Improving 2-m Temperature Spread in the Next NCEP Global Ensemble

A suite of three stochastic physics components, stochastically perturbed physics tendencies (SPPT), stochastic humidity perturbation (SHUM), and stochastic kinetic energy backscatter (SKEB), will likely replace stochastic total tendency perturbation (STTP) as the model of model uncertainty in the next NCEP global ensemble. While this stochastic physics suite has improved ensemble spread in the free atmosphere in many systems, including pre-production tests of the next NCEP global ensemble, 2-m temperatures remain substantially underspread. This underspread is due to the lack of direct perturbation of the land surface, which heavily influences the 2-m temperature. ESRL and EMC are exploring different approaches to address this issue ahead of the next implementation: ESRL is attempting to introduce spread through stochastic parameter perturbation, while EMC  is attempting to introduce spread by perturbing soil temperature directly as part of the SPPT. Here we discuss the different impact of these methods on 2-m temperature spread.