Empirical estimation of the annual range of monthly-mean temperatures

Summary ?This is a sequel to a study of the empirical estimation of the annual-mean temperature at any location on land, using only geographical information – latitude, elevation, distance from the nearer ocean shore at the same latitude – and coastal sea-surface temperature. Here long-term mean station data and NCAR/NCEP (National Center for Atmospheric Research/National Centers for Environmental Prediction) global re-analysis data are used to describe and estimate spatial patterns of annual range of monthly-mean temperatures. The two key influences on annual range are the latitude and the distance from the upwind shore. Secondary factors include mountain barriers, shape of the local topography, elevation, and vicinity to large bodies of water. An empirical relationship is derived, based on the two key factors alone, assuming zonal winds and adjustments for the effects of mountain barriers and for the proximity to a sea to the north or south. An independent test of this relationship yields errors around 1.0 K. The range estimates yield January and July average temperatures when combined with annual-mean temperatures. Such estimates also carry an uncertainty of about 1.0 K. The procedure can be inverted, i.e. knowledge of the annual mean and range can be used to infer location. Received August 23, 2001; accepted June 17, 2002