Abstract

Making use of the extensive shipboard and aircraft measurements of aerosol properties over the oceanic regions surrounding the Indian peninsula, under the Integrated Campaign for Aerosols, gases and Radiation Budget (ICARB) field experiment during the premonsoon season (March–May), supplemented with long-term satellite data and chemical transport model simulations, investigations are made of the east–west and north–south gradients in aerosol properties and estimated radiative forcing, over the oceans around India. An eastward gradient has been noticed in most of the aerosol parameters that persisted both within the marine atmospheric boundary layer and above up to an altitude of ~6 km; the gradients being steeper at higher altitudes. It was also noticed that the north–south gradient has contrasting patterns over the Bay of Bengal and the Arabian Sea on the either side of the Indian peninsula. The aerosol-induced atmospheric heating rate increased from a low value of ≤0.1 K day−1 in the southwestern Arabian Sea to as high as ~0.5 K day−1 over the northeastern Bay of Bengal. The simulations of species-resolved spatial gradients have revealed that the observed gradients are the result of the strong modulations by anthropogenic species over the natural gradients, thereby emphasizing the role of human activities in imparting regional forcing. These large spatial gradients in aerosol forcing induced by mostly anthropogenic aerosols over the oceanic regions around the Indian peninsula can potentially affect the regional circulation patterns.

1. Introduction

Atmospheric absorption and surface dimming of solar radiation due to atmospheric aerosols and the resulting heating of the lower atmosphere and cooling at the surface are being increasingly investigated over South Asia in the context of regional and global climate implications (Chung and Zhang 2004; Lau et al. 2006; Satheesh et al. 2008; Lawrence and Lelieveld 2010; Bollasina et al. 2011). Several recent studies have highlighted the far-reaching impacts of absorbing aerosols on regional climate, which include (i) the elevated heat pump (EHP) effect, leading to early onset and intensification of monsoon over the Indo-Gangetic plain (Lau et al. 2006); (ii) global dimming, leading to slowing down of the hydrological cycle (Ramanathan et al. 2001); (iii) reduction in tropical cloudiness (Ackerman et al. 2000); (iv) change in precipitation pattern (Chung et al. 2002); and (v) reduction in winter monsoon rainfall over Bay of Bengal (BoB) due to the aerosol loading over the Arabian Sea (AS) (Krishnamurti et al. 2009). However, most of these climate impact assessments were made mainly based on the Indian Ocean Experiment (INDOEX) data (Menon et al. 2002; Chung et al. 2002; Chung and Zhang 2004) or using the chemical transport model simulations (Lau et al. 2006). While the former did not make spatially well-resolved measurements over the Arabian Sea [but rather followed a north–south transect; Moorthy et al. (2001)] and made very little measurements over the Bay of Bengal, the latter (models) simulated regionally unrealistic aerosol fields that deviated largely from the measurements (Menon et al. 2010; Nair et al. 2012). Subsequently, even though several measurements of aerosols have been made over the oceanic regions around India using short-term cruises, studies addressing the horizontal, vertical, and temporal distributions and radiative impacts with a high degree of spatial resolution remained sparse until the Integrated Campaign for Aerosols, gases and Radiation Budget (ICARB) experiment (Moorthy et al. 2009) and its outcome (e.g., Nair et al. 2008a; Moorthy et al. 2009; Lawrence and Lelieveld 2010).

Spatial heterogeneities in aerosol properties, resulting from the heterogeneous continental outflows into the Arabian Sea and the Bay of Bengal have been extensively examined in recent years under various field campaigns such as INDOEX (Ramanathan et al. 2001), the Arabian Sea Monsoon Experiment (ARMEX) (Moorthy et al. 2005), the Joint Global Ocean Flux Study (JGOFS) (Johansen and Hoffmann 2004), and ICARB (Moorthy et al. 2008). The meridional gradients in aerosol properties over the oceans adjacent to India were investigated using shipborne and satellite measurements by several investigators (e.g., Satheesh et al. 1998; Ramanathan et al. 2001; Satheesh et al. 2006a; Nair et al. 2008a). Lawrence and Lelieveld (2010) extensively reviewed the attempts and the outcomes, including the gap areas. It is well known that the spatial gradients (zonal and meridional) in warming by the greenhouse gases are small because of their homogeneous spatial distribution, whereas the heterogeneity in atmospheric forcing due to aerosols significantly influences the regional climate via dynamical/thermodynamical feedbacks (Matsui and Pielke 2006). Recently, Ming and Ramaswamy (2011) have shown that spatially heterogeneous aerosol forcing could alter the regional-scale circulation pattern through thermodynamic coupling. In the backdrop of all the above, we have analyzed the ICARB data, which provided spatially well-resolved and accurate data on several aerosol parameters from concurrent and collocated measurements over the Indian region during March–May 2006, to examine the spatial gradients in these parameters and its implications on the regional climate.

2. Database and analysis

a. The ICARB

The ICARB was an exhaustive field experiment conducted over the Indian subcontinent and surrounding oceanic regions, following an “integrated and segmented” approach (Moorthy et al. 2008). Extensive measurements of several aerosol properties (optical, physical, and chemical) were made from fixed land and island stations (land segment), on board ship cruises (ocean segment), and using instrumented aircraft (air segment) during 18 March–10 May 2006, and several results have been published (for, e.g., Moorthy et al. 2008, 2009; Nair et al. 2008a,b, 2009, 2010; Satheesh et al. 2008; Lawrence and Lelieveld 2010). The ocean segment of ICARB had two legs: the first (SK 223A) focusing mainly on the Bay of Bengal and the northern Indian Ocean during March–April 2006 and the second (SK 223B) over the Arabian Sea during April–May 2006. Concurrent measurements of mass concentrations, mass size distributions, scattering and absorption coefficients, bulk aerosol collection, spectral aerosol optical depth, columnar water vapor, and vertical profiling of atmospheric thermodynamical parameters were made on board the ship from a specially configured laboratory, following well-documented measurement protocols (Moorthy et al. 2008; Nair et al. 2009). Simultaneous measurements of the vertical distribution of aerosol black carbon (BC) mass concentration and composite aerosol extinction coefficients were made from four coastal base stations using an instrumented aircraft (Babu et al. 2008; Satheesh et al. 2008). The cruise tracks and the base stations for aircraft sorties are shown in Fig. 1. In this study, we have used the shipborne measurements of spectral aerosol optical depth (AOD) using a Microtops sun photometer (Moorthy et al. 2008), and spectral scattering and absorption coefficients using an integrating nephelometer (TSI 3563) and Aethalometer (Magee Scientific AE-31) (Nair et al. 2009). Aircraft measurements of spatial gradients of BC mass concentrations at two altitudes (500 and 1500 m) were made during daytime using another intercompared Aethalometer. A detailed description of these instruments, database, analysis, and error budget are given in earlier papers (Nair et al. 2009; Moorthy et al. 2009; Satheesh et al. 2008) and hence are not repeated.

Fig. 1.

Spatial distribution of AOD at 500 nm over the AS and BoB during the ICARB field experiment (adopted from Moorthy et al. 2009). Cruise track of Sagar Kanya over the BoB and the AS is also shown (solid line). Base stations of aircraft measurements—Bhubaneshwar (BBR), Chennai (CHN), Trivandrm (TVM), Goa (GOA), and Hyderabad (HYD)—are indicated by aircraft symbols.

Fig. 1.

Spatial distribution of AOD at 500 nm over the AS and BoB during the ICARB field experiment (adopted from Moorthy et al. 2009). Cruise track of Sagar Kanya over the BoB and the AS is also shown (solid line). Base stations of aircraft measurements—Bhubaneshwar (BBR), Chennai (CHN), Trivandrm (TVM), Goa (GOA), and Hyderabad (HYD)—are indicated by aircraft symbols.

b. Satellite data

To supplement the temporally limited ICARB data, we have used the Moderate Resolution Imaging Spectroradiometer (MODIS)-derived AOD (for the March–May period) over a decade (2000–2010) and vertical profiles of aerosols from the Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) to examine the persistence and vertical structure of aerosols around India. The MODIS sensors, on board Aqua and Terra satellites, have a long history of retrieving the aerosol properties since 2000. The MODIS algorithm has been extensively refined in recent years to provide higher accuracy of the retrieved AODs, especially over oceans (Remer et al. 2005). We have used the average of level 3 (1° × 1° gridded products: collection 5.1) AODs from both the Aqua and Terra satellites for the period 2000–10 (11 years of data) as supplementary data in this analysis, especially to extend the period of study well beyond ICARB. Uncertainties in the AOD retrieved from MODIS over the ocean and land are well documented, and over oceans it is 0.03 ± 0.05AOD (Remer et al. 2005).

Vertical distribution of aerosols over the Indian region is investigated using the CALIPSO data for the years 2007 and 2008 (Winker et al. 2009). We have averaged the individual nighttime profiles (version 2) into 2.5° × 2.5° to ensure a statistically significant number of observations in each grid cell. To have a high level of confidence in CALIPSO measurements, we have filtered all the profiles whose cloud–aerosol discrimination (CAD) score is less than −75. The mean profiles of CALIPSO and MODIS-derived AOD over each grid in the Bay of Bengal and the Arabian Sea are used to get a realistic extinction profile of aerosols. Since there were not many attempts to validate the extinction profiles and the optical parameters retrieved from the CALIPSO over the Indian region and the surrounding oceanic regions, the extinction profiles of CALIPSO were normalized for the corresponding MODIS-AOD to get the realistic aerosol vertical profiles of the extinction coefficient.

c. GOCART model

The Goddard Chemistry Aerosol Radiation and Transport (GOCART) model is a widely used chemical transport model that simulates the three-dimensional distributions of sulfate, BC, organic carbon (OC), dust (fine and coarse), and sea salt (fine and coarse) aerosols (Chin et al. 2002). The GOCART model uses the meteorological data from the Goddard Earth Observing System Data Assimilation System (GEOS DAS) to simulate the aerosol fields. The GOCART simulates the aerosols at a spatial resolution of 2° × 2.5° or 1° × 1° with 20–55 vertical sigma levels. We have used 2° × 2.5° AODs of sulfate, BC, OC, dust, and sea salt aerosols over the Indian region obtained from the Giovanni website of the Goddard Earth Sciences Data and Information Services Center (GES DISC). More details on the model configuration, parameterizations, and validation are available in Chin et al. (2002).

3. Results and discussions

a. ICARB: Regional gradients in aerosol properties

During the campaign period, extensive measurements were made along the longitudes separated by 1° latitude bins. Instantaneous AOD measurements made during the ICARB were used to generate the spatial image of AOD over the oceans around India. The spatial distribution of AOD (at 500 nm) is shown in Fig. 1. In general, spatial variations are large, with AOD values ranging from as low as 0.05 in the central Arabian Sea to as high as 1.0 in the northern Bay of Bengal. High AODs are seen over the northern Bay of Bengal compared to those over the southeastern and northern Arabian Sea and the central and southern Bay of Bengal, whereas low AOD regions were seen over the southeastern Bay of Bengal and the central Arabian Sea. A pocket of high AODs seen over the northern Bay of Bengal, which lasted only for few days, was associated with the change in a large-scale circulation pattern and wind convergence over that region (Aloysius et al. 2008).

The latitudinal and longitudinal variations of columnar AOD from the sun photometric measurements and aerosol extinction coefficients at 550 nm (in the marine atmospheric boundary layer) estimated from the concurrent measurements of scattering and absorption coefficients and BC mass concentrations within and above the marine atmospheric boundary layer (based on the aircraft measurements) are shown in Fig. 2 from top to bottom. Longitudinal variations of BC mass concentrations made at altitudes of 500 and 1500 m above the ground level, obtained by combining the entire data collected on board the aircraft (from the four coastal bases shown in Fig. 2) into as far as ~500 km over the adjoining oceanic regions, are shown in the top panel of Fig. 2. Notwithstanding the clustering of data points over ocean with a gap over the peninsular landmass where no airborne measurements were made, a clear eastward gradient is revealed within (500 m) and above (1500 m) the marine atmospheric boundary layer, with the concentrations increasing from west to east. Despite the averaging over a wide latitudinal range, both the extinction coefficient (within the atmospheric boundary layer) and AOD (columnar) also showed conspicuous longitudinal gradients, increasing eastward, as represented by the lines joining the points in the middle panel of the same figure. Nearly threefold to fourfold increase occurs in the columnar AOD and near-surface extinction coefficients as we move from ~57° to 93°E over the oceanic regions (separated by the Indian peninsula). Such a feature has not been reported from any of the earlier field experiments primarily because of the absence of such exhaustive spatial coverage. In the bottom panel of Fig. 2, we show the latitudinal gradients of AOD and extinction coefficients separately for the Arabian Sea and the Bay of Bengal, in which the measurements are averaged over the respective longitudes covered by the cruise. Even though some of the earlier cruises (such as INDOEX) have examined the latitudinal gradients in aerosol properties (Moorthy et al. 2001; Ramanathan et al. 2001) over the western Indian Ocean, the ICARB experiment clearly delineated for the first time the gradients in all three dimensions: latitudinal, longitudinal, and vertical. This is important in the backdrop of several modeling studies because the gradients in aerosol loading over the oceanic regions around India will have a large influence on the circulation pattern and precipitation over this region (Chung et al. 2002; Chung and Zhang 2004).

Fig. 2.

Longitudinal and latitudinal variations of aerosol properties over the AS and the BoB measured during ICARB. (a) Longitudinal variation of BC mass concentration at 500 and 1500 m measured on board aircraft, (b) longitudinal variation of AOD and extinction coefficient measured on board a ship, and (c) latitudinal variation of AOD and extinction coefficient measured on board a ship.

Fig. 2.

Longitudinal and latitudinal variations of aerosol properties over the AS and the BoB measured during ICARB. (a) Longitudinal variation of BC mass concentration at 500 and 1500 m measured on board aircraft, (b) longitudinal variation of AOD and extinction coefficient measured on board a ship, and (c) latitudinal variation of AOD and extinction coefficient measured on board a ship.

The optical properties of aerosols, and hence the radiative forcing, strongly depend on the ambient relative humidity (RH) (Im et al. 2001; Jeong et al. 2007). So, it is important to understand the contribution of hygroscopic growth of particles to the observed gradients in aerosol optical properties. To examine this, we have analyzed the columnar water vapor content measured using a sun photometer and the RH measurements made on board the ship during ICARB 2006. The longitudinal variations of water vapor and RH (thus obtained) are shown in Fig. 3. Even though the column water vapor and RH showed an eastward increasing trend west of 80°E, the variations are not one to one; on several occasions the variations are out of phase, indicating a change in the water vapor scale height. The frequency of occurrence of RH values during ICARB peaks at ~72.5%, and RH values greater than 85% were less than 2% of the total number of observations. The longitudinal variation of RH varied from 66% to 78% with a mean value of 72 ± 3%, and the columnar water vapor content varied from 1.8 to 4.0 cm with a mean value of 2.6 ± 0.6 cm. In general, low values are seen over the far west (<65°E) and east (>90°E). Nevertheless, to estimate the effect of this RH change on the observed longitudinal gradients in aerosol optical depth, we have estimated the change in optical depth attributable to RH [ΔAOD(RH)] for the aerosol model derived from the ICARB measurements (Moorthy et al. 2009). More details about the aerosol model developed using the physical and optical properties of aerosols are described in Moorthy et al. (2009) and hence not repeated here. The longitudinal variation of RH (averaged over 0°–22°N) from 66% to 78% leads to an AOD increase of 0.02 only. Hence, the contribution of relative humidity to the threefold to fourfold increase in AOD observed in the present study from west to east is negligibly small. From the longitudinal variation of AOD (Fig. 2b) and the variation in AOD estimated using the observed RH variation (Fig. 3) it is clearly indicated that the contribution to the observed longitudinal gradients in AOD due to hygroscopic growth of aerosols is inconspicuous.

Fig. 3.

Longitudinal variations of columnar water vapor and RH measured during ICARB, and change in AOD attributable to RH estimated for the ICARB aerosol model.

Fig. 3.

Longitudinal variations of columnar water vapor and RH measured during ICARB, and change in AOD attributable to RH estimated for the ICARB aerosol model.

b. Persistent feature: MODIS data (2000–10)

Despite its inherent advantages and measurements at a high spatial resolution carried out during a short yet finite duration, shipborne measurements are not fully adequate for delineating the spatial and temporal heterogeneities, because these get coupled during the finite length of the campaign period. As such, continuous (time series) measurements of spectral AOD were being made from two island stations, Port Blair (11.64°N, 92.71°E) and Minicoy (8.2°N, 73°E), as part of a long-term monitoring program under the Aerosol Radiative Forcing over India (ARFI) project, and these data provided information on the trend in the temporal variations of aerosols during ICARB. These island measurements supplemented with the spatial data from the ship (Nair et al. 2008a; Moorthy et al. 2009) were useful to delineate the temporal variations. Nevertheless, the spatially resolved measurements on board the ship were available only for year 2006. With a view to examine the long-term nature of the gradients revealed during the ICARB (to examine whether the gradients were a persistent feature or specific to 2006), we have used the AOD (at 550 nm) derived from the MODIS for the premonsoon (March–May) period of 2000–10. To validate the MODIS-retrieved AOD over the region with the in situ measurements, we have used shipborne measurements of the AOD made during several campaigns over the oceanic regions surrounding the Indian peninsula during 2003–10. Figure 4 depicts the scatterplot of AOD between the shipborne measurements using a calibrated Microtops sun photometer with concurrent MODIS-derived AOD values. This exercise involved ~171 daily mean AODs from the shipborne measurements (the total number of individual measurements exceeded more than 9000), spanning over different seasons over the Arabian Sea, the Bay of Bengal, and the northern Indian Ocean, and it is perhaps the most exhaustive compilation over South Asia. Even though these measurements were made during different seasons at multiple locations over oceans, the very good agreement between the measured and MODIS-derived AOD (with a slope of 0.95) values demonstrate the usefulness of MODIS data for examining the spatial gradients.

Fig. 4.

Intercomparison of AOD measured during various ship cruises using a Microtops sunphotometer with MODIS-derived AOD. Shipborne measurements include data collected during (i) the second phase of ARMEX (ARMEX II), 2003 (Moorthy et al. 2005); (ii) ARMEX IIA, 2005 (Nair et al. 2008c); (iii) ICARB-2006, (Moorthy et al. 2009); (iv) Winter-ICARB, 2009 (Moorthy et al. 2010); (v) unpublished data (collected during the very short cruise experiments over the BoB). Shown is the linear best fit between MODIS and Microtops AODs (dotted line) and the ideal 1:1 line (solid line).

Fig. 4.

Intercomparison of AOD measured during various ship cruises using a Microtops sunphotometer with MODIS-derived AOD. Shipborne measurements include data collected during (i) the second phase of ARMEX (ARMEX II), 2003 (Moorthy et al. 2005); (ii) ARMEX IIA, 2005 (Nair et al. 2008c); (iii) ICARB-2006, (Moorthy et al. 2009); (iv) Winter-ICARB, 2009 (Moorthy et al. 2010); (v) unpublished data (collected during the very short cruise experiments over the BoB). Shown is the linear best fit between MODIS and Microtops AODs (dotted line) and the ideal 1:1 line (solid line).

The mean latitudinal and longitudinal variations of MODIS-derived AOD (at 550 nm) for the premonsoon (March–May) season are shown in Fig. 5 in the top and bottom panels, respectively, for the 11-yr period. Each data point is the average of more than 10 000 individual AOD values, and the vertical bars are the standard deviations, depicting the interannual variations of the AOD. Examination of the bottom panel reveals that the latitudinal gradients, though generally show a northward increase in AOD, are quite steeper over the Bay of Bengal compared to that over the Arabian Sea. In the case of longitudinal variations, AOD values increases gradually from west to east with a gradient of ~0.08 increase in every 10° longitude. Even though the spatial pattern of AOD variations remains the same, longitudinal gradients are weaker in MODIS data compared to ICARB observations, probably because of the averaging over multiple years. It is seen that even though there exists large interannual variabilities in the gradient of AOD, the west–east increasing pattern remained consistent throughout the last decade. Based on the four years of MODIS data, Satheesh et al. (2006a) have reported that north–south gradients in AOD over the Arabian Sea maximize during summer monsoon (June–September) in association with the dust transport from the west Asian regions over the northern Arabian Sea. In contrast, over the Bay of Bengal, the northward increasing trend in AOD is seen in all the seasons (Satheesh et al. 2006b). While all these pertained to the columnar AOD, its vertical structure remained unexplored over this region and it was not clear whether the gradients persist above the atmospheric boundary layer and if so, were they steeper or shallower?

Fig. 5.

Latitudinal and longitudinal variations of AOD at 550 nm derived from the MODIS measurements for the period 2000–10. Vertical bars on the mean indicate the standard deviation of the data.

Fig. 5.

Latitudinal and longitudinal variations of AOD at 550 nm derived from the MODIS measurements for the period 2000–10. Vertical bars on the mean indicate the standard deviation of the data.

c. Vertical structure

The altitude variation has been examined using the vertical profiles of aerosol extinction retrieved from the Cloud–Aerosol Lidar with Orthogonal Polarization (CALIOP) on board CALIPSO. We have estimated the layer AODs for 0–1, 1–2, and 2–3 km and averaged along the latitudes (0°–22°) for examining the longitudinal variations as shown in the top panel of Fig. 6 and along the longitudes (50°–100°E) for examining the latitudinal variations (bottom panel of Fig. 6). It is seen that, while aerosol extinction increased with latitudes (bottom panel of Fig. 6) at all three altitude bins, which is in line with the columnar features (Fig. 2), the longitudinal gradients in aerosol extinction (increasing from west to east) becomes steeper above the boundary layer. This indicated the presence of higher abundance of aerosols in the lower free-tropospheric altitudes as we moved eastward, especially over the far eastern Bay of Bengal.

Fig. 6.

Latitudinal and longitudinal variations of layer AOD (weighted with MODIS AOD) derived from CALIPSO at altitudes (i) 0–1, (ii) 1–2, and (iii) 2–3 km for the period 2007–08. Values given in the labels indicate the corresponding regression coefficient of the linear fit to the gradients at three heights.

Fig. 6.

Latitudinal and longitudinal variations of layer AOD (weighted with MODIS AOD) derived from CALIPSO at altitudes (i) 0–1, (ii) 1–2, and (iii) 2–3 km for the period 2007–08. Values given in the labels indicate the corresponding regression coefficient of the linear fit to the gradients at three heights.

Based on airborne lidar measurements during ICARB, Satheesh et al. (2008) have reported an increased aerosol extinction in the 2–4-km-altitude region over the east coast of peninsular India compared to the west coast. They also have reported a northward increase in the extinction associated with “elevated layers” of aerosols, the altitude and magnitude of which increased from south to north Satheesh et al. (2008). Similar gradients (eastward and northward) over Bay of Bengal have been observed during winter season also (Sreekanth et al. 2011). Analyzing the CALIPSO profiles over the Arabian Sea, Rajeev et al. (2010) have also reported a south–north gradient in aerosol vertical distribution during the premonsoon and summer monsoon season.

d. Aerosol radiative forcing and heating rates

Making use of extensive measurements of aerosol optical properties [aerosol optical depth, single scattering albedo (SSA), and asymmetry parameter] made on board ship and aircraft (Table 1), Moorthy et al. (2009) have estimated the spatial distribution of diurnally averaged clear-sky aerosol radiative forcing at the top of the atmosphere and surface over the oceanic regions around India. The aerosol radiative forcing is estimated from the radiative fluxes simulated using the Santa Barbara Discrete Ordinate Radiative Transfer model (DISORT) Atmospheric Radiative Transfer (SBDART) model with and without aerosol conditions. For both the cases, we have used measured meteorological conditions such as the vertical profiles of temperature, pressure and humidity, column-integrated values of ozone, and water vapor. For aerosol simulations, measured values of spectral AOD and SSA were used along with semiempirically estimated phase functions as shown in Table 1. Aerosol phase functions were estimated from the assumed aerosol chemical model, which reproduces the measured optical properties (spectral AOD and SSA) exactly. More information regarding the estimation of aerosol radiative forcing using the measured aerosol parameters is described in Moorthy et al. (2009). The spatial gradients seen in the aerosol properties result in similar gradients in the aerosol radiative forcing (ARF): the aerosol-induced atmospheric heating and surface cooling. The heating rate (K day−1) due to aerosols was estimated as the ratio of atmospheric radiative forcing (ARFatm) to the layer depth of (ΔP) multiplied by the dry adiabatic lapse rate (Moorthy et al. 2009). Based on the vertical profiling during the campaign, it is found that most of the aerosols are confined in the first 3–4-km region (Satheesh et al. 2009) and hence we have used a steady value of ΔP = 300 hPa for the entire analysis.

Table 1.

Instruments used and parameters measured during ICARB.

Instruments used and parameters measured during ICARB.
Instruments used and parameters measured during ICARB.

The average latitudinal and longitudinal variations of aerosol-induced heating rate and surface forcing (ARFsur) are shown in Fig. 7. Over the Bay of Bengal, the aerosol heating rate values increased northward from 0.1 K day−1 at 9°N to as high as 0.5 K day−1 at 19°N followed by a weak decrease. A similar trend is observed in the magnitude of surface cooling also. In contrast to the Bay of Bengal, the atmospheric heating rate over the Arabian Sea was lower in magnitude and showed an opposite gradient (increases southward) because of the higher abundance of absorbing aerosols over the southern latitudes. In contrast, however, a consistent and systematic increasing trend is seen longitudinally in the atmospheric heating rate and surface cooling rate. This, thus, creates a very interesting situation above the Indian peninsula in which the aerosol-induced atmospheric heating (and surface cooling) increases steadily from ~58° to 93°E, which showed an opposite latitudinal structure on either side of the peninsular landmass. The heating rate gradient reversed, being southward to the west and northward to the east of the peninsula. This is believed to modify the lower-level circulation system with implications to the regional climate.

Fig. 7.

(bottom) Latitudinal and (top) longitudinal variations of ARFsur and aerosol-induced atmospheric heating rate over the AS and BoB during ICARB.

Fig. 7.

(bottom) Latitudinal and (top) longitudinal variations of ARFsur and aerosol-induced atmospheric heating rate over the AS and BoB during ICARB.

As such, the large-scale heterogeneity in aerosol-induced heating rate might lead to the meso-/synoptic-scale perturbations to the atmospheric circulations (Ming and Ramaswamy 2011) by modulating the horizontal pressure gradient forcing. Matsui and Pielke (2006) have also reported that spatial heterogeneity in aerosol forcing could influence the atmospheric circulations via modulating the horizontal pressure gradients. This could lead to altering the regional climate through the dynamical feedbacks (Chung et al. 2002; Menon et al. 2002; Lau et al. 2006). Recently, Bollasina et al. (2011) have shown that spatially heterogeneous anthropogenic aerosol forcing dynamically induce circulation changes and that could explain the decrease in monsoon precipitation over central-northern India. Slowdown of tropical circulation in association with anthropogenic forcing was reported earlier by Vecchi et al. (2006). Following Chung and Ramanathan (2006), the meridional gradient in aerosol forcing and its influence on the weakening of sea surface temperature gradient over the Indian Ocean and north of it led to weakening of the monsoon system. Present investigations have shown strong eastward gradients in aerosol-induced atmospheric heating not only in the marine atmospheric boundary layer but in the free-tropospheric heights as well. Based on the airborne lidar measurements of extinction profiles over the Arabian Sea and Bay of Bengal, Satheesh et al. (2008) have reported a meridional gradient of heating rate over the Indian region. Compared to the longitudinal gradients, the latitudinal gradients are stronger over the Bay of Bengal, which formerly persisted over a much wider spatial extent. Large-scale climate simulations incorporate these realistic gradients in three dimensions to understand the regional circulation pattern and rainfall.

e. Natural or anthropogenic?

From previous discussions, it is clear that aerosol properties and their radiative impacts over the oceanic regions around the Indian peninsula depict systematic and significant longitudinal gradients, superposed on the well-known latitudinal gradients. This results in a significant eastward gradient in the diabatic heating of the lower atmosphere (up to 6 km) and cooling at the surface. However, it remained unresolved whether these gradients are associated with natural or anthropogenic processes. To investigate this, we have analyzed the GOCART model simulations for the period March–May 2006 (ICARB period). As a first step, we examined the AOD (at 550 nm) simulated by GOCART against those retrieved from MODIS, in Fig. 8. It clearly emerges that (i) the nature of spatial variations in the AOD agree fairly well between GOCART and MODIS and (ii) the GOCART-simulated AODs are considerably lower than the MODIS values (by a factor of 4 in the western Arabian Sea, decreasing down to a factor of 2.5 in the eastern Bay of Bengal). Earlier studies have also reported the significant underestimation of aerosol properties over the Indian region by chemical transport models and regional and global climate models (Menon et al. 2010; Nair et al. 2012).

Fig. 8.

Intercomparison of longitudinal variations of GOCART AOD with MODIS AOD at 550 nm averaged over 0°–22°N latitudes.

Fig. 8.

Intercomparison of longitudinal variations of GOCART AOD with MODIS AOD at 550 nm averaged over 0°–22°N latitudes.

To delineate the possible source mechanisms behind the longitudinal and latitudinal gradients, we have analyzed the columnar AOD of the different aerosol species, such as black carbon, organic carbon, sulfate, dust, and sea salt simulated using GOCART for the premonsoon period of 2006. The latitudinal variations, averaged for 50°–100°E, of aerosol species are shown in the bottom panel of Fig. 9. Though all the aerosol species, except sea salt, showed a southward decreasing trend (since most of the aerosol sources are located on the continental landmass), the rate of decrease varied with species: dust and organics depicted steeper gradients. Even though several earlier cruise measurements also reported a north–south decreasing trend in composite aerosol properties over this region (Moorthy et al. 2001; Ramanathan et al. 2001; Satheesh et al. 2006b; Nair et al. 2008a, 2010), the gradients of different aerosol species, except BC, are still not examined in detail (Nair et al. 2010). Based on the ICARB and INDOEX measurements, several investigators have reported a significant dominance of anthropogenic aerosols over the oceans around India (Ramanathan et al. 2001; Kumar et al. 2008; Nair et al. 2010). Assuming that most of the BC, OC, and sulfate are anthropogenic and that dust and sea salt are natural, it emerges that the anthropogenic aerosols reinforce the latitudinal gradient induced by the dust (natural) aerosols. The longitudinal variations of sea salt and dust (natural components) AOD are shown in top panel of Fig. 9 and AOD of BC, OC, and sulfate (mostly anthropogenic components) are shown in the middle panel. The longitudinal variations of natural (sea salt and dust) aerosols are distinctly different from that of the anthropogenic (BC, OC, and sulfate) species as well as what emerged from the measurements (Figs. 2, 5, 6, 7, 8). In general, the longitudinal variations of the anthropogenic aerosols well resemble the pattern seen from the measurements of composite aerosol properties. Notwithstanding this, while sulfate aerosols show very high values over the Indian peninsula, BC and OC concentrations peak in the East Asian region probably because of the open biomass and residual burning activities being stronger there. The longitudinal variations of the anthropogenic species appear to dominate the gradients of natural (by dust aerosols) components, so that the longitudinal gradients seen in composite aerosol properties are in general agreement with those of anthropogenic components. It appears that the anthropogenic activities during the past few decades reversed the longitudinal gradients and reinforced the latitudinal gradients.

Fig. 9.

Longitudinal (averaged over 0°–22°N) and latitudinal (averaged over 50°–100°E) variations of AOD simulated using GOCART model for different chemical species. (a) Longitudinal variation of AOD of sea salt and dust; (b) longitudinal variation of AOD of BC, OC, and sulfate; and (c) latitudinal variations of AOD of dust, sea salt, BC, OC, and sulfate.

Fig. 9.

Longitudinal (averaged over 0°–22°N) and latitudinal (averaged over 50°–100°E) variations of AOD simulated using GOCART model for different chemical species. (a) Longitudinal variation of AOD of sea salt and dust; (b) longitudinal variation of AOD of BC, OC, and sulfate; and (c) latitudinal variations of AOD of dust, sea salt, BC, OC, and sulfate.

4. Conclusions

Spatially resolved measurements of optical and physical properties of aerosols, made over the oceanic regions around India during premonsoon period, revealed strong gradients in the latitudinal and longitudinal distribution of aerosol loading within the marine atmospheric boundary layer, free troposphere, and in the entire column. Supplementing these in situ observations with long-term multisatellite data, the three-dimensional distribution of aerosol-induced radiative effects is examined. This study has brought out several interesting findings that have strong implications to regional climate:

  • Significant gradients are seen in all the climate-sensitive parameters of aerosols (scattering coefficient, absorption coefficient, columnar AOD, etc.). While all these parameters increased steeply with latitude over the Bay of Bengal, the variations were highly subdued, in general, over the Arabian Sea and in some instances showed an opposite trend.

  • An increase in the AOD at the rate of ~0.08 per every 10° longitude is observed from west to east across the domain from ~58° to 100°E. This gradient, superposed over the latitudinal variation, has been a persistent feature, as revealed by the long-term (11 years) satellite data.

  • Over the Bay of Bengal, the aerosol heating rate values increased northward from 0.1 K day−1 at 9°N to as high as 0.5 K day−1 at 19°N followed by a weak decrease. Aerosol-induced atmospheric heating rates (surface dimming) increased eastward from the southwestern Arabian Sea to the northeastern Bay of Bengal, across the peninsula landmass.

  • The GOCART simulations of species-resolved spatial gradients have revealed that the observed gradients are a result of the strong modulations by anthropogenic species over the natural gradients, thereby emphasizing the role of human activities in imparting regional forcing.

These gradients in atmospheric heating or surface cooling could significantly influence the synoptic circulations over the regions when the winds are in transition from northeasterly to southwesterly over the oceanic regions around India. Extensive studies using the regional climate models are essential to resolve the effect of meridional and zonal gradients in aerosol-induced atmospheric warming and surface cooling on the regional circulation pattern and on large-scale monsoon.

Acknowledgments

The study was carried out as a part of the ICARB project of ISRO-GBP. We acknowledge NASA for providing MODIS aerosol data online (at ftp://ladsftp.nascom.nasa.gov). The CALIPSO data were obtained from the NASA Langley Research Center Atmospheric Sciences Data Center. Analyses and visualizations of GOCART data used in this article were produced with the Giovanni online data system, developed and maintained by NASA GES DISC.

REFERENCES

REFERENCES
Ackerman
,
A. S.
,
O. B.
Toon
,
D. E.
Stevens
,
A. J.
Heymsfield
,
V.
Ramanathan
, and
E. J.
Welton
,
2000
:
Reduction of tropical cloudiness by soot
.
Science
,
288
,
1042
1047
.
Aloysius
,
M.
,
M.
Mohan
,
S. S.
Babu
,
V. S.
Nair
,
K.
Parameswaran
, and
K. K.
Moorthy
,
2008
:
Influence of circulation parameters on the AOD variations over the Bay of Bengal during ICARB
.
J. Earth Syst. Sci.
,
117
,
353
360
.
Babu
,
S. S.
,
S. K.
Satheesh
,
K. K.
Moorthy
,
C. B. S.
Dutt
,
V. S.
Nair
,
D. P.
Alappattu
, and
P. K.
Kunhikrishnan
,
2008
:
Aircraft measurements of aerosol black carbon from a coastal location in the north-east part of peninsular India during ICARB
.
J. Earth Syst. Sci.
,
117
,
263
271
.
Bollasina
,
M.
,
Y.
Ming
, and
V.
Ramaswamy
,
2011
:
Anthropogenic aerosols and the weakening of the South Asian summer monsoon
.
Science
,
334
,
502
505
,
doi:10.1126/science.1204994
.
Chin
,
M.
, and
Coauthors
,
2002
:
Tropospheric aerosol optical thickness from the GOCART model and comparisons with satellite and sun photometer measurements
.
J. Atmos. Sci.
,
59
,
461
483
.
Chung
,
C. E.
, and
G. J.
Zhang
,
2004
:
Impact of absorbing aerosol on precipitation: Dynamic aspects in association with convective available potential energy and convective parameterization closure and dependence on aerosol heating profile
.
J. Geophys. Res.
,
109
,
D22103
,
doi:10.1029/2004JD004726
.
Chung
,
C. E.
, and
V.
Ramanathan
,
2006
:
Weakening of north Indian SST gradients and the monsoon rainfall in India and the Sahel
.
J. Climate
,
19
,
2036
2045
.
Chung
,
C. E.
,
V.
Ramanathan
, and
J.
Kiehl
,
2002
:
Effects of the South Asian absorbing aerosol haze on the northeast monsoon and surface–air heat exchange
.
J. Climate
,
15
,
2462
2476
.
Im
,
J. S.
,
V. K.
Saxena
, and
B. N.
Wenny
,
2001
:
An assessment of hygroscopic growth factors for aerosols in the surface boundary layer for computing direct radiative forcing
.
J. Geophys. Res.
,
106
(D17),
20 213
20 224
.
Jeong
,
M.-J.
,
Z.
Li
,
E.
Andrews
, and
S.-C.
Tsay
,
2007
:
Effect of aerosol humidification on the column aerosol optical thickness over the Atmospheric Radiation Measurement Southern Great Plains site
.
J. Geophys. Res.
,
112
,
D10202
,
doi:10.1029/2006JD007176
.
Johansen
,
A. M.
, and
M. R.
Hoffmann
,
2004
:
Chemical characterization of ambient aerosol collected during the northeast monsoon season over the Arabian Sea: Anions and cations
.
J. Geophys. Res.
,
109
,
D05305
,
doi:10.1029/2003JD004111
.
Krishnamurti
,
T. N.
,
A.
Chakraborty
,
A.
Martin
,
W. K.
Lau
,
K.-M.
Kim
,
Y.
Sud
, and
G.
Walker
,
2009
:
Impact of Arabian Sea pollution on the Bay of Bengal winter monsoon rains
.
J. Geophys. Res.
,
114
,
D06213
,
doi:10.1029/2008JD010679
.
Kumar
,
A.
,
A. K.
Sudheer
, and
M. M.
Sarin
,
2008
: Chemical characteristics of aerosols in MABL of Bay of Bengal and Arabian Sea during spring inter-monsoon: A comparative study. J. Earth Syst. Sci.,117, 325–332.
Lau
,
K. M.
,
M. K.
Kim
, and
K. M.
Kim
,
2006
:
Asian summer monsoon anomalies induced by aerosol direct forcing: The role of the Tibetan Plateau
.
Climate Dyn.
,
26
,
855
864
,
doi:10.1007/s00382-006-0114-z
.
Lawrence
,
M. G.
, and
J.
Lelieveld
,
2010
:
Atmospheric pollutant outflow from southern Asia: A review
.
Atmos. Chem. Phys.
,
10
,
11 017
11 096
,
doi:10.5194/acp-10-11017-2010
.
Matsui
,
T.
, and
R. A.
Pielke
Sr.
,
2006
:
Measurement-based estimation of the spatial gradient of aerosol radiative forcing
.
Geophys. Res. Lett.
,
33
,
L11813
,
doi:10.1029/2006GL025974
.
Menon
,
S.
,
J.
Hansen
,
L.
Nazarenko
, and
Y.
Luo
,
2002
:
Climate effects of black carbon aerosols in China and India
.
Science
,
297
,
2250
2253
.
Menon
,
S.
,
D.
Koch
,
G.
Beig
,
S.
Sahu
,
J.
Fasullo
, and
D.
Orlikowski
,
2010
:
Black carbon aerosols and the third polar ice cap
.
Atmos. Chem. Phys.
,
10
,
4559
4571
.
Ming
,
Y.
and
V.
Ramaswamy
,
2011
: A model investigation of aerosol-induced changes in tropical circulation. J. Climate,24, 5125–5133.
Moorthy
,
K. K.
,
A.
Saha
,
B. S. N.
Prasad
,
K.
Niranjan
,
D.
Jhurry
, and
P. S.
Pillai
,
2001
:
Aerosol optical depths over peninsular India and adjoining oceans during the INDOEX campaigns: Spatial, temporal, and spectral characteristics
.
J. Geophys. Res.
,
106
(D22),
28 539
28 554
.
Moorthy
,
K. K.
,
S. S.
Babu
, and
S. K.
Satheesh
,
2005
:
Aerosol characteristics and radiative impacts over the Arabian Sea during the intermonsoon season: Results from ARMEX field campaign
.
J. Atmos. Sci.
,
62
,
192
206
.
Moorthy
,
K. K.
,
S. K.
Satheesh
,
S. S.
Babu
, and
C. B. S.
Dutt
,
2008
:
Integrated Campaign for Aerosols, Gases and Radiation Budget (ICARB): An overview
.
J. Earth Syst. Sci.
,
117
,
243
262
.
Moorthy
,
K. K.
,
V. S.
Nair
,
S. S.
Babu
, and
S. K.
Satheesh
,
2009
:
Spatial and vertical heterogeneities of aerosol radiative forcing over the oceanic regions surrounding the Indian peninsula: Climate implications
.
Quart. J. Roy. Meteor. Soc.
,
135
,
2131
2145
.
Moorthy
,
K. K.
, and
Coauthors
,
2010
:
Optical and physical characteristics of Bay of Bengal aerosols during W-ICARB: Spatial and vertical heterogeneities in the marine atmospheric boundary layer and in the vertical column
.
J. Geophys. Res.
,
115
,
D24213
,
doi:10.1029/2010JD014094
.
Nair
,
V. S.
,
S. S.
Babu
, and
K. K.
Moorthy
,
2008a
: Aerosol characteristics in the marine atmospheric boundary layer over the Bay of Bengal and Arabian Sea during ICARB: Spatial distribution and latitudinal and longitudinal gradients. J. Geophys. Res.,113, D15208,
doi:10.1029/2008JD009823
.
Nair
,
V. S.
,
S. S.
Babu
, and
K. K.
Moorthy
,
2008b
:
Spatial distribution and spectral characteristics of aerosol single scattering albedo over the Bay of Bengal inferred from shipborne measurements
.
Geophys. Res. Lett.
,
35
,
L10806
,
doi:10.1029/2008GL033687
.
Nair
,
V. S.
,
S. S.
Babu
,
S. K.
Satheesh
, and
K. K.
Moorthy
,
2008c
:
Effect of sea surface winds on marine aerosol characteristics and impacts on longwave radiative forcing over the Arabian Sea
.
Atmos. Chem. Phys. Discuss.
,
8
,
1
45
.
Nair
,
V. S.
,
K. K.
Moorthy
,
S. S.
Babu
, and
S. K.
Satheesh
,
2009
:
Optical and physical properties of atmospheric aerosols over the Bay of Bengal during ICARB
.
J. Atmos. Sci.
,
66
,
2640
2658
.
Nair
,
V. S.
,
S. K.
Satheesh
,
K. K.
Moorthy
,
S. S.
Babu
,
P. R.
Nair
, and
S. K.
George
,
2010
:
Surprising observation of large anthropogenic aerosol fraction over the “near-pristine” southern Bay of Bengal: Climate implications
.
J. Geophys. Res.
,
115
,
D21201
,
doi:10.1029/2010JD013954
.
Nair
,
V. S.
,
F.
Solmon
,
F.
Giorgi
,
L.
Mariotti
,
S. S.
Babu
, and
K. K.
Moorthy
,
2012
:
Simulation of South Asian aerosols for regional climate studies
.
J. Geophys. Res.
,
117
,
D04209
,
doi:10.1029/2011JD016711
.
Rajeev
,
K.
,
K.
Parameswaran
,
B. V.
Thampi
,
M. K.
Mishra
,
A. K. M.
Nair
, and
S.
Meenu
,
2010
:
Altitude distribution of aerosols over southeast Arabian Sea coast during pre-monsoon season: Elevated layers, long-range transport and atmospheric radiative heating
.
Atmos. Environ.
,
44
,
2597
2604
.
Ramanathan
,
V.
, and
Coauthors
,
2001
:
Indian Ocean Experiment: An integrated analysis of the climate forcing and effects of the great Indo-Asian haze
.
J. Geophys. Res.
,
106
(D22),
28 371
28 398
.
Remer
,
L. A.
, and
Coauthors
,
2005
:
The MODIS aerosol algorithm, products and validation
.
J. Atmos. Sci.
,
62
,
947
973
.
Satheesh
,
S. K.
,
K. K.
Moorthy
, and
B. V. K.
Murthy
,
1998
: Spatial gradients in aerosol characteristics over the Arabian Sea and Indian Ocean. J. Geophys. Res.,103 (D20), 26 183–26 192.
Satheesh
,
S. K.
,
K. K.
Moorthy
,
Y. J.
Kaufman
, and
T.
Takemura
,
2006a
:
Aerosol optical depth, physical properties and radiative forcing over the Arabian Sea
.
Meteor. Atmos. Phys.
,
91
,
45
62
,
doi:10.1007/s00703-004-0097-4
.
Satheesh
,
S. K.
,
J.
Srinivasan
, and
K. K.
Moorthy
,
2006b
:
Spatial and temporal heterogeneity in aerosol properties and radiative forcing over Bay of Bengal: Sources and role of aerosol transport
.
J. Geophys. Res.
,
111
, D08202,
doi:10.1029/2005JD006374
.
Satheesh
,
S. K.
,
K. K.
Moorthy
,
S. S.
Babu
,
V.
Vinoj
, and
C. B. S.
Dutt
,
2008
:
Climate implications of large warming by elevated aerosols over India
.
Geophys. Res. Lett.
,
35
, L19809,
doi:10.1029/2008GL034944
.
Satheesh
,
S. K.
, and
Coauthors
,
2009
:
Vertical structure and horizontal gradients of aerosol extinction coefficients over coastal India inferred from airborne lidar measurements during the Integrated Campaign for Aerosols, Gases and Radiation Budget (ICARB) field campaign
.
J. Geophys. Res.
,
114
,
D05204
,
doi:10.1029/2008JD011033
.
Sreekanth
,
V.
,
K. K.
Moorthy
,
S. K.
Satheesh
,
S. S.
Babu
,
V. S.
Nair
, and
K.
Niranjan
,
2011
:
Airborne measurements of aerosol scattering properties above the MABL over Bay of Bengal during W_ICARB—Characteristics and spatial gradients
.
Ann. Geophys.
,
29
,
895
908
.
Vecchi
,
G. A.
,
B. J.
Soden
,
A. T.
Wittenberg
,
I. M.
Held
,
A.
Leetmaa
, and
M. J.
Harrison
,
2006
:
Weakening of tropical Pacific atmospheric circulation due to anthropogenic forcing
.
Nature
,
441
,
73
76
,
doi:10.1038/nature04744
.
Winker
,
D. M.
,
M. A.
Vaughan
,
A.
Omar
,
Y.
Hu
,
K. A.
Powell
,
Z.
Liu
,
W. H.
Hunt
, and
S. A.
Young
,
2009
:
Overview of the CALIPSO mission and CALIOP data processing algorithms
.
J. Atmos. Oceanic Technol.
,
26
,
2034
2050
.