerosols over Delhi during pre-monsoon months: Characteristics and effects on surface radiation forcing, Geophys. Res. Lett., 32, L13808, doi:10.1029/2005GL023062.
  • Sinyuk, A., O. Torres, and O. Dubovik (2003), Combined use of satellite and surface observations to infer the imaginary part of the refractive index of Saharan dust, Geophys. Res. Lett., 30(2), 1081, doi:10.1029/2002GL016189.
  • Smirnov, A., B. N. Holben, T. F. Eck, O. Dubovik, and I. Slutsker (2000), Cloud-screening and quality control algorithms for the AERONET database, Rem. Sens. Env., 73(3), 337349.
  • Sokolin, I. N., and O. B. Toon (1996), Direct forcing by air borne mineral aerosol, Nature, 381, 681683.
  • Sokolik, I. N., and O. B. Toon (1999), Incorporation of mineralogical composition into models of the radiative properties of mineral aerosol from UV to IR wavelengths, J. Geophys. Res., 104, 94239444.
  • Sokolik, I., A. Andronova, and T. C. Johnson (1993), Complex refractive index of atmospheric dust aerosols, Atmospheric Environment — Part A General Topics, 27A(16), 24952502.
  • Stamnes, K., S.-C. Tsay, W. Wiscombe, and K. Jayaweera (1988), Numerically stable algorithm for discrete-ordinate-method radiative transfer in multiple scattering and emitting layered media, Appl. Opt., 27, 25022509.
  • Tanré, D., M. Herman, and Y. Kaufman (1996), Information on the aerosol size distribution contained in the solar reflected spectral radiances, J. Geophys. Res., 101, 19,04319,060.
  • Tegen, I., and I. Fung (1994), Modeling of mineral dust in the atmosphere: sources, transport, and optical thickness, J. Geophys. Res., 99(D11), 22,89722,914.
  • Tegen, I. (2003), Modeling the mineral dust aerosol cycle in the climate system, Quaternary Science Reviews, 22(18–19), 18211834.
  • Tegen, I., A. A. Lacis, and I. Fung (1996), The influence on climate forcing of mineral aerosols from disturbed soils, Nature, 380(6573), 419422.
  • Tegen, I., M. Werner, S. P. Harrison, and K. E. Kohfeld (2004), Relative importance of climate and land use in determining present and future global soil dust emission, Geophys. Res. Lett., 31, L05105, doi:10.1029/2003GL019216.
  • Wang, J., and S. A. Christopher (2003), Intercomparison between satellite-derived aerosol optical thickness and PM2.5 mass: implications for air quality studies, Geophys. Res. Lett., 30(21), 2095, doi:10.1029/2003GL018174.
  • World Meteorological Organization (WMO) (1983), Radiation commission of IAMAP meeting of experts on aerosols and their climatic effects, Rep. WCP55, Geneva, Switzerland.
  • Yoon, S.-C., S.-W. Kim, J. Kim, B.-J. Sohn, A. Jefferson, S.-J. Choi, D.-H. Cha, and R. J. Weber (2006), Enhanced water vapor in Asian dust layer: Entrainment processes and implication for aerosol optical properties, Atmos. Environ., 40(13), 24092421.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data Used
  5. 3. Optical Characteristics of Dust Storms Over the Indo-Gangetic Plains
  6. 4. Hysplit Air Mass Back-Trajectories and Trans-Continental Sources of Major Dust Storms
  7. 5. Spatial Extent of Major Dust Storms and Health Risks
  8. 6. Enhanced Total Column Water Vapor With Dust Storms and Implications for Climatic Conditions Over the IG Plains
  9. 7. Conclusions
  10. Acknowledgments
  11. References
  12. Supporting Information
FilenameFormatSizeDescription
jgrd13376-sup-0001-t01.txtplain text document1KTab-delimited Table 1.
jgrd13376-sup-0002-t02.txtplain text document1KTab-delimited Table 2.
jgrd13376-sup-0003-t03.txtplain text document2KTab-delimited Table 3.
jgrd13376-sup-0004-t04.txtplain text document1KTab-delimited Table 4.
jgrd13376-sup-0005-t05.txtplain text document1KTab-delimited Table 5.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.