During a typical winter, freezing rain can occur several times, usually associated with synoptic weather systems, over the northeastern United States and southwestern Canada. Impact of a freezing rain event depends on its intensity and duration. Over the past several decades, a countable number of major freezing events have struck the region; the most damaging by far in recent years was the large ice storm that occurred in early January 1998. Ice Storm ′98 has been classified as one of the worst weather-related disasters in Canada. Large accumulations of ice on trees caused branches and small trunks to break, resulting in tree mortality in some cases damage to forest land and harvestable timber stands are considerable.

Low streamflows constitute an important component of hydro-climatic extremes. This is particularly true for Canada Where reduced flows can affect several economic and environmental activities ranging from less hydro-electric production and increased freshwater transportation costs, to reduced water quality and ecological habitat destruction. This paper reviews past research regarding the impacts of large-scale circulation patterns on streamflow variability (including low flows) over Canada. Results from the various Studies reveal that streamflow responses are generally consistent with those observed for large-scale climate. For western Canada, this includes a higher frequency of low-flow events in association with the warmer/drier conditions during El Nino events and positive phases of the Pacific Decadal Oscillation (PDO) and the Pacific North American (PNA) pattern. Relationships in northern/northeastern regions of the country are less robust but in general, reduced streamflows occur during positive phases of the North Atlantic Oscillation (NAO) and Arctic Oscillation (AO). However, it is clearly evident that the spatial and temporal aspects of these relationships are greatly influenced by hydro-climatic complexities associated with individual watersheds, especially in the cordilleran areas of Canada. Future research requires more in-depth analyses into the spatial and temporal aspects of relationships between Circulation variability and low-flow occurrences over critical watersheds within Canada including the combination of climatic patterns and associations with regional, synoptic-scale circulation. The incorporation of fully coupled climate and hydrologic models to assess the impacts of projected climate change on future low-flow Occurrences is also needed. This research would result in a better Understanding, and enhanced prediction of low streamflow events that is critical for the current and future efficient management of water resources throughout the current

The influence of the modal structures of the El Nino-Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO) on the wintertime intraseasonal temperatures and temperature extremes over Canada has been analyzed for the 1900-2008 period. Results demonstrate that intraseasonal temperature variability (longer than 11 days) dominates with a high degree of persistence in western Canada, whereas synoptic-scale variability (shorter than 11 days) prevails in eastern Canada. The PDO has been found to exert a significant modulating effect on ENSO-related intraseasonal temperature impacts in western Canada. In eastern Canada, temperature anomalies are generally weaker, often with ENSO and the PDO producing opposite effects. Spectral analysis shows reddening of the temperature spectra with significant peaks near the 20-day period in western Canada. In contrast, the power of the spectra is weaker with significant peaks occurring near the 13-day period in eastern Canada.

This study examines intercontinental linkages between late spring and early summer Eurasian snow cover extent (SCEss) anomalies and the following winter temperature anomalies over Canada for the 1972-2006 period. The structure of the second interannual mode of Canadian winter temperatures variability captures the SCEss related modulation. The North Atlantic winter atmospheric circulation changes associated with the SCEss, resembling the negative phase of the North Atlantic Oscillation (NAO), suggest a possible pathway for the SCEss influences on the Canadian winter temperatures. Regression and composite analyses show that the SCEss relate robustly to the Canadian winter climate. Larger-than-normal SCEss is associated with below normal winter temperatures in south-central Canada and above normal temperatures over northeastern Canada. Predictive skill of Canadian winter temperatures based on a cross-validated regression model shows that the SCEss offers the predictive potential over regions of Canada where El Nino-Southern Oscillation (ENSO) related skill is weak or nonexistent. Analysis of winter extreme minimum temperatures, by a non-stationary generalized extreme value model, with the SCEss as a covariate, exhibits statistically significant changes over Canada resembling a pattern similar to that of winter mean temperatures. Wavelet analysis shows significant coherence between the SCEss and the second mode of winter temperature variability in the 812-year band. Copyright (c) 2011 Crown in the right of Canada. Published by John Wiley & Sons, Ltd

Two non-linear, machine-learning/statistical methods, i.e., Bayesian neural network (BNN) and support vector regression (SVR), plus multiple linear regression (MLR), were used to forecast surface wind speeds at lead times of 12, 24, 48 and 72 h. Three different schemes, a statistical downscaling model (Scheme 1) using daily reforecast data from the National Centers for Environmental Prediction (NCEP) Global Forecasting System (GFS), an autoregressive model (Scheme 2) based on past wind observations, and a full model (Scheme 3) combining the two, were investigated in this study for the October–March winds from two meteorological stations in the Canadian Arctic (Clyde River and Paulatuk). At very short lead times, Scheme 2 provides better wind speed prediction than Scheme 1, but its forecast scores decrease rapidly with lead time. Scheme 3 generally performs best, especially at shorter lead times. All the linear and non-linear downscaling methods have significantly higher forecast scores at the two stations than the GFS reforecast. The non-linear methods tended to have slightly better forecast scores than linear methods (MLR and the linear version of SVR). There is particular interest in high-wind events, defined as having wind speeds over 22 knots (11.3 m s−1). After rescaling, the continuous wind predictions from Scheme 3 were classified into two types — high-wind event or non-event. For high-wind event forecasting, the non-linear methods have marginally better binary forecast scores than the linear methods for Clyde River but not for Paulatuk. The alternative approach of using support vector classification (SVC) did not perform better, but weighting the high-wind events more heavily than the non-events during model training improved the binary forecast scores.

Droughts are among the world's most costly natural disasters and collectively affect more people than any other form of natural disaster. The Canadian Prairies are very susceptible to drought and have experienced this phenomenon many times. However, the recent 1999-2005 Prairie drought was one of the worst meteorological, agricultural and hydrologic droughts over the instrumental record. It also had major socio-economic consequences, adding up to losses in the billions of dollars. This recent drought was the focus of the Drought Research Initiative (DRI), the first integrated network focusing on drought in Canada. This article addresses some of the key objectives of DRI by providing a collective summary, understanding and synthesis of the 1999-2005 drought. Bringing together the many datasets used in this study was in itself a major accomplishment. This drought exhibited many important, and sometimes surprising, features. This includes, for example, (1) a non-steady large-scale atmospheric circulation (and sea surface temperature) pattern that mainly resulted in subsidence over the region but also cold and warm periods in its evolution; such features have typically not occurred in previous droughts; (2) large spatial gradients between wet and dry areas that, in some instances, were linked with major precipitation events; and (3) many impacts at and below the earth's surface that occurred with varying temporal lags from the meteorological conditions and, in response, these impacts would have fed back onto the character of the drought (e.g., the surface-convection feedback). The drought's complexity poses enormous challenges for its simulation and prediction at all temporal scales. High-resolution models coupled with the surface are needed to address these and many other issues identified in this article.

This study documents and assesses the atmospheric and oceanic variability associated with growing season (May to August) droughts on the Canadian Prairies. For comparison, extreme wet seasons or pluvials are also examined. Using the Palmer Z-Index as a drought indicator, extreme dry and wet seasons are first identified for the period 1950 to 2007. Interrelationships among several atmospheric parameters including large-to synoptic-scale circulation patterns, low-level moisture transport, moisture convergence, precipitable water content and cyclone frequency are then assessed during extreme drought and pluvial periods. In addition, links to the previous winter's global sea surface temperature (SST) patterns are identified using the multivariate technique of singular value decomposition.Results show that moisture from the Gulf of Mexico is notably decreased during the identified drought seasons. Stronger than normal subsidence associated with anomalously high pressure over northwestern North America also leads to weakened moisture transport from the Pacific Ocean. Conversely, during pluvial seasons, low-level flow aided by the circulation associated with increased cyclone frequency over western North America brings abundant moisture northward into the southern Prairies. These circulation patterns over western North America and their associated moisture transport anomalies into the Prairies show some linkages to previous winter SST patterns both globally and in the Pacific Ocean where the SSTs are similar to those associated with interannual El Nino Southern Oscillation (ENSO) events and ENSO-like interdecadal North Pacific variability. This is the first study to examine several interconnected atmospheric and oceanic processes at various scales as they relate to the occurrence of growing season extreme climate over the Canadian Prairies. Results provide a better understanding of the physical mechanisms responsible for the initiation and perpetuation of these extremes.

For forecasting the maximum 5-day accumulated precipitation over the winter season at lead times of 3, 6, 9 and 12 months over Canada from 1950 to 2007, two nonlinear and two linear regression models were used, where the models were support vector regression (SVR) (nonlinear and linear versions), nonlinear Bayesian neural network (BNN) and multiple linear regression (MLR). The 118 stations were grouped into six geographic regions by K-means clustering. For each region, the leading principal components of the winter maximum 5-d accumulated precipitation anomalies were the predictands. Potential predictors included quasi-global sea surface temperature anomalies and 500 hPa geopotential height anomalies over the Northern Hemisphere, as well as six climate indices (the Niño-3.4 region sea surface temperature, the North Atlantic Oscillation, the Pacific-North American teleconnection, the Pacific Decadal Oscillation, the Scandinavia pattern, and the East Atlantic pattern). The results showed that in general the two robust SVR models tended to have better forecast skills than the two non-robust models (MLR and BNN), and the nonlinear SVR model tended to forecast slightly better than the linear SVR model. Among the six regions, the Prairies region displayed the highest forecast skills, and the Arctic region the second highest. The strongest nonlinearity was manifested over the Prairies and the weakest nonlinearity over the Arctic.

An empirical scheme for predicting the meteorological conditions that lead to summer forest fire severity for Canada using the multivariate singular value decomposition (SVD) has been developed for the 1953-2007 period. The levels and sources of predictive skill have been estimated using a cross-validation design. The predictor fields are global sea surface temperatures (SST) and Palmer drought severity index. Two consecutive 3-month predictor periods are used to detect evolving conditions in the predictor fields. Correlation, mean absolute error, and percent correct verification statistics are used to assess forecast model performance. Nationally averaged skills are shown to be statistically significant, which suggests that they are suitable for application to forest fire prediction and for management purposes. These forecasts average a 0.33 correlation skill across Canada and greater than 0.6 in the forested regions from the Yukon, through northern Prairie Provinces, northern Ontario, and central Quebec into Newfoundland. SVD forecasts generally outperform persistence forecasts. The importance of the leading two SVD modes to Canadian summer forest fire severity, accounting for approximately 95% of the squared covariance, is emphasized. The first mode relates strongly to interdecadal trend in global SST. Between 1953 and 2007 the western tropical Pacific, the Indian, and the North Atlantic Oceans have tended to warm while the northeastern Pacific and the extreme Southern Hemisphere oceans have shown a cooling trend. During the same period, summer forest fire exhibited increased severity across the large boreal forest region of Canada. The SVD diagnostics also indicate that the El Nino-Southern Oscillation and the Pacific decadal oscillation play a significant role in Canadian fire severity. Warm episodes (El Nino) tend to be associated with severe fire conditions over the Yukon, parts of the northern Prairie Provinces, and central Quebec. The linearity of the SVD manifests opposite response during the cold (La Nina) events.

Contrary to the impacts ascribed to strong cold El Niño–Southern Oscillation (ENSO) events (La Niña) during the last half of the twentieth century, the 2‐year‐long La Niña during 1998–2000 was accompanied by abnormally warm winter conditions over most of North America. In this study, climate anomalies associated with the 1998–2000 La Niña are compared with the composite of the previous seven strong La Niña events since 1950. Despite the presence of a moderately strong La Niña, analyses of atmospheric circulation and thermal features show that the Pacific–North American sector climate was associated with a zonally elongated ridge in the extratropical North Pacific and warmer‐than‐normal sea surface temperatures (SSTs) in the western tropical Pacific that extended into the North Pacific midlatitudes during 1998–2000 event. Relative to the composite of the previous seven strong La Niñas, anomalous Walker circulation was considerably stronger and shifted westward with the ascending branch of the cell located near 110°E during the 1998–2000 event. This was associated with stronger convection and hence positive precipitation anomalies in the western Pacific. The relationship between SSTs and 500‐hPa geopotential heights over the North Pacific shows that the differences, in thermal and circulation features, between the 1998–2000 La Niña and the seven strong La Niñas bear a close resemblance to a coupled mode of extratropical ocean‐atmosphere variability as revealed by singular value decomposition analysis. This non‐ENSO coupled mode of variability, which relates the interannual variability in SSTs with 500‐hPa heights, was prominent during the winter of 1998–2000. The processes related to this mode are different from those involved in the establishment of extratropical atmospheric circulation and SST anomalies associated with typical La Niña events.

This study provides further evidence of the impacts of tropical Pacific interannual [El Ni (n) over tildeo-Southern Oscillation (ENSO)] and Northern Pacific decadal-interdecadal [North Pacific index (NPI)] variability on the Pacific-North American (PNA) sector. Both the tropospheric circulation and the North American temperature suggest an enhanced PNA-like climate response and impacts on North America when ENSO and NPI variability are out of phase. In association with this variability, large stationary wave activity fluxes appear in the mid- to high latitudes originating from the North Pacific and flowing downstream toward North America. Atmospheric heating anomalies associated with ENSO variability are confined to the Tropics, and generally have the same sign throughout the troposphere with maximum anomalies at 400 hPa. The heating anomalies that correspond to the NPI variability exhibit a center over the midlatitude North Pacific in which the heating changes sign with height, along with tropical anomalies of comparable magnitudes. Atmospheric heating anomalies of the same sign appear in both the tropical Pacific and the North Pacific with the out-of-phase combination of ENSO and NPI. Both sources of variability provide energy transports toward North America and tend to favor the occurrence of stationary wave anomalies.

In Bayesian neural network (BNN) models, the regularization parameter(s) are automatically determined to prevent overfitting. However, such nonlinear BNN models can still underperform linear regression (LR) models during forecast cross-validation. The suspects are the test data predictors which are outliers relative to the training data used in building the models, as the nonlinear model can do wild extrapolations with outlying predictors, thereby decreasing the forecast skills. A new scheme is proposed: If a test datum is considered an outlier, the nonlinear model forecast is replaced by the LR forecast. Methods employing the Mahalanobis distance (MD) and the robust distance (RD) are used to detect " outliers " , with these models referred to as BNN-MD and BNN-RD. BNN models were built to forecast North American surface air temperature (SAT) and precipitation (PRCP) over all seasons at lead times of 3, 6, 9 and 12 months, with cross-validated forecast skills. For the SAT forecasts, without correcting for outliers, the skills of the BNN model are lower than the LR skills in most situations, esp. when using more hidden neurons in the BNN. When outliers are corrected, the skills are generally improved. Both BNN-MD and BNN-RD models show comparable skills to LR at 3-month lead, and higher skills than LR at 12-month lead, though they cannot outperform LR at lead times of 6 and 9 months when the skill is calculated over all seasons. The nonlinear models tend to have higher skills than LR in the eastern half of U.S.A. and central Canada. For spring, the corrected nonlinear models can outperform LR at all lead times. For PRCP forecasts, BNN gives better skills than LR even without correcting for outliers. Correcting for outliers adds skill at lead times of 3 and 12 months. The nonlinear models tend to have higher skills than LR in the western half of Canada, Alaska and western U.S.A., with highest skills in winter, followed by autumn, at 3 and 6-month lead, and vice versa at 9 and 12-month lead.

Neural network models are used to reveal the nonlinear winter atmospheric teleconnection patterns associated with the El Niño‐Southern Oscillation (ENSO) and with the Arctic Oscillation (AO) over the N. Hemisphere. The nonlinear teleconnections (for surface air temperature, precipitation, sea level pressure and 500 hPa geopotential height) are found to relate quadratically to the ENSO and AO indices. Relative to linear teleconnections, nonlinear teleconections appear to propagate perturbations farther, into regions where classical linear teleconnections are insignificant.

The question of the impact of the Atlantic on North American (NA) seasonal prediction skill and predictability is examined. Basic material is collected from the literature, a review of seasonal forecast procedures in Canada and the United States, and some fresh calculations using the NCEP-NCAR reanalysis data.The general impression is one of low predictability (due to the Atlantic) for seasonal mean surface temperature and precipitation over NA. Predictability may be slightly better in the Caribbean and the (sub) tropical Americas, even for precipitation. The NAO is widely seen as an agent making the Atlantic influence felt in NA. While the NAO is well established in most months, its prediction skill is limited. Year-round evidence for an equatorially displaced version of the NAO (named ED_NAO) carrying a good fraction of the variance is also found.In general the predictability from the Pacific is thought to dominate over that from the Atlantic sector, which explains the minimal number of reported Atmospheric Model Intercomparison Project (AMIP) runs that explore Atlantic-only impacts. Caveats are noted as to the question of the influence of a single predictor in a nonlinear environment with many predictors. Skill of a new one-tier global coupled atmosphere-ocean model system at NCEP is reviewed; limited skill is found in midlatitudes and there is modest predictability to look forward to.There are several signs of enthusiasm in the community about using "trends" (low-frequency variations): (a) seasonal forecast tools include persistence of last 10 years' averaged anomaly (relative to the official 30-yr climatology), (b) hurricane forecasts are based largely on recognizing a global multidecadal mode (which is similar to an Atlantic trend mode in SST), and (c) two recent papers, one empirical and one modeling, giving equal roles to the (North) Pacific and Atlantic in "explaining" variations in drought frequency over NA on a 20 yr or longer time scale during the twentieth century.

Nonlinear projections of the Arctic Oscillation (AO) index onto North American winter (December–March) 500-mb geopotential height (Z500) and surface air temperature (SAT) anomalies reveal a pronounced asymmetry in the atmospheric patterns associated with positive and negative phases of the AO. In a linear view, the Z500 anomaly field associated with positive AO resembles a positive North Atlantic Oscillation pattern with statistically significant positive and negative anomalies stretching zonally into central-eastern USA and Canada, respectively, resulting in a cold climate anomaly over northeastern and eastern Canada, Alaska and the west coast of USA, and a warm climate anomaly over the rest of the continent. By contrast, the nonlinear behavior, mainly a quadratic association with AO, which is most apparent when the amplitude of the AO index is large, has the same spatial pattern and sign for both positive and negative values of the index. The nonlinear pattern reveals negative Z500 anomalies over the west coast of USA and the North Atlantic and positive Z500 anomalies at higher latitudes centered over the Gulf of Alaska and northeastern Canada accompanied by cooler than normal climate over the USA and southwestern Canada and warmer than normal climate over other regions of the continent. A similar analysis is conducted on the data from the Canadian Center for Climate Modelling and Analysis second generation coupled general circulation model. The nonlinear patterns of North American Z500 and SAT anomalies associated with the AO in the model simulation are generally consistent with the observational results, thereby confirming the robustness of the nonlinear behavior of North American winter climate with respect to the AO in a climate simulation that is completely independent of the observations.

In this study, the temporal structure of the variation of North Atlantic Oscillation (NAO) and its impact on regional climate variability are analyzed using various datasets. The results show that blocking formations in the Atlantic region are sensitive to the phase of the NAO. Sixty-seven percent more winter blocking days are observed during the negative phase compared to the positive phase of the NAO. The average length of blocking during the negative phase is about 11 days, which is nearly twice as long as the 6-day length observed during the positive phase of the NAO. The NAO-related differences in blocking frequency and persistence are associated with changes in the distribution of the surface air temperature anomaly, which, to a large extent, is determined by the phase of the NAO. The distribution of regional cloud amount is also sensitive to the phase of the NAO. For the negative phase, the cloud amounts are significant, positive anomalies in the convective zone in the Tropics and much less cloudiness in the mid latitudes. But for the positive phase of the NAO, the cloud amount is much higher in the mid-latitude storm track region. In the whole Atlantic region, the cloud amount shows a decrease with the increase of surface air temperature. These results suggest that there may be a negative feedback between the cloud amount and the surface air temperature in the Atlantic region.