Climate Change & Toronto

[Memph]

Tornado and hurricane season have been pretty quiet this year. Looks like the tornado season in the United States has been in the bottom 10% in terms of low activity.

https://www.spc.noaa.gov/wcm/adj.html

Normally we'd be at the peak of hurricane season right now, but still not much activity. Last time we had so little activity so late into the season seems to have been 28 years ago in 1994.

1994 Atlantic hurricane season - Wikipedia

en.wikipedia.org

 

[Memph]

Not usually in pipes. Usually water is diverted long distances using canals or channels.

I suspect it would make more sense to do a better job retaining run-off in the west in reservoirs than to try to pump water uphill. Lake Superior is 183m above sea level. Calgary is at 1045. Saskatoon is 482. Lifting water almost a kilometer is a lot of energy. Lifting a cubic m/1000 KG of water by 860m to Calgary is 8.4 MJ of potential energy, or 2.3 kwh.

Yeah and the canals are of course not moving water uphill. Canadian agriculture consumes around 3000 million cubic meters of irrigated water (ie not including rain) per year compared to 150 million cubic meters of oil consumed and 270 million cubic meters of oil produced. Most of the oil is piped only about 300m uphill from Fort McMurray to Edmonton, and then it's all downhill from there heading across the Prairies and into the US Midwest. Even the oil heading into the Trans Mountain pipeline is only reaching a peak elevation of about 1150m, which is 100m higher than downtown Calgary, but 100m lower than Calgary's NE suburbs.

Municipal water in Alberta costs about $2.5/m^3. The cost to access water for agriculture is probably much less. Crude oil costs around $700/m^3, and about double that at the pump.

The size of pipelines required and water consumed is immense compared to oil. And it's much more difficult to warrant the expense considering how cheap the water is (compared to oil). The Prairies are better off just making do with the water they have available locally from their rivers and aquifers. Use drought resistant crops/landscaping, good agricultural practices, etc. I don't think the impact of climate change on the Prairies has been that drastic. Yes, there have been droughts in recent years, but the Prairies are a semi-arid climate so droughts have always been a thing.

 

[Memph]

More likely Calgary or Saskatoon would siphon water from the Rockies. Like how the Romans did...

I mean they're kinda doing that already. It's called the North Saskatchewan and Bow/Elbow River. AFAIK aqueducts were mainly used because Rome was built on hills so it was easier to bring water down from surrounding mountains than up from the Tiber River? Probably the Tiber was also polluted by human waste and they didn't have the means to clean it whereas the water from the mountains was clean.

 

[Memph]

Toronto (Downtown) precipitation by year.

 

[Memph]

Months with the highest precipitation totals in Toronto (Downtown) history

1. 1843/09: 248.2mm
2. 1894/05: 238.5mm
3. 1986/09: 217.8mm
4. 1841/07: 206.8mm
5. 1915/08: 206.8mm

6. 1870/06: 205.5mm
7. 1996/09: 202.4mm
8. 1878/09: 195.6mm
9. 1868/05: 195.1mm
10. 1986/08: 191.7mm

11. 2013/07: 191.0.mm
12. 1985/11: 186.2mm
13. 2015/06: 183.9mm
14. 1878/08: 180.1mm
15. 1870/03: 177.8mm

16. 1932/01: 173.7mm
17. 1870/09: 173.0mm
18. 1942/05: 172.5mm
19. 1847/09: 169.7mm
20. 2018/08: 169.5mm

21. 1968/08: 167.1mm
22. 1981/08: 166.2mm
23. 2000/06: 162.2mm
24. 1858/05: 162.1mm
25. 1842/09: 159.0mm

July 2013 was quite wet for Downtown Toronto, but there were still 10 months that were wetter.

Ranked by date.

1841/07: 206.8mm
1842/09: 159.0mm
1843/09: 248.2mm
1847/09: 169.7mm
1858/05: 162.1mm

1868/05: 195.1mm
1870/03: 177.8mm
1870/06: 205.5mm
1870/09: 173.0mm
1878/08: 180.1mm
1878/09: 195.6mm

1894/05: 238.5mm

1915/08: 206.8mm

1932/01: 173.7mm

1942/05: 172.5mm

1968/08: 167.1mm

1981/08: 166.2mm
1985/11: 186.2mm
1986/08: 191.7mm
1986/09: 217.8mm
1996/09: 202.4mm

2000/06: 162.2mm
2013/07: 191.0.mm
2015/06: 183.9mm
2018/08: 169.5mm

The 2000s were pretty active for heavy precipitation, but the early period of the history of the Downtown's weather station operation (1840-1878) had a lot of intense precipitation too, as did the 1980s.

 

[Memph]

Mean annual temperature at Queens Park weather station from 1840 to present with 11 year moving average.

1840-1890 were very cool, at least compared to the present. When compared to a smaller city like Guelph, which is actually a touch south of Toronto, Toronto in 1840-1890 was about as warm as Guelph in 1980-2010.

Then very rapid warming occurred at the end of the 19th century of about 1C in a single decade. My best guess is due to urbanization, as Toronto expanded from being mostly south of Dundas, to spreading north of Bloor, leading to an increase in the urban heat island. Global temperatures rose by about 0.1C during this period, so a 10x greater increase is out of sync with global patterns.

From 1900 to 1950, temperatures rose by another 1.5C or so, again much more than the 0.3C that took place globally during the same time period.

From 1950 to 1980, there was a slight dip in temperatures, which is consistent with the pattern beyond Toronto.

Then another 1C of warming from about 1980 to 2000, which is more than the 0.4C of warming that occurred globally.

And finally a levelling off from 2000 to 2021, whereas the rest of the world warmed by another 0.3C.

 

[Northern Light]

Mean annual temperature at Queens Park weather station from 1840 to present with 11 year moving average.

1840-1890 were very cool, at least compared to the present. When compared to a smaller city like Guelph, which is actually a touch south of Toronto, Toronto in 1840-1890 was about as warm as Guelph in 1980-2010.

Then very rapid warming occurred at the end of the 19th century of about 1C in a single decade. My best guess is due to urbanization, as Toronto expanded from being mostly south of Dundas, to spreading north of Bloor, leading to an increase in the urban heat island. Global temperatures rose by about 0.1C during this period, so a 10x greater increase is out of sync with global patterns.

From 1900 to 1950, temperatures rose by another 1.5C or so, again much more than the 0.3C that took place globally during the same time period.

From 1950 to 1980, there was a slight dip in temperatures, which is consistent with the pattern beyond Toronto.

Then another 1C of warming from about 1980 to 2000, which is more than the 0.4C of warming that occurred globally.

And finally a levelling off from 2000 to 2021, whereas the rest of the world warmed by another 0.3C.

It would be interesting to break this into overnight lows vs daytime highs, or the diurnal range.

Urban heat island effect notoriously lifts night lows more than daytime highs.

This has a notable adverse effect on ability to sleep in the absence of air conditioning.

****

I would also be interested to see if there's been variation in humidity levels in Toronto over time; as high humidity increases one's perception of heat.

 

[Memph]

It would be interesting to break this into overnight lows vs daytime highs, or the diurnal range.

Urban heat island effect notoriously lifts night lows more than daytime highs.

This has a notable adverse effect on ability to sleep in the absence of air conditioning.

****

I would also be interested to see if there's been variation in humidity levels in Toronto over time; as high humidity increases one's perception of heat.

The diural range was actually increasing a bit during the early part of urbanization (pre-1915).

But then during 1920 to 1980 the daytime highs haven't really increased, while the lows increased a fair bit, resulting in a significant decrease in the diurnal range from about 9.5C to 7C.

Surprisingly, during the past two decades, the diurnal range increased a bit, with low temps decreasing slightly and high temps increasing slightly. We'll see if that's a long term trend or not.

 

[Memph]

It would be interesting to break this into overnight lows vs daytime highs, or the diurnal range.

Urban heat island effect notoriously lifts night lows more than daytime highs.

This has a notable adverse effect on ability to sleep in the absence of air conditioning.

****

I would also be interested to see if there's been variation in humidity levels in Toronto over time; as high humidity increases one's perception of heat.

As for the humidity, this data is for Pearson, where the adjacent urbanization/heat island formation occurred more recently (1960-2010 roughly) than in Downtown Toronto.

https://www.reddit.com/r/TorontoWeatherRecords/comments/xe56nw

It seems like the absolute humidity has been essentially constant (maybe slightly increasing) but relatively humidity, which is the measurement that's used most often, has decreased, I guess because the temperatures (and therefore moisture holding capacity) have increased.

 

[Northern Light]

As for the humidity, this data is for Pearson, where the adjacent urbanization/heat island formation occurred more recently (1960-2010 roughly) than in Downtown Toronto.

https://www.reddit.com/r/TorontoWeatherRecords/comments/xe56nw

It seems like the absolute humidity has been essentially constant (maybe slightly increasing) but relatively humidity, which is the measurement that's used most often, has decreased, I guess because the temperatures (and therefore moisture holding capacity) have increased.

Brilliant contributions on this, Thank You.

 

[Memph]

Here's some stats on weather extremes for Downtown Toronto/Queen's Park.

So I looked at the daily records for minimum temperature, maximum temperature, and maximum precipitation. Then counted the number of records that were set for any given year.

For example, the record for the most precipitation on Sept 22 was set on 2021, with a total of 49.9mm. So that counts as one precipitation record set in 2021. 2021 also set four high temperature records - March 11 (18.8C), May 25 (30.8C), December 11 (16.0C), and December 16 (15.2C). There were no low temperature records set in 2021.

Given there are 365 days per year, and 183 years of climate records, you would expect 2 precipitation, 2 high temperature and 2 low temperature records (4 overall temperature records) set per year. Then I calculated a 13 year central moving average (ex the data point for 2000 would look at the average of 1994-2006).

There have been a fair bit of precipitation records set in this new century. Top years include 2013 (7), 2022 (4), 2011 (4), 2010 (4), 2006 (4) with 12 of the remaining years setting 2-3 records and only 6 years setting 0-1 records. So there has been an increase in extreme precipitation relative to the 1900s.

However, there was a time looking further back when even more records were being set, from about 1858 to 1878. Top years then include 1869 (6), 1870 (6), 1871 (5), 1866 (5), 1878 (5), 1864 (4), 1862 (4), with 11 of the remaining years setting 2-3 records and only 3 years setting 0-1 records.

The "Temperature" line on the graph represents the combined sum of daily high and daily low temperature records. As Toronto's climate has warmed, there have been fewer cold records set, and more warm records set, but the total number of records set has decreased, so at least by this metric, the climate has not gotten more extreme on this 180+ year timescale. Although Feb 2015 had us talking about the Polar Vortex an awful lot, Jan 1857, Feb 1875 and Feb 1885 were even colder.

Days with lows of -20C or less
Jan 1857: 12
Feb 1875: 11
Feb 1885: 11
Feb 2015: 6

Coldest temp
Jan 1857: -28.9C
Feb 1875: -26.7C
Feb 1885: -26.1C
Feb 2015: -25.1C

Although the past two decades have seen an increase in extreme temperatures compared to the 1960s-1980s, the greatest number of temperature records were set in the 1800s.

 

[Northern Light]

Although the past two decades have seen an increase in extreme temperatures compared to the 1960s-1980s, the greatest number of temperature records were set in the 1800s.

This is precisely what one would expect.

Records were relatively new in the 1800s, when one has only 10 historic measures for 'x', there is a high probability in any given year of a result outside that 10-year dataset.

Once you have more than 10x the amount of data, you will have recorded both greater natural swings and one-off anomalies ( a 25-year storm event, a 100-year storm event, a drought, a flood etc.)

This in turn makes an additional anomaly or record less likely.

****

Urban heat island effect, as discrete from climate change writ-larger (Though the phenomenon interact), would be expected raise overall temperatures over time, but particularly those associated with night-time lows.

As night-time lows are more likely to be the record-setting colds of winter, fewer of these are to be expected as the City grows/sprawls.

On the other hand, growth in daytime high records is to be expected

Current research suggests Urban Heat Islands typically result in increased levels of summer precipitation as well:

NASA Satellite Confirms Urban Heat Islands Increase Rainfall Around Cities

NASA researchers have for the first time used a rainfall-measuring satellite to confirm that "urban heat-islands" create more summer rain over and downwind of major cities, including Atlanta, Dallas, San Antonio and Nashville.

www.sciencedaily.com

*****

To sort out the impact on Toronto of Climate Change, vs Sprawl/Urban Heat Island, one needs some control samples.

The ideal would be places nearby, which historically have similar weather/climate, but which are much smaller than Toronto and produce considerably less of the latter effect.

South of the Barrie snowbelt, south of the shield, north side of Lake Erie/Ontario.

****

Not trying to give you extra work, LOL, Just something to consider.

Good efforts on the data collation btw.

 

[Memph]

This is precisely what one would expect.

Records were relatively new in the 1800s, when one has only 10 historic measures for 'x', there is a high probability in any given year of a result outside that 10-year dataset.

Once you have more than 10x the amount of data, you will have recorded both greater natural swings and one-off anomalies ( a 25-year storm event, a 100-year storm event, a drought, a flood etc.)

This in turn makes an additional anomaly or record less likely.

Sorry, "record set" may not have been the best way to put it. I meant set and held. There are more temperature records that are still standing today that were set pre-1900 compared to today (in per year terms).

As for heat island affecting precipitation, my annual precipitation graph does not seem to show any such increase. If anything, there was a bit more precipitation in the 1850s-1870s. There does seem to have been an increase in extreme precipitation in 2000-present compared to 1880-2000, although I'd be hesitant to attribute that to UHI since the area around Downtown Toronto probably began to see an urban heat island effect more than 100 years ago, and I'm not sure if the more recent urbanization in far flung suburbs like Milton, Brampton, Vaughan and Whitby would have much of an effect on Downtown Toronto.

And yes, I am considering looking at a smaller city/rural location as a control. It might be hard to find a good one though since large old cities are the ones that are the most likely to have already had a weather station 100+ years ago, and there's even issues with finding recent data since Environment Canada has closed a lot of weather stations in the past few decades, as well as changing the location of weather stations from one end of town to another.

 

[Memph]

Here's the breakdown of the years that the record highs and record lows date to.

Given global warming + UHI warming + UHI caused diurnal range reduction, it's not surprised that there are much fewer record lows now than it the past, although the extent to which the record lows are skewed towards the 1800s and away from post-1950 is still pretty impressive.

But it's also interesting how the period of 1910-1960 holds many of the record highs, with a comparable number of record highs to 2000-present, and about 50% more than for 1960-2000.

 

[Memph]

These are the heat records for Toronto broken down by season (more or less). The 2020s is based on the nearly 3 years of the decade so far, and assumes the the per year rate of records broken for 2020-2022 continues into 2029. There have been 9 daily high records broken in 2020-present, evenly split between each of the three temperature groupings (hot, cold and transitional).

So 1910-1960 had a lot of summer time heat records associated with extreme heat, these are typically records in the 30-40C range, then dropped dramatically in the 60s-80s, but have begun to rise again, though they are still considerably fewer of those dating to recent decades compared to 1910-1960.

For the transitional seasons, there is a peak in records held in the 1920s-1940s, with a less consistent trend of decreasing in the 50s-80s, and then increasing again. Recent decades have seen almost as many records in these transitional seasons as in the 1920-1950 peak, but not quite. These records are typically in the 18-30C range.

For cold season records, typically in the 10-18C range, they've been increasing more or less steadily since records began.