[1] Airflow trajectories were used to create a long-term (40-year) air mass climatology for the lower peninsula of Michigan. The climatology provides a necessary baseline for evaluating the impact of changes in airflow on historical and potential future variations in temperature and precipitation. Five-day back trajectories were calculated four times per day at the 925 hPa level using wind data from the NCEP/NCAR reanalysis fields. A Geographic Information System (GIS) was used to manage and display the large (58,440) volume of trajectories. The analysis revealed spatially coherent airflow pathways, and six resulting air mass source regions were defined. Also evident were large monthly variations in the frequency of trajectories from the major source regions. A residence time analysis performed within a GIS proved to be an efficient and effective means for summarizing the trajectories. Finally, a comparison of the results presented here to those of previous climatological analyses suggests that the choice of methodology has considerable influence on identification of the primary source regions for a location.

A climatology of springtime lower tropospheric airflow, as represented by 850 hPa observed winds and 1000 hPa geostrophic winds, is presented for central and eastern North America. The climatology is based on twice‐daily grid‐point values for 1969–1989 from the operational analyses for the Northern Hemisphere prepared by the USA National Weather Service's National Meteorological Centre. Sixteen‐point wind roses are used to show the spatial and temporal distributions by direction category of wind frequency, average speed, and wind run. These analyses are supplemented by resultant wind vector and streamline maps for each month. Evident in the climatological distributions is the declining strength of the circulation from March to May, the northward migration and strengthing of the North Atlantic subtropical anticyclone in late spring, and the nocturnal strengthening and veering of southerly airflow over the southern Great Plains. Additionally, areas of cyclonic and/or anticyclonic resultant flow correspond well with earlier identified centres of cyclogenesis and anticyclogenesis. The results presented here may provide a useful climatological baseline for future studies involving lower tropospheric airflow.