Abstract
A series of observing system experiments for the two‐month period January–February 2007 was carried out to assess the impact of radiosonde and aircraft data available over North America, as well as the impact of satellite data available over the North Pacific Ocean, on the global data assimilation and forecast system of the Meteorological Service of Canada (MSC). The impact is estimated by comparing two data assimilation and forecast cycles: one (control) assimilates all the observations operationally available while the other (experimental) is identical to the control except that the data from the observing network being examined is denied. In the first part of this study, presented in this paper, particular attention is given to the propagation and magnitude of the impact of the data from these observing networks. It is found that the impact on the accuracy of forecasts over the North American continent is not uniform. Radiosonde and aircraft data combined are the main contributors to the skill of short‐range forecasts over North America. However, as the effect of satellite observations over the North Pacific Ocean moves downstream over the continent, their impact on forecasts becomes dominant for forecast lengths longer than 36 h over western North America, and longer than 72 h over the eastern part of the continent. The impact of these satellite observations is more important over the continental United States than over Canada. In data‐denial experiments, the impact of the aircraft and radiosonde observing networks collocated over southern Canada and the United States, used individually, is much weaker than their combined impact. For short‐range forecasts, the effect of aircraft observations is more important than radiosonde data over eastern North America. Finally, the quality of the forecasts over the Canadian Arctic relies heavily on the radiosonde network.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have