A free, data-light utility panel for daily life in Mongolia. It puts four things on one fast page: live PM2.5 and PM10 for four Ulaanbaatar reference points (Bayankhoshuu, Zuragt, the city centre and Zaisan), with the numbers translated into multiples of the WHO guideline so they actually mean something; weather and computed wind chill for seven aimag centres from Khovd to Choibalsan, with a 7-day outlook; live tugrug exchange rates (USD, CNY, RUB, KRW against MNT) and a converter; and a set of herder tools — a private on-device livestock log plus plain-English explainers of dzud and index-based livestock insurance (IBLI). No app install, no account, no ads. Everything loads from free public APIs and the page works on a cheap phone over a slow connection, in English or with Mongolian interface labels.
Roughly half of Ulaanbaatar’s population lives in the ger districts — sprawling hillside neighbourhoods of fenced plots, small houses and traditional felt gers that ring the city’s north and west. Most plots are not connected to the central heating grid, so through a winter that regularly sits below −25 °C each household burns raw coal (and, in the hardest times, whatever else burns) in a simple stove, every day, from October to April. Tens of thousands of stoves lighting up at dusk act like one enormous, unfiltered chimney. On the coldest January evenings, monitors in districts such as Bayankhoshuu have recorded PM2.5 concentrations dozens of times above the WHO 24-hour guideline — episodes that rank among the worst urban air readings recorded anywhere. The switch to refined briquettes after 2019 helped, but winter pollution remains a defining fact of city life.
Geography makes the coal problem dramatically worse. Ulaanbaatar sits in the valley of the Tuul River at about 1,300 m, hemmed in by four mountains. In winter, dense cold air drains down the slopes and pools on the valley floor while a layer of warmer air sits on top — a classic temperature inversion. The warm lid stops vertical mixing, so smoke from stoves, power plants and traffic cannot rise and disperse; it simply accumulates in the air people breathe. Inversions are strongest on clear, calm, extremely cold nights — exactly the nights when everyone burns the most fuel. That is why the same city can show moderate air on a windy afternoon and hazardous air twelve hours later: the emissions barely changed, but the lid closed.
PM2.5 is particulate matter smaller than 2.5 micrometres — small enough to pass from the lungs into the bloodstream. The World Health Organization’s 2021 guideline values are 5 µg/m³ as an annual mean and 15 µg/m³ as a 24-hour mean. Mongol Panel divides the live reading by the 24-hour guideline and shows the multiple, because “12× WHO” communicates more than “180 µg/m³”. A rough decoder:
| PM2.5 (µg/m³) | vs WHO 24-h | What it feels like in UB terms |
|---|---|---|
| 0–15 | ≤1× | Clean — a windy summer day on the steppe |
| 15–35 | 1–2× | Noticeable haze; sensitive people limit exertion |
| 35–75 | 2–5× | A typical UB winter afternoon; masks sensible outdoors |
| 75–150 | 5–10× | A bad winter day; keep children indoors when possible |
| 150–250 | 10–17× | A cold January evening in a ger district |
| 250+ | 17×+ | Peak inversion nights; hazardous for everyone |
A single “Ulaanbaatar AQI” number hides how uneven the city’s air really is, because pollution concentrates where the stoves are and where the cold air pools. The four reference points in this panel are chosen to span that gradient. They use approximate coordinates and model data, so treat them as indicative, and use the official agaar.mn monitors for decisions:
| Point | Character | Typical winter pattern |
|---|---|---|
| Bayankhoshuu | Ger district, north-west | Often the worst readings in the city, peaks after dusk |
| Zuragt | North of centre, mixed ger/apartment | High evenings, somewhat better middays |
| City centre | Apartment core, heavy traffic | Moderate-to-high; traffic adds NO₂ to the smoke drifting south |
| Zaisan | South of the Tuul, foot of Bogd Khan | Usually the cleanest of the four, but not exempt on still nights |
Dzud (зуд) is Mongolia’s signature climate disaster: a winter so severe that livestock die en masse. The mechanism is a one-two punch across two seasons. First a dry summer stunts pasture growth, so animals enter November without the fat reserves they need and herders cannot cut enough hay. Then a hard winter — deep snow, ice crusts, or weeks of −40 °C — locks away whatever grazing remains. Animals that would survive either half easily die when the two combine, not of cold directly but of slow starvation. Because roughly a quarter of Mongolians depend on herding, a bad dzud is an economic and social shock, driving migration to Ulaanbaatar’s ger districts and, indirectly, the city’s air problem.
Herders distinguish several kinds of dzud by the specific way winter blocks access to pasture. The 1999–2002 run of consecutive dzuds killed roughly 11 million animals; the 2009–10 dzud killed about 10 million — around a fifth of the national herd — and the 2023–24 winter was estimated at over 7 million head lost. The main types:
| Type | Mechanism |
|---|---|
| White dzud (цагаан зуд) | Snow too deep for animals to dig through to grass |
| Black dzud (хар зуд) | No snow at all in a cold winter — no drinking water on the steppe |
| Iron dzud (төмөр зуд) | Rain or thaw refreezes into an ice sheet that seals off pasture |
| Storm dzud (шуурган зуд) | Prolonged blizzards scatter herds and stop grazing |
| Hoof dzud (туурайн зуд) | Overcrowded pasture is trampled and exhausted before spring |
Classic livestock insurance barely works on the steppe: verifying individual animal deaths across thousands of kilometres is impossible, and moral hazard is real. Mongolia’s answer, piloted with the World Bank in 2006, is index-based livestock insurance (IBLI). The trigger is not your losses but the soum-level adult mortality rate for each species, computed from the National Statistics Office’s annual livestock census. If your soum’s mortality for, say, goats crosses the threshold (historically around 6%), every insured goat herder in that soum is paid automatically — whether their own animals lived or died. Catastrophic losses beyond the commercial layer’s cap have been backstopped by a public safety-net layer. The design rewards good husbandry: saving your animals in a dzud never reduces your payout. Sales run through commercial insurers; ask an agent or your soum office for current terms before each winter.
Air temperature understates the danger of steppe winter, because wind strips heat from skin far faster than still air. The table below uses the standard North American wind-chill formula — the same one this panel computes live — for common Mongolian winter conditions. Values are “feels like” °C; exposed skin can freeze in under 10 minutes below roughly −40 wind chill, and in a couple of minutes below −55:
| Air temp | 10 km/h wind | 20 km/h | 30 km/h | 40 km/h |
|---|---|---|---|---|
| −10 °C | −15 | −18 | −20 | −21 |
| −20 °C | −27 | −30 | −33 | −34 |
| −30 °C | −39 | −43 | −46 | −48 |
| −40 °C | −51 | −56 | −59 | −61 |
The tugrug has depreciated against the US dollar over most of the past two decades, and Ulaanbaatar’s economy has quietly adapted: landlords of newer apartments often quote rent in USD (payable in MNT at the day’s rate), some employers advertise salaries as dollar-equivalents, and big-ticket items like cars are mentally priced in dollars. This creates a real household exposure — if your income is fixed in MNT but your rent tracks USD, a 10% depreciation is effectively a 10% rent increase. The TUGRUG tab shows the daily mid-market rate for USD, CNY (the trade currency), RUB and KRW (remittances from workers in Korea are significant), with a converter. Bank counter rates will differ from mid-market by a spread; the mid-market number is your fair-value reference.
Dzud losses are decided long before the first blizzard, which is why the useful question is never “how cold will January be?” but “what did I do in August?”. The calendar below maps the decision points that recur every year in Mongolian herding — the moments when acting early is cheap and acting late is ruinous. Exact timing shifts with the aimag and the year, but the sequence does not:
| Season | What happens | The decision that matters |
|---|---|---|
| Spring (Mar–May) | Birthing season; animals at their weakest after winter | Shelter and feed for newborns; spring storms kill more weak stock than midwinter does |
| Summer (Jun–Aug) | Pasture growth — or, in a dry year, the seed of the next dzud | Judge the grass honestly; a poor summer means destocking and hay purchases start now |
| Autumn (Sep–Nov) | Animals fatten; markets are liquid; IBLI sales window | Sell surplus while prices hold, stack hay above a normal year’s need, insure before winter |
| Winter (Dec–Feb) | Survival phase; feed is whatever was stored or reachable | Ration fodder, agree otor movement routes with neighbouring soums before snow blocks them |
The pattern generalises: every catastrophic dzud on record was preceded by a summer whose warnings were visible by September. The NAMEM risk maps, the autumn body-condition of the herd and the hay stack are the three instruments on the dashboard — the winter itself only reads out what they already said.
On hazardous days the goal is to cut the dose, not to be perfect. Outdoors, a well-fitted KF94/N95-class respirator filters most PM2.5; surgical masks and scarves do very little for fine particles. Indoors matters more than people assume: UB apartments and gers leak smoky air, so a HEPA purifier sized to the room (or a DIY box-fan filter) in the sleeping room is the highest-value purchase, run on recirculate with doors closed during evening peaks. Time your ventilation for midday, when the inversion is weakest. In ger households, stove discipline — dry refined briquettes, correct lighting technique, a sealed flue — cuts both indoor smoke and carbon-monoxide risk, which is a separate and deadly winter hazard.
No. The PM2.5/PM10 numbers come from Open-Meteo’s free air-quality API, which serves atmospheric model data (CAMS) for approximate district coordinates. It is good for spotting trends and bad episodes, but the official reference is the national monitoring network at agaar.mn.
December through February, and within a day the worst hours are evening and night, when stoves are stoked and the inversion lid is firmly closed. Calm, clear, extremely cold nights are the worst combination; windy days flush the valley and can be surprisingly clean.
Everything. A dry summer means poor pasture, so animals enter winter thin and herders enter it with too little hay. The winter itself only finishes what the summer started — that is the dzud mechanism, and why risk can be assessed months before the first snowstorm.
By an index, not by inspecting your animals. When the annual census shows a soum’s adult mortality rate for a species crossed the threshold, every insured herder of that species in the soum is paid the same rate automatically. Your individual losses — larger or smaller — do not change the payout.
Cold-weather closures are a real and regularly used practice in Mongolia, and −40 is the number everyone quotes, but the actual thresholds vary by year, by grade level and by aimag directive. Treat any threshold in this panel as folklore-with-a-basis and confirm with your school or local education authority.
No. The livestock log is stored only in your browser’s localStorage and never leaves your device. Clearing browser data deletes it; there are no accounts and no server.
Both, for different jobs. This page shows the mid-market rate — the midpoint between global buy and sell prices — which is the fair-value reference for judging any quote. The bank or exchange-office counter rate is what you will actually get, and it always sits a spread away from mid-market. If the counter rate is more than a couple of percent off the mid-market number shown here, shop around.
Barely. Fine particles slip around the loose edges of a surgical mask and straight through the weave of a scarf. Protection outdoors requires a respirator built to a filtration standard — KF94, N95 or FFP2 — worn with a tight seal to the face. On the coldest days a scarf over the respirator keeps it from icing up.