Large-eddy simulations of a Salt Lake Valley cold-air pool
Graphical abstract
Introduction
Stable and persistent cold-air pools in the atmospheric boundary-layer are observed in many topographic basins throughout the world (e.g., Western United States, Whiteman et al., 2001, Reeves et al., 2011; United Kingdom, Sheridan et al. 2014; and the Austrian Alps, Dorninger et al. 2011). Utah's Salt Lake Valley is frequented in winter by cold air pools that last between 2 days and a week (Lareau et al. 2013), resulting in poor air quality as pollutants accumulate within the boundary-layer (Silcox et al., 2012, Whiteman et al., 2014, Crosman et al., 2017). The need for improved understanding of the meteorological forcing mechanisms that impact cold-air pools was the goal of the Persistent Cold Air Pool Study (PCAPS, 1 December 2010–7 February 2011, Lareau et al. 2013). The PCAPS research effort has resulted in a number of observational and modelling studies on cold-air pools (Silcox et al., 2012, Hall et al., 2014, Lu and Zhong, 2014, Whiteman et al., 2014, Whiteman and Hoch, 2014, Lareau and Horel, 2015a, Lareau and Horel, 2015b, Young and Whiteman, 2015, Crosman and Horel, 2016, Foster et al., 2017). This study represents the first simulation of a PCAPS cold-air pool using a large-eddy simulation (LES).
A number of recent numerical studies have discussed the inadequacy of numerical weather prediction (NWP) models to simulate aspects of the cold-air pool lifecycle, resulting in poor temperature and air quality forecasts in some situations (Reeves and Stensrud, 2009, Baklanov et al., 2011, Price et al., 2011, Reeves et al., 2011, Ahmadov et al., 2015, Neemann et al., 2015, Lareau and Horel, 2015b). The following meteorological aspects of cold-air pools are particularly difficult to model: vertical temperature and humidity structure, cloudiness, turbulent mixing and boundary-layer flows (Holtslag et al., 2013, Neemann et al., 2015, Lareau and Horel, 2015b, Foster et al., 2017). Consequently, weather and air quality forecasters have difficulty providing timely and accurate guidance regarding cold-air pool metrics such as their intensity, duration and decay.
The planetary boundary layer (PBL) parameterizations used in mesoscale NWP models are typically over-dispersive in the presence of synoptic flow aloft and result in too much vertical mixing and inaccurate moisture and temperature profiles within the stable boundary layer (Teixeira et al., 2008, Fernando and Weil, 2010, Baklanov et al., 2011, Baker et al., 2011, Shin and Hong, 2011, Jiménez et al., 2012, Huang et al., 2013, Reeves et al., 2011, Holtslag et al., 2013). In contrast, during periods of calm winds, these same PBL schemes may be under-dispersive, leading to limited vertical mixing at night (Foster et al. 2017). Improving understanding of cold-air pool turbulent erosion and momentum transport in the presence of variations in stability and other meteorological parameters has been the topic of several recent observational and idealized modelling investigations during PCAPS (Lareau, 2014, Lareau and Horel, 2015a, Lareau and Horel, 2015b). These studies found that complex terrain features such as those found in the Salt Lake Valley interact with large-scale pressure gradients and associated synoptic flows to enhance the downward movement of strong winds toward the surface. The wind shear between the relatively calm low-level cold-air pool and the higher winds at the top of the cold-air pool result in wind shear-driven mixing, or ‘turbulent erosion’ processes that act to remove the cold-air pool from the top down over time (Lareau and Horel, 2015a, Lareau and Horel, 2015b). Modelling studies of several persistent Salt Lake Valley cold-air pools during PCAPS were also conducted by Wei et al. (2013), Lu and Zhong (2014), and Foster et al. (2017), but these simulations were run at typical mesoscale model resolutions (1–3 km Δx) and did not occur during a prolonged period of turbulent erosion on the upper boundary of the cold-air pool such as occurred during the last two days of the cold-air pool simulated in this study.
As discussed by Teixeira et al. (2008), LES can provide improved representation of stable boundary-layer physics. Despite their advantages, however, LES are computationally expensive and can suffer from numerical instabilities (Moeng et al., 2007, Talbot et al., 2012). Advances in computational speed and parallelization have allowed for recent increases in LES studies (e.g., Crosman and Horel, 2012, Mirocha et al., 2013, Joe et al., 2014, Zhang et al., 2014, Matheou and Chung, 2014, Schalkwijk et al., 2015, Falasca et al., 2016). Zhou and Chow (2013) used LES to accurately simulate nighttime flow dynamics during stable conditions during the Terrain-Induced Rotor Experiment in California's Owens Valley while Zhou and Chow (2014) analyzed the formation of shallow nocturnal cold pools and turbulent interactions in shallow topography in Kansas. Hughes et al. (2015) used LES to illustrate the importance of sub-grid scale humidity variations on diurnal cold pools in the United Kingdom. Research is also currently underway to simulate stable slope flows with LES over complex terrain on the Dugway Proving Grounds in Utah (Fernando et al. 2015).
In addition to modelling turbulent erosion processes, the initialization of the land surface and atmospheric state is an important parameter for numerical simulations in northern Utah (Massey et al., 2014, Massey et al., 2016, Neemann et al., 2015, Blaylock et al., 2017, Foster et al., 2017). For example, Neemann et al. (2015) and Foster et al. (2017) found that cold-air pool simulations in Utah's Uintah and Salt Lake Basins were sensitive to initial snow depth and albedo, while Blaylock et al. (2017) found that summertime Great Salt Lake breezes were impacted by lake temperature. Within the Great Salt Lake Basin, both land surface state (e.g., snow cover, soil moisture, land use typology) and Great Salt Lake surface state (lake temperature and salinity) influence the surface energy balance and hence cold-air pool characteristics. In this study, we conduct several LES sensitivity studies to understand the impact on numerical simulations of variations in snow cover and Great Salt Lake surface temperature.
This is the first LES study to our knowledge to simulate the evolution of a long-duration multi-day stable wintertime cold pool in a large mountain valley. The primary goals of this study are to 1) demonstrate the increase in model skill obtained by running the Weather Research and Forecasting (WRF) model in ‘LES mode’ and to 2) quantify the sensitivity of cold-air pools to variations in Great Salt Lake temperature and snow cover. This paper is organized as follows: In Section 2 the WRF model setup, numerical simulations and PCAPS observations are described. In Section 3 a meteorological overview of the 27–30 January 2011 persistent cold-air pool is followed by a comparison between mesoscale and LES simulations. The sensitivity of cold-air pools to variations in Great Salt Lake lake temperature and snow cover are also presented. A summary and discussion of future work are given in Section 4.
Section snippets
WRF model setup
The WRF model is nonhydrostatic and uses a pressure-based, terrain-following vertical coordinate system (Skamarock and Klemp 2008). Table 1 (Fig. 1) summarizes the model set-up and physics options (model domains) used for this study. The WRF model version 3.4.1 was configured as both a large-eddy simulation (hereafter referred to as WRF-LES) and as a mesoscale model (herafter referred to as WRF-MESO). The simulations evaluated the core period of PCAPS intensive observational period 9 (IOP9;
27–30 January 2011 persistent cold-air pool
The persistent cold-air pool strengthened beneath an amplifying 500 hPa shortwave ridge on 27 and 28 January, with 500 hPa heights rising to 575 dm (Fig. 2a). A descending temperature inversion between 2000 and 2500 m above sea level (ASL), a common characteristic of cold-air pools during their intensification stage, was noted during the first half of the episode (Figs. 3a, c and 4a). Beneath the shortwave ridge, 500 hPa wind speeds below 5–10 m s− 1 were observed over the intermountain west (Fig. 2c),
Discussion and future work
In this study we have demonstrated that the WRF model run in a coarse LES mode (ΔX = 250 m) was able to better simulate the evolution and turbulent erosion of a persistent cold-air pool compared to a mesoscale WRF model configured with a standard PBL scheme (ΔX = 1333 m). It is well-known that PBL schemes struggle to correctly represent the turbulent erosion of cold-air pools, although more testing of the various PBL schemes in stable conditions is needed (Banks et al. 2016). For the cold-air pool
Acknowledgements
An allocation of computer time from the Center for High Performance Computing at the University of Utah is gratefully acknowledged. We would like to acknowledge Neil Lareau for the Great Salt Lake time-height data and C. David Whiteman for leading the PCAPS field campaign. This research was supported by the National Science Foundation [grants # ATM-0938397 and ATM-1252315].
References (64)
- et al.
Remote sensing of the surface temperature of the Great Salt Lake
Remote Sens. Environ.
(2009) - et al.
Impacts of anthropogenic emissions and cold air pools on urban to montane gradients of snowpack ion concentrations in the Wasatch
Utah. Atmos. Environ.
(2014) - et al.
Implementation of a high-resolution source-oriented WRF/Chem model at the port of Oakland
Atmos. Environ.
(2014) - et al.
Investigation of model parameters for high-resolution wind energy forecasting: case studies over simple and complex terrain
J. Wind Eng. Ind. Aerodyn.
(2014) - et al.
Wintertime PM2.5 concentrations in Utah's Salt Lake Valley during persistent multi-day cold-air pools
Atmos. Environ.
(2012) - et al.
A time-split nonhydrostatic atmospheric model for weather research and forecasting applications
J. Comput. Phys.
(2008) - et al.
Relationship between particulate air pollution and meteorological variables in Utah's Salt Lake Valley
Atmos. Environ.
(2014) - et al.
Understanding high wintertime ozone pollution events in an oil and natural gas producing region of the western US
Atmos. Chem. Phys.
(2015) - et al.
Orographic influences on a Great Salt Lake-effect snow-storm
Mon. Weather Rev.
(2013) - et al.
Near-term acceleration of hydroclimatic change in the western U.S
J. Geophys. Res. Atmos.
(2013)
Challenges to modelling “cold pool” meteorology associated with high pollution episodes
Environ. Sci. Technol.
The nature, theory, and modeling of atmospheric planetary boundary layers
B. Am. Meteorol. Soc.
Sensitivity of boundary layer variables to PBL schemes in the WRF model based on surface meteorological observations, lidar, and radiosondes during the HygrA-CD campaign
Atmos. Res.
Impact of lake breezes on summer ozone concentrations in the Salt Lake Valley
J. Appl. Meteorol. Climatol.
Coupling an advanced land surface–hydrology model with the Penn State– NCAR MM5 modeling system. Part I: model implementation and sensitivity
Mon. Weather Rev.
Idealized large-eddy simulations of sea and lake breezes: sensitivity to lake diameter, heat flux, and stability
Boundary-Layer Meteorol.
Winter lake breezes near the Great Salt Lake
Boundary-Layer Meteorol.
A novel approach for monitoring vertical profiles of boundary-layer pollutants: utilizing routine news helicopter flights
Atmospheric Pollution Research
A new vertical grid nesting capability in the Weather Research and Forecasting (WRF) model
Mon. Weather Rev.
Meteorological events affecting cold-air pools in a small basin
J. Appl. Meteorol. Climatol.
Numerical study of the daytime planetary boundary layer over an idealized urban area: influence of surface properties, anthropogenic heat flux, and geostrophic wind intensity
J. Appl. Meteorol. Climatol.
The MATERHORN: unraveling the intricacies of mountain weather
Bull. Am. Meteorol. Soc.
Simulations of a cold-air pool in Utah's Salt Lake Valley: sensitivity to land use and snow cover
Boundary-Layer Meteorol.
Improving large-eddy simulation or neutral boundary layer flow across grid interfaces
Mon. Weather Rev.
Techniques for using MODIS data to remotely sense lake water surface temperatures
J. Atmos. Ocean. Technol.
Stable atmospheric boundary layers and diurnal cycles: challenges for weather and climate models
B. Am. Meteorol. Soc.
A new vertical diffusion package with explicit treatment of entrainment processes
Mon. Weather Rev.
Evaluation of the WRF PBL parameterizations for marine boundary layer clouds: cumulus and stratocumulus
Mon. Weather Rev.
Assessment of valley cold pools and clouds in a very high-resolution numerical weather prediction model
Geosci. Model Dev.
Radiative forcing by long-lived greenhouse gases: calculations with the AER radiative transfer models
J. Geophys. Res.
A revised scheme for the WRF surface layer formulation
Mon. Weather Rev.
Cited by (22)
Diode effects on street canyon ventilation in valley city: Temperature inversion and calm geostrophic wind
2023, Building and EnvironmentTurbulence parameterizations for dispersion in sub-kilometer horizontally non-homogeneous flows
2019, Atmospheric ResearchCitation Excerpt :In addition, the snow on the valley floor also caused a discontinuity in the surface fluxes among neighbouring upper and lower valley regions, contributing to the development of the down-valley wind. Indeed, snow cover initialization in high-resolution meteorological simulations over complex terrain turns out to be a very delicate issue (Crosman and Horel, 2017): not even initial and boundary conditions with a 9-km resolution are fully reliable for the correct initialization of this variable. As no snow was actually present over the study area during the day of the release (except for the highest mountains peaks, which are far from the focus of this study), for a second model run the initialization of the snow cover in the WRF model was modified (as similarly done in Tomasi et al., 2017), completely removing the snow in the domain.
Assessing PM<inf>2.5</inf> model performance for the conterminous U.S. with comparison to model performance statistics from 2007-2015
2019, Atmospheric EnvironmentCitation Excerpt :These relatively large negative MBs are driven by underpredictions of nitrate in mountain valleys during ammonium nitrate episodes associated with strong meteorological temperature inversions in winter (e.g., Chen et al., 2012; Chow et al., 2006; Franchin et al., 2018). Modeling stagnant meteorology in complex terrain often requires finer grid resolution than is currently possible in national-scale modeling (e.g., <1–4 km; Crosman and Horel, 2017), and wintertime nitrate episodes have been simulated reasonably well in the western U.S. for higher-resolution CMAQ simulations (e.g., Chen et al., 2014; Kelly et al., 2018). Since model performance can differ for PGM simulations at different grid resolutions (e.g., Zakoura and Pandis, 2018), comparisons of performance statistics among simulations to assess operational model performance are most meaningful when a consistent grid resolution is used.
Characteristics of roll and cellular convection in a deep and wide semiarid valley: A large-eddy simulation study
2019, Atmospheric ResearchCitation Excerpt :For instance, S17 introduced a measure of roll organization ℛ that relies solely on horizontal LES fields of resolved vertical velocity, enabling them to better quantify the transition from cells to rolls. Nested LES over both flat (Talbot et al., 2012) and various complex terrain areas has been conducted, including gentle slopes (Muñoz-Esparza et al., 2017; Rai et al., 2017b), isolated mountains (Michioka and Chow, 2008; Xue et al., 2014) and basins (Crosman and Horel, 2017). Previous attempts at nested LES in valleys exist as well.
WRF-based assessment of the Great Lakes’ impact on cold season synoptic cyclones
2018, Atmospheric ResearchFour dimensional data assimilation (FDDA) impacts on WRF performance in simulating inversion layer structure and distributions of CMAQ-simulated winter ozone concentrations in Uintah Basin
2018, Atmospheric EnvironmentCitation Excerpt :This approach, however, is expensive to apply on the relatively large domain (∼350 × 250 km) of the entire Uintah Basin. Besides, finer resolution itself still poses certain limitations in fully-simulated stable PBL characteristics due to inadequate basic understanding of stable PBL physics (Dabberdt et al., 2005; Ralph et al., 2005; Reen et al., 2006; Kang and Davis, 2008; Crosman and Horel, 2017; Pagès et al., 2017). Another approach to consider for improving model performance is four dimensional data assimilation (FDDA) or “nudging,” especially in areas with adequate observational data to independently nudge and evaluate model performance.