Search

The roughness sublayer (RSL) is one compartment of the surface layer (SL) where turbulence deviates from Monin–Obukhov similarity theory. As the computing power increases, model grid sizes approach the gray zone of turbulence in the energy-containing range and the lowest model layer is located within the RSL. From this perspective, the RSL has an important implication in atmospheric modeling research. However, it has not been explicitly simulated in atmospheric mesoscale models. This study incorporates the RSL model proposed by Harman and Finnigan (2007, 2008) into the Jiménez et al. (2012) SL scheme. A high-resolution simulation performed with the Weather Research and Forecasting model (WRF) illustrates the impacts of the RSL parameterization on the wind, air temperature, and rainfall simulation in the atmospheric boundary layer. As the roughness parameters vary with the atmospheric stability and vegetative phenology in the RSL model, our RSL implementation reproduces the observed surface wind, particularly over tall canopies in the winter season by reducing the root mean square error (RMSE) from 3.1 to 1.8 m s -1. Moreover, the improvement is relevant to air temperature (from 2.74 to 2.67 K of RMSE) and precipitation (from 140 to 135 mm per month of RMSE). Our findings suggest that the RSL must be properly considered both for better weather and climate simulations and for the application of wind energy and atmospheric dispersion. [ABSTRACT FROM AUTHOR]

On 21 September 2010, heavy rainfall with a local maximum of 259 mm d occurred near Seoul, South Korea. We examined the ability of the Weather Research and Forecasting (WRF) model in reproducing this disastrous rainfall event and identified the role of two physical processes: planetary boundary layer (PBL) and microphysics (MPS) processes. The WRF model was forced by 6-hourly National Centers for Environmental Prediction (NCEP) Final analysis (FNL) data for 36 hours form 1200 UTC 20 to 0000 UTC 22 September 2010. Twenty-five experiments were performed, consisting of five different PBL schemes-Yonsei University (YSU), Mellor-Yamada-Janjic (MYJ), Quasi Normal Scale Elimination (QNSE), Bougeault and Lacarrere (BouLac), and University of Washington (UW)-and five different MPS schemes-WRF Single-Moment 6-class (WSM6), Goddard, Thompson, Milbrandt 2-moments, and Morrison 2-moments. As expected, there was a specific combination of MPS and PBL schemes that showed good skill in forecasting the precipitation. However, there was no specific PBL or MPS scheme that outperformed the others in all aspects. The experiments with the UW PBL or Thompson MPS scheme showed a relatively small amount of precipitation. Analyses form the sensitivity experiments confirmed that the spatial distribution of the simulated precipitation was dominated by the PBL processes, whereas the MPS processes determined the amount of rainfall. It was also found that the temporal evolution of the precipitation was influenced more by the PBL processes than by the MPS processes. [ABSTRACT FROM AUTHOR]

Arctic tundra is undergoing a rapid transition due to global warming and will be exposed to snow-free conditions for longer periods under projected climate scenarios. Regional climate modeling is useful for understanding and predicting climate change in the Arctic tundra, however, the lack of in-situ observations of surface energy fluxes and the planetary boundary layer (PBL) structure hinders accurate predictions of local and regional climate around the Arctic. In this study, we investigate the performance of the Polar-optimized version of the Weather Research and Forecasting model (PWRF) in the Arctic tundra on clear days in summer. Based on simultaneous observations of surface fluxes and the PBL structure in Cambridge Bay, Nunavut, Canada, our validation shows that the PWRF simulates a drier environment, leading to a larger Bowen ratio and a warmer atmosphere compared to observations. Further sensitivity analyses indicate that the model biases are mainly from the uncertainties in physical parameters such as surface albedo and emissivity, the solar constant, and the model top height, rather than structural flaws in the model physics. Importantly, the PWRF reproduces the observations more accurately when the observed soil moisture is fed into the simulation. This indicates that there must be improvements in simulations of the land-atmosphere interaction at the Arctic tundra, not only in the accuracy of the initial soil moisture conditions but also in soil hydraulic properties and drainage processes. The mixing diagram analysis also shows that the entrainment process between the PBL and the overlying atmosphere needs to be improved for better weather and climate simulation. Our findings shed light on modeling studies in the Arctic region by disentangling the model error sources from uncertainties by parameters and physics package options. • Arctic tundra is exposed to warmer conditions with a warming global climate. • Polar WRF has bias in simulating climate near the surface on clear summer days. • Polar WRF reproduces the observations accurately with the observed soil moisture. • Regional climate simulation improves with proper numerical design over the Arctic. [ABSTRACT FROM AUTHOR]