92nd American Meteorological Society Annual Meeting (January 22-26, 2012)

Monday, 23 January 2012: 1:45 PM
Canadian Experiences Applying Drought Indices and Drought Monitoring to Agriculture [INVITED]
Room 352 (New Orleans Convention Center )
Allan Howard, National Agroclimate Information Service, Regina, SK, Canada

Severe droughts in Canada have resulted in significant costs, both direct costs in support to the agriculture industry and indirect costs as reduced Gross Domestic Product. Canada is increasingly using drought monitoring as an indicator of need for agriculture disaster assistance, meaning that the reliability of drought indices must be validated with impact in the field. The diversity of Canada's climatic regions, combined with the diversity of agricultural operations adds considerable complexity to this interpretation.

Canada, the United States and Mexico use the same indices for preparing the North American Drought Monitor. While the use of the same indicators simplifies the analysis and comparison of drought characteristics across the diverse landscapes (extent, severity, intensity and duration), the assumptions behind some of the indices (e.g. the Palmer drought Index, the percentage of total precipitation) are such that the magnitude of indices in different climatic regions of Canada do not always mean the same thing when indices are examined in relation to agriculture impacts.

The National Agroclimate Information Service (NAIS) conducted initial studies of some drought indices in Canada, including the Palmer Drought Severity Index (PDSI), Standardized Precipitation Index (SPI) and the Percentage of average precipitation, and a Multi-Index drought model (MID) and found that impacts to agriculture are not similar in all climatic zones when the drought index reaches a certain threshold. For example, the SPI was found to be well correlated to wheat yields in a Mixed Grassland site on the prairies however it was poorly correlated with wheat yields at a site in the Lake Erie Lowland.

These results suggest that there are regional sensitivities that are probably not well captured by existing indices. While a system to monitor drought on a continental or global scale is needed, a regional system linking indices, triggers and impacts should be considered as part of the way forward for advancing our understanding of drought and it effects on agriculture.

Supplementary URL: