Investigations on Stubble-Burning Aerosols over a Rural Location Using Ground-Based, Model, and Spaceborne Data

Abstract

Agriculture crop residue burning has become a major environmental problem facing the Indo-Gangetic plain, as well as contributing to global warming. This paper reports the results of a comprehensive study, examining the variations in aerosol optical, microphysical, and radiative properties that occur during biomass-burning events at Amity University Haryana (AUH), at a rural station in Gurugram (Latitude: 28.31° N, Longitude: 76.90° E, 285 m AMSL), employing ground-based observations of AERONET and Aethalometer, as well as satellite and model simulations during 7–16 November 2021. The smoke emissions during the burning events enhanced the aerosol optical depth (AOD) and increased the Angstrom exponent (AE), suggesting the dominance of fine-mode aerosols. A smoke event that affected the study region on 11 November 2021 is simulated using the regional NAAPS model to assess the role of smoke in regional aerosol loading that caused an atmospheric forcing of 230.4 W/m². The higher values of BC (black carbon) and BB (biomass burning), and lower values of AAE (absorption Angstrom exponent) are also observed during the peak intensity of the smoke-event period. A notable layer of smoke has been observed, extending from the surface up to an altitude of approximately 3 km. In addition, the observations gathered from CALIPSO regarding the vertical profiles of aerosols show a qualitative agreement with the values obtained from AERONET observations. Further, the smoke plumes that arose due to transport of a wide-spread agricultural crop residue burning are observed nationwide, as shown by MODIS imagery, and HYSPLIT back trajectories. Thus, the present study highlights that the smoke aerosol emissions during crop residue burning occasions play a critical role in the local/regional aerosol microphysical and radiation properties, and hence in the climate variability.

1. Introduction

Biomass burning (BB) is a significant source of atmospheric aerosols and trace gas emissions, affecting local, regional, and global climates while posing serious risks to human health [1,2]. The composition, size, and mixing state of BB aerosols play a crucial role in determining the optical and microphysical properties of smoke plumes that are released into the atmosphere, significantly influencing the energy balance in the Earth system. During long-range transport, smoke aerosol plumes can change their chemical composition, potentially affecting local and regional air quality when they descend into the planetary boundary layer (PBL) [3]. The processes underlying the smoke aerosol production, transformation, and transport on local and regional scales have been extensively studied over the past decade [4,5,6,7,8,9]. The production and transport of smoke aerosols play a crucial role in the radiation balance of the earth–atmosphere system [10]. Hence, the uncontrolled emission of black carbon (BB) aerosols has a dangerous impact on human health and when combined can also contribute to climate change [11]. The biomass fire events in India vary seasonally, with the highest forest-fire counts reported during pre-monsoon (March to May), whereas crop residue burning occurs in both the pre-monsoon (April and May) and post-monsoon (October and November) seasons. On a global scale, the major sources of BB mass are Africa (~52%), South America (~15%), Equatorial Asia (~10%), boreal forests (~9%), and Australia (~7%) [12]. But, in India, 5% of the biomass fires originate from the irrigated croplands, while 14% originate from semideciduous forest fires [13]. The widespread burning of agricultural crop residue contributes a quantitative quantity of greenhouse gas emissions and spreads over the Indo-Gangetic Plains (IGPs) from the post-monsoon to winter seasons in particular. Farmers in the States of Punjab and Haryana use fire to quickly and cheaply clear and fertilize fields before planting winter wheat. The IGP is a very important agro-ecological region in South-Asia, which occupies almost one-fifth of the whole geographic area in four countries (Pakistan, India, Nepal, and Bangladesh). Every year, in the month of November, satellites detect large plumes of smoke and heightened fire activity in northwestern India as farmers burn off excess paddy straw after the rice harvest [8]. Thus, it is possible to observe how ‘clean’ remote areas have been affected by urbanization and industrialization. BB aerosols, along with long-range air mass transport over remote areas, are crucial for understanding regional climate change and global air quality. They also serve as reference values for estimating pollutant concentrations in surrounding environments.

Measurements over remote stations yield background levels of aerosol concentrations [14,15]. Hence, it would be possible to observe and examine the extent to which the ‘clean’ remote areas have been affected by growing urbanization/industrialization. BB aerosols, in conjunction with the long-range transport of air mass over remote areas, play a pivotal role in understanding regional climate change and global air quality, as well as in serving as reference values for the quantitative estimation of pollutant concentration over surrounding environments. In this study, we conduct a thorough analysis of the optical, microphysical, and radiative properties of the aerosols produced and their interplay during biomass burning in an essential area of Amity University Haryana (AUH), Manesar, using multiple data sets from 7 to 16 November 2021.

A recent study by Che et al. [16] reported that biomass-burning aerosols will have a significant impact on the future climate. In this study, we analyze the optical, microphysical, and radiative properties of aerosols during the biomass-burning period over Amity University Haryana (AUH), Manesar, using multiple data sets from 7 to 16 November 2021. Furthermore, the backward trajectories of air parcel analysis using HYSPLIT, surface smoke concentration images from the NAAPS model, and CALIPSO satellite images have been carried out to explain the variation in AOD.

2. Experimental Site

The study site, Amity University Haryana (AUH) (latitude 28.31° N, longitude 76.90° E, 285 m above mean sea level), is a rural location. This site receives pollution from surrounding small industries whenever the wind blows in the northeast direction. The site is covered by an agricultural field, 25 km north of Amity University. The site is approximately 5 km from the Delhi–Jaipur National Highway (NH8) in the northeast, surrounded by Aravalli hillocks averaging 200 m in elevation. Thus, its complex topography features a valley-like terrain. The area often experiences pollution, especially at night from heavy vehicles on NH8, and wind patterns can further impact air quality. Additionally, construction activities on the campus contribute to observations. The site is also affected by long-range dust pollution from the northeast Thar Desert. Although primarily rural with few residential buildings and vegetation, it faces occasional pollution due to these natural and human activities. More details can be found in [17].

3. Data and Methodology

3.1. AERONET

Over the last 30 years, the AERosol RObotic NETwork (AERONET) network has collected data on the characteristics of aerosols from approximately 500 sites around the world. These data sets are used for the validation and bias correction of satellite AOD data sets [18,19]. The indirect parameters are derived during almucantar measurements, typically conducted in the early morning and late afternoon under clear skies. Direct sun measurements occur every 3 to 5 min across all available wavelengths, while sky radiances in almucantar, principal plane, and hybrid scenarios are generally measured hourly at several wavelengths (380, 440, 500, 675, 870, 1020, and 1640 nm). Depending on their quality, different levels are assigned to the AERONET data. All measurements are initially level 1, for instance. Since the device is self-operating, a special automated algorithm of cloud screening automatically screens the data immediately when they are received [20]. The data that pass the cloud screening procedure are automatically assigned to level 1.5. After the final calibration and some additional checks, the data are assigned to the final level, 2.0. In this study, we used level 2.0 data sets from 7 to 16 November 2021.

3.2. Aethalometer

The “Aethalometer” is the most-used instrument in the world for the real-time measurements of aerosol black carbon (BC), aerosol mass concentration (in µg per cubic meter), and biomass burning (BB) in percent. These measurements at Amity University [21] were made with a model AE-33 Aethalometer (Magee Scientific, Berkeley, CA, USA). The analysis is carried out at seven optical wavelengths spanning the spectrum from 370 nm to 950 nm at 1 min temporal resolution. BC concentration was measured at 880 nm, where light absorption is solely due to BC. The AE-33′s dual-spot technology eliminates the need for shadowing effect correction, and it automatically corrects for multiple scattering with C = 2.14 [22] and deposited aerosols was suggested to be wavelength independent [23]. The absorption Angstrom exponent (AAE) was derived from the loading-corrected BC data at seven wavelengths of the Aethalometer, representing the slope of the linear regression of absorption coefficient against the wavelength on a logarithmic scale. In the present study, we used the data obtained from 7 to 16 November 2021.

3.3. NAAPS Model

For determining whether smoke or dust aerosols in the troposphere contributed to individual extreme event days, we rely on the Navy Aerosol Analysis and Prediction System (NAAPS) [24]. The model has 25 vertical levels and provides data at every 6 h intervals out to 120 h (5 days) with 1° × 1° spatial resolution. It is designed based on the Danish Eulerian Hemispheric Model, finite element horizontal diffusion, and finite element vertical transport. The Navy Operational Global Atmospheric Prediction System (NOGAPS) [25] provides meteorological data to drive individual sulfate, smoke, and dust emissions models. We specially used data for the surface mass concentrations associated with smoke in this work for assigning extreme events to specific sources.

3.4. CALIPSO

Ground-based systems effectively view local aerosol levels and estimate their impact on the radiation budget. To analyze global cloud and aerosol vertical distributions, NASA and CNES launched a satellite with a lidar system [26,27]. The Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) mission, developed by NASA Langley in collaboration with CNES, was launched in 2006. This study uses CALIPSO’s vertical feature mask and aerosol subtype images to identify aerosol types and confirm smoke layer altitudes in potential source regions using HYSPLIT model back trajectories.

3.5. Air Mass Back-Trajectories

The characteristics of aerosols at the surface during a biomass-burning event can significantly differ from those observed in the columnar measurements, as various aerosol types exhibit variations based on their scale heights and lifetimes [28]. A back-trajectory analysis was conducted at two specific altitudes: 500 m, to identify the origins of air masses close to the surface, and 1500 m, which represents a height within the boundary layer where biomass-burning aerosols are typically transported [8]. The Air Resources Laboratory’s Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model is commonly used in back trajectory applications to identify the origin of air masses using previously gridded meteorological data [29]. In addition, the reanalysis data set of GDAS (Global Data Assimilation System, http://ready.arl.noaa.gov/HYSPLIT.Php, accessed on 22 April 2024) was used to further validate our conclusions obtained with trajectories based on CALIPSO and AERONET data.

4. Results and Discussion

4.1. The MODIS Fire Image and Columnar AODs During an Aerosol-Laden Day

The MODIS-Aqua (afternoon, ~1330 h) satellite true color image of fires over the Punjab State on 11 November 2021 is shown in Figure 1. Red color dots indicate fire detection along the Himalayas (Indo Gangetic Plains) from VIIRS satellite images. The blue-colored asterisk indicates the observational site, Gurugram (Amity University Haryana). The MODIS fire image showed intense BB activities over the IGP region, particularly in Punjab, due to paddy crop residue burning. IGP is significantly impacted by BB during the two growing seasons and must deal with the smoke aerosols emitted by BB. This affects the weather and climate of the IGP region. Punjab, a state in northwest India, has two growing seasons: one from May to September and another from November to April. With November being in between the two, many farmers will typically sow crops such as wheat and vegetables at this time. Winter is the coldest season in Punjab, with temperatures reaching as low as −10 degrees Celsius. Farmers who have yet to sow crops during the winter months will likely start doing so in November. They light fires to clear their fields before planting, a practice known as stubble burning. Although the smoke from this recent fire appeared to primarily be coming from fires in rural areas, other factors such as urban and industrial smog may also have played a role. A smoke plume extended far to the northeast, obscuring from 7 to 16 November 2021, is plotted in Figure 2. The variations show larger values of AOD from 9 to 14 November 2021.

It is clear from the figure that the AOD shows maximum in the wavelength region between 11 and 12 November 2021 during the peak smoke-event period between 11 and 12 November 2021. Another smaller peak in AOD due to the presence of clouds prevailing over the site can also be seen around 15 November 2021. Such high values in AOD over the region during biomass burning due to agricultural crop residue burning were earlier reported [30].

4.2. Day-to-Day Variation in Different Types of AODs (FM and CM, 500 nm), AE (440–870 nm), and FMF (500 nm)

The day-to-day variation in the AERONET-measured fine-mode and coarse-mode fractions of total AOD at 500 nm is depicted in Figure 3. It is interesting to note that the variations in coarse-mode and fine-mode aerosol fractions are opposite, which is consistent. It can also be seen that the fine-mode fraction dominates the coarse-mode particles on 11 and 12 November 2021. This indicates that the concentration of smaller particles dominates that of coarse-mode particles. Furthermore, the higher (~15.27%) AE (fine-mode aerosols) values together with high AOD values are attributed to the presence of smoke aerosols over the study region.

4.3. Aerosol Size Distribution (ASD)

Figure 4 displays the mean aerosol radius versus their volume size distribution during 7–16 November 2021, which encompasses the dust storm period on 11 and 12 November 2021. There are two size ranges, wherein one region has particles of size up to 1 micron, while the other ranges between 2 and 7 microns, which is the prominent size range noticed during the smoke-event peak period. Aerosol size distributions from biomass burning evolve after emission, with size distributions tending to shift to larger sizes and to decrease in radius due to condensation/evaporation and coagulation [31,32,33]. Coagulation reduces particle number, shifts the distribution to larger sizes, and changes the modification of radius of the size distribution [34]. Hence, the concentrations in the smoke plume affect the coagulation rate; and because the reduction in relatively cleaner, background air into smoke plumes lowers number concentrations, the plume reduction rate also impacts the coagulation rate [35].

4.4. Single-Scattering Albedo (SSA)

Single-Scattering Albedo (SSA) is one of the important parameters that is useful for separating scatters from absorbers in an aerosol ensemble. Thus, it is important to make the estimation of radiative forcing due to scattering/absorbing aerosols. Figure 5 portrays the spectral distribution of SSA on different days of the smoke event of November 2021 over the study region. It is known that an SSA value close to one indicates that the particle is a pure scatter, and low SSA values indicate the absorbing nature of the particle. It is very interesting to know that the SSA values are maximum and close to 1 (0.94, on average), which reveals that the aerosols over the study region during the smoke event are of the scattering type. Even the minimum SSA value (~0.86) supports this behavior of aerosols.

4.5. Radiative Forcing (RF, Wm⁻²)

Decreasing and increasing greenhouse forcing agents such as aerosols strongly impact the impacts of human-made greenhouse gases on climate. The daily mean Aerosol Radiative Forcing (ARF) estimated using the SBDART (Santa Barbara DISORT Atmospheric Radiative Transfer) FORTRAN computer code designed for the analysis of a wide variety of radiative transfer model at Bottom-of-the-Atmosphere (BOA), in the Atmosphere (ATM), and at Top-of-the Atmosphere (TOA) during the smoke event period is shown in Figure 6. The radiation forcing at BOA, ATM and TOA are related by the type of aerosol. Normally, the BOA and TOA forcing is mostly negative during all months, indicating a cooling effect, while the ATM forcing is positive during all months, indicative of the heating of the atmosphere. Moreover, the ATM forcing becomes more pronounced during the pre-monsoon season (April–June). However, Figure 6 shows (i) negative forcings at BOA and TOA, which indicate cooling, while ATM indicates the warming of the atmosphere, and (ii) larger values of forcing in the BOA, ATM, and TOA during the peak active period (11–12 November 2021). Higher ATM values (230.4 W/m²) appear on 11–12 November 2021. This is due to the atmosphere, which was apparently attributable to smoke aerosols, which not only heated the atmosphere through strong absorption but also simultaneously reduced the radiation reaching the TOA and surface levels. These larger values could be ascribed to the excess particle density during the smoke event period.

4.6. Variations in BC Mass Concentration and Percentage of Biomass Burning (BB, %)

The day-to-day variations in the black carbon (BC) mass concentration (microgram per cubic meter), along with absorption Angstrom exponent (AAE) and (b) associated biomass burning (in percent) during the period from 7 to 16 November 2021, are plotted in Figure 7. All three parameters, BC, AAE, and BB, are found to be high at the beginning and then gradually decrease, but with higher values of BC and BB, and lower values of AAE during the peak intensity of the smoke event on 10–12 November 2021. This also indicates that the BC mass concentration is contributed to by bigger size carbonaceous aerosols. Moreover, these smoke aerosols were entrained into the boundary layer during the growth of the planetary boundary layer (PBL) and accumulated near the surface when the PBL collapsed [36]. This also increases the concentration of aerosol particles in the smoke discharge event.

4.7. Surface Concentration of Smoke in Aerosol-Laden Day

Figure 8 displays the contours of smoke aerosol concentrations at different times of the day in the study region on the peak smoke-event day, i.e., 11 November 2021, at different hours: 00:00, 06:00, 12:00, and 18:00 UTC. Figure 8a,c,d show that the maximum smoke aerosol concentration is 64–128 μg/m³, while Figure 8b shows a higher smoke aerosol concentration of 128–256 μg/m³ covering Punjab, which is where most crop burning activity is found. Overall, the larger concentration of smoke aerosol concentration in the afternoon, which corroborates the observations. Early studies by Shaik et al. [8] examined the spatial, seasonal, and yearly variation in biomass burning in Northern India and its impact on regional aerosol optical properties using multi-satellite aerosol observations and satellite and ground-based observations. They pointed out that there were a lot of smoke aerosols traveling during the post-monsoon season, with the majority going to Eastern India, Central India, and the neighboring areas. Recently, Tariq et al. [37] also reported studies on the long-term climatology of biomass-burning aerosols over different environments using AERONET data. They found that the aerosols of optical, microphysical, and radiative properties are changed due to dominance of biomass-burning aerosols.

4.8. CALIPSO Images During Smoke-Event Period

The CALIPSO satellite-derived vertical profiles of smoke aerosols are shown in Figure 9. It can be clearly seen that the polluted dust and smoke aerosols from the satellite sub-type time–height cross-sections (upper frame) and from the vertical feature mask height–time cross-sections (bottom frame) on 11 November 2021 are consistent with the surface-based observations from synchronous NASA–AERONET observations. It clearly shows a layer of dense smoke extending from the surface to an altitude of about ~3 km. Crop residue burning is a major contributor to harmful air pollution, which we can now identify from satellite measurements coming from CALIPSO satellites. In Northern India (19.21–31.44° N, 74.90–77.94° E), the long-range transport of smoke aerosols from agriculture crop residue burning is observed in the downwind direction [3]. The observational data clearly shows a dramatic increase in polluted dust and smoke aerosols over the study region on the day. The combined effect of these aerosols seems to be a higher cloud cover with an increasing aerosol concentration [38].

4.9. Detection and Transport of Biomass-Burning Aerosols

To assess the flow paths for the air and to establish a link between synoptic air masses and hourly variation in aerosol loading over the experimental region, we used the HYSPLIT model. The HYSPLIT trajectories are calculated using the Global Data Assimilation System (GDAS), a data-harvesting system from the National Weather Service of the National Centre for Environmental Prediction (NCEP). Figure 10 depicts two frames, (a) and (b), explaining the spread of smoke aerosols over the experimental site as viewed from the MODIS satellite (Figure 10a) and the HYSPLIT-estimated long-range transport of aerosols (Figure 10b) that contribute to those prevailing over the experimental location on 11 November 2021 (peak storm day), respectively. As the short-length back-trajectories may be from closer sources, they are enlarged and shown as sub-sets in the figure. This feature reveals that the enhancement in the aerosol concentration around 11 November 2021 could be not only due to local activities but also due to long-range transport from nearby regions. So, the increase in fine-mode particle loading, due to the long-range transport of aerosols from BB was reflected in an increase in aerosol optical depth (AOD) on 11 and 12 November 2021 (Figure 2). In addition, other aerosol optical, microphysical, and radiative properties are also enhanced due to the long-range transport of smoke aerosols.

5. Conclusions

Aerosol transport is a very complex issue, not just for pollution control in densely populated areas but also for effects on the overall climate system. In this study, the transport characteristics of smoke aerosols were investigated utilizing ground-based, model and spaceborne data archived over AUH, Gurugram during 7–16 November 2021, corresponding to the peak period of agriculture crop residue burning in the IGP region, particularly in the state of Punjab. The main results are summarized as follows:

• The smoke intrusion caused an increase in the AOD and Ångström exponent, suggesting the dominance of fine-mode aerosols.
• A significant increase in aerosol size distribution over experimental sites due to the long-range transport of smoke aerosols was observed.
• Higher values of BC and maximum percentage of BB aerosols were observed in the region during the study period.
• The observed higher atmospheric forcing (230.4 W/m²) appears during the aerosol-laden day due to the biomass-burning activity.
• CALIPSO data also revealed that, during the study period, more smoke and polluted dust aerosols were present up to 3 km in the atmosphere.
• The NAAPS model images of smoke-surface concentrations have been studied.
• Winds at different altitudes (back-trajectories from the HYSPLIT model reanalysis) contribute additional aerosols from the biomass-burning activity at the experimental site due to long-range transport.