Impact of the Arctic observing systems on the ECCC global weather forecasts

This study examines the impact of the conventional (in situ) and satellite observing systems in the Arctic on forecasts in the Northern Hemisphere polar and midlatitude regions using the Canadian global forecast system. It is part of an international cooperation to understand the role and the relative importance of the polar observing systems in numerical prediction systems in order to make recommendations for future optimal observing network design in this area. Both Observing System Experiment and Forecast Sensitivity to Observation Impact methods are used to assess the relative importance of the observing systems north of 60°N. The experiments are conducted during winter 2017–2018 and summer 2018, which include the first two Special Observing Periods of the Year of Polar Prediction. The impact of supplementary radiosondes launched during these periods is also assessed. It is found that the joint impact of all Arctic satellite data on forecasts is two to four times larger than the impact of all conventional data in this area. The microwave sounders and the radiosonde network have the largest impact among the satellite and convention observing systems, respectively. The impact of the conventional data is larger than that of microwave sounders in both winter and summer seasons. The impact of the supplementary radiosondes on short‐range forecast in the Arctic is positive, but becomes negligible beyond 24 hr forecast lead‐time. The impact of observations in the Arctic on medium‐range forecasts in the midlatitudes is larger over eastern North America and northern Asia for both seasons examined. Observing system experiments were conducted to examine the impact of the Artic observing systems on forecasts in the Northern Hemisphere polar and midlatitude areas. It is found that the joint impact of all satellite data in the Arctic is much larger than the impact of all conventional data. The microwave sounders and the radionsonde network have the largest impact among the satellite and convention observing systems, respectively. The impact on medium‐range forecasts in the midlatitudes is larger over eastern North America and northern Asia.