Agriculture and Agri-Food Canada have produced daily precipitation, minimum and maximum temperature across Canada (south of 60°N) for climate related application purpose using thin-plate smoothing splines, as implemented in the ANUSPLIN climate modeling software (Hutchinson et al., 2009; McKenney et al., 2011). The so-called ANUSPLIN data uses ground-based observations and generates daily gridded data from 1951 to 2017 on a Lambert conformal conic projection with 5’ arc minutes spacing (equivalent to a resolution of about 10 km). The key strength of this spatial interpolation method is its global dependence on all data, permitting robust and stable determination of spatially varying dependences on elevation. Hutchinson et al. (2009) have shown that while ANUSPLIN fall month’s absolute errors were remarkably small, those of winter months were quite large due to rather difficult observation and measurement conditions. However, the recent comparison between various (ANUSPLIN and other) gridded observed products shows compatible values for mean total precipitation, wet days occurrence and mean/extreme intensity of daily precipitation, over both southern Canada and eastern USA in winter (e.g. Diaconescu et al., 2016).

The initially 1-hour – 0.125° x 0.125° precipitation and temperature used as a forcing data in the North America Land Data Assimilation System (NLDAS2, Xia et al., 2012) is involved in the present study as observed rainfall over the USA. The so called NLDAS precipitation results from daily accumulation of the 1 hourly gauge-only precipitation analysis of the Climate Prediction Center, performed directly on the NLDAS grid.

ANUSPLIN and NLDAS were combined and regridded on a common North America grid (south of 60°N) at 5’ arc minutes spacing using a conservative mapping algorithm (as described in Diaconescu et al. ,2015) allowing minimizing interpolation errors. We calculated indices of extreme weather based on those daily values of temperature and precipitation to create a climatological atlas over north america over 1981-2010 normal period. Please choose a variable and a frequency bellow to display climatology.