Evaluation of a New Land Surface Package in the Canadian Numerical Weather Prediction System Using Scree-Level Analyses

Objective evaluation of a new land surface package for implementation in Environment and Climate Change Canada (ECCC)’s medium-range global deterministic numerical weather prediction (NWP) system has been performed in this study using screen-level analyses of air temperature and humidity.

Global medium range model forecasts are compared with screen-level analyses to investigate the impact of the new land surface package featuring new modifications of i) a new database to specify soils and vegetation characteristics, ii) improved initialization of land surface variables by the assimilation of space-based remote sensing observations, and iii) a more sophisticated land surface scheme.

The evaluation is performed by generating the lead-time-averaged maps of the standard deviation error (STDE) and temporal correlation between forecasts and analysis, as well as plots of area-averaged lead-time evolution of these parameters.

The objective evaluation confirms the positive impact of the new land surface package on the medium-range forecast of near-surface air temperature and humidity for a summertime period. Comparing with the current operational configuration, results from the new package resulted in smaller STDEs and larger temporal correlation for both variables. Greater improvement was achieved for humidity than for air temperature.

The maximum impact was found around 7-day lead-time, with substantial gains in absolute and relative values for STDE and temporal correlation.