Upper-air Soundings Research Articles

Atmospheric and surface observations during the Saint John River Experiment on Cold Season Storms (SAJESS)

Abstract. The amount and the phase of cold-season precipitation accumulating in the upper Saint John River (SJR) basin are critical factors in determining spring runoff, ice jams, and flooding. To study the impact of winter and spring storms on the snowpack in the upper SJR basin, the Saint John River Experiment on Cold Season Storms (SAJESS) was conducted during winter–spring 2020–2021. Here, we provide an overview of the SAJESS study area, field campaign, and data collected. The upper SJR basin represents 41 % of the entire SJR watershed and encompasses parts of the US state of Maine and the Canadian provinces of Quebec and New Brunswick. In early December 2020, meteorological instruments were co-located with an Environment and Climate Change Canada station near Edmundston, New Brunswick. This included a separate weather station for measuring standard meteorological variables, an optical disdrometer, and a micro rain radar. This instrumentation was augmented during an intensive observation period that also included upper-air soundings, surface weather observations, a multi-angle snowflake camera, and macrophotography of solid hydrometeors throughout March and April 2021. During the study, the region experienced a lower-than-average snowpack that peaked at ∼ 65 cm, with a total of 287 mm of precipitation (liquid-equivalent) falling between December 2020 and April 2021, a 21 % lower amount of precipitation than the climatological normal. Observers were present for 13 storms during which they conducted 183 h of precipitation observations and took more than 4000 images of hydrometeors. The inclusion of local volunteers and schools provided an additional 1700 measurements of precipitation amounts across the area. The resulting datasets are publicly available from the Federated Research Data Repository at https://doi.org/10.20383/103.0591 (Thompson et al., 2023). We also include a synopsis of the data management plan and a brief assessment of the rewards and challenges of conducting the field campaign and utilizing community volunteers for citizen science.

Spatiotemporal distribution of air pollutants during a heat wave-induced forest fire event in Uttarakhand.

Prevailing dry conditions and rainfall deficit during the spring season in North India led to heat wave conditions which resulted in widespread and intense forest fire events in the Himalayan state of Uttarakhand during April 16-30, 2022. A total of 7589 active fires were detected by VIIRS during the second half of April 2022 compared to 1558 during the first half. The TROPOMI observed total column values of CO and NO2 increased by 4.4% and 11.7%, respectively during April 16-30, 2022 with respect to April 1-15, 2022. A noticeable increase in surface level concentration of trace gases was also observed at Dehradun. In situ measurements of CO, NOx, and O3 during April 16-30, 2022 show an increase of 133, 35, and 6%compared to previous year observations during the same period. Weather Research and Forecasting model with chemistry (WRF-Chem) is utilized to quantitatively estimate the contribution of this event on the distribution of air pollutants over this state. The model results were evaluated against ERA5 reanalysis, upper air soundings, and TROPOMI-retrieved total column density (TCD) of CO, NO2, and O3. Two simulations with (Fire) and without (NoFire) biomassburning emissions input were performed to quantify the contribution of forest fires to the concentration of trace gases and particulates. The CO, NO2, and O3 emitted/produced from forest fire over Uttarakhand during April 2022 contributed approximately 39.95, 35.73, and 9.97% to the surface concentration of respective gas. In the case of aerosols, it was around 71.20, 71.44, and 33.62% for PM2.5, PM10, and BCrespectively. The vertical profile analysis of pollutants revealed that extreme forest fire events can perturb the distribution of air pollutants from the surface up to 450 hPa.

Verification of Rapid Refresh and High-Resolution Rapid Refresh Model Variables in Tornadic Tropical Cyclones

Abstract Tropical cyclone tornadoes (TCTORs) are a hazard to life and property during landfalling tropical cyclones (TCs). The threat is often spread over a wide area within the TC envelope and must be continually evaluated as the TC moves inland and dissipates. To anticipate the risk of TCTORs, forecasters may use high-resolution, rapidly updating model analyses and short-range forecasts such as the Rapid Refresh (RAP) and High-Resolution Rapid Refresh (HRRR), and an ingredients-based approach similar to that used for forecasting continental midlatitude tornadoes. Though RAP and HRRR errors have been identified in typical midlatitude convective environments, this study evaluates the performance of the RAP and the HRRR within the TC envelope, with particular attention given to sounding-derived parameters previously identified as useful for TCTOR forecasting. A sample of 1730 observed upper-air soundings is sourced from 13 TCs that made landfall along the U.S. coastline between 2017 and 2019. The observed soundings are paired with their corresponding model gridpoint soundings from the RAP analysis, RAP 12-h forecast, and HRRR 12-h forecast. Model errors are calculated for both the raw sounding variables of temperature, dewpoint, and wind speed, as well as for the selected sounding-derived parameters. Results show a moist bias that worsens with height across all model runs. There are also statistically significant underpredictions in stability-related parameters such as convective available potential energy (CAPE) and kinematic parameters such as vertical wind shear.

Revisiting Diagrams of Ice Growth Environments

Abstract Ice habit diagrams published prior to 2009—and many since—do not accurately describe in situ observations of ice shapes as a function of temperature and moisture. Laboratory studies and analysis of field observations by Bailey and Hallett in a series of papers in 2002, 2004, and 2009 corrected several errors from earlier studies, but their work has not been widely disseminated. We present a new, simplified diagram based on Bailey and Hallett’s work that focuses on several ice growth forms arising from the underlying surface processes by which mass is added to a crystal: tabular, columnar, branched, side branched, two types of polycrystalline forms, and a multiple growth regime at low ice supersaturations. To aid interpretation for a variety of applications, versions of the ice growth diagram are presented in terms of relative humidity with respect to water as well as the traditional formats of relative humidity with respect to ice and vapor density excess. These diagrams are intended to be understandable and useful in classroom settings at the sophomore undergraduate level and above. The myriad shapes of pristine snow crystals can be described as the result of either a single growth form or a sequence of growth forms. Overlays of data from upper-air soundings on the ice growth diagrams aid interpretation of expected physical properties and processes in conditions of ice growth.

Biases in the thermodynamic structure over the Southern Ocean in ERA5 and their radiative implications

AbstractThe thermodynamic structure of the lower troposphere in the 37 standard levels ERA5 reanalysis has been evaluated against 2,186 high‐resolution upper air soundings collected over the Southern Ocean (SO). The reanalysis, which incorporated these soundings, was found to be skilled in depicting the general synoptic meteorology and thermodynamic structure as defined by the cluster analysis of Truong et al. (2020) Journal of Geophysical Research: Atmospheres, 125, e2020JD033214. Using dew‐point depression as a proxy for cloud, however, we found a significant reduction in the number of inferred cloud layers, which is inherited from a bias in the specific humidity in the ERA5 reanalysis, most notably over the high latitudes of the SO, where a multilayer cloud structure is frequently observed. The reanalysis was also found to have thinner inferred cloud geometric layer and shallower cloud top heights. Further analysis showed that the reanalysis displays a greater percentage of soundings having no inversion with this bias being more pronounced at high latitudes that tends to be associated with the colder sea surface temperature. While the statistics of the main inversion height are largely consistent, the average inversion strength in the ERA5 reanalysis is found to be weaker than the observations. We anticipate the 137‐level ERA5 reanalysis simulation yields a smoothed vertical structure, from which the 37 standard levels ERA5 reanalysis is linearly interpolated. An examination of the sensitivity of the radiative transfer to cloud macrophysics suggests that the correct representation of thin multiple cloud layers can help reduce the amount of downward shortwave surface radiation over the SO.

Shape of hail and its thermodynamic characteristics related to records in Catalonia

Staring at a hailstone helps to determine various characteristics, such as shape or size. However, from the physics point of view, one of the most important characteristic is the type of growth. Each of the stone’ elements (nucleus, lobes, rings) is conductive to explaining the environmental behaviour of its formation and development stages. The thermodynamic variables derived from the upper-air soundings or the numerical weather prediction simulations can help us understand the differences between the different growth modes. In this way, the hail type has been analyzed according to its shape and size. The study has comprised 151 hailstorms that occurred in Catalonia (NE of the Iberian Peninsula) between 2017 and 2020. We have characterized the events according to different properties. The size of the stones (small, large and, huge), the shape (embryo, spherical, and irregular), and, the season of occurrence (spring, summer, and autumn) are the selected characterizing elements. Maximum vertical updraft velocity, vertical wind shear, and precipitable water content are the variables that explain better the behaviour of each category.

Data Fusion of GEO FY-4A GIIRS and LEO Hyperspectral Infrared Sounders with Surface Observations: A Hong Kong Case Study

Abstract Hyperspectral infrared satellite observations from geostationary platforms allow for the retrieval of temperature and water vapor measurements with higher temporal and vertical resolution than was previously available. The Chinese satellite Fengyun-4A (FY-4A) includes the Geostationary Interferometric Infrared Sounder (GIIRS), which has the ability to measure vertical profiles of temperature and water vapor from space at times when ground-based upper-air soundings are not available and can fill an important need in short-range weather prediction. In this study, convective available potential energy (CAPE) and lifted index (LI), which are used for forecasting atmospheric instability, were computed using the SHARPpy algorithm used by the NWS Storm Prediction Center. However, remote infrared and microwave sensing is lacking detailed information in the boundary layer, so the addition of the NOAA Meteorological Assimilation and Data Ingest System (MADIS) surface data may be necessary in order to get accurate temperature and moisture measurement near the surface. This study uses 10–16 May 2019 in the coastal region near Hong Kong for evaluating the use of hourly surface observations combined with satellite soundings from FY-4A GIIRS at 2-h intervals. The GIIRS plus MADIS surface-based CAPE and LI estimates are compared to estimates derived from low-Earth-orbiting (LEO) SNPP and NOAA-20 from NOAA, MetOp from EUMETSAT, NWP reanalysis, and local radiosondes. In the case study, the 2-h sampling interval of the GIIRS geostationary sounder was able to capture the rapid transition (16 h) from a stable to an unstable atmosphere in both CAPE and LI. The use of surface observations with satellite soundings gave mixed results, possibly due to the complex terrain near Hong Kong.

Examination of Optical Processes in The Atmosphere During Upper Air Soundings

AbstractImproving the quality of weather forecasts and the reliability of climate research requires increasing the reliability of measurements. This paper presents results for optical sensors attached to radiosondes. These sensors can measure cloud-top height (CTH) with high accuracy, determine the presence of precipitation particles in clouds and the height of the boundary between the tropospheric and stratospheric air masses. These research findings are especially important in the Arctic, where the reliability of cloud data is poor, especially during polar nights. With the help of a visible range optical sensor, during the daytime, it is possible to measure CTH with an accuracy of 50 meters. Using data from an IR sensor it is possible to measure CTH both day and night. The paper also discusses the possibility of using optical sensors in an observational network. The results from this study could be useful for both weather forecasting and climate research.

Representation of low-tropospheric temperature inversions in ECMWF reanalyses over Europe

Despite the fact that tropospheric temperature inversions are thought to be an important feature of climate as well as a significant factor affecting air quality, low-level cloud formation, and the radiation budget of the Earth, a quantitative assessment of their representation in atmospheric reanalyses is yet missing. Here, we provide new evidence of the occurrence of low-tropospheric temperature inversions and associated uncertainties in their parameters existing among reanalyses produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) and upper-air soundings for Europe covering the period 2001–2010. The reanalyses utilized here include (1) surface-input reanalyses represented by ERA-20C and CERA-20C as well as (2) full-input reanalyses represented by ERA-Interim and ERA5. The upper-air soundings were derived from the Integrated Global Radiosonde Archive (IGRA), version 2. The data consists mainly of air temperature and geopotential height from the model levels (ModLev) and pressure levels (PresLev) of ECMWF reanalyses. The results show that the frequency of surface-based inversions (SBI) and elevated inversions (EI) is largely in agreement among the reanalyses. The quality of their representation depends, however, on the inversion type, season, and region. Over the vast majority of IGRA upper-air stations, SBI frequency is overestimated and EI frequency is underestimated by ECMWF reanalyses. Substantially larger uncertainties arise from the selection between the data of ModLev and PresLev of the reanalyses—the differences in the frequency of the temperature inversions are particularly large for summertime SBI suggesting that PresLev are not capable of resolving the main features of shallow and weak SBI.

Evidence of a Diurnal Cycle in Precipitation over the Southern Ocean as Observed at Macquarie Island

Due to a lack of observations, relatively large discrepancies exist between precipitation products over the Southern Ocean. In this manuscript, surface hourly precipitation observations from Macquarie Island (54.62 ° S, 158.85 ° E) are analysed (1998–2016) to reveal a diurnal cycle. The precipitation rate is at a maximum during night/early morning and a minimum in the afternoon at Macquarie Island station. Seasonally, the diurnal cycle is strongest in summer and negligible over winter. Such a cycle is consistent with precipitation arising from marine boundary layer clouds, suggesting that such clouds are making a substantial contribution to total precipitation over Macquarie Island and the Southern Ocean. Using twice daily upper air soundings (1995–2011), lower troposphere stability parameters show a stronger inversion at night, again consistent with precipitation arising from marine boundary layer clouds. The ERA-Interim precipitation is dominated by a 12 hourly cycle, year around, which is likely to be a consequence of the twice-daily initialisation. The implication of a diurnal cycle in boundary layer clouds over the Southern Ocean to derived A-Train satellite precipitation products is also discussed.

Assessment of Different WRF Configurations Performance for a Rain Event over Panama

A set of 15 combinations of microphysics and cumulus parameterizations for the WRF numerical model were tested in the forecast of a rain event on January 16 2018 over Panama. GPM satellite, upper air soundings and ground weather stations were used to evaluate the performance of each configuration. Different metrics were used to obtain the configurations that produced the best forecasts. The analysis showed a strong fluctuation of the diurnal precipitation cycle from one region to another. From the set of tests carried, BMJ with Morrison performed slightly above the other combinations in representing well the diurnal cycle for some domain areas, followed by Thom-Gr, WSM6-BMJ and WSM6-KF. While the experiments had a better performance forecasting over sea, they were not able to match the highest values of total precipitation. Regarding spatial rain distribution, all BMJ cumulus combinations showed the best skills mainly over the area comprising the Panama Canal while the others tended to overestimate the rain. Finally, most BMJ experiments showed the highest correlation factor and performed quite well representing the vertical profiles of relative humidity, temperature and wind.

Analysis of a Sea Fog Episode at King George Island, Antarctica

In this study, a marine fog episode at King George Island off the Antarctic Peninsula from 26–30 January 2017 was investigated using surface observations, upper-air soundings, and re-analysis data as well as the air mass backward trajectory method. The marine fog episode resulted from an approaching low-pressure system, was maintained at high wind speeds, and quickly dissipated when the low-pressure system passed the observation site. During this episode, cloud lay existed above the fog and stratus, the atmosphere was stably stratified for 1600 m, and the air close to the surface was more mixed than air in the upper layer. The air-sea temperature difference (ASTD) of 1–2 °C and a strong surface wind parallel to the gradient of SST were two important factors in the formation and maintenance of the marine fog near the Antarctic region. The convergence of flux for both water vapor and heat during the fog episode was also discussed.

A very-high resolution (20m) measurement-based dust emissions and dispersion modeling approach for the Oceano Dunes, California

This study shows the results from very high-resolution (20 m) dust emissions and transport simulations for the Oceano Dunes State Vehicular Recreation Area (ODSVRA), a coastal sand dune complex located in San Luis Obispo County, California. Field data from an enhanced observation period carried out in May–July 2013 helped estimate the emissions and flow conditions over the dune field. Emissions are based on a comprehensive emissions grid developed from in-situ measurements using the Portable In-Situ Wind ERosion Lab (PI-SWERL). PI-SWERL estimates the potential for a soil surface to produce PM10 dust emissions for a range of wind speeds. This approach provided a well-determined PM10 emissions field as a function of time and space. Wind and turbulence fields were estimated using the CALMET diagnostic meteorological model constrained with surface stations, upper air soundings, buoys, and the North American Reanalysis data. Hourly, three-dimensional wind flow and instability objective analysis fields were developed at 20 m resolution in order to consider the complex flow over realistic dune morphology, land use/land cover and terrain characteristics over and around the Oceano Dunes. The dust dispersion simulations were performed using a computationally efficient and vectorized Lagrangian Stochastic Particle Dispersion Model driven by the CALMET output and the PI-SWERL time-space variable emissions. The dispersion model is based on the Langevin formulation and includes the turbulent diffusion and stochastic particle motion (of millions of particles) in the inertial sub-range, and assuming particles as discrete units neglecting deposition. The model estimates diffusion of particles from an initial particle releases that scale according to the PI-SWERL time-variable emissions estimates. Results were then tested at two independent-downwind locations, with positive correlations for flow conditions (R2 = 0.89) and similar receptor PM10 concentrations (R2 = 0.85). Evaluations against those observations during mean flow conditions as well as for elevated dust events suggest that the model framework can capture the spatial and temporal characteristics of mean day-to-day and diurnal PM10 variability. In this study we describe the details of the model framework and its performance as well as its implementation to locate the dust sources that have the strongest impact in the receptor sites and to evaluate the impact of different dust reduction strategies used at the ODSVRA to mitigate PM10 at downwind receptors.

Climatological evaluation in a Central Anatolian city and indirect effects of climatological variation on air quality

Climate is defined as the statistics of weather over a long period of time. Climate has a dynamic structure, and climate change is caused either by natural or anthropogenic effects. Meteorological parameters are routinely recorded by national weather stations. In this study, climate variability was analyzed in a mid-populated city in the middle of the Anatolian Peninsula. Humidity, temperature, precipitation, open surface evaporation, and solar radiation records over 57-year period from 1960 to 2016 were considered. The increase in temperature and solar radiation was obvious. The increase rates were 0.05 °C and 0.62 W/m2 for temperature and solar radiation, respectively. Relative humidity showed a declining trend from 64 to 53%. This study also aimed to evaluate the climate change of planetary boundary layer development, which influences air quality level. AERMET, which is the AERMOD meteorological preprocessor was exploited in order to simulate the planetary boundary layer height, and twice daily upper air soundings, wind speed, wind direction, ambient temperature, and cloud cover have been used as minimum required parameters. The complete meteorological data available were between 2006 and 2016; for that reason, the boundary layer was calculated for 11 years. The determined average boundary layer height was 1018 m in 2006 and was reduced to 889 m in 2016.

CHOCO-JEX: A Research Experiment Focused on the Chocó Low-Level Jet over the Far Eastern Pacific and Western Colombia

AbstractThe ChocoJet Experiment (CHOCO-JEX) is an interinstitutional research program developed by the Universidad Nacional de Colombia, the General Maritime Directorate of the Ministry of National Defense of Colombia, the Colombian Air Force, and the Desert Research Institute. The main goal of CHOCO-JEX is to characterize the vertical structure of the low-level Chocó jet (ChocoJet) through observations and modeling. Thus, four 7-day intensive observation periods (IOPs) took place during different seasons in 2016, two over land and two over the far eastern Pacific off the coast of Colombia, including the deployment of upper-air soundings four times per day to monitor the predominant diurnal cycle and the synoptic and seasonal variability. Preliminary results show deeper westerly moisture flow and a stronger diurnal cycle over land than over ocean. IOP4 provides the first observational evidence of the southwesterly ChocoJet with mean winds of 5 m s–1. Diurnal cycles of zonal wind are coherent with mountain–valley and sea–land breezes at low levels and the easterly flow is predominant at midlevels. Potential temperature anomalies appear to be related to gravity waves that modulate the diurnal cycle of precipitation in the region.

Estimating Monthly Energy Fluxes Using Observations of Near-Surface Air Temperature, Humidity and Radiosonde Profiles

A new boundary-layer method is proposed to estimate sensible and latent heat fluxes on a monthly scale, based on hourly surface observations of temperature and specific humidity, as well as their vertical profile in the morning. It assumes that advective effects may be neglected over such a time scale, so that temporal tendencies of temperature and specific humidity are controlled solely by the vertical flux convergence over the mixed layer. The growth of the mixed layer in the model is driven by the sensible heat flux and opposed by the mean thermal stratification observed by upper-air soundings over that month. The method is tested for three sites, where the flux estimates are compared to eddy-covariance observations. No filtering of days based on synoptic condition, cloudiness or other external conditions is applied, providing realistic monthly estimates. In the three cases, the flux estimates approach closely the eddy-covariance observations, indicating that the method may be applied to estimate the fluxes at places where only weather station data are available.

Kinematic and thermodynamic conditions related to convective systems with a bow echo in Poland

Severe wind events are often related to the occurrence of mesoscale convective systems with arch-shaped radar reflectivity, i.e., a bow echo. In this research, the kinematic and thermodynamic conditions associated with 91 bow echo cases which occurred in the warm season (i.e., from early April until late September) in Poland (2007–2014) were analyzed. The environmental conditions were determined primarily based on the upper air soundings, and additionally on data obtained from ERA-Interim reanalysis. The results indicate that there is a relatively wide range of shear and instability environments associated with bow echoes over Poland. The identified cases occurred both in weakly forced environments, and as well developed in dynamic synoptic patterns with low instability. We have also found cases with strong instability and significantly increased shear values. The combination of a moist boundary layer and steep mid-tropospheric lapse rate usually resulted in moderate to high CAPE values for identified bow echo cases. The median of surface-based CAPE was equal to 1594 J/kg (Mean Layer CAPE = 1038 J/kg) for soundings, and to 1622 J/kg (Mean Layer CAPE = 1275 J/kg) for ERA-Interim. Bow echo environments also showed significantly increased potential for strong downdrafts and damaging outflow winds (the median Downdraft CAPE reached 849 J/kg for soundings and 734 J/kg for ERA-Interim). Bow echoes were usually associated with the occurrence of strong air flow in the troposphere. The presence of a jet stream in the middle and upper troposphere contributed to the development of increased vertical wind shear values. The median of 0–6-km shear exceeded 15 m/s, whereas for 0–3-km shear, it was approximately equal to 12.5 m/s and to 7 m/s for 0–1-km shear.

Environmental Conditions Associated with Observed Snowband Structures within Northeast U.S. Winter Storms

AbstractNortheast U.S. winter storms commonly exhibit multiple meso-β-scale (L < 200 km) bands of enhanced radar reflectivity and precipitation. We use radar observations, upper-air soundings, and reanalyses from 108 cases of cool season (October–April) storms from 1996 to 2016 that occurred within the coastal corridor from Delaware to Maine to identify and assess various banding structures and environments. Banding can occur in several configurations among storms, and banding characteristics can differ at different times within the same storm. We classified 6-h storm periods as containing long (>200 km) single bands, single bands co-occurring with sets of mesoscale multibands, multibands only, and radar echoes without any bands using a combination of automated and manual methods. Use of radar reflectivity data at 0.5-dB precision and a variable rather than a fixed threshold showed that the occurrence of long single bands without any mesoscale multibands was rare, occurring in only 5 of 113 6-h periods. The most frequently occurring band configuration (55%) was concurrent single bands and multibands, which usually were present in the northwest quadrant of mature cyclones. Sets of multibands without a nearby single band usually occurred in the northeast quadrant of a cyclone poleward of weak midlevel forcing along a warm front. Overall, mesoscale single and multibands more commonly occurred after the cyclone occluded than in the developing stages. Multibands occurred in a wide range of frontogenesis and moist potential vorticity environments.

Convective Boundary Layer Depth Estimation from S-Band Dual-Polarization Radar

AbstractThis study investigates Bragg scatter signatures in dual-polarization radar observations, which are defined by low differential reflectivity values, as a proxy for convective boundary layer (CBL) depth. Using data from the WSR-88D in Twin Lakes, Oklahoma (KTLX), local minima in quasi-vertical profiles of are found to provide a reasonable estimate of CBL depth when compared with depth estimates from upper-air soundings from Norman, Oklahoma (KOUN), during 2014. The 243 Bragg scatter and upper-air sounding CBL depth estimates have a correlation of 0.90 and an RMSE of 254 m. Using Bragg scatter as a proxy for CBL depth was expanded to other seasons and locations—performing well in Wilmington, Ohio; Fairbanks, Alaska; Tucson, Arizona; Minneapolis, Minnesota; Albany, New York; Portland, Oregon; and Tampa, Florida—showing its potential usefulness in monitoring CBL depth throughout the year in a variety of geographic locations and meteorological conditions.

Characteristics of Wintertime Daily Precipitation over the Australian Snowy Mountains

Abstract The relationship between orographic precipitation, low-level thermodynamic stability, and the synoptic meteorology is explored for the Snowy Mountains of southeast Australia. A 21-yr dataset (May–October, 1995–2015) of upper-air soundings from an upwind site is used to define synoptic indicators and the low-level stability. A K-means clustering algorithm was employed to classify the daily meteorology into four synoptic classes. The initial classification, based only on six synoptic indicators, distinctly defines both the surface precipitation and the low-level stability by class. Consistent with theory, the wet classes are found to have weak low-level stability, and the dry classes have strong low-level stability. By including low-level stability as an additional input variable to the clustering method, statistically significant correlations were found between the precipitation and the low-level stability within each of the four classes. An examination of the joint PDF reveals a highly nonlinear relationship; heavy rain was associated with very weak low-level stability, and conversely, strong low-level stability was associated with very little precipitation. Building on these historical relationships, model output statistics (MOS) from a moderate resolution (12-km spatial resolution) operational forecast were used to develop stepwise regression models designed to improve the 24-h forecast of precipitation over the Snowy Mountains. A single regression model for all days was found to reduce the RMSE by 7% and the bias by 75%. A class-based regression model was found to reduce the overall RMSE by 30% and the bias by 85%.