Publications·November 30, 2009
Ulaanbaatar (UB) has earned the unwelcome title of one of the world’s most polluted capitals. Rapid population growth—especially in peri-urban ger districts—has combined with harsh winters, an inversion-prone valley setting and universal reliance on coal to produce extreme particulate-matter (PM) episodes. The World Bank’s Air-Monitoring and Health-Impact Baseline (AMHIB) project therefore set out to:
compile the best available monitoring data,
build a preliminary emissions inventory,
model the city-wide dispersion of PM₁₀ and PM₂․₅,
quantify population-weighted exposure (PWE) and health costs, and
test the benefits of realistic abatement packages for household stoves, heat-only boilers (HOBs) and road dust control.
2 How the assessment was done
Component Key inputs & tools
Monitoring audit Pre-2008 datasets from CLEM, NAMHEM, NUM (1 PM and 4 SO₂/NO₂ stations) + seven new AMHIB PM stations (Jun 2008–May 2009). Instrument inter-comparisons exposed flow-rate drift and filter artefacts that under-reported PM₂․₅ by ≈30 %.
Emissions inventory (2007 base year) 130 000 ger stoves, ±250 coal-fired HOBs, 3 CHPs, traffic counts on 80 road links, suspension factors for paved/unpaved dust; open burning and industrial stacks treated qualitatively.
Dispersion modelling 1 km × 1 km CALPUFF grid, year-long hour-by-hour meteorology, inversion depth from radiosonde, valley wind-rose, back-trajectories.
Health valuation WHO concentration-response coefficients, UB-specific baseline incidence, willingness-to-pay (income elasticity 0.5–1.0) for mortality and chronic bronchitis.
3 What was found
Chronic exceedance – Annual PM₁₀ at NUM rose from 141 µg m⁻³ (2006) to 279 µg m⁻³ (2008); winter peaks exceeded 1 000 µg m⁻³ daily and 2 300 µg m⁻³ hourly—up to 14 × WHO 24-h guidance.
Source split (population-weighted, 2007):
Ger stoves 50–60 % of PM₂․₅;
HOBs 15–20 %;
Suspended road & soil dust 20 %;
CHP stacks < 10 % thanks to 100–250 m stacks.
Exposure gradient – Northern ger districts record the highest annual mean; apartment cores are ~30 % lower, yet still 5–7 × Mongolian standards.
Health burden – Base-year modelling attributes ≈ 600 premature deaths, 380 chronic-bronchitis cases and 1 180 hospital admissions annually to PM in UB—monetised at US $147 million yr⁻¹ (≈ 8 % of city GDP). Lower VSL assumptions still yield 4 % of GDP.
4 Abatement scenarios tested
Scenario Emission cuts modelled PM₂․₅ ↓ (PWE) Avoided deaths/yr Notes
“30 / 50 / 80” (ger/HOB/dust) 30 % stove upgrade + 50 % HOB retrofit + 80 % paved/unpaved dust control –59 µg m⁻³ ~244 Meets Mongolian 24-h PM₁₀ limit in most districts.
75 % ger / 83 % HOB / 50 % dust Deep stove replacement + high-efficiency HOB ESP + sweeping & paving –75 µg m⁻³ ~308 First pathway to approach WHO interim-3 for PM₂․₅.
Fuel switch only (80 % ger) Coal → clean coal/LPG/electric –273 deaths saved but still >3× WHO annual guideline Highlights dominance of household heating.
Benefits outrun costs in all multi-sector packages once fuel-savings, avoided morbidity and visibility gains are counted; household measures give the highest benefit-cost ratios.
5 Six strategic messages
Household coal is the core problem. Even wholesale retrofitting of HOBs and street paving cannot meet standards without a ger-heating transition.
Monitoring overhaul is urgent. A credible AQM system demands calibrated reference BAMs, co-located PM₂․₅/NO₂/SO₂/O₃ analysers, QA/QC SOPs and data transparency.
Short-term quick win: promote smokeless ignition practices (portable fire-pots & top-down lighting) that cut evening PM peaks 15-25 %.
Dust matters. Paving and mechanical sweeping of unpaved lanes in ger areas halves coarse-PM and is cost-effective compared with end-of-pipe retrofits.
Health costs dwarf abatement bills. Even conservative valuations peg current damage at 4-8 % of UB GDP—ample fiscal space for stove subsidies and road maintenance.
Integrated governance needed. A National Clean Air law, stable stove-credit funds, and ger-area energy-service concessions can align city, ministry and donor programmes.
6 Key words
PM₂․₅; ger-stove emissions; heat-only boilers; winter inversion; dispersion modelling; population-weighted exposure; health-cost valuation; dust control; Ulaanbaatar Clean Air Program.
The discussion paper presents preliminary findings for dissemination to initiate a discussion on the short-term and long-term emission reduction strategies for reducing Ulaanbaatar's (UB) air pollution, given its changing demographics growing population and a growing urbanization. This discussion paper assesses the current air quality situation in UB, based on available data from 2006 to 2008, and estimates effects of pollution abatement options on ambient concentrations of particulate matter (PM). Population exposure estimates are used to assess current health damage attributable to air pollution in the city and the health benefits that can be achieved from implementing the abatement options. This World Bank study launched a collection of PM data in seven monitoring stations in UB in July 2008. The final paper is due in early 2010 and will incorporate more recent pollution measurements.