The City of Toronto commissioned Toronto's Future Weather & Climate Driver Study - the 2012 results indicate projected changes in extreme rainfall for a few durations and return periods. Results of the Outcomes Report are here https://www.toronto.ca/wp-content/uploads/2018/04/982c-Torontos-Future-Weather-and-Climate-Drivers-Study-2012.pdf. The baseline period for the study is 2000-2009 and statistics are predicted out to 2040-2049.
The Engineering Climate Datasets have been updated in early 2019, including for two Toronto-area climate stations with long records called "Toronto City" and "Toronto International Airport". The following tables compares the predicted increase in extreme rainfall in the 2012 study with trends in the same statistics from 1990 to 2017 at these two Toronto-area stations.
A key take away is that the Future Weather & Climate Driver Study does not agree with the direction and magnitude of changes in the actual statistics, which are based on real observations (not modelling predictions). Some actual statistics have been decreasing since 1990, not increasing as predicted int eh study. When a statistic is increasing, it is at a significantly lower rate that what is predicted in the study.
The following chart compares the past 100 year daily data to the study predictions - the Toronto study seems to have a hockey stick shape, jumping significantly upward by the 2040's which does not match the past trends.
The next chart shows changes in 10 year hourly rainfall. The Toronto study significantly understates the value today, suggesting it will double by the 2040's - the predicted future value has already been in place since the 1990's however.
It is questionable whether the City of Toronto should consider any changes to design criteria for municipal infrastructure considering these future predictions - best to follow ASCE's approach and incorporate flexibility in future design and wait and see with the 'observational method'? - if observations show that there is no change in the statistics, there should be no significant driver in changing design criteria, especially based on models that do not match the magnitude or trend in actual extreme rainfall statistics.
***
Bonus: here are a few more predicted 24 Hour 100 Year rainfall values, also going against the observed trends:
