Indices
1. Overview of INDICES Dataset
Purpose: Provides statistically downscaled climate indices to assess extreme climate events and their impacts.
Developed by: Environment and Climate Change Canada (ECCC).
Geographic Coverage: Canada (land mass only).
Time Coverage:
Historical period: 1951–2005
Future projections: 2006–2100
Resolution: 10 km x 10 km spatial grid
Emission Scenarios (RCPs):
RCP2.6 (low emissions, strong mitigation)
RCP4.5 (moderate emissions stabilization)
RCP8.5 (high emissions, worst-case scenario)
Key Climate Indices Modeled
The dataset provides a wide range of climate indices, categorized into:
Temperature-Based Indices
Precipitation-Based Indices
Agroclimatic Indices
Temperature-based indices track extreme heat events, frost days, and growing season changes.
Precipitation-based indices measure heavy rainfall events, droughts, and wet/dry spell lengths.
Agroclimatic indices assess crop viability, growing degree days, and frost-free periods.
2. Data and Processing
2.1 Multi-Model Downscaled Data
CMIP5 climate model outputs were statistically downscaled to 10 km resolution.
Daily minimum and maximum temperature (Tmin, Tmax) and precipitation were downscaled.
Bias Correction/Constructed Analogues with Quantile Mapping (BCCAQv2) was used for downscaling.
Bias Correction/Constructed Analogues with Quantile Mapping (BCCAQv2)
Combines two statistical downscaling methods:
Bias Correction/Constructed Analogues (BCCA) (Maurer et al., 2010)
Quantile Mapping (QM) (Cannon et al., 2015)
BCCA corrects large-scale climate biases and QM adjusts daily variability.
Reference period for downscaling: 1951–2005.
2.2 Climate Models Used in INDICES
The dataset is based on 24 CMIP5 climate models, including:
CMIP5 Model Name
Institution
CanESM2
Environment Canada
CCSM4
NCAR (USA)
HadGEM2-ES
UK Met Office
MPI-ESM-LR
Max Planck Institute (Germany)
MIROC5
JAMSTEC (Japan)
IPSL-CM5A-LR
IPSL (France)
NorESM1-M
Norwegian Climate Centre
GFDL-ESM2M
NOAA-GFDL (USA)
(Full model list available in Table 2 of INDICES documentation.)
3. Definitions of Climate Indices
INDICES includes several types of climate indices, grouped into temperature, precipitation, and agroclimatic categories.
3.1 Temperature-Based Indices
Index Name
Definition
Hot Days (TX90p)
Days when maximum temperature exceeds the 90th percentile.
Cold Nights (TN10p)
Days when minimum temperature is below the 10th percentile.
Growing Season Length (GSL)
Number of days between the last spring and first fall frost.
Frost Days (FD0)
Number of days with Tmin < 0°C.
Heatwave Duration Index (HWD)
Maximum duration of consecutive days with Tmax > 30°C.
3.2 Precipitation-Based Indices
Index Name
Definition
Very Wet Days (R95p)
Annual precipitation from days exceeding the 95th percentile.
Maximum 1-Day Precipitation (Rx1day)
Wettest day of the year (mm).
Maximum 5-Day Precipitation (Rx5day)
Maximum precipitation over any 5-day period.
Dry Spell Length (CDD)
Maximum number of consecutive dry days (Precip < 1 mm).
Heavy Precipitation Days (R20mm)
Number of days with rainfall > 20 mm.
3.3 Agroclimatic Indices
Index Name
Definition
Growing Degree Days (GDD)
Sum of daily temperatures above a base threshold (e.g., 5°C).
Frost-Free Period (FFP)
Number of days between the last and first frost.
Accumulated Freezing Degree Days (AFDD)
Sum of negative daily temperatures.
Indices based on the Expert Team on Climate Change Detection and Indices (ETCCDI).
Agroclimatic indices developed in collaboration with the Canadian adaptation community.
4. Model Weighting & Ensemble Percentiles
4.1 Equal Model Weighting
Each climate model is given equal weight ("one model, one vote").
Different physical parameterizations are treated as distinct models.
4.2 Ensemble Percentiles
To account for uncertainty, percentiles are provided:
5th percentile (low change)
25th percentile
50th percentile (median response)
75th percentile
95th percentile (high change)
Range captures model spread but not full uncertainty.
5. Uncertainty in Climate Projections
Temperature projections are highly consistent across models.
Precipitation projections have higher uncertainty.
By 2.1°C global warming, models disagree on precipitation changes, but at 4.5°C warming, they align.
Projected precipitation extremes scale with temperature increases.
6. Best Practices for Using INDICES Data
Use multi-model ensembles rather than individual models.
Analyze multiple emission scenarios (RCP2.6, RCP4.5, RCP8.5).
Compare median projections with percentile ranges.
Downscaling does not eliminate all uncertainties—regional bias correction may be required.
Limitations
Does not include wind, humidity, or soil moisture indices.
Extreme precipitation projections remain uncertain due to model spread.
Best suited for regional-scale climate impact studies rather than site-specific assessments.
7. Data Access & References
7.1 Where to Access INDICES Data
ECCC Climate Indices Portal:
Pacific Climate Impacts Consortium (PCIC):
7.2 References
Cannon, A.J. (2015). Quantile Mapping Bias Correction of Precipitation Extremes. Journal of Climate.
Li, G., et al. (2018). Indices of Canada’s Future Climate for Adaptation Applications. Climatic Change.
Sillmann, J., et al. (2013a, b). Climate Extremes Indices in CMIP5. Journal of Geophysical Research: Atmospheres.
8. Conclusion
The Statistically Downscaled Climate Indices (INDICES) dataset provides high-resolution (10 km) projections of temperature, precipitation, and agroclimatic indices, supporting regional climate adaptation strategies.
By using CMIP5 climate models, BCCAQv2 statistical downscaling, and multi-model ensembles, INDICES offers a powerful tool for analyzing future climate extremes, aiding policy planning, and improving climate resilience across Canada.
Last updated
Was this helpful?