A Thesis submitted to Goa University for the Award of the Degree of
DOCTOR OF PHILOSOPHY
In
Marine Sciences
By
Shilpa Shirodkar
Research Guide Prof. H.B.Menon
Department of Marine Sciences
Goa University, Taleigao Goa
2014
Dedicated to
“My beloved family”
As required by the university ordinance 0.19.8(vi), I state that the present thesis entitled
“Aerosol characterization and radiative forcing over regions of different optical domains” is my original contribution and the same has not been submitted on any pervious occasion. To the best of my knowledge the present study is the first comprehensive work of its kind from area mentioned.
The literature related to the problem investigated has been cited. Due acknowledgments have been made wherever facilities and suggestions have been availed of.
Shilpa Shirodkar
This is to certify that the thesis entitled “Aerosol characterization and radiative forcing over regions of different optical domains”, submitted by Ms. Shilpa Shirodkar for the award of Doctor of Philosophy in Marine Sciences is based on her original studies carried out by her under my supervision. The thesis or any part thereof has not been previously submitted for any degree or diploma in any universities or Institutions.
Prof. H. B. Menon
Department of Marine sciences Goa University
Taleigao Plateau, Goa- 403206, India
Shilpa Shirodkar
Department of Marine Sciences Tel:09823314632
Goa University, Taleigao-403206-Goa email:md161shilpa@gmail.com Education
2000 – 2004 B.Sc, in Physics, Goa University, Goa, India
2005 – 2007 M.Sc, in Physical Oceanography, Goa University, Goa, India.
2009 – 2014 Ph.D in Marine sciences, Goa University, Goa, India
Thesis title: Aerosol characterization and radiative forcing over regions of different optical domains.
Advisor: Prof.H.B.Menon.
Oceanography cruise experience:
1) Marion Dufrene: 11th May 2007 to 28th May 2007- Bay of Bengal.
2) Sagar Manjusha: 8th April 2008 to 17th April 2008-Eastern Arabian Sea.
3) Sagar Kanya: 27th December 2008 to 30th January 2009- Bay of Bengal.
4) Sagar Sampada: 4th April 2009 to 14th April 2009- Eastern Arabian sea.
5) Sagar Sampada: 20th April 2010 to 5th May 2010- Eastern Arabian Sea.
Presentation
(1) Oral presentation at AOGS meeting held from 5th to 9th July 2010 at Hyderabad on
a) Variation of Aerosol Optical depth over coastal station Goa, along west coast of India.
Shilpa Shirodkar and H.B.Menon
b) Temporal analysis of atmospheric black Carbon over a coastal station in Goa. Shilpa Shirodkar and H.B.Menon
(2) Oral presentation at Pan Ocean Remote Sensing Conference (PORSEC), from 05-09 November 2012, at Kochi on paper entitled “Interannual aerosol radiative properties over Goa”
Shilpa Shirodkar and H.B.Menon
(3) Poster Presentation at OCHAMP (Opportunities and Challenges in Monsoon Prediction in a Changing Climate) held at Indian institute of Tropical Meteorology,Pune –India, February 21- 25 ,2012, on “Temporal variation of aerosol optical depth and associated shortwave radiative forcing over a coastal site along the west of coast of India”.
Publication
Menon. H.B., Shirodkar, S., Kedia, S., Ramachandran, S., Babu, S.S., Krishnamoorthy, K., 2013. “Temporal variation of aerosol optical depth and associated shortwave radiative forcing over a coastal site along the west of coast of India”. Accepted in Science of the Total Environment .
Shilpa Shirodkar and H.B.Menon “Microphysical characteristics of aerosol over a coastal site in Goa, along the west coast of India”. Under review in Journal of Atmospheric and Solar Terrestrial Physics.
Submitted Masters Dissertation
Topic: “Air Temperature Inversion and its link with southwest monsoon along the west coast of India”.
I would like to express my gratitude to all people who helped me to bring this thesis to fruition.
First and foremost, I thank Prof. H.B.Menon, for introducing me to this interesting topic, imposing wide questions in aerosol research and thus moulding me into a researcher from a layman in this field. I appreciate all the support and freedom given by him to carry out this work.
Without the support of my parents, this work would have never been completed. I am indebted to them and my elder brother, for everything they did for me. I am fortunate to have Dibakar as my husband, who selflessly supported and encouraged me throughout my thesis, Thanks for being there for me! I take this opportunity to thank my lovely in-laws, for being kind and supportive. I am blessed with wonderful family and I thank you all for being part of my life.
I would like to thank, Prof. G.N.Nayak (Head, Department of Marine Sciences), Dr. S.
Upadhyay , Dr. V.M.M.Matta, Dr. Rivonkar and Dr. Aftab Can, for their encouragement, right from Master degree. The valuable suggestion by V.C’s nominee, Dr. Ramesh Kumar (Scientist- NIO) helped me to improve my research understanding, I would like to thank him for being there whenever needed. I would like to thank Dr. Ramchandran (PRL, Ahmedabad), for proving necessary guidance and lab facilities at PRL. Despite his busy schedule, it was kind of him to spend time on my research problem. At this juncture I cannot forget the help rendered by Mr. Rohit Srivastava and Ms. Sumita Kedia, throughout my stay at PRL. I would like to thank Prof. S.K.Satheesh (IISC, Bangalore) and Vinoj (IISC), for giving a primary training on radiative transfer modeling. I appreciate, Dr. Satheesh Shenoi Director INCOIS, who was
helped me in running and understanding the satellite data processing software (SeaDas). It was kind of Prakash Mehra (NIO) for providing, the meteorological data. Also, I wish to acknowledge Indian Meteorological data, Panjim for proving the rainfall data over Goa. All the financial support given under ISRO-ARFI project is deeply acknowledged and I thank Dr.
K.Krishnamoorty and Dr Suresh Babu, (SPL, Trivandrum) for providing necessary support. As a part of ISRO IGBP project, I got an opportunity to participate in WICARB cruise, where I improved my knowledge on various aspects of aerosol research. I take this opportunity to thank Dr. CBS Dutt (ISRO Bangalore), Scientist-in-chief, for his encouragement throughout the cruise. Stay on cruise was made worthwhile by Vijaykumar (SPL), Naseema (SPL) Sumita (PRL), Rohit (PRL), Srinivas (PRL), Sumit (IITM), Puna Ram Sinha (Tfir) and many others. I thank them for being kind and helping me whenever needed.
During the course of my PhD, I made many friends, who contributed immensely to my personal and professional time at Goa University. This fun group has been a source of friendship as well as wise advice. Thanks to you, Sweety, Nutan, Renosh, Deepti, Vinay, Aneesh, Reachel, Tomchou, Lina, Santosh, Anant, Mahesh, Maria, Abhilash, Soniya, Vineel, Shrivardhan, Veloisa, Vijaylaxmi, Vineet, Satyam, Neela, Samida, Cheryl, Kalpana, Cynthia and Nandini for sharing wonderful memories. I would like to express my deep gratitude for Dinesh, who helped me in finalizing this thesis, Thanks! Sharing some lighter moment makes one happy, and I am delighted that I found people with whom I can share this moments. Many thanks to you Mabu, Ratna and Trupti for not only being part of lighter moments but also for being supportive.
and Mr.Amonkar. Thanks are also due to Mr.Martin and Mr.Serrro, Department of Marine Biotechnology, for all the help they gave.
At last but not the least, I would like to thank almighty God for giving me strength and courage in all my pursuit!
Shilpa Shirodkar
List of Figures i-iv
List of Tables v-vi
List of Abbreviation vii-ix
Notation and symbols x-xi
Chapter 1. Introduction 1-12
1.1. General Introduction 1
1.2. Classification of Aerosols 2
1.2.1. Based on source 1.2.2. Based on size 1.2.3. Based on shape
1.3. Residence time 6
1.4. Transformation Processes 6
1.4.1. Coagulation 1.4.2. Condensation 1.4.3. Cloud processing
1.5. Aerosol sinks 7
1.6. Role of aerosols in climate 8
1.7. Rationale for the study 9
2.1.1. Coastal site in Goa, along the West Coast of India
2.1.2. Eastern Arabian Sea (EAS)
2.1.3. Bay of Bengal
2.2. In-situ measurements 22
2.2.1. MICROTOPS II Sunphotometer (a) Principle of Operation
(b) Theory
(c) Error estimation
2.2.2. Aethalometer
(a) Principle of Operation (b) Theory
2.3. Remote sensing of aerosol optical depth 29
2.3.1. Characteristics of MODIS 2.3.2. Data processing
2.4. Methodology 33
2.4.1. Ångström parameters (α, β) and Second order derivative 2.4.2. Radiative forcing
2.4.2 (a) (i) Aerosol components and types 2.4.2 (a) (ii) Aerosol model in the present work 2.4.2 (b) Surface Reflectance
2.5.1. Theory
2.5.2. Back trajectory analysis
2.5.2 (a) Over a coastal site in Goa
2.5.2 (b) Over Eastern Arabian Sea
Chapter 3. Microphysical properties of aerosols over Goa 50-68 and along the Eastern Arabian Sea
3.1. Introduction 50
3.2. Results and Discussion 52
3.2.1. Coastal site in Goa
3.2.1 (a) Spectral variation of AOD 3.2.1 (b) Seasonal variation of AOD
3.2.1 (c) Variability of Ångström parameters (α, β) 3.2.1 (d) Wavelength dependence of AOD
3.2.2. Eastern Arabian Sea
3.2.2 (a) Daily variation of AOD and α 3.2.2 (b) Second order Ångström exponent
Chapter 4. Shortwave radiative forcing over regions of different 69-93 optical domains
4.1. Introduction 69
4.2. Results and discussion 70
4.2.1 (b) Derived parameters: Single Scattering albedo (SSA - ω), Asymmetry parameter (g) and Aerosol Optical Depth (AOD) 4.2.1 (c) Direct Shortwave Radiative Forcing
4.2.1 (d) Forcing efficiency and heating rate 4.2.2. Eastern Arabian Sea (EAS)
4.2.2 (a) Percentage contribution of aerosol species
4.2.2 (b) Single scattering albedo (ω), Asymmetry parameter (g) and aerosol optical depth (AOD)
4.2.2 (d) Heating rate and forcing efficiency
Chapter 5. Synoptic analysis of MODIS derived aerosol 94-105 optical depth over Arabian Sea and Bay of Bengal
5.1. Introduction 94
5.2. Results and Discussion 95
5.2.1. Arabian Sea 5.2.2. Bay of Bengal
5.2.3 Comparison between AOD distribution over Arabian Sea and Bay of Bengal
6.1. Summary of the thesis 106
6.1.1. Microphysical properties of aerosols
6.1.1 (a) Coastal site in Goa
6.1.1 (b) Eastern Arabian Sea
6.1.2. Shortwave Radiative forcing over region of different optical domain
6.1.2 (a) Coastal site in Goa 6.1.2 (b) Eastern Arabian Sea
6.1.3. Satellite data MODIS AOD mapping
6.2 Conclusion 110
6.3. Scope for future work
111
Bibliography 112-130
i
Fig. 1.1. Pictorial representation of different sources and sinks of atmospheric aerosols (www.ems.psu.edu)
Fig. 1.2. The global mean radiative forcing of the climate system by GHG’s and aerosols between 1750 to 2005 (IPCC, 2007).
Fig.2.1. (a) Network of ARFI stations across India (b) Sampling site at a coastal station in Goa.
Fig. 2.2. Synoptic wind vector (m/s at 850 hPa) over India during (a) WMS (b) SIMS (c) SMS and (d) FIMS.
Fig. 2.3. (a) Shows the cruise track of CRV Sagar Sampada (SS) ,blue color indicates cruise track of CRV SS-265 during the year 2009 and red indicates CRV SS- 274. (b) Shows, Upper zone (4th -6th April 2009 & 2nd - 4th May 2010), Middle zone (9th - 10th April 2009 & 29th April 2010), Lower zone (13th - 14th April 2009), Lower zone-1 (21st – 22nd April 2010) and Lower zone-2 (25th -26th April 2010).
Fig. 2.4. Synoptic wind vector (a) during 2009 and (b) during 2010.
Fig. 2.5. The cruise track of Sagar Kanya 254 during W-ICARB, 27 December 2008 – 30 January 2009.
Fig.2.6. Correlation between in-situ and MODIS-A derived AOD at 0.500μm
Fig. 2.7 (a) MODIS derived Surface reflectance values at 0.469, 0.555, 0.645, 0.859, 1.24, 1.64 and 2.13 µm are correlated with surface reflectance of August estimated from a model with mixture of 5% sand, 40%, vegetation and 55% water. (b) Seasonal mean variation of surface reflectance.
Fig. 2.8. Mean five days back trajectory using HYSPLIT model at 500 m and 1500 m over the study area.
ii
hrs (a) Trajectories at 500 m during 2009 (b) Trajectories at 500 m during 2010.(c) Trajectories at 1500 m during 2009 and (d) Trajectories at 1500 during the year 2010.
Fig. 3.1. Intra-annual variability of mean AOD spectra over a coastal site in Goa for the period from 2008 to 2010. The vertical bar denotes ±1σ.
Fig. 3.2. Seasonal variation of mean Aerosol Optical Depth (AOD) at 0.500 µm, during (a) WMS (b) SIMS (c) SMS and (d) FIMS for the period of 2008- 2010. The vertical bars denote ±1σ.
Fig. 3.3. Variability of (a) Ångström exponent (α) and (b) turbidity coefficient β during different seasons (black square represents mean value of α and open circle is the mean value of β. Standard deviation is shown by vertical bar).
Fig. 3.4. Correlation between AOD (0.500 µm) and α computed at short (α0.440-0.500), long (α0.675- 0.870) and whole (α0.440-0.870) wavelength spectra during (a) WMS (b) SIMS (c) SMS and (d) FIMS
Fig. 3.5. Correlation between Ångström exponent α and difference in the coefficient of the polynomial fit (α2- α1) during different seasons.
Fig. 3.6. Daily variation of mean spectral AOD during (a) 2009 and (b) 2010. Standard deviation is shown by vertical bar.
Fig. 3.7. Daily mean Ångström exponent (α) computed in short-wavelength range (0.440 – 0.500 μm), long-wavelength range (0.675 – 0.870 μm) and full wavelength range (0.440 – 0.675 μm), during the year (a) 2009 and (b) 2010. Vertical bar represents the standard deviation
iii and fall inter-monsoon season (FIMS).
Fig. 4.2. Monthly mean variation of BC mass concentration in 2008, 2009 and 2010.
Vertical bar denote the standard deviation.
Fig. 4.3. Seasonal mean variation of BC mass concentration during the years 2008, 2009 and 2010.
Fig. 4.4. Seasonal variation of (a) single scattering albedo (ω) and (b) aerosol optical depth (AOD)
Fig. 4.5. Seasonal variation of direct aerosol radiative forcing during the year 2008, 2009 and 2010.
Fig. 4.6. Spectral variation of single scattering albedo over EAS during (a) 2009 and (b) 2010
Fig. 4.7. Spectral variation of asymmetry parameter over EAS during (a) 2009 and (b) 2010.
Fig. 4.8. Region wise variation in shortwave radiative forcing (0.25 μm to 4.0 μm) computed using Santa Barbara DISORT Atmospheric Radiative Transfer (SBDART) during the year (a) 2009 and (b) 2010.
Fig. 5.1. Spatial and temporal variations in monthly mean AOD over Arabian Sea and Bay of Bengal (BoB) during (a) 2008, (b) 2009 and (c) 2010.
Fig. 5.2. Monthly mean synoptic wind pattern at 850 hPa representing (a) January, (b) February, (c) March, (d) April, (e) September, (f) October, (g) November and (h) December during the year 2008.
Fig. 5.3. Monthly mean synoptic wind pattern at 850 hPa representing (a) January, (b) February, (c) March, (d) April, (e) September, (f) October, (g) November and (h) December during the year 2009.
iv (h) December during the year 2010.
v
Table. 2.1. Seasonal variation of mean meteorological parameters during WMS, SIMS, SMS and FIMS for the period between 2008 and 2010.
Table. 2.2. Daily variation of meteorological parameters during the respective cruises in 2009 and 2010
Table. 2.3. Mean percentage contribution of air mass back trajectory at 500 m and 1500 m over the study area. The air mass trajectory from northwest India crossing central and east coast of India, from central India and from northeast India represents continental source. Area crossing Bay of Bengal, Arabian Sea and west coast of India represents maritime source. Trajectories originating from Arabian peninsula crossing Arabian Sea and northwest India are identified and termed as West Asia.
Table. 3.1. Seasonal variation of Ångström exponent (α) and turbidity coefficient (β) over different regions of Indian Subcontinent and adjoining Sea.
Table. 3.2. Percentage representation of seasonal variation of mean positive and negative values of second order Ångström exponent ά and coefficient of the polynomial fit (α1 and α2) grouped into α2- α1 ≤ 1, 1 < α2- α1 < 2 and α2- α1 ≥ 2 during different seasons.
Table. 3.3. Percentage representation of positive and negative values of second order Ångström exponent ά and coefficient of the polynomial fit (α1 and α2) grouped into α2-α1 ≤ 1, 1 < α2-α1 < 2 and α2-α1 ≥ 2 during different seasons.
Table. 4.1. Seasonal variation in forcing efficiency and heating rate
Table. 4.2. Region wise contribution of water soluble (WASO), SOOT, sea salt accumulation mode (SSAM), sea salt coarse mode (SSCM) and Mineral transported components of aerosol generated using OPAC.
vi
vii
Full form Acronym
Aerosol Characterization Experiment ACE
Aerosol Optical Depth AOD
Aerosol Radiative Forcing over India ARFI
Aerosol Radiative Forcing ARF
Arabian Sea Monsoon Experiment ARMEX
Arabian Sea AS
Atmosphere ATM
Atmospheric Boundary Layer ABL
Automatic Weather Station AWS
Bay of Bengal BoB
Black Carbon BC
Central Arabian Sea CAS
Cloud Condensation nuclei CCN
Coastal Research vessel CRV
Dimethylsulphide DMS
Discrete Ordinate Radiative Transfer DISORT
Earth’s Observation System EOS
East coast of India ECI
Eastern Arabian Sea EAS
Fall Inter-Monsoon Season FIMS
viii
Global Positioning System GPS
Graphical User Interface GUI
Greenhouse Gases GHGs
HYbrid Single-Particle Langrangian Integrated Trajectory HYSPLIT
Indian Meteorological Department IMD
Indian Middle Atmosphere Program IMAP
Indian Ocean Experiment INDOEX
Indian Space Research Organization ISRO
Integrated Campaign of Aerosols, gases and Radiation Budget ICARB
Liters Per Minute LPM
Moderate Resolution Imaging Spectroradiometer MODIS
Multiangle Imaging Spectroradiometer MISR
National Center for Atmospheric Research NCAR
National Center for Prediction NCEP
Nocturnal Boundary layer NBL
North Arabian Sea NAS
Optical Properties of Aerosols and Clouds OPAC
Organic Carbon OC
Santa Barbara DISORT Atmospheric Radiative Transfer SBDART
Sea-Viewing Wide Field-of-View Sensor SeaWiFS
SeaWiFS Data Analysis System SeaDas
Single Scattering Albedo SSA
ix
Standard Mapped Images SMI
Summer Monsoon Season SMS
Surface SUR
Top of the Atmosphere TOA
Troposphere Aerosol Radiative Forcing Observational Experiment TARFOX
West coast of India WCI
Winter - Integrated Campaign of Aerosols, gases and Radiation Budget W-ICARB
Winter Monsoon Season WMS
x
Absorption coefficient babs
Absorption due to ozone τo3λ
Absorption due to water vapour τwλ
Acceleration due to gravity g
Airmass m
Attenuation coefficient bATN
Direct solar flux F (λ)
Electrical output signal V (λ)
Extinction coefficients be
Extinction due to aerosols τpλ
Extinction due to molecules τmλ
Extraterrestrial flux F0 (λ)
Filter spot area A
Instantaneous Sun Earth distance d Intensity of the incoming light I0
Light attenuation ATN
Light intensity after passing through a medium I Mass specific absorption cross section σabs
Mass specific attenuation cross section σATN
Mean Sun Earth distance do
Pressure difference ∆P
Radiometer output V0 (λ)
xi
Volumetric flow rate V
Wavelength λ
1 1.1. General Introduction
Earth’s atmosphere comprises of a thin envelope consisting of mixture of life supporting gases.
Among them the major components are Nitrogen (N2 = 78%), Oxygen (O2 = 21%) and very minute percent of Argon (Ar = 0.9%). Four principle gases emitted due to human activities are Carbon dioxide (CO2), Methane (CH4), Nitrous oxide (N2O) and the Halocarbons. These gases together with water vapor and ozone are termed as Greenhouse Gases (GHGs), constitute for remaining part of the atmosphere. Since an increased concentration of GHGs during the industrial era perturbed the Earth’s radiation budget, all out efforts were made by the International scientific community to understand them better for remedial measures. Outcome of the research carried out by group of scientists had revealed the potential of aerosols in destabilizing the equilibrium of Earth’s global temperature (Charlson et al., 1992). Aerosols not only affect the Earth’s climate directly by scattering or absorbing the downward irradiance but also modulate the cloud microphysics indirectly by acting as cloud condensation nuclei (CCN) (Hadley and Kirchstettes, 2012; Bond et al 2013). Large spatial and temporal variability in aerosol concentration with short residence time hinder the understanding of radiative effects on Earth’s radiation budget. They also have a role in the abundance and distribution of atmospheric trace gases through heterogeneous chemical reaction. Moreover, airborne particles play a vital role in spreading biological organisms, reproductive materials, and pathogens (pollen, bacteria, spores, viruses) which can cause or enhance respiratory, cardiovascular, infectious, and allergic diseases (Finlayson-Pitts and Pitts, 2000).
The word aerosol was coined more than 80 years ago as an analogy to the term hydrosol, a stable liquid suspension of solid particles (Hinds, 1999). Atmospheric aerosols are tiny particles
2 in solid or liquid phase, suspended in air, associated with phenomena such as dust, fume, smoke, mist, fog, haze and smog and varying in size ranging from few nanometers (nm) to tens of micrometers (μm) in diameter (Seinfeld and Pandis, 1998). Based on the process of formation, they are classified into two; primary particles, emitted directly as solid or liquid from the sources and secondary particles, formed as a result of gas-to-particle conversion in the atmosphere, due to process of nucleation and condensation of gaseous precursors such as sulphur dioxide, nitrogen oxide and volatile organic compounds (VOC’s). Particles in the atmosphere are either of natural sources such as windborne dust, forest fires, sea spray and/or from anthropogenic sources such as fuel combustion, industrial processes and automobile exhaust. On a global scale, aerosols from natural processes are assumed to be 4 to 5 times more than those from anthropogenic activities. However, on a regional scale this ratio may change significantly, particularly in the Earth’s industrialized northern hemisphere. Studies showed that there exists a large uncertainty in the aerosol direct climate forcing estimate in the range from -0.1 to -0.9 W/m2 while the uncertainty in the indirect forcing varied between -0.3 to -1.8 W/m2. Thus a better understanding of the formation, composition and transformation of aerosols in the atmosphere is of vital importance to quantify the radiative forcing for a better climate analysis.
1.2. Classification of Aerosols
1.2.1. Based on source
Depending on the process of production, sources of aerosols are broadly classified into two categories. They are primary/direct sources of natural origin produced by mechanical
3 disintegration processes and transport of inter-planetary aerosols, terrestrial aerosols and sea- salt aerosols. Secondary sources are those generated by man–made activities and by the chemical reaction of natural atmospheric gaseous species resulting in solid or liquid particles in the troposphere. This process of phase change is termed as gas-to-particle conversion.
Mineral dust aerosols
Human activities like construction, deforestation and agriculture are the source of mineral dust which is lifted to the atmosphere by the wind from the area where the vegetation cover is sparse (Tegen and Fung, 1994).
Marine aerosols
Marine aerosols are of two types, viz: primarily those produced by the action of wind over the seas which includes sea salt aerosols and secondary, those produced by gas-to-particle conversion over the maritime environment, which includes dimethylsulfide (DMS) emitted by the marine phytoplankton which converts into sulfate aerosols (Hoppel et al., 1979).
Volcanic aerosols
Volcanic aerosols are due to volcanic eruption (e.g. Mt. Pinatubo eruption in 1991). Once the eruption has taken place, a layer of aerosol is formed in the stratosphere. Aerosol layer is formed by sulfur dioxide gas which is converted to droplet of sulfuric acid in the stratosphere, after a week or a month once the eruption has occurred. This is followed by the transport of aerosols by wind in the stratosphere, which can remain in the atmosphere as long as two years (Hunten, 1975).
4 Biological aerosols
Biological aerosols includes ubiquitous component of aerosol covering a large size from virus to large pollens. Some of the examples are pollen, virus, insect part and cellulose (Lohmann et al., 2005).
Biomass burning aerosols
Aerosols formed due to burning of biomass are made up of two major elements namely Black Carbon (BC) (which mainly absorbs the incoming solar radiation) and Organic Carbon (OC), which scatters the radiation. Sources of biomass burning aerosols are forest fire, burning of agricultural waste and also substance burned for fuel such as wood, dung and peat (Seinfeld and Pandis, 1998).
Extraterrestrial particles
Aerosols originating from comet and meteoric debris reach the troposphere through scavenging process (mostly sedimentation and dry deposition). They have residence time ranging from months to years (Seinfeld and Pandis, 1998).
Sulfate aerosols
Sulfate aerosols are fine solid particles of a sulfate or tiny droplet of a solution of sulfate or of sulfuric acid. Sulfate aerosols have the potential to scatter the incoming solar radiation and they also act as CCN (Seinfeld and Pandis, 1998).
Nitrate aerosols
The most common nitrate present in the atmosphere is the ammonium nitrate (Pruppacher and Klett, 1978). Atmospheric nitrate aerosols are formed if sulfate aerosols are neutralized completely in excess ammonia.
5 1.2.2. Based on size
Based on the diameter, aerosols are classified into Nucleation mode (10-3 μm – 10-2 μm), Aitkin mode (10-2 μm – 0.1 μm), Accumulation mode (0.1 μm – 1 μm) and Coarse mode (>1 μm).
Nucleation mode aerosols are produced by condensation of hot vapors or freshly formed particles by the gas-to-particle conversion process. By means of rapid coagulation and/or condensation they get converted to accumulation mode particles. The Aitkin mode particles are formed by means of gas-to-particle conversion as well as condensation of hot vapours during combustion process. These particles act as nuclei for condensation of low vapor pressure gaseous species, thus converting to Accumulation mode. Aitkin and accumulation mode particles are sometimes called as fine mode particles. Coarse mode aerosols are produced by mechanical processes and can be from both natural and anthropogenic sources. For example, sea-spray generated due to the action of wind on the ocean surface produces sea-salt particles, similarly dust, soil and biological particles are also lifted by wind. Abrasion of machinery in various ways such as industrial and agricultural processes, contribute to anthropogenic coarse particles. Owing to their large size, coarse mode particles tend to settle faster due to gravity, thus reducing the residence time in the atmosphere.
1.2.3. Based on shape
The shape critically controls the particle’s optical properties (Mc Cartney, 1976) and hence radiative forcing. Scanning electron microscopy (SEM) studies have shown that liquid aerosol particles are nearly spherical, but solid particles usually have variable shapes (Hinds, 1999).
Depending upon the particle shape, aerosols are divided into three, such as (1) Isometric particles (2) Platelets and (3) Fibers.
6 1.3. Residence time
The residence time of aerosols at steady state is defined as the ratio of the aerosol concentration at that time to the production/loss rate. Residence time is mainly influenced by processes of formation, coagulation, removal and transformation (Jaenicke, 1993). Residence time of the nucleation mode aerosols is about a day in the atmosphere where as accumulation mode stay in the atmosphere for about 1 to 2 weeks, thus being amenable for long range transport. Moreover, coarse mode particles are easily removed by means of sedimentation.
1.4. Transformation Processes
Aerosol particles in the atmosphere are transformed from one size to another mainly through coagulation, condensation and cloud processing. These processes do not remove aerosol mass from the atmosphere but are capable of producing changes in the aerosol number density, size distribution and hence change in the optical and radiative properties.
1.4.1. Coagulation
Due to random motion, aerosols collide and coalesce to form larger chains made up of many particles. The Brownian motion of particles, turbulence and external forces such as gravity and electrical forces are believed to be responsible for coagulation process. Thus coagulation can be mainly viewed as a process for removing smaller atmospheric particles from the atmosphere.
7 1.4.2. Condensation
Process of forming aerosol mass from the condensation of a gaseous precursor, especially vapour is termed as nucleation. Net condensation of the vapour requires supersaturation, a partial pressure greater than its vapour pressure. Thus in condensational growth, aerosol leaves the nucleation mode.
1.4.3. Cloud processing
Once the particle is airborne, it undergoes various physical and chemical transformations such as change in particle size, structure and composition. Most efficient particle change occurs in clouds which are formed by condensation of water vapour on pre-existing aerosol particles.
Most clouds re-evaporate and modified aerosol particles are again released from the evaporating cloud droplets or ice crystals.
1.5. Aerosol sinks
Aerosol particles are removed from the atmosphere by means of dry and wet deposition.
During dry deposition, particles are removed from the atmosphere by sedimentation wherein large particles are affected. In wet deposition, aerosols are removed by collision with raindrops and by snowflakes and by uptake into cloud droplet or ice crystals which are washed out from the atmosphere by means of precipitation (also called as scavenging). Wet deposition, considered to be the most efficient removal processes, consists of three processes. These are (1) Rainout: As an indirect effect of aerosols, it acts as condensation nuclei for cloud droplet and during rainout some of the cloud droplet grow lager in size and reach ground in the form of
8 raindrop; (2) Washout: In washout process aerosols coexist with already existing cloud drop and when the drop expand the cloud droplet fall in the form of rain; (3) Sweep out: If the aerosols present at the base of the raining cloud are impacted by the rain drops, leading to the incorporation of aerosol in the rain drops, which eventually wash out in the form of rain (Seinfeld and Pandis, 1998). Different sources and sinks of aerosols are pictorially presented in figure.1.1.
Fig. 1.1. Pictorial representation of different sources and sinks of atmospheric aerosols (www.ems.psu.edu)
1.6. Role of aerosols in climate
Aerosols which scatter solar irradiance such as the sea-salt and sulfate aerosols enhances the planetary albedo and exerts negative climate forcing. Most species of phytoplankton, abundant in the ocean, excrete dimethylsulphide (DMS), which reacts in the air to form sulphate and methane sulphonate aerosol. Such a non sea-salt sulfate aerosol is found everywhere within the
9 boundary layer over marine environment and act as CCN (Korhonen et al., 2008; Woodhouse et al., 2010). Marine aerosols also are believed to play a significant role in the atmospheric sulfur cycle. In the lower troposphere, aerosols are important in the formation of fog, mist and clouds besides affecting the visibility. In the upper atmosphere, they produce chemical and electrical effect. The large atmospheric absorption by BC and its consequent potential to alter the radiation budget are well recognized (Jacobson, 2001; Babu et al., 2004, Nair et al., 2014).
It has also been suggested that BC light absorption alters the precipitation pattern and cloud lifetime thus decreases the reflectivity and increases the melting of snow and ice (Menon et al., 2002; Ackerman et al., 2000). The role of air mass trajectory in changing aerosol optical depths/composition/physical characteristics at distant location is well documented (Smirnov et al., 2002)
1.7. Rationale for the study
IPCC (2007) revealed that there has been considerable warming of the atmosphere since 1750 by GHGs. However, aerosols from different sources such as burning of fossil fuel and forest fires have either warmed or cooled the atmosphere over the time. Boucher and Haywood (2001) pointed out that climate forcing by aerosols offset as much as 25% of global warming caused by GHGs. Interestingly the long-lived GHGs have gained high level of understanding while climate forcing by aerosol is poorly characterized in climate models which is partly due to lack of comprehensive database on concentration, chemistry and optical properties of aerosols, which is clearly noticeable by large standard deviation (Fig. 1.2). Thus, presence of
10 both scattering and absorbing aerosols makes the understanding of atmospheric forcing more challenging which brings out the need for continuous and reliable global aerosol database.
Fig. 1.2. The global mean radiative forcing of the climate system by GHG’s and aerosols between 1750 to 2005 (IPCC, 2007).
Initial study by Langner and Rodhe (1991) by calculating global atmospheric aerosol fields and subsequent work by Charlson et al. (1992) on radiative properties of aerosols, lead to a number of dedicated field experiments in order to generate in-situ data for a better understanding. Some examples are Aerosol Characterization Experiment (ACE 1&2) (Bates et al., 1998; Raes et al., 2000), Smoke, Clouds, Aerosols, Radiation-Brazil (SCAR-B) (Kaufman et al., 1998), Troposphere Aerosol Radiative Forcing Observational Experiment (TARFOX) (Russell et al., 1999), Indian Ocean Experiment (INDOEX) (Ramanathan et al., 2001) and Arabian Sea Monsoon EXperiment (ARMEX) (Moorthy et al., 2004; Babu et al., 2004). Similarly as an outcome of the Indian Middle Atmosphere Program (IMAP) Indian Space Research Organization (ISRO) initiated Geosphere Biosphere Program (GBP) in the year 1990. One of the important components under this program is the estimation of aerosol radiative forcing over India. In this context, ISRO had
11 initiated network stations across the country as a part of Aerosol Radiative Forcing over India (ARFI) program. Under ARFI, several field campaigns were performed integrating air borne, ship borne and land based survey and results of the same are already documented (Niranjan et al., 2007; Moorthy et al., 2008; Kedia and Ramanathan, 2008; Nair et al., 2008; Gogoi et al., 2009; Beegum et al., 2009; Kumar et al., 2010; Ramachandran and Kedia, 2010; Pathak et al., 2010; Aloysius et al., 2011; Niranjan et al., 2011; Menon et al., 2011; Gularia et al., 2012;
Kompalli et al., 2014). Goa was identified as one of the network stations along west coast of India and aerosol measurements were initiated since 2008. Prior to this, studies were concentric to Trivandrum in the state of Kerala, and this was the only station along southwest coast of India carrying out aerosol studies extensively. Moreover, no concentrated efforts were initiated to study aerosol characterization and radiative forcing over Goa. Study carried out by Suresh et al. (2005) was an offshoot of their efforts to develop an algorithm to retrieve ocean color components from an optical sensor. Hence continuous measurements were performed on the AOD and mass concentration of BC over Goa since January 2008. In addition, field experiments were also carried out over Eastern Arabian Sea (EAS). Thus, with an aim to analyze aerosol both by in-situ and remotely sensed data, present work has been carried out with the following objectives,
1) To delineate the coarse and fine mode fraction of aerosol over a station in Goa and Eastern Arabian Sea (EAS)
2) To analyse the seasonal variation of black carbon and its effect on incoming solar radiation 3) Analysis of aerosol optical depth and aerosol dynamics over Arabian Sea and Bay of Bengal using OCM/ Moderate Resolution Imaging Spectroradiometer (MODIS) data.
12 Chapter 1 documents the details of the literature review on aerosols till the initiative of the study. In this chapter details of aerosol sources, sinks, and classification and production mechanisms are also discussed. In addition rationale for the study and subsequently objectives of the thesis are mentioned in this chapter. Chapter 2 provides an overview of the instruments used to generate data in the present work, details of the radiative transfer model and satellite data processing, different methodology adopted and brief introduction about the study areas namely a coastal site in Goa, Eastern Arabian Sea and Bay of Bengal constitutes part of this chapter. Chapter 3 describes microphysical properties of aerosols analysed for the period 2008 - 2010 over a coastal site in Goa and Eastern Arabian Sea. Detailed analysis of aerosol optical depth (AOD) and associated variation in Ångström parameters (α, β) (Ångström, 1961) are presented in this chapter. When the columnar AOD results from multitude sources a curvature in ln(AOD) and ln(λ) relationship is seen which is accounted by applying second order polynomial fit to the Ångström equation. Thus the chapter deals with detailed analysis of second order Ångström exponent (ά) and coefficient of the polynomial fit (α1 and α2) to understand the dominant aerosol type. Chapter 4 details the analysis of black carbon (BC), wherein the diurnal variation comprises of morning peak (~08:00 hrs) and afternoon minimum are investigated. Subsequently, seasonal variation of BC mass concentration between the years 2008 to 2010 has been discussed. Further, shortwave radiative forcing computed for the period 2008 – 2010 has been investigated with respect to local meteorological factors and remote forcing. Chapter 5 explains time series analysis of AOD using MODIS–Aqua data. Chapter also explains the spatial variability over Arabian Sea and Bay of Bengal for the period 2008 - 2010. Finally Chapter 6 integrates all the results. Summary, conclusion and scope for the future work is explained in this chapter.
13 2.1. Study area and general meteorology
Since the study has been carried out to understand the aerosol optical depth and microphysical properties and the associated radiative forcing over areas of different optical domains, study areas were carefully chosen so that they include coastal areas (subjected to different atmospheric phenomenon such as land-sea breeze and seasonal wind, coastal waters) and open ocean. Details of environmental set up are given below.
2.1.1. Coastal site in Goa, along the West Coast of India.
Goa is a tiny state along the west coast of India, surrounded by Arabian Sea (AS) at the west and Western Ghats at the east (Fig.2.1a).
Fig.2.1 (a) Network of ARFI stations across India (b) Sampling site at a coastal station in Goa
14 The selected study areas experience summer monsoon during June to September and winter monsoon during December to March. April/May and October/November are the transition periods between northeast to southwest monsoon season and vice-versa. Thus, the study areas experience four distinct seasons, namely winter monsoon season (December, January, February and March - WMS); spring inter-monsoon season (April and May - SIMS); summer monsoon season (June, July, August and September - SMS) and fall inter-monsoon season (October and November - FIMS). The sampling site located at Goa university campus (15.460N and 73.830E) is ~0.05 km above the mean sea level, ~0.7 km from the AS and ~5 km from the capital city of Goa, Panjim. There are no polluting industries located in the close proximity of the sampling site. Mining is a routine activity of Goa and Murmagoa port, which is closer to the sampling site, is busy in handling transportation of mining (Fig.2.1b). Coastal areas in India are inhabited by more than 50% of its over one billion population. Therefore, the associated anthropogenic activities contribute significantly to the aerosol loading, which is further modified by natural processes (Moorthy et al., 2007). One of the significant process affecting aerosol distribution in the coastal area is land-sea breeze. Thus a mixture of aerosol from natural processes and anthropogenic aerosols from land prevail on either side of the coastline.
Along with land-sea breeze, aerosol spatial and temporal variability are governed by meteorological factors such as wind speed, wind direction, relative humidity and air temperature. These parameters were measured using Automatic Weather Station (AWS).
Rainfall data for the study period were obtained from Indian Meteorological Department (IMD) situated ~4 km from the coastal study site. Seasonal variation of mean meteorological factors is shown in Table 2.1. Wind speed increased from 1.09 m/s in WMS to 1.37 m/s in SIMS and
15 attaining a peak to 2.10 m/s during SMS, further it was reduced to 1 m/s during FIMS. Wind blows mostly from south/southeast during WMS and southwest during SIMS. During SMS, wind direction changes to southerlies/southwesterlies which changes to southerlies/easterlies during FIMS. Relative humidity showed an increasing trend from WMS to SMS, thereafter decreased as FIMS approached. SMS received above normal rainfall over the study region which was ~725 mm. Considerable amount of precipitation was also observed during FIMS (161 mm), which is a transition phase. Rainfall during April/May constituted to a seasonal mean of 30 mm. Negligible amount of precipitation (~17 mm) was noticed in December during WMS. Synoptic wind at 850 hPa using the data from National Center for Prediction (NCEP) revealed that during WMS, winds were moderate and easterlies, while during SIMS weak northwesterlies were observed. Further, during SMS strong winds originating from southwest while moderate northeasterlies winds were noticed during FIMS (Fig. 2.2).
Table. 2.1. Seasonal variation of mean meteorological parameters during WMS, SIMS, SMS and FIMS for the period between 2008 and 2010.
Meteorological parameters WMS(D,J,F,M) SIMS(A,M) SMS(J,J,A,S) FIMS(O,N) Wind speed (m/s) 1.09±0.07 1.37±0.02 2.10±0.45 1.00±0.11 Wind Direction (deg) 171±14.91 232±2.70 226±6.95 153±3.42
Relative Humidity (%) 70±1.13 73±2.11 88±5.2 80±9.7
Temperature (0C) 26.34±0.67 29.32±0.49 26.77±0.14 27.31±0.24
Rainfall (mm) 17±22.22 30±33.10 725±106 161±138
16 Fig. 2.2. Synoptic wind vector (m/s at 850 hPa) over India during (a) WMS (b) SIMS (c) SMS and (d) FIMS.
2.1.2. Eastern Arabian Sea (EAS)
Arabian Sea (AS) constitutes one of the components of northern India Ocean, bounded by the Indian subcontinent along the east, Pakistan and Iran at the north, and Saharan region along the west (Fig.2.3a). Area experiences seasonally reversing wind, which leads to a mixture of clean southern hemispheric air and polluted air from the continents.
17 Fig. 2.3. (a) Shows the cruise track of CRV Sagar Sampada (SS) ,blue color indicates cruise track of CRV SS-265 during the year 2009 and red indicates CRV SS-274. (b) Shows, Upper zone (4th -6th April 2009 & 2nd - 4th May 2010), Middle zone (9th - 10th April 2009 & 29th April 2010), Lower zone (13th - 14th April 2009), Lower zone-1 (21st – 22nd April 2010) and Lower zone-2 (25th -26th April 2010).
In the present study, EAS covered the region between Goa to Cochin wherein the study period was SIMS of 2009 and 2010. Coastal research vessel (CRV) Sagar Sampada was deployed to carry out the survey. The first cruise CRV Sagar Sampada (CRV-265) sailed from Goa (150N
&730E) on 4th April 2009 and reached Cochin (90N &740E) on 14th April 2009, while the second cruise (CRV - 274) sailed from Cochin on 21st April 2010 and reached Goa on 4th May 2010 (Fig.2.3b). These measurements were planned and executed in such a way that the cruise was initiated in the first half of April 2009 while that in 2010 was in the second half of April.
18 This criterion was evolved to understand precisely the effect of receding northeast monsoon on aerosol dynamics of the area. Since there was a large spatial heterogeneity in the aerosol distribution, the area was divided into three zones, namely upper zone, middle zone and lower zone. Upper zone was the region between the coastal waters of Goa and Karwar; middle zone was the coastal waters off Mangalore while lower zone was the region from Cannanore to Cochin. Since this zone covered quite a large distance, this was further subdivided into lower zone-1 (regions off Cannanore and Bepore) and lower zone-2 (regions off Cochin and Ponnani). Meteorological observations such as wind speed, wind direction, relative humidity and temperature were measured using AWS during the study period and the details are shown in Table 2.2. During the observation in 2009, the mean atmospheric temperature along the cruise track was ~300C. Relative humidity varied between 72% and 83%. The variation of RH resembled a wavy like appearance between Goa and Cochin with comparatively high values off Goa and a low off Mangalore coast with a subsequent high off Cochin.
In general, the wind direction during the period was from easterly to southeasterly, with wind speed varying between 2 m/s and 6 m/s. Similarly during the observation in 2010, atmospheric temperature varied between 290C and 310C. RH values were higher (~79%) off Goa (2nd May to 4th May) and also off Cochin (21st April and 22nd April). The highest wind speed of 6.3 m/s was observed on 2nd May and the lowest of 0.4 m/s was encountered on 25th April. The direction of the wind was north and northwest. The NCEP/NCAR reanalysis revealed that during 2009, wind direction was predominantly easterly with moderate wind speed, which decreased further over the central AS, making significant advection of continental aerosol.
However, during 2010, strong anticyclonic motion was observed north of AS and a weak
19 westerly towards southeast (Fig. 2.4).
Fig. 2.4. Synoptic wind vector (a) during 2009 and (b) during 2010.
20 Temperature (0C) Relative Humidity (%) Wind Speed (m/s) Wind direction (Deg)
2009
4th April 29.89±0.78 83.80±3.32 2.86±1.31 215.31±117.05 5th April 29.96±0.83 81.19±2.31 3.53±2.23 129.74±117.12 6th April 29.98±0.64 79.80±0.64 6.02±3.79 53.34 ±12.84 7th April 30.30±0.53 80.80±2.27 4.22±2.99 163.05±118.26 8th April 30.16±0.43 78.23±4.46 3.11±1.71 147.86±118.10 9th April 30.47±0.35 72.10±3.35 4.74±1.59 258.99±103.47 10th April 30.00±0.42 74.00±1.87 5.50±1.74 72.77±18.15 11th April 29.24±1.26 77.17±3.66 4.69±2.32 150.50±104.54 12th April 29.17±0.58 79.99±2.34 3.70±1.01 157.21±103.24 13th April 29.63±0.65 78.35±4.52 4.78±1.73 136.60±114.60 14th April 30.32±0.45 75.06±3.29 3.60±1.05 236.56±93.80
2010
21st April 30±1.00 79±1.0 3.8±0.9 325±100
22nd April 31±0.23 79±2.0 3.23±1.08 320±85
25th April 29±0.35 70±4.0 0.4±0.2 270±110
26th April 29±2.00 78±3.0 4±2.0 312±112
29th April 31±0.70 75±4.0 3.5±1.2 302±54
2nd May 29±2.00 79±1.0 6.3±2.5 350±25
3rd May 30±1.00 80±2.0 3.4±1.3 334±65
4th May 31±2.0 79±1.0 4.1±0.5 337±73
Table. 2.2. Daily variation of meteorological parameters during the respective cruises in 2009 and 2010
21 2.1.3. Bay of Bengal
Bay of Bengal (BoB) is the sector of north Indian Ocean between 80°E and 100°E and 5°N and 22°N which is surrounded by densely populated continental landmasses (Fig.2.5). It is surrounded by Indian subcontinent and Sri Lanka along west, Myanmar and Andaman and Nicobar Island towards the east and Bangladesh towards North. During WMS, BoB is influenced by the anthropogenic transport of air from the Indian subcontinent and Southeast Asia, particularly from Indonesia. A deep air flow ~2.5 m thick, originating from Kolkata and Bangladesh extends south and east over BoB, which subsequently merges with the Asian plume and move south and west (Niranjan et al., 2006). Thus aerosol study gains significance over this region due to the influx of large amount of anthropogenic aerosols. To have a better understanding of such aerosols over the region, Indian Space Organization (ISRO) launched a Winter Integrated campaign for Aerosol gases and Radiation Budget (WICARB) from 27th December 2008 to 30th January 2009 over entire BoB. The ship sailed on 27th December from Chennai port (13.10N, 80.20E), cruising entire BoB and eventually arrived at Cochin port on 30th January 2009 after crossing Sri Lanka on 28th January 2009.
22 Fig. 2.5. The cruise track of Sagar Kanya 254 during W-ICARB, 27 December 2008 – 30 January 2009.
2.2. In-situ measurements
2.2.1. MICROTOPS II Sunphotometer (a) Principle of Operation
Aerosol Optical Depth (AOD) has been estimated using MICROTOPS II Sunphotometer at five wavelengths namely 0.380 μm, 0.440 μm, 0.500 μm, 0.675 μm and 0.870 μm following the principle of Bourguer Lambert-Beer law. Instrument consists of narrow-band interface and a photodiode suitable for these bands. The instrument has five accurately aligned optical collimators with a field of view of 2.50. A quartz window provides access to collimator tubes
23 for the five photodiodes and a sun alignment target. The offset of each channel is adjusted to zero while the photodiode view the inner black surface of the window door.
The optical block determines the field of view, filters the incoming radiation, and detects and enhances sun-targeting. The optical block is made from a cast aluminum plate for assuring mechanical stability for a long-term measurement. Internal baffles and low reflectance lining in each collimator prevent internal reflection. The signals from the photodiode are amplified, converted to digital form and processed in the signal-processing block. A sun target and pointing assembly which is laser-aligned to within 0.10 of the optical axis of the block, ensures accurate alignment with the optical bands. As the sun‟s image is centered in the bull‟s eye of sun target, all optical bands are oriented at the solar disk. A self-contained microcomputer computes AOD and the irradiance at each wavelength, wherein a nonvolatile memory stores upto 800 scans consisting of time and date, internal temperature, barometric pressure and the geographic coordinates. The time (universal), date of observation and geographical coordinates have to be entered manually or automatically using the Global Positioning System (GPS) (Morys et al., 2001). Daily observations were carried out from 0900 to 1730 hrs local time at 30 minutes interval, avoiding the period of obstruction of the sun by passing clouds.
(b) Theory
For a nearly monochromatic radiation of wavelength „λ‟, the ground reaching direct solar flux
„F(λ)‟, is related to the extraterrestrial flux „F0(λ)‟ {at the Top of the atmosphere(TOA)}
through Lambert-Bourguer Beer law.
F (λ) = F0(λ) {do/d}2 exp (-m τλ) (2.1)
24 Where,
do and d are the mean and instantaneous sun earth distances. τλ is the total columnar optical depth of the atmosphere at wavelength λ and m is the relative air mass.
A radiometer gives an electrical output signal V (λ), which is directly proportional to F (λ).
Therefore Equation (2.1) becomes
V (λ) = V0(λ) {do/d}2 exp (-m τλ) (2.2) Where the radiometer output V0 (λ) is corresponding to F0 (λ).
In logarithmic form, equation (2) becomes log V (λ) = logV0 (λ) +2 log {do/d} -m τλ.
Total optical depth is given by
τλ = τmλ + τpλ + τO3λ+ τwλ (2.3)
τmλ = extinction due to molecules τpλ = extinction due to aerosols τo3λ = absorption due to ozone τwλ = absorption due to water vapour
If continuous time series measurements of V (λ) are made during a day, plot of log (V (λ)) against m, will show the points very close to a straight line. The slope of the straight line yields the total optical depth. This procedure is widely known as Langley technique (Shaw et al., 1973). From the columnar total optical depth, AOD can be deduced by accounting for the effect
25 of molecules.
(c) Error estimation
The error in AOD estimation typically ranges between 0.005 and 0.02, which is primarily due to inaccurate calibration, error in correction of molecular scatter and absorption, and faulty instrumentation (Porter et al., 2001). Ichoku et al. (2002) suggested that as long as Microtops is calibrated regularly (at least once a year) and cleaned frequently, the calibration coefficient are stable and vary slowly with time. Sunphotometers are calibrated by using the Langley plot method following Beer-Lambert law on a clear sky day (see section 2.2.1b). The Mauna Loa Observatory on Hawaii Island is normally chosen for calibration of sunphotometer owing to low and stable AOD, throughout the year, and due to the location which lies above the trade wind inversion. Moreover, over this region it takes less than 2 hrs to cover the air mass from 5 to 2, sufficient for good calibration. Bias towards higher optical depth is caused due to error in pointing towards the sun and hence certain criteria are adopted while applying Langley method.
These are 1) fit a line to the data, 2) eliminate any data points that are more than 0.1% below the line, and 3) refit the line (Morys et al., 2001).
2.2.2. Aethalometer
(a) Principle of Operation
Continuous and near real time measurements of the Black Carbon (BC) mass concentration were performed using a seven channel Aethalometer namely 0.370 μm, 0.450 μm, 0.520 μm, 0.590 μm, 0.660 μm, 0.880 μm and 0.950 μm. This instrument uses continuous filtration and optical transmission technique to measure concentration of BC. The major component of
26 Aethalometer is the optical head, consisting of aerosol inlet, the optical source assembly, the light guides for the photo-detectors and the filter tape support. The instrument aspirates ambient air through a cyclone type inlet that pre-segregates all the particles larger than 1 μm. The particles in the incoming airflow are deposited on the quartz fiber filter tape of the Aethalometer. Such kind of filter tape which are thermally stable to high temperature, has a deep mat of optically scattering fibers within which the aerosol particles are collected so as to nullify any effect on optical transmission by light scattering from the deposited particles (Hansen et al., 1984). The instrument was operated throughout the year at a flow rate of 2 liters per minute (LPM) and a time base of 15 minutes, so that BC mass concentration was available round the clock. The inlet of the instrument was placed ~10 m above the ground. Any uncertainty in the readings due to varying relative humidity was taken care by connecting a heated sampler line to the inlet of the Aethalometer. Moreover, data recorded during isolated events such as festivals and events of fire were filtered out.
(b) Theory
The absorption coefficient (babs) is defined using Beer-Lambert‟s law as follows,
I=I0 b abs
e
(2.4)
Where I0 is the intensity of the incoming light and I the remaining light intensity after passing through a medium with thickness X. The light attenuation (ATN) is given by,
ATN ln (I0/I). (2.5)
27 Light attenuation is measured through a quartz filter matrix, wherein the fiber filter acts as a perfect diffuse scattering matrix in which the light absorbing particles are embedded. Further, light transmission through the filter is monitored by two detectors such that one detector measures light passing through sample spot and the other measures the light passing through a reference spot (which is blank). Such monitoring ensures correction of variation in light intensity and drift in electronics. Absorption coefficient of the filtered aerosol particles bATN, known as the attenuation coefficient is defined as
bATN
A ATN
Q t , (2.6)
Where A is the filter spot area, Q the volumetric flow rate and ATN is the change in attenuation during the time interval t. bATN and babs may be significantly different and hence calibration factors C and R (ATN) are introduced as follows,
babs = bATN 1
( )
C R ATN , (2.7)
Where C and R describe the effect, which change the optical properties of particle embedded on the filter with respect to properties of the same particle while in air. Multiple scattering of the light at the filter fibers in the unloaded filter causes C being greater than unity thereby enhancing the optical depth and hence light absorption of the deposited particles. Ballach et al.
(2001) demonstrated that these scattering effects can be minimized by immersion of the particle loaded filters in oil with a refracting index similar to the filter fibers. Other effects caused by deposited particles are described by empirical function R (ATN) which varies with (1) the
