Plate II · Atmospheric remote sensing · June 2023

Smoke from Canada over New York City

A plume that travelled a thousand kilometres in two days, followed from orbit and measured at street level.

Monitoring Cities · NYU · Prof. Masoud Ghandehari | Google Earth Engine · Sentinel-5P · EPA AirData

Aerosol index over NYC: 0.11 → 6.67
Carbon monoxide: 4× normal
Peak daily PM2.5: 201.7 µg/m³ (Queens)
City's own NO₂: 13.2 vs 13.3 ppb

On the afternoon of 7 June 2023, New York's air quality index reached 484, the worst reading officials had recorded since the 1960s. The sky turned orange. The cause was about 1,000 km to the north: wildfires burning in Quebec.

I wanted to trace that smoke from the fires to the city, measure how much it changed the air above New York, and test whether the pollution really came from outside rather than from the city's own traffic. The satellite is Sentinel-5P, whose TROPOMI instrument passes over once a day at about 1:30 pm; the street-level check comes from EPA monitors in the Bronx and Queens.

Aerosol index on 5 June 2023: thick smoke over Ontario and western Quebec — **Fig. 1** · The smoke's journey, one satellite pass per day. Darker is thicker smoke; New York City is drawn in black, and grey areas have no valid data. On 6 June the plume reached the city after the 1:30 pm pass. −1 2+

Aerosol index on 6 June 2023: the plume moving south over upstate New York and Pennsylvania — **Fig. 1** · The smoke's journey, one satellite pass per day. Darker is thicker smoke; New York City is drawn in black, and grey areas have no valid data. On 6 June the plume reached the city after the 1:30 pm pass. −1 2+

Before and during

To measure the change, I compared a normal week before the smoke (T1, 30 May to 4 June) with 7 June (T2), the only day the satellite's pass found thick smoke over the city itself. Every city-wide value is the median over the five boroughs.

Over New York, the aerosol index went from 0.11 to 6.67, about sixty times higher. The rise sits in a single band running from Pennsylvania across New Jersey to the city, the same path the smoke followed in the daily maps, while most of the region outside it barely changed.

Aerosol index on 7 June 2023, the worst smoke day — **Fig. 2** · Drag the arrow right towards 7 June, left towards the normal week. Sentinel-5P aerosol index, on one scale for both. −1 2+

Average aerosol index for the normal week, 30 May to 4 June 2023 — **Fig. 2** · Drag the arrow right towards 7 June, left towards the normal week. Sentinel-5P aerosol index, on one scale for both. −1 2+

Carbon monoxide, the fingerprint of fire

Haze alone could be dust or cloud. Burning forests also release carbon monoxide, so if the haze was smoke, CO should rise on the same day along the same path. It did: over the city it went from 41,700 to 169,700 (mol/m² × 10⁶), about four times higher, following the same band out over the Atlantic. In the normal week the highest CO was already over Quebec, where the fires were releasing it before the wind turned south.

Change in aerosol index, 7 June minus the normal week — **Fig. 3** · Change, 7 June minus the normal week. Both rise along one band from Pennsylvania to New York City. Grey areas have no valid data. fellrose

Change in carbon monoxide, 7 June minus the normal week — **Fig. 3** · Change, 7 June minus the normal week. Both rise along one band from Pennsylvania to New York City. Grey areas have no valid data. fellrose

The control, and what people breathed

Nitrogen dioxide comes mainly from the city's own traffic and breaks down within about a day. If the pollution came from outside, NO₂ over the city should have stayed normal. The smoke was so thick on 7 June that the satellite's quality check removed all of its NO₂ data over the city, so this test fell to four EPA ground stations: 13.2 ppb on 7 June against 13.3 ppb in the normal week, well within the usual 5 to 26.

Fine particles told the opposite story. Both PM2.5 stations averaged about 11 µg/m³ in the normal week, then stayed above the EPA's daily limit of 35 for three days running. On 7 June the daily average reached 176.5 in the Bronx and 201.7 in Queens, five to six times the limit, with a worst single hour of 410.

Fig. 4 · Four measurements over New York City, one value per day, 15 May to 30 June. Hover to read any day; the orange band is the smoke, 6–8 June. Smoke, CO and fine particles spike together while traffic NO₂ (grey band: range across four stations) doesn't move. A second, smaller wave returns on 29–30 June. QueensBronx

What it adds up to

The daily maps show the plume moving about 1,000 km south and covering the city on 7 June. On that day the satellite saw smoke and carbon monoxide rise sharply over New York, while the city's own traffic pollution, measured on the ground, stayed at its usual level. The particles people breathed reached five to six times the EPA limit. A study in JAMA found asthma emergency-room visits in the city rose by 44% during the smoke.