This page has been archived on the Web

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

Skip booklet index and go to page content

Threats to Water Availability in Canada

3. Droughts

Barrie Bonsal,1 Grace Koshida,2 E.G. (Ted) O’Brien3 and Elaine Wheaton4

1 Environment Canada, National Water Research Institute, Saskatoon, SK
2 Environment Canada, Adaptation and Impacts Research Group, Toronto, ON
3 Agriculture and Agri-Food Canada, Prairie Farm Rehabilitation Administration, Regina, SK
4 Saskatchewan Research Council, Saskatoon, SK


Current Status

Since most human activities and ecosystem health are dependent on reliable, adequate water supplies, droughts present a serious national threat to Canada. Large-area droughts have major impacts on a wide range of water-sensitive sectors including agriculture, industry, municipalities, recreation, and aquatic ecosystems. They often stress water supplies by depleting soil moisture reserves, reducing streamflows, lowering lake and reservoir levels, and diminishing groundwater supplies. This in turn affects several economic activities: for example, decreased agricultural production, less hydroelectric power generation, and increased marine transportation costs. In addition, droughts have major environmental implications such as reduced water quality, wetland loss, soil erosion and degradation, and ecological habitat destruction.

Droughts are complex phenomena with no standard definition. Simply stated, drought is a prolonged period of abnormally dry weather that depletes water resources for human and environmental needs (AES Drought Study Group, 1986). However, each drought is different depending on factors such as area affected, duration, intensity, antecedent conditions, and a region’s capability to adapt to water shortages. Droughts also differ from other threats (e.g., floods) since they have long durations, and lack easily identified onsets and terminations. Furthermore, their recurrence in drought-prone areas is practically certain since drought is characteristic of dry environments (Maybank et al., 1995). Droughts occur on a variety of temporal and spatial scales with their impacts dependent on timing and sequencing of dry periods. For example, a shortage of water and soil moisture at a critical time for crop growth may initiate agricultural drought, but hydropower generation would not be affected if reservoirs have adequate supplies. Climate anomalies that last from a month to years are the root of most droughts; however, human impacts on resources and climate and changing demands for water are also major contributing factors (McKay et al., 1989).

Droughts in Canada

Although most regions of Canada have experienced drought, the Canadian Prairies (and to a lesser extent, interior British Columbia) are more susceptible mainly due to their high variability of precipitation in both time and space. During the past two centuries, at least 40 long duration droughts have occurred in western Canada. In southern regions of Alberta, Saskatchewan, and Manitoba, multi-year droughts were observed in the 1890s, 1930s, and 1980s (Phillips, 1990; Wheaton, 2000). Droughts in eastern Canada are usually shorter, smaller in area, less frequent, and less intense; nonetheless, some major droughts have occurred during the 20th century. In 1963/64 in southern Ontario, for instance, several wells ran dry, necessitating shipping of water from other areas. Great Lakes’ water levels also fell to extreme lows with major losses incurred by the shipping industry (Gabriel and Kreutzwiser, 1993; Brotton, 1995). Droughts in the Atlantic Provinces occur even less frequently, but reduced occurrence results in lower adaptive capacity, making the region more susceptible to drought impacts (Nova Scotia Department of Agriculture and Fisheries, 2001). Droughts are less of a concern for northern Canada mainly due to their lower population densities; nevertheless, increased frequencies of forest fires during drought years can have serious economic impacts.

The recent 2001/02 drought was unusual in terms of its vast spatial extent. Intense dry conditions encompassed most of southern Canada extending from British Columbia, through the Prairies, into the Great Lakes-St. Lawrence region and even the Atlantic Provinces. Over much of the Prairies, several consecutive seasons of below average precipitation have led to one of the most severe prairie droughts on record, devastating many water-related resources in 2001 and 2002. In 2001, the aggregate level of the Great Lakes plunged to its lowest point in more than 30 years, with lakes Superior and Huron displaying near record lows (Mitchell, 2002). Over Atlantic Canada, three consecutive years of drought conditions have forced Nova Scotia to seek advice from the Prairie Farm Rehabilitation Administration (PFRA) on procedures to augment on-site water supplies for agricultural communities.


Droughts are the result of disruptions to an expected precipitation pattern and can be intensified by anomalously high temperatures that increase evaporation. The major factor in the onset and perpetuation of drought involves circulation patterns in the upper atmosphere. Over Canada, the most extreme warm-season droughts are associated with a persistent upper-air ridge of large amplitude over the affected area. This flow pattern creates 'blocking conditions' that displace the jet stream, cyclonic tracks, and moist air masses and fronts (Chakravarti, 1976; Dey, 1982; AES Drought Study Group, 1986). Droughts can also be initiated and/or perpetuated during the cold season when a lack of precipitation results in lower than normal spring runoff and, thus, in reduced stream flow and reservoir and soil moisture replenishment. These precipitation deficiencies are also caused by anomalous upper-atmospheric circulation patterns and, in particular, a split in the jet stream over North America (e.g., Shabbar et al., 1997).

Several studies have found relationships between sea surface temperatures (SSTs) over various regions of the globe and large-scale atmospheric patterns with associated temperature and precipitation anomalies over Canada. For example, significant relationships between El Niño - Southern Oscillation (ENSO) and winter/early spring temperature and precipitation patterns for several regions of the country have been identified (Shabbar and Khandekar, 1996; Shabbar et al., 1997). Associations between North Pacific SSTs and atmospheric ridging over the Prairies leading to more intense droughts during the growing season have also been shown (Bonsal et al., 1993; Bonsal and Lawford, 1999). However, these summer relationships are much less robust as compared to winter. Relationships between Canadian temperature and precipitation and other large-scale oscillations such as the Pacific Decadal Oscillation (PDO) and the North Atlantic Oscillation (NAO) are also evident during the winter season (e.g., Bonsal et al., 2001a). Droughts tend to persist in that warm, dry springs are followed by hot, dry summers. In addition, there appears to be a tendency for warm summers to follow other warm summers, and so on. Reasons for this are not clear but are likely related to feedback processes that enhance or prolong drought situations (e.g., soil moisture anomalies) (Maybank et al., 1995).

Monitoring, Modelling, and Prediction

Real-time reports of lake and reservoir levels, stream flows, snowpack accumulations, water-supply volume forecasts, dugout water levels (for the Prairies), and precipitation anomalies are currently used for drought monitoring in Canada. The status of these water supplies is critical to activities such as irrigation, water apportionment, storage, flood forecasting, hydroelectric power generation, navigation, fisheries, and wetland habitat. In the Canadian Prairies, provincial water resource agencies have been publishing monthly reports of stream, lake, reservoir, and groundwater levels since the late 1970s. Pasture conditions, on-farm surface water supplies, and seasonal precipitation accumulations are monitored by Agriculture and Agri-Food Canada (AAFC). AAFC maintains the Drought Watch web site that provides real-time information on prairie drought conditions, and promotes practices to reduce drought vulnerability. The Canada-wide drought of 2001 prompted the expansion of Drought Watch to monitor the risk and status of drought over the major agricultural regions of the country. A national map illustrating precipitation accumulations is now prepared in collaboration with the Meteorological Service of Canada (MSC).

Numerous indices that are measures of drought severity are also used for monitoring and modelling drought conditions. These range from simple approaches that only consider precipitation, to more complex indices incorporating a water balance approach using precipitation, potential evapotranspiration, antecedent soil moisture, and runoff (e.g., the Palmer Drought Severity Index [PDSI]; Palmer, 1965). Various soil moisture indices have also been used to monitor and model soil moisture changes from daily precipitation and actual evapotranspiration. A problem with these more complex indices is that evapotranspiration is difficult to compute since it relies on meteorological measurements that are generally not readily available (net radiation, vapour pressure deficit, wind speed). The high spatial variability of summer convective rainfall and the difficulties in modelling snowmelt and blowing snow also hinder regional-scale moisture modelling (Maybank et al., 1995). There are currently several meteorological and surface water indices under investigation and/or consideration for use over all of Canada. Plans are underway to incorporate these indices to monitor near real-time drought conditions across the entire country, similar to the Drought Monitor project in the United States (Svoboda et al., 2002). Satellite and radar measurements can potentially provide solutions to the spatial-scale problems associated with drought monitoring and modelling. MSC currently uses Special Sensor Microwave Imager (SSMI) to produce snow water equivalent maps for the Prairie Provinces, available to water resource agencies.

Drought prediction involves anticipating climatic anomalies that produce unusually dry conditions for an extended period of time; however, at present, there is no completely satisfactory method that can routinely predict Canadian climate over the month to season time frame required for drought analysis. Environment Canada currently produces seasonal forecasts for temperature and precipitation for lead times of 3, 6, 9, and 12 months using both statistical and numerical weather modelling techniques. The forecasts are updated quarterly at the national scale but this is often too infrequent for regional and local drought analyses.


Adaptation involves adjusting to climate change, variability, and extremes to avoid or alleviate negative impacts and benefit from opportunities (Watson et al., 2001). Drought adaptations include short- to long-term actions, programs, and policies implemented both during and in advance of drought to help reduce risks to human life, property, and productive capacity (Wilhite, 2000). Canadians have a great deal of experience in adapting to droughts; however, their adaptation strategies vary by sector and location. Areas with a greater risk of droughts are often better prepared to deal with dry conditions. Drought adaptation decisions are made at a variety of levels ranging from individuals, to groups and institutions, to local and national governments. There are various adaptation processes or strategies including sharing and/or bearing the loss, modifying drought effects, research, education, behavioural changes, and avoidance (Burton et al., 1993). Adaptive drought measures include soil and water conservation, improved irrigation, and construction of infrastructure, including wells, pipelines, dugouts and reservoirs, and exploration of groundwater supplies. The usefulness of each set of strategies varies with location, sector, and the nature and timing of the drought. Better management responses may be made with improved drought and drought impacts monitoring and advanced prediction. Adjustments that occur after drought are generally less effective than planned anticipatory adaptation.

Drought adaptation research and planning strategies are in their early stages although risk management plans for drought-prone regions of the country have been established (e.g., the Agriculture Drought Risk Management Plan for Alberta). Many adaptive strategies have been devised and tested for their effectiveness in reducing drought impacts (Maybank et al., 1995). However, intense, large-area droughts that persist for several years still result in severe hardship, even to those regions used to coping with droughts. An improved capability to estimate the numerous impacts associated with drought is required for enhanced adaptation. In addition, future national, provincial, and municipal level coordinated and proactive drought planning is needed, since vulnerability to future droughts could be exacerbated by increasing development, as well as by increased summer drying and risk of drought projected to occur over most mid-latitude continental interiors as a result of climate change (Watson et al., 2001).


Droughts present a serious threat to water quantity in Canada and thus impact a wide range of water-sensitive sectors including agriculture, industry, municipalities, recreation, and aquatic ecosystems.

Trends and Variability

There has been some effort to define large-scale trends and variability in Canadian temperature and precipitation, and, to a lesser extent, various drought-related indices during the period of instrumental record. With regard to the latter, results have generally shown substantial decadal-scale variability with no consistent trends in terms of frequency, duration, or severity of droughts during the 20th century. A problematic issue for most of these trend analyses is that they have been carried out independently with limited attempts to derive comprehensive results for the entire country. Also, they often differ in terms of starting dates for trend calculation, and with respect to initial conditions for determining drought indices. Furthermore, the limitation of the instrumental record to approximately the last 100 years, combined with sparse high-resolution paleo-climatic information in areas most prone to drought, makes inference into long-term trends in Canadian droughts very difficult. Selected examples of trends and variability in various drought-related parameters are provided below.

High surface temperatures can intensify drought conditions through enhanced evaporation in summer and increased sublimation and melting of the snowpack during winter. Several studies have shown significant trends in temperature and various temperature-related indices over Canada during the 20th century. Mean annual air temperature has increased by an average of 0.9°C over southern Canada for the period 1900-98 (Fig. 1). The greatest warming was observed in the West and the largest rates occurred during winter and especially spring (Zhang et al., 2000). Much of the country has also experienced significant trends toward longer frost-free periods (Bonsal et al., 2001b). This could affect drought occurrence since these trends translate into a longer ice-free season for lakes and rivers, thus increasing the potential for open-water evaporation. From 1900-98, annual precipitation has significantly increased over most of southern Canada, with the exception of southern Alberta and Saskatchewan (Fig. 1). This pattern is also generally evident during all seasons within the year (Zhang et al., 2000). The period 1915-97 was associated with substantial interdecadal variability in North American snow cover, including lowest snow cover anomalies in the 1920s and 1930s and highest during the late 1970s and early 1980s. Coincident with the large increases in spring temperature, the 1980s/90s were characterized by rapid reductions in snow cover during the second half of the snow season and especially in April (Brown, 2000).

Figure 1: Trends in mean annual temperature and total annual precipitation over southern Canada from 1900-98

Fig. 1 Trends in mean annual temperature (°C/99-year period) and total annual precipitation (% change/99-year period) over southern Canada from 1900-98. Grid squares with trends statistically significant at the 5% level are denoted by crosses (taken from Zhang et al., 2000).

Examples of 20th century PDSI time series for various regions of the country are provided in Fig. 2 (Skinner, 2002). Negative PDSI represent drought-like conditions. The series show considerable decadal-scale variability with no long-term trends discernible in any portion of the country. All four graphs, however, do show the large-area drought conditions observed over much of Canada during late 1990s to early 2000s. Sauchyn and Skinner (2001) reconstructed July PDSI for the southwestern Canadian Plains using tree ring chronologies dating back to 1597. Results showed that the 20th century lacked the prolonged droughts of the 18th and 19th centuries when the PDSI was consistently below zero for decades at a time. Clusters of drought years in the series suggest the existence of a 20- to 25-year periodicity over this region.

Figure 2: Annual PDSI values for a) Kamloops, BC, b) Saskatoon, SK, c) Sherbrooke, QC, and d) Yarmouth, NS.

Fig. 2 Annual PDSI values for a) Kamloops, BC, b) Saskatoon, SK, c) Sherbrooke, QC, and d) Yarmouth, NS. Solid lines represent 10-year running means (source: Climate Research Branch, Meteorological Service of Canada, Environment Canada, Downsview, ON).

In terms of large-scale circulation, Skinner et al. (1999) identified an increasing trend in 500 hPa heights over much of Canada with an amplification of the western Canadian ridge and an eastward shift of the Canadian Polar Trough for the period 1953-1995. Many large-scale atmospheric and oceanic oscillations such as the PDO and NAO generally revealed considerable inter-annual and inter-decadal variability during the last century. El Niño events, however, have tended to be more frequent and intense in the last 20 to 30 years and some models are projecting more El Niño-like conditions in the future (e.g., Timmermann et al., 1999). This could affect winter drought conditions since El Niño has been shown to be associated with warmer, drier winters over most of southern Canada (Shabbar and Khandekar, 1996; Shabbar et al., 1997).

There have been some analyses of trends and variability in various water-related drought indicators over Canada, but these records tend to be much shorter. Over the last 30 to 50 years, mean stream flow has decreased in many parts of Canada with significant reductions in southern regions of the country (Zhang et al., 2001). Great Lakes’ water levels have shown considerable variability during the 20th century. For example, Fig. 3 reveals several decadal-scale periodicities in Lake Huron levels with no evidence of any long-term trend. Lower levels coincided with the droughts of the 1930s, early 1960s, and the most recent 1999-2001 dry period. Over the Prairies, the numbers and water levels of wetlands have shown no clear trend over the last 40 to 50 years (Conly and van der Kamp, 2001).

Figure 3: Annual lake levels for Lake Huron for the period 1900-2001

Fig. 3 Annual lake levels for Lake Huron for the period 1900-2001 (source: Environment Canada, Water Issues Division, Burlington, ON).

All Global Climate Models are projecting future increases of summer continental interior drying and associated risk of droughts. The increased drought risk is ascribed to a combination of increased temperature and potential evaporation not being balanced by precipitation (Watson et al., 2001). However, considerable uncertainty exists with respect to future precipitation, particularly on a regional and intra-seasonal basis. Furthermore, relatively little is known regarding changes to large-scale circulation and, since these patterns have a significant impact on temperature and precipitation over Canada, the occurrence of future drought remains a huge knowledge gap.

Knowledge Gaps and Program Needs

There are several gaps in the knowledge of droughts that limit our ability to understand their occurrence, monitor/model their status, and adapt to their negative effects. The following identifies major research and program needs regarding droughts in Canada.

Occurrence of Droughts

A better understanding of the physical causes and characteristics of past droughts including their spatial and temporal variability is required. This understanding will provide improved insight to short-term (seasonal to annual) and long-term (decade to century) projections of future droughts in Canada. In particular, we require:

  • improved knowledge of drought trends and variability prior to the instrumental record. This requires more research into reliable proxy indicators to reconstruct drought occurrence over various regions of Canada for the last few hundred years.
  • improved understanding of the physical causes of drought initiation, persistence, and termination during the last few hundred years. This includes:
    1. the role of large-scale atmospheric and oceanic oscillations in the initiation and persistence of anomalous circulation patterns responsible for drought, particularly during the summer season
    2. impacts of soil moisture anomalies on the perpetuation and migration of drought
    3. physical causes of multi-year droughts and their recurrence on decadal time scales
    4. atmospheric circulation patterns associated with unusually large spatial-scale droughts (e.g., the 2001 drought over most of southern Canada)
    5. atmospheric conditions responsible for the termination of a drought including aspects such as convective rainfall, precipitation trigger mechanisms, and moisture sources
  • better knowledge regarding the occurrence of future droughts in terms of likely areas to be affected and potential changes to their frequency, duration, and severity. This requires:
    1. more reliable future climate simulations (particularly precipitation) from Global and Regional Climate Models
    2. improved downscaling methods for application of climate model data to appropriate spatial and temporal scales
    3. knowledge of future changes to large-scale circulation patterns and oscillations such as ENSO, PDO, and the NAO.

Monitoring, Modelling, and Prediction

The ability to predict drought onset, intensity, and termination more accurately requires improvements in modelling and monitoring of current drought conditions, as well as better short-term (seasonal) climate forecasts. The following are needed to improve our capability to monitor, model, and predict droughts in Canada:

  • improved accessibility to past and near real-time meteorological data
  • restoration and expansion of the climate station network to provide adequate spatial coverage of meteorological observations over the country
  • development of a total water supply database including, for example, improved data of streamflow records, wetland numbers, and groundwater supplies
  • development of an index or combination of indices to monitor past and near real-time drought conditions and to aid in recognition of drought sufficiently in advance. Standard indices would allow for national-scale evaluations of drought
  • better understanding of the amount and distribution of groundwater resources including linkages to climate and surface water supply
  • development of better methodologies to incorporate remote sensing and ground-level radar for drought monitoring and management (to augment the climate station network). The geospatial and temporal capacity of satellite imagery offers many opportunities for advanced monitoring capabilities
  • incorporation of existing Geographical Information System (GIS) techniques to provide better spatial representations of drought. For example, the migration patterns of drought and its associated synoptic circulation patterns could be tracked on a variety of temporal scales
  • better hydrologic modelling techniques and, in particular, improved methodologies to estimate evapotranspiration
  • improved integration of Global and Regional Climate Models with distributed water balance models in order to model future drought conditions
  • more reliable short-term (seasonal) forecasts of temperature and precipitation at the appropriate spatial scales to aid in prediction of drought onset, intensity, persistence, and termination.

Impacts and Adaptation

Droughts are certain to recur in the future. As a result, more effective short- and long-term adaptation strategies are required to defend against these future droughts including improved technological, monitoring, and predictive capabilities. Additional requirements include:

  • more rapid updates of potential drought conditions to activate drought response and adaptation options
  • identification of ecosystem thresholds to determine at what point during a drought adaptation options should be activated to avoid serious or irreversible losses. The same applies to economic thresholds
  • more research into understanding and modelling of drought adaptation measures, including their effectiveness, practicality, costs, and benefits
  • improved knowledge regarding adaptation to prolonged droughts including those that may result from climate change
  • better abilities to assess adequately the socio-economic consequences of alternative drought adaptation strategies.


AES Drought Study Group. 1986. An applied climatology of drought in the Canadian Prairie Provinces. Report No. 86-4, Canadian Climate Centre, Downsview, ON. 197 p.

Bonsal, B.R., A.K. Chakravarti and R.G. Lawford. 1993. Teleconnections between north Pacific SST anomalies and growing season extended dry spells on the Canadian Prairies. Int. J. Climatol. 13: 865-878.

Bonsal, B.R. and R.G. Lawford. 1999. Teleconnections between El Niño and La Niña events and summer extended dry spells on the Canadian Prairies. Int. J. Climatol. 19: 1445-1458.

Bonsal, B.R., A. Shabbar and K. Higuchi. 2001a. Impacts of low frequency variability modes on Canadian winter temperature. Int. J. Climatol. 21: 95-108.

Bonsal, B.R., X. Zhang, L.A. Vincent and W.D. Hogg. 2001b. Characteristics of daily and extreme temperatures over Canada. J. Climate 14: 1959-1976.

Brotton, J. 1995. Causes and impacts of 1960s low water levels on Canadian Great Lakes interests: phase 2 report. Report prepared for Environment Canada, Environmental Adaptation Research Group, Burlington, ON. 60 p.

Brown, R.D. 2000. Northern hemisphere snow cover variability and change, 1915-97. J. Climate 13: 2339-2355.

Burton, I., R.W. Kates and G.F. White. 1993. The environment as a hazard. The Guilford Press, New York. 290 p.

Chakaravarti, A.K. 1976. Precipitation deficiency patterns in the Prairie Provinces of Canada. Prairie Forum – The Journal of the Canadian Plains Research Centre 1: 95-110.

Conly, F.M. and G. van der Kamp. 2001. Monitoring the hydrology of Canadian Prairie wetlands to detect the effects of climate change and land use changes. J. Environ. Monitor. Assess. 67: 195-215.

Dey, B. 1982. Nature and possible causes of droughts on the Canadian Prairies - case studies. J. Climatol. 2: 233-249.

Gabriel, A.O. and R.D. Kreutzwiser. 1993. Drought hazard in Ontario: a review of impacts, 1960-1989, and management implications. Can. Water Resour. J. 18: 117-132.

Maybank, J., B.R. Bonsal, K. Jones, R.G. Lawford, E.G. O’Brien, E.A. Ripley and E. Wheaton. 1995. Drought as a natural disaster. Atmosphere-Ocean 33: 195-222.

McKay, G.A., R.B. Godwin and J. Maybank. 1989. Drought and hydrological drought research in Canada. Can. Water Resour. J. 14: 71-84.

Mitchell, J.G. 2002. Down the drain? The incredible shrinking Great Lakes, p. 37-51. National Geographic, September 2002.

Nova Scotia Department of Agriculture and Fisheries. 2001. Estimated impact of 2001 drought on agriculture in Nova Scotia. Nova Scotia Department of Agriculture and Fisheries, Halifax, NS.

Palmer, W.C. 1965. Meteorological drought. U.S. Weather Bureau, Research Paper 45, Washington, DC. 58 p.

Phillips, D. 1990. The climates of Canada. Environment Canada, Ottawa, ON. 176 p.

Sauchyn, D.J. and W.R. Skinner. 2001. A proxy record of drought severity for the southwestern Canadian Plains. Can. Water Resour. J. 26: 253-272.

Shabbar, A., B.R. Bonsal and M. Khandekar. 1997. Canadian precipitation patterns associated with the Southern Oscillation. J. Climate 10: 3016-3027.

Shabbar, A. and M. Khandekar. 1996. The impact of El Niño-Southern Oscillation on the temperature field over Canada. Atmosphere-Ocean 34: 401-416.

Skinner, W.R. 2002. Update on western Canada drought: the Palmer Drought Severity Index. Unpublished report.

Skinner, W.R., B.J. Stocks, D.L. Martell, B.R. Bonsal and A. Shabbar. 1999. The association between circulation anomalies in the mid-troposphere and area burned by wildland fire in Canada. Theoret. Appl. Climatol. 63: 89-105.

Svoboda, M., D. LeComte, M. Hayes, R. Heim, K. Gleason, J. Angel, B. Rippey, R. Tinker, M. Palecki, D. Stooksbury, D. Miskus and S. Stephens. 2002. The drought monitor. Bull. Amer. Meteorolog. Soc. 83: 1181-1190.

Timmermann, A., J. Oberhuber, A. Bacher, M. Esch, M. Latif and E. Roeckner. 1999. Increased El Nino frequency in a climate model forced by future greenhouse warming. Nature 398: 694-696.

Watson, R.T. and the Core Writing Team. 2001. Climate change 2001: synthesis report. Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK. 34 p.

Wheaton, E.E. 2000. Canadian prairie drought impacts and experiences, p. 312-330. In D. Wilhite (ed.), Drought: a global assessment, Volume I. Routledge Press, London, UK.

Wilhite, D. 2000. Drought as a natural hazard: concepts and definitions, Part I. In D. Wilhite (ed.), Drought: a global assessment, Volume I. Routledge Press, London, UK.

Zhang, X., K.D. Harvey, W.D. Hogg and T.R. Yuzyk. 2001. Trends in Canadian streamflow. Water Resour. Res. 37: 987-998.

Zhang, X., L.A. Vincent, W.D. Hogg and A. Niitsoo. 2000. Temperature and precipitation trends in Canada during the 20th century. Atmosphere-Ocean 38: 395-429.

Date modified: