Observational study of land-surface-cloud-atmosphere coupling on daily timescales.

Our aim is to provide an observational reference for the evaluation of the surface and boundary layer parameterizations used in large-scale models using the remarkable long-term Canadian Prairie hourly dataset. First we use shortwave and longwave data from the Baseline Surface Radiation Network (BSRN) station at Bratt's Lake, Saskatchewan, and clear sky radiative fluxes from ERA-Interim, to show the coupling between the diurnal cycle of temperature and relative humidity and effective cloud albedo and net longwave flux. Then we calibrate the nearby opaque cloud observations at Regina, Saskatchewan in terms of the BSRN radiation fluxes. We find that in the warm season, we can determine effective cloud albedo to ±0.08 from daytime opaque cloud, and net long-wave radiation to ±8 W/m2 from daily mean opaque cloud and relative humidity. This enables us to extend our analysis to the 55 years of hourly observations of opaque cloud cover, temperature, relative humidity, and daily precipitation from 11 climate stations across the Canadian Prairies. We show the land-surface-atmosphere coupling on daily timescales in summer by stratifying the Prairie data by opaque cloud, relative humidity, surface wind, day-night cloud asymmetry and monthly weighted precipitation anomalies. The multiple linear regression fits relating key diurnal climate variables, the diurnal temperature range, afternoon relative humidity and lifting condensation level, to daily mean net longwave flux, windspeed and precipitation anomalies have R2-values between 0.61 and 0.69. These fits will be a useful guide for evaluating the fully coupled system in models.