SPEI

1. Overview of SPEI Dataset

  • Purpose: Quantifies surface water availability, identifying wet and dry periods over time.

  • Developed by: Environment and Climate Change Canada (ECCC).

  • Geographic Coverage: Canada (land mass only).

  • Time Coverage:

    • Historical period: 1900–2005

    • Future projections: 2006–2100

  • Spatial Resolution: 1° x 1° global grid (~110 km)

  • Temporal Resolution: Monthly

Key Features

Feature

Description

Drought Index

SPEI measures precipitation minus evapotranspiration (P–PET)

Climate Models Used

29 CMIP5 models

Downscaling Approach

Bias Correction & Multivariate Statistical Adjustment (MBCp)

Emission Scenarios

RCP2.6, RCP4.5, RCP8.5

Time Scales

SPEI-1, SPEI-3, SPEI-12 (1, 3, and 12-month averages)

Reference Period for Anomalies

1950–2005


2. Data and Processing

2.1 Multi-Model Climate Projections

  • 29 CMIP5 climate models were used to generate SPEI projections.

  • Equal weighting applied to all models ("one model, one vote").

  • Ensemble median (50th percentile) used as the central estimate, with uncertainty ranges (25th–75th percentiles).

2.2 Bias Correction & Observational Target

  • Bias correction applied to all climate variables (temperature, precipitation).

  • Reference dataset: Canadian Gridded Climate Dataset (CANGRD).

  • Time Period for Bias Correction: 1950–2005.

2.3 Potential Evapotranspiration (PET) Calculation

  • Method Used: Modified Hargreaves Equation.

  • Why Modified? It includes precipitation as an input, improving accuracy in cold and dry regions.

  • Alternative PET Methods:

    • Penman-Monteith (more complex but requires more data).

    • Thornthwaite (less accurate, requires only temperature data).

2.4 Log-Logistic Fit & Parameter Estimation

  • Log-logistic probability distribution used for standardization.

  • Parameters estimated using Probability Weighted Moments (PWM).

  • Applied to SPEI data over the full period (1900–2100).


3. SPEI Time Scales & Applications

3.1 SPEI-1 (Short-Term Drought)

  • Captures rapid-onset drought conditions (flash droughts).

  • Useful for detecting month-to-month rainfall deficits.

  • Common applications: Agriculture, water supply monitoring.

3.2 SPEI-3 (Seasonal Drought)

  • Represents cumulative moisture deficit over three months.

  • Captures seasonal drought conditions impacting crops, forests, and reservoirs.

  • Common applications: Wildfire risk assessment, energy demand forecasting.

3.3 SPEI-12 (Long-Term Drought)

  • Measures persistent droughts lasting over a year.

  • Best suited for analyzing climate change-driven trends.

  • Common applications: Hydrological planning, long-term water resource management.


4. Interpretation of SPEI Values

SPEI Value

Drought/Wetness Category

≥ 2.0

Extremely Wet

1.5 to 1.99

Very Wet

1.0 to 1.49

Moderately Wet

-0.99 to 0.99

Near Normal

-1.0 to -1.49

Moderately Dry

-1.5 to -1.99

Very Dry

≤ -2.0

Extremely Dry (Severe Drought)

  • SPEI is a standardized index, meaning a value of zero indicates "normal" conditions relative to the 1950–2005 baseline.

  • Negative values indicate dryness, while positive values indicate wetness.


5. Uncertainty Representation

  • SPEI projections include ensemble percentiles (25th, 50th, 75th).

  • This range does not capture all uncertainty (e.g., future emissions, local hydrological changes).

  • Multi-model ensembles are recommended instead of single-model projections.


6. Applications of SPEI in Climate Studies

6.1 Drought Monitoring

  • Tracking seasonal and long-term drought severity.

  • Assessing the impact of climate change on Canadian drought risk.

  • Example: Analyzing multi-year drought patterns in the Prairies.

6.2 Water Resource Management

  • Planning for municipal and agricultural water supplies.

  • Assessing reservoir and groundwater recharge trends.

  • Example: Predicting future water shortages in Ontario and Alberta.

6.3 Wildfire Risk Forecasting

  • SPEI-3 and SPEI-12 used for wildfire risk assessment.

  • Drought conditions influence fire spread, intensity, and duration.

  • Example: Correlating dry periods with wildfire outbreaks in British Columbia.

6.4 Agricultural Impact Assessments

  • SPEI-1 and SPEI-3 used to assess soil moisture conditions.

  • Helps determine crop viability and irrigation needs.

  • Example: Predicting wheat and corn yield losses during dry years.


7. Best Practices for Using SPEI Data

  • Use multi-model ensembles instead of individual model outputs.

  • Compare results across different emission scenarios (RCP2.6, RCP4.5, RCP8.5).

  • Be cautious when applying SPEI in snow/glacier-dominated regions.

  • Combine SPEI with other hydrological data (streamflow, groundwater) for better decision-making.

Limitations

  • SPEI does not account for snowpack, glaciers, or permafrost dynamics.

  • May not fully capture short-term hydrological droughts influenced by local factors.

  • Best used as an indicator of meteorological drought rather than hydrological drought.


8. Data Access & References

8.1 Where to Access SPEI Data

  • Environment Canada SPEI Portal:

    • Climate Data Access

  • CMIP5 SPEI Datasets:

    • Earth System Grid Federation (ESGF)

8.2 References

  • Vicente-Serrano, S. M., et al. (2010). A Multiscalar Drought Index Sensitive to Global Warming. Journal of Climate.

  • Cannon, A. J. (2016). Multivariate Bias Correction of Climate Model Outputs. Journal of Climate.

  • Tam, B. Y., et al. (2018). CMIP5 drought projections in Canada based on SPEI. Canadian Water Resources Journal.

  • Droogers, P. and Allen, R. G. (2002). Estimating Reference Evapotranspiration Under Inaccurate Data Conditions. Irrigation and Drainage Systems.


9. Conclusion

The SPEI dataset is a powerful tool for tracking and predicting drought in Canada, integrating temperature, precipitation, and evapotranspiration for multi-scalar drought assessments.

With multi-model ensemble projections, bias-corrected datasets, and long-term historical records (1900-2100), SPEI supports climate adaptation strategies in agriculture, water management, and disaster risk reduction.

Last updated

Was this helpful?