Rasouli K.,University of British Columbia |
Hsieh W.W.,University of British Columbia |
Cannon A.J.,Meteorological Service of Canada
Journal of Hydrology | Year: 2012
Weather forecast data generated by the NOAA Global Forecasting System (GFS) model, climate indices, and local meteo-hydrologic observations were used to forecast daily streamflows for a small watershed in British Columbia, Canada, at lead times of 1-7. days. Three machine learning methods - Bayesian neural network (BNN), support vector regression (SVR) and Gaussian process (GP) - were used and compared with multiple linear regression (MLR). The nonlinear models generally outperformed MLR, and BNN tended to slightly outperform the other nonlinear models. Among various combinations of predictors, local observations plus the GFS output were generally best at shorter lead times, while local observations plus climate indices were best at longer lead times. The climate indices selected include the sea surface temperature in the Niño 3.4 region, the Pacific-North American teleconnection (PNA), the Arctic Oscillation (AO) and the North Atlantic Oscillation (NAO). In the binary forecasts for extreme (high) streamflow events, the best predictors to use were the local observations plus GFS output. Interestingly, climate indices contribute to daily streamflow forecast scores during longer lead times of 5-7. days, but not to forecast scores for extreme streamflow events for all lead times studied (1-7. days). © 2011 Elsevier B.V.
Fleming S.,BC Hydro |
Fleming S.,University of British Columbia |
Whitfield P.,Meteorological Service of Canada |
Whitfield P.,Simon Fraser University
Atmosphere - Ocean | Year: 2010
We assessed the impacts of some key Pacific ocean-atmosphere circulation patterns on annual cycles of temperature and precipitation across British Columbia, Yukon, and southeast Alaska. The El Niño-Southern Oscillation (ENSO), the Pacific Decadal Oscillation (PDO), and ENSO conditional on PDO states were considered in composite analyses of 71 long, high-quality datasets from surface meteorological stations. Month-by-month, station-by-station Monte Carlo bootstrap tests were employed to assess statistical significance. The results trace precipitation and temperature responses as a function of location, season, and climate mode. In summary, temperature responses were relatively uniform, with higher (lower) temperatures during the warm (cool) phases of these circulation patterns. Nevertheless, strength and seasonal persistence varied considerably with location and climate mode. Impacts were generally most consistent in winter and spring but could extend through most of the year. Overall spatiotemporal patterns in precipitation response were decoupled from those in temperature and were far more heterogeneous. Complexities in precipitation signals included north-south inverse teleconnectivity along the Pacific coast, with a zero-response hinge point in the approximate vicinity of northern Vancouver Island; seasonally opposite anomalies in several interior regions, which might conceivably reflect contrasting effects of Pacific climate modes on wintertime frontal storms versus summertime convective storms; and a consistent lack of substantial response in northwestern British Columbia and possibly southwestern Yukon, conjectured to reflect complications associated with the Icefield Ranges. The product is intended primarily as a basic-level set of climate response maps for hydrologists, biologists, foresters, and others who require empirical assessments of relatively local-scale, year-round ENSO and PDO effects across this broad region.[Traduit par la rédaction] Nous avons évalué les répercussions de certaines configurations de circulation océan Pacifique-atmosphère clés sur les cycles annuels de température et de précipitations en Colombie-Britannique, au Yukon et dans le sud-est de l'Alaska. Nous avons examiné l'El Niño-oscillation australe (ENSO), l'Oscillation décennale du Pacifique (ODP) et la sensibilité de l'ENSO à l'égard des états de l'ODP dans des analyses composites de 71 ensembles de données longs et de bonne qualité de stations météorologiques de surface. Nous avons employé des tests d'amorçage de Monte Carlo mois par mois, station par station, pour mesurer la signification statistique. Les résultats tracent les réponses des précipitations et de la température en fonction de l'endroit, de la saison et du mode climatique. En résumé, les réponses de la température étaient relativement uniformes, avec les températures les plus élevées (basses) durant les phases chaudes (froides) de ces configurations de circulation. Néanmoins, la force et la persistance saisonnière variaient considérablement selon l'endroit et le mode climatique. Les répercussions étaient généralement plus cohérentes en hiver et au printemps mais pouvaient se faire sentir durant la majeure partie de l'année. Généralement, les configurations spatiotemporelles générales dans la réponse des précipitations étaient dissociées de celles de la température et étaient beaucoup plus hétérogè nes. Les complexités dans les signaux de précipitations comprenaient une téléconnexion nord-sud inverse le long de la côte du Pacifique, avec un point charnière de réponse nulle dans le voisinage approximatif du nord de l'île de Vancouver; des anomalies saisonnièrement opposées dans plusieurs régions intérieures, qui pourraient refléter des effets contrastants de modes climatiques du Pacifique sur les tempêtes frontales hivernales par rapport aux tempêtes convectives estivales; et un manque constant de réponse marquée dans le nord-ouest de la Colombie-Britannique et possiblement le sud-ouest du Yukon, que l'on croit attribuables aux complications liées aux chaînons des Glaciers. Le produit est principalement destiné à servir d'ensemble de base de cartes de réponse climatique pour les hydrologistes, les biologistes, les forestiers, etc. qui ont besoin d'évaluations empiriques des effets de l'ENSO et de l'ODP à une échelle assez locale, toute l'année, dans cette grande région.
Bitar L.,Dalhousie University |
Duck T.J.,Dalhousie University |
Kristiansen N.I.,Norwegian Institute For Air Research |
Stohl A.,Norwegian Institute For Air Research |
Beauchamp S.,Meteorological Service of Canada
Journal of Geophysical Research: Atmospheres | Year: 2010
The eruption of Kasatochi volcano on 7-8 August 2008 injected material into the troposphere and lower stratosphere of the northern midlatitudes during a period of low stratospheric aerosol background concentrations. Aerosols from the volcanic plume were detected with a lidar in Halifax, Nova Scotia (44.64°N, 63.59°W) 1 week after the eruption and for the next 4 months thereafter. The volcanic origin of the plume is established using the FLEXPART Lagrangian particle transport model for both the stratosphere and troposphere. The stratospheric plume descended 47.1 ± 2.8 m/d on average as it dispersed, corresponding to a cooling rate of 0.60 ± 0.07 K/d. The descent rate was the same for the tropopause (within statistical uncertainties). The top of the plume remained steady at about 18 km altitude and was likely sustained by vertical eddy diffusion from large-scale horizontal mixing. The lower boundary of the plume descended with the tropopause. The integrated aerosol backscatter between 15 and 19 km altitude was relatively constant at about 8 × 10 -5 sr-1 for 532 nm wavelength. Observations and modeling of Kasatochi aerosols in the middle and lower troposphere indicate a possible ground impact. The volcanic contribution to surface PM2.5 did not exceed 5 g/m3 at the measurement site. Copyright © 2010 by the American Geophysical Union.
Ma J.,Environment Canada |
Cao Z.,Meteorological Service of Canada
Environmental Science and Technology | Year: 2010
A perturbed air-surface coupled model has been developed to simulate and predict perturbations of POPs concentrations in various environmental media under given climate change scenarios. By introducing the perturbations in air temperature and precipitation induced by climate change in the model, we have examined the corresponding perturbations in the concentration of POPs in the closed air-soil and air-water systems. Numerical experiments for several POPs have been conducted based on the possible future climate change scenarios. It was found that hexachlorobenzene (HCB), hexachlorocyclohexanes (HCHs), and a polychlorinated biphenyl (PCBs) congener, PCB-153, exhibit strong response to specified climate change scenarios as shown by their high concentrations perturbations in air. In the air-soil system the coupled model predicts 4-50% increases in the air concentrations of these chemicals corresponding to an increase of 0.05-0.1 K yr-1 in the air temperature. Based on our simulations, a 20% increase/decrease in precipitation can result in a 53% and 4% decrease/increase in perturbed air concentration of γ-HCH and α-HCH, respectively. Also, the model can be used to determine the direction of air-surface exchange of POP perturbations induced by climate change. © Published 2010 by the American Chemical Society.
Pickart R.S.,Woods Hole Oceanographic Institution |
Pratt L.J.,Woods Hole Oceanographic Institution |
Torres D.J.,Woods Hole Oceanographic Institution |
Whitledge T.E.,University of Alaska Fairbanks |
And 5 more authors.
Deep-Sea Research Part II: Topical Studies in Oceanography | Year: 2010
The flow of summer and winter Pacific water masses through Herald Canyon is investigated using data from a high-resolution hydrographic/velocity survey conducted in summer 2004. The survey was part of the Russian-American Long Term Census of the Arctic (RUSALCA) program, and consisted of four cross-canyon transects occupied over a 2-day period. At the time of the survey dense winter water was entering the western side of the canyon from the Chukchi Sea, flowing alongside a poleward jet of summer water on the canyon's eastern flank. As the dense water progressed northward it switched sides of the canyon and underwent a sudden increase in layer thickness. This coincided with vertical mixing near the interface of the winter and summer water, producing a new water mass mode exiting the canyon. All of these features are consistent with the notion of hydraulic activity occurring in the canyon. A three-layer hydraulic theory is applied to the flow, which suggests that it is supercritical and that hydraulic control is likely. A lock-exchange formulation accurately predicts the northward transport of the winter water. The origin of the winter water and the manner in which it drains into the canyon is investigated using satellite ice-concentration data, atmospheric re-analysis fields, historical in-situ data, and a simple circulation model. Finally, the fate of the Pacific water exiting the canyon, and its connection to the Chukchi shelfbreak current, is discussed. © 2009 Elsevier Ltd. All rights reserved.