Air Pollution — Sources, Standards & Control

1. Types of Air Pollutants

1.1 Primary vs Secondary Pollutants

1.2 Classification by Physical State

2. Sources of Air Pollution

3. National Ambient Air Quality Standards (NAAQS 2009)

India’s NAAQS are prescribed by the Central Pollution Control Board (CPCB) under the Air (Prevention and Control of Pollution) Act, 1981. The current standards were revised in 2009 to add PM₂.₅ and tighten PM₁₀ limits.

India’s PM₂.₅ annual standard of 40 μg/m³ is 4× the WHO guideline (10 μg/m³, 2021). Most Indian cities have annual PM₂.₅ averages of 50–120 μg/m³ — well above even the relaxed national standard. Delhi’s annual average PM₂.₅ was ~96 μg/m³ in 2023.

4. Health and Environmental Effects

4.1 Health Effects by Pollutant

4.2 Acid Rain

4.3 Photochemical Smog

Photochemical smog forms when VOCs and NO_x react in sunlight to form ground-level ozone, PAN, and aerosols. It causes the hazy, yellowish-brown smog seen in cities with heavy traffic in sunny weather. The Los Angeles smog of the 1940s was the first recognised photochemical smog. Delhi’s winter smog (Nov–Jan) is a combination of photochemical smog and primary PM₂.₅ from crop burning and vehicle emissions, trapped by temperature inversions.

5. Atmospheric Dispersion — Gaussian Plume Model

The Gaussian plume model describes the steady-state concentration of a pollutant downwind of a continuous point source (stack). It assumes: continuous, uniform emission rate Q; wind is steady and uniform at speed u; the plume disperses as a Gaussian (normal) distribution in both horizontal (y) and vertical (z) directions; no chemical reactions or deposition.

5.1 Gaussian Plume Equation

5.2 Maximum Ground-Level Concentration (C_max)

6. Effective Stack Height and Plume Rise

6.1 Minimum Stack Height (CPCB)

7. Atmospheric Stability Classes

Atmospheric stability determines how much the atmosphere mixes pollutants vertically. Pasquill-Gifford stability classes (A to F) are the standard classification used with the Gaussian model.

Temperature inversion: A temperature inversion occurs when warm air overlies cool air — the normal lapse rate is reversed (temperature increases with height instead of decreasing). Inversions trap pollutants below the inversion layer, leading to extreme pollution episodes (Class E/F conditions). The Delhi winter smog is largely due to nocturnal inversions trapping PM₂.₅ and other pollutants.

7.1 Fumigation

Fumigation occurs when an elevated plume (from a tall stack) encounters unstable air below an inversion layer — the plume is mixed rapidly downward to ground level, causing very high short-duration ground-level concentrations. It is the most dangerous dispersion scenario for stack emissions and occurs typically in the morning when the mixed layer grows upward from the heated ground and reaches the elevated plume.

8. Particulate Matter Control

8.1 Collection Efficiency

8.2 Cyclone — Cut Size

9. Gaseous Pollutant Control

9.1 Flue Gas Desulphurisation (FGD)

10. Air Quality Index (AQI)

India’s Air Quality Index (AQI) was launched by CPCB in 2014 as part of the Swachh Vayu (Clean Air) initiative to communicate ambient air quality to the public in a simple, standardised way.

10.1 AQI Calculation

10.2 AQI Categories

11. Worked Examples (GATE CE Level)

Example 1 — Ground-Level Concentration from Gaussian Model (GATE CE 2022 type)

Problem: A stack emits SO₂ at Q = 50 g/s with effective stack height H = 80 m. At 2 km downwind, σ_y = 150 m and σ_z = 50 m. Wind speed u = 5 m/s. Find the ground-level concentration at the plume centreline (y = 0, z = 0).

Answer: C(2km, 0, 0) = 118 μg/m³ (exceeds NAAQS 24-hr SO₂ standard of 80 μg/m³)

Example 2 — Effective Stack Height (GATE CE type)

Problem: A stack is 60 m tall with internal diameter d = 1.5 m. Stack gas exits at v_s = 8 m/s and temperature T_s = 400 K. Ambient temperature T_a = 293 K. Wind speed u = 4 m/s. Atmospheric pressure P = 1000 mb. Find the effective stack height using Holland’s formula.

Answer: Δh = 7.73 m; Effective stack height H = 67.7 m ≈ 68 m

Example 3 — Effect of Stack Height on C_max (GATE CE type)

Problem: A stack emits pollutant Q = 100 g/s. At x_max downwind, σ_y = σ_z = H/√2 = 35 m. Wind speed u = 5 m/s. Current effective stack height H = 50 m. Find (a) C_max at H = 50 m and (b) C_max if stack height is doubled to H = 100 m (assuming σ_y = σ_z = H/√2 = 70 m at new x_max).

Answer: C_max(H=50) = 1905 μg/m³; C_max(H=100) = 476 μg/m³; doubling H reduces C_max by 4×

Example 4 — ESP Collection Efficiency (GATE CE type)

Problem: An electrostatic precipitator has an inlet particulate loading of 10 g/m³ and outlet loading of 0.08 g/m³. Find (a) the collection efficiency and (b) if a second ESP unit is placed in series with 95% efficiency, find the overall collection efficiency.

Answer: η₁ = 99.2%; Combined η = 99.96%; Final outlet = 4 mg/m³

Example 5 — Minimum Stack Height from CPCB Formula (GATE CE type)

Problem: An industry emits SO₂ at 120 kg/hour. Find the minimum stack height required using the CPCB formula H = 14Q^0.3.

Answer: Minimum stack height = 60 m

Example 6 — AQI Calculation (GATE CE type)

Problem: A monitoring station records: PM₂.₅ = 75 μg/m³ (24-hr average); PM₁₀ = 90 μg/m³. Using India’s AQI breakpoints (PM₂.₅: 0–30 = 0–50, 30–60 = 51–100, 60–90 = 101–200; PM₁₀: 0–50 = 0–50, 50–100 = 51–100, 100–250 = 101–200), find the AQI sub-indices and overall AQI.

Answer: PM₂.₅ sub-index = 151; PM₁₀ sub-index = 90; AQI = 151 (Moderate category)

12. Common Mistakes

Mistake 1 — Using Physical Stack Height Instead of Effective Stack Height H in Gaussian Model

Error: Substituting the physical stack height h (e.g., 60 m) instead of H = h + Δh (e.g., 68 m) in the Gaussian plume equation.
Root Cause: The plume doesn’t start at the physical stack top — it rises further due to momentum and buoyancy before dispersing horizontally. This additional rise (Δh) significantly raises the effective emission height H, which reduces ground-level concentrations.
Fix: Always compute H = h + Δh using Holland’s formula (or Briggs’ formula) before applying the Gaussian model. H > h always (Δh is always positive for hot stack gases).

Mistake 2 — Using the Wrong Form of the Gaussian Equation for Ground-Level Concentration

Error: Using C(x,y,z) = Q/(2πσ_yσ_zu) × exp[–y²/(2σ_y²)] × exp[–(z–H)²/(2σ_z²)] for the ground (z=0), forgetting the ground reflection term.
Root Cause: The full Gaussian equation has two exponential terms in z — one for the direct plume and one for the reflected plume (from the ground). At ground level (z = 0), both terms combine: exp[–(H)²/(2σ_z²)] + exp[–(H)²/(2σ_z²)] = 2exp[–H²/(2σ_z²)]. The factor of 2 in the ground reflection causes the simplified formula C(x,0,0) = Q/(πσ_yσ_zu) × exp[–H²/(2σ_z²)] (note: 2π→π denominator accounts for the reflected plume).
Fix: For ground-level centreline (y=0, z=0): C = Q/(πσ_yσ_zu) × exp[–H²/(2σ_z²)]. For elevated receptor: use full equation with reflection term. The π in denominator (not 2π) is correct for ground-level.

Mistake 3 — Confusing Overall Efficiency with Series Device Combination

Error: Adding efficiencies instead of multiplying penetrations: η_total = 99.2 + 95 = 194.2% (impossible).
Root Cause: Efficiencies cannot be added because each device acts on the outlet of the previous device, not on the original inlet stream. Series calculation must track remaining pollutant (penetration).
Fix: Series devices: P_total = P₁ × P₂ × … then η_total = 1 – P_total. Never add efficiencies directly. For P₁ = 0.008, P₂ = 0.05: P_total = 0.0004; η = 99.96%.

Mistake 4 — Using Mass Loading Instead of Concentration in AQI Calculation

Error: Plugging PM₂.₅ in kg/day (mass loading) into AQI sub-index formula instead of 24-hour average concentration in μg/m³.
Root Cause: AQI requires ambient air concentration (μg/m³) from monitoring, not emission mass loading from sources.
Fix: AQI uses 24-hour average ambient concentration in μg/m³ (or ppm for CO, O₃). Source emissions (in g/s, kg/day) are used in dispersion modelling — the dispersion model converts emission rate to ambient concentration.

Mistake 5 — Treating PM₂.₅ as Less Harmful Than PM₁₀ Due to Smaller Size

Error: Concluding “PM₁₀ is more dangerous because it is larger” — mixing up particle size and health risk.
Root Cause: Counter-intuitive: smaller particles are MORE dangerous because they penetrate deeper into the respiratory system. PM₁₀ (≤10 μm) penetrates the upper respiratory tract but is largely filtered by the nose/throat/bronchi. PM₂.₅ (≤2.5 μm) penetrates the alveoli (deepest lung tissue) where gas exchange occurs and can enter the bloodstream — causing cardiovascular and systemic effects. PM₀.₁ (ultrafine, ≤0.1 μm) can enter the brain via the olfactory nerve.
Fix: Health impact: PM₀.₁ > PM₂.₅ > PM₁₀. Regulations are correspondingly stricter for finer particles: NAAQS annual PM₂.₅ = 40 μg/m³ vs PM₁₀ = 60 μg/m³ (PM₂.₅ more stringent).

13. Frequently Asked Questions

Q1. Why is the Gaussian plume model only valid under specific atmospheric conditions, and what are its main limitations?

The Gaussian plume model assumes: (1) steady-state conditions — wind speed, direction, and turbulence are constant; (2) the emission rate Q is continuous and uniform; (3) the plume disperses as a normal distribution in both horizontal and vertical directions; (4) no chemical reactions or deposition of pollutants; and (5) flat, homogeneous terrain. These assumptions make the model analytically tractable but deviate significantly from reality. Real atmospheric flow is turbulent and time-varying; wind direction shifts; the terrain (buildings, hills) distorts the plume; chemical reactions transform pollutants; and wet/dry deposition removes them. The model breaks down near the stack (within ~1 km), in complex terrain (hills, valleys), during calm or near-calm winds, during fumigation conditions, and for reactive species (ozone, NO₂). Despite these limitations, the Gaussian model remains the standard regulatory tool globally (and specifically under India’s Environment Impact Assessment notifications for industrial projects) because it is computationally simple, requires minimal input data, and gives conservative estimates suitable for worst-case screening. More complex models (AERMOD, CALPUFF, CFD-based) are used for detailed impact assessments where regulatory decisions require higher accuracy.

Q2. What is the difference between an ESP and a fabric filter (baghouse) for particulate control, and which is better for Indian power plants?

Both ESPs and baghouses achieve very high collection efficiency (95–99.9%) but via different mechanisms. The ESP uses an electrostatic charge: a corona discharge ionises particles, which then migrate to oppositely charged collecting plates under the electric field. Particles are removed by mechanical rapping (plates vibrate, shedding collected dust). ESPs handle very large gas volumes (millions of m³/hour from large boilers) at low pressure drop (0.5–1.5 kPa), making them energy-efficient for high-volume applications. Their weakness is sensitivity to particle resistivity (highly resistive fly ash, common from some Indian coals, reduces ESP efficiency significantly because charge leaks through the collected layer). Baghouses use fabric filtration: flue gas passes through tightly woven or felted fabric bags; particles are captured on the bag surface and periodically shaken/pulse-jetted off. Baghouses achieve higher efficiency for fine particles (PM₂.₅) and are less sensitive to particle properties. Their weakness is higher pressure drop (1–3 kPa), heat sensitivity (fabrics limited to 250–300°C), and inability to handle very sticky or hygroscopic particles. For Indian power plants, ESPs have historically been standard, but baghouses (or ESP-baghouse hybrids) are increasingly required to meet the stricter 2015 PM emission norms (100 mg/Nm³ for existing plants; 50 mg/Nm³ for new plants) because Indian coal fly ash often exhibits high resistivity that reduces ESP performance.

Q3. How does the Air (Prevention and Control of Pollution) Act 1981 relate to the Environment Protection Act 1986 in controlling air pollution in India?

The Air Act 1981 was India’s first specific legislation for air pollution control, establishing the Central and State Pollution Control Boards, empowering them to prescribe ambient air quality standards, set emission standards for industries, and grant/refuse consent to operate facilities. However, it was limited to industrial sources. The Environment Protection Act 1986 (EPA) is an umbrella legislation enacted after the Bhopal gas tragedy, giving the Central Government broad powers to take all measures to protect and improve environmental quality. The EPA covers all environmental media (air, water, soil) and all sources (including vehicles — though vehicle emissions are now primarily under the Motor Vehicles Act and BS emission norms). The EP Rules 1986 Schedule 1 lists ambient standards and Schedule 2 lists industrial effluent standards, superseding some Air Act provisions. In practice: the Air Act 1981 → SPCBs regulate industrial consent-to-operate (stack emission standards). EPA 1986 → Central Government issues notifications for specific sectors (thermal power plants, cement, paper mills) with stricter emission standards. BS norms (Bharat Stage, equivalent to Euro standards) → Ministry of Road Transport and Highways → regulate vehicle tailpipe emissions. All three operate simultaneously, with EPA generally taking precedence in case of conflict.

Q4. Why does India have significantly higher ambient PM₂.₅ concentrations than developed countries, despite having similar emission density in some regions?

India’s consistently high PM₂.₅ levels (national average ~55–65 μg/m³ annual mean vs EU average ~12 μg/m³) result from several compounding factors beyond just emission density. First, India’s meteorology is unfavourable: the Indo-Gangetic Plain (where most of India’s largest cities are located) has frequent temperature inversions (winter, post-monsoon), low wind speeds during winter anticyclonic conditions, and bounded by the Himalayas to the north — limiting dispersal of pollutants. Second, India’s high ambient temperature and strong solar radiation accelerate photochemical production of secondary PM₂.₅ and ozone from precursor emissions. Third, the emission mix includes large proportions of primary PM₂.₅ and black carbon from biomass combustion (open burning, cooking stoves) which is not adequately captured in official emission inventories. Fourth, India’s dust environment is particularly severe — both mineral dust from Thar Desert transport and construction/road dust contribute significantly to coarse PM. Fifth, rapid urbanisation without adequate air quality planning and enforcement has allowed emission sources to grow faster than control measures. The National Clean Air Programme (NCAP 2019) targets 20–30% PM reduction by 2024 in 102 non-attainment cities (cities where NAAQS are not met) — the first national-level multi-year PM reduction target in India’s history — using a combination of emission source control, green cover expansion, public transport promotion, and stringent enforcement.