New schemes to perturb sea‐surface temperature and soil moisture content in MOGREPS

Abstract

This article investigates two schemes that perturb sea‐surface temperatures (SSTs) and soil moisture content (SMC) in the Met Office Global and Regional Ensemble Prediction System (MOGREPS), to address a known deficiency of a lack of ensemble spread near the surface. Results from a two‐month‐long trial during the Northern Hemisphere summer show positive benefits from these schemes. These include a decrease in the spread deficit of surface temperature and improved probabilistic verification scores. SST perturbations exhibit a stronger impact than SMC perturbations but, when combined, the increased spread from the two schemes is cumulative. A regional ensemble system driven by the global ensemble members largely reflects the same changes seen in the global ensemble but cycling fields, like SMC, between successive regional forecasts does show some benefit.

Inclusion of sea‐surface temperature variation in the U.S. Navy ensemble‐transform global ensemble prediction system

“…Meanwhile, the Met Office of the United Kingdom has introduced random, spatially correlated SST analysis perturbations within the local ensemble transform Kalman filter (ETKF) of its operational Met Office Global and Regional Ensemble Prediction System (MOGREPS). These random SST analysis perturbations provide a beneficial increase in the ensemble spread of surface temperature and wind [ Tennant and Beare , 2011; Tennant , 2012]. In addition to the efforts at ECMWF and the Met Office, Kunii and Miyoshi [2012]have recently examined SST initial‐state uncertainty in an ensemble Kalman filter (EKF).…”

Operational setup of the soil-perturbed, time-lagged Ensemble Prediction System at the Institute of Meteorology and Water Management – National Research Institute.

“…A larger EPS spread implies more uncertainty in the forecast mean and hence in the ensemble members. The EPS spread is typically underestimated near the surface (Tennant, Beare 2014) where the relations between prognostic parameters are controlled with higher fidelity parameterizations, not accounting enough for observation errors. Thus, in general EPS leads to overconfident forecasts of near-surface weather elements, especially in winter conditions where the natural variability of soil parameters is lower.…”

“…As model(s) resolution increases, the stochastic nature of many weather phenomena becomes more evident. So far, uncertainties of surface conditions have been rarely considered in operational EPS implementations, however several recent surveys show sensitivity of weather forecasts to perturbations of the land surface state within the range of their uncertainty (Sutton et al 2006;Quintanar, Mahmood 2012;Tennant, Beare 2014;Bonanno, Loglisci 2017). Sutton et al demonstrated large sensitivity of the spread of ensemble surface temperature and the timing and location of convective precipitation -in certain warmseason situations -to changes in soil moisture.…”

Evaluation of the Plant–Craig stochastic convection scheme (v2.0) in the ensemble forecasting system MOGREPS-R (24 km) based on the Unified Model (v7.3)

“…While progress is continually being made in improving the accuracy of single forecasts -through improvements in the model formulation as well as increases in grid resolution -a complementary approach is the use of ensembles in order to obtain an estimate of the uncertainty in the forecast (Buizza et al, 2005;Montani et al, 2011;Buizza et al, 2007;Bowler et al, 2008;Thirel et al, 2010;Yang et al, 2012;Zhu, 2005;Abhilash et al, 2013;Roy Bhowmik and Durai, 2008;Clark et al, 2011;Tennant and Beare, 2013). Of course, ensemble forecasting systems themselves remain imperfect, and one of the most important problems is insufficient spread in ensemble forecasts, where the forecast tends to cluster too strongly around rainfall values that turn out to be incorrect.…”

“…The mass-flux variance produced by the PC scheme is inversely proportional to the grid box area being used, and so it is not obvious from the results of Groenemeijer and Craig (2012) whether the stochastic variations of PC will contribute significantly to variability within an ensemble system operating at the scales of MOGREPS-R. Nonetheless, MOGREPS-R has been shown, in common with most ensemble forecasting systems, to produce insufficient spread relative to its forecast error in precipitation (Tennant and Beare, 2013), suggesting that there is scope for the introduction of a stochastic convection parameterization to be able to improve its performance.…”