Toronto climate data for building simulations with urban heat island effects and nature-based solutions

As cities face rising temperatures, increased frequency of extreme weather events, and altered precipitation patterns, buildings are subjected to increasing energy demand, heat stress, thermal comfort issues, and decreased service life. Therefore, evaluating building performance under changing climate conditions is essential for building sustainable and resilient communities. Unique climate characteristics of cities, such as the urban heat island effect, are not well simulated by global or regional climate models, and is therefore often not included in typical building analyses. Consequently, a computationally efficient approach is used to generate “urbanized” climate data, derived from regional climate models, to prepare building simulation climate data that incorporate urban effects. We demonstrate this process using existing climate data for Toronto airport’s weather station and extend it to prepare projections for scenarios where nature-based solutions, such as increased greenery and albedo, were implemented. We find significant improvements in the representation of the urban heat island and subsequent cooling effects of nature-based solutions in the urbanized climate data. This dataset allows building practitioners to evaluate building performance under historical and potential future changes in climate, considering the complex interactions within the urban canopy and the implementation of mitigation efforts such as nature-based solutions.

This dataset contains hourly historical and future weather files for use in building simulations for the city of Toronto, Canada. While similar weather files are usually based on measurements taken at a city's nearby airport, the current dataset utilizes a novel statistical-dynamical downscaling technique which involves the use of the dynamical Weather Research and Forecasting (WRF) model combined with a statistical approach and climate projections from an ensemble of 15 Canadian Regional Climate Model 4 (CanRCM4) to generate urban climate data which includes the effects of the urban heat island and different nature-based solutions (NBS) as mitigation strategies (such as increasing surface albedo and greenery). Additionally, different levels of implementation of these mitigation strategies were produced, for example, when the albedo is increased to 0.40 (ALBD40) and 0.80 (ALBD80), and similarly for the green and combined scenarios, GRN40, GRN80, COMB40, and COMB80. The URBAN scenario is considered the control case where the urban heat island effects are accounted for in the data, but the NBS scenarios are not yet implemtned. 

The data are stored in large CSV files, where the rows consists of all 15 realizations of the CanRCM4 ensemble and the variables make up the columns. For example, each 31-year period is repeated 15 times, once for each of the RCM realizations. Therefore, there are 4,073,400 (15x31x8760) rows in each file. We recommend viewing the data using packages from Python or R.  

The historical and future global warming thresholds and their corresponding time periods are as follows:

Global Warming Scenario

Time Period

Historical

1991-2021

Global Warming 0.5ºC

2003-2033

Global Warming 1.0ºC

2014-2044

Global Warming 1.5ºC

2024-2054

Global Warming 2.0ºC

2034-2064

Global Warming 2.5ºC

2042-2072

Global Warming 3.0ºC

2051-2081

Global Warming 3.5ºC

2064-2094

The following variables are included in the files:

Variable
Description

RUN
Run number (R1-R15) of Canadian Regional Climate Model, CanRCM4 large ensemble associated with the selected reference year data

YEAR
Year associated with the record

MONTH
Month associated with the record

DAY
Day of the month associated with the record

HOUR
Hour associated with the record

YDAY
Day of the year associated with the record

DRI_kJPerM2
Direct horizontal irradiance in kJ/m2 (total from previous HOUR to the HOUR indicated)

DHI_kJperM2
Diffused horizontal irradiance in kJ/m2 (total from previous HOUR to the HOUR indicated)

DNI_kJperM2
Direct normal irradiance in kJ/m2 (total from previous HOUR to the HOUR indicated)

GHI_kJperM2
Global horizontal irradiance in kJ/m2 (total from previous HOUR to the HOUR indicated)

TCC_Percent
Instantaneous total cloud cover at the HOUR in % (range: 0-100)

RAIN_Mm
Total rainfall in mm (total from previous HOUR to the HOUR indicated)

WDIR_ClockwiseDegFromNorth
Instantaneous wind direction at the HOUR in degrees (measured clockwise from the North)

WSP_MPerSec
Instantaneous wind speed at the HOUR in meters/sec

RHUM_Percent
Instantaneous relative humidity at the HOUR in %

TEMP_K
Instantaneous temperature at the HOUR in Kelvin

ATMPR_Pa
Instantaneous atmospheric pressure at the HOUR in Pascal

SnowC_Yes1No0 
Instantaneous snow-cover at the HOUR (1 - snow; 0 - no snow)

SNWD_Cm
Instantaneous snow depth at the HOUR in cm