North American Mesoscale Model Research Articles

Weather Research and Forecasting model simulations of a rare springtime bow echo near the Great Salt Lake, USA

The semiarid climate and rugged terrain in the interior west of the United States do not favour the development of bow echoes, a type of convective storm associated with intense, damaging winds. However, on 21 April 2011, a bow echo associated with a fast-moving midtropospheric perturbation formed across the Great Salt Lake (GSL) in Utah, producing damaging winds along its path. Intrigued by the rarity of this bow echo and the inability of the North American Mesoscale model (NAM) to forecast it, this event was studied by using available observations and conducted simulations with the Advanced Research Weather Research and Forecasting (WRF) model. Sensitivity to the microphysics schemes (MPSs), horizontal grid spacing, intensity of moisture content, and a physical lake model in the WRF model were examined. It was found that: (a) reduction in grid spacing from 12 and 4 km to 1 km along with improved depiction of low-level moisture substantially improved the bow echo simulation, (b) the presence of GSL did not impact bow echo development, and (c) the WRF model appeared to inherit a phase error in the passage of the midtropospheric perturbation from the NAM initial and lateral boundary conditions. The phase error resulted in a 1–2 h delay in the bow echo passage. These results highlight the difficulties in simulating such a bow echo event, and suggest similar challenges future faced by subsequent regional climate downscaling studies on future extreme weather in the western United States.

Forecast value considering energy pricing in California

In this study, production forecast value is investigated using day-ahead market (DAM) and real-time market (RTM) locational marginal prices (LMP) at 63 sites in California. Using the North American Mesoscale (NAM) Model, day-ahead global horizontal irradiance (GHI) forecasts are established and converted to power assuming that a 1MW solar photovoltaic plant is co-located at each observation site. Using this forecast, energy is hypothetically sold in the DAM. As the RTM occurs, deviations between forecast and observation are settled by hypothetically purchasing or selling energy at the RTM price. Total revenue is calculated by the sum of these two transactions.Comparison of NAM forecast revenue to perfect day-ahead forecast revenue shows that perfect forecast revenue is always greater. However, yearly NAM forecast revenue is as much as 98% of the perfect forecast revenue for some sites. After a bias-correction is applied to NAM forecasts, NAM forecast revenue decreases. This demonstrates that based on the observed DAM-RTM price spread, biased forecasts can have a higher forecast value than more accurate forecasts. However, when a deviation penalty is assessed, the most accurate forecasts always yield the highest total revenue.

A numerical study of a persistent cold air pool episode in the Salt Lake Valley, Utah

The Weather Research and Forecasting (WRF) model is used to simulate a persistent cold air pool (CAP) episode observed in December 2010 during the Persistent Cold Air Pool Study (PCAPS) field campaign in Salt Lake Valley, Utah. The availability of intensive observations from PCAPS, especially the added vertical profiles, allows for an in-depth analysis of CAP structure and evolution and model evaluation beyond what was done in similar work before. Comparisons of the WRF simulation with surface observations and upper air soundings indicate that WRF is capable of simulating the observed spatial and temporal variation of the valley atmosphere during the CAP episode. The model also successfully captures the formation and removal of the persistent temperature inversion. Process analysis helps quantify the contribution of various physical processes to the buildup and breakup of the inversion. The inversion developed due primarily to a rapid warming above the valley by synoptic subsidence and warm advection and to strong radiative cooling within the valley under clear sky conditions. The inversion was lifted when strong southerly winds entered the valley from the gap mixing out the cold air in the valley. Synoptic-scale cold advection above the valley contributed to the inversion removal by weakening the inversion from aloft. Sensitivity experiments suggested that the WRF simulation of the CAP episode is more sensitive to large-scale initialization fields with North American Mesoscale Model outperforming North American Regional Reanalysis and National Centers for Environmental Prediction (NCEP) Eta analysis than to the choice of boundary layer parameterizations and land surface models.

A New Scheme for Daily Peak Wind Gust Prediction Using Machine Learning

A major challenge in meteorology is the forecasting of winds owing to their highly chaotic nature. However, wind forecasts, and in particular daily peak wind gust forecasts, provide the public with a general sense of the risks associated with wind on a given day and are useful in decision making. Additionally, such knowledge is critical for wind energy production. Currently, no operational daily peak wind gust product exists. As such, this project will seek to develop a peak wind gust prediction scheme based on output from an operational numerical weather prediction model. Output from the North American Mesoscale (NAM) model will be used in a support vector regression (SVR) algorithm trained to predict daily peak wind gusts for ten cities commonly impacted by hazardous wind gusts (cities in the Midwest and central Plains) and with interests in wind energy. Output from a kernel principal component analysis of the fully three-dimensional atmosphere as characterized by the NAM forecasts will be used to predict peak wind gusts for each location at 36hours lead time. Ultimately, this initial product will lead to the development of a more robust prediction scheme that could one day transition into an operational forecast model.

Analyzing predictability and communicating uncertainty: Lessons from the post-Groundhog Day 2009 storm and the March 2009 “megastorm”

Forecasting winter storms in the northeastern United States during the 2008–2009 season was very challenging owing to large uncertainty in the numerical weather prediction guidance prior to each storm. Forecasts for the February 2009 post-Groundhog Day event and the March 2009 megastorm featured significant spatial and timing errors in storm track, precipitation type, and areal extent. Each storm’s impacts were communicated with considerable certainty, leading to confusion and misunderstanding of the actual uncertainty in each event. Both cases can serve as instructional examples for the forecast community to improve interpretation of levels of uncertainty, along with communication of uncertainty, during potentially high-impact events. Examples of spatial and temporal uncertainties associated with both storms are presented. These uncertainties are illustrated using ensemble data from the National Centers for Environmental Prediction Global Ensemble Forecast System, Short Range Ensemble Forecast, and the higher-resolution deterministic models such as the North American Mesoscale model, Global Forecast System, and European Center for Medium-Range Weather Forecasts. In addition to standard ensemble output, forecast anomalies are presented because highly anomalous situations frequently have large societal impacts. Techniques for analyzing ensemble probabilities for quantitative precipitation forecast (QPF) threshold values and ensemble plume QPF diagrams are demonstrated. In addition, various combinations of deterministic and ensemble means and spreads for mean sea level pressure and 500-hPa height will be presented to evaluate the predictability of surface low-pressure tracks. Several experimental techniques are presented to promote better understanding of predictability and better communication of uncertainty to users.

Verification of Convection-Allowing WRF Model Forecasts of the Planetary Boundary Layer Using Sounding Observations

Abstract This study evaluates forecasts of thermodynamic variables from five convection-allowing configurations of the Weather Research and Forecasting Model (WRF) with the Advanced Research core (WRF-ARW). The forecasts vary only in their planetary boundary layer (PBL) scheme, including three “local” schemes [Mellor–Yamada–Janjić (MYJ), quasi-normal scale elimination (QNSE), and Mellor–Yamada–Nakanishi–Niino (MYNN)] and two schemes that include “nonlocal” mixing [the asymmetric cloud model version 2 (ACM2) and the Yonei University (YSU) scheme]. The forecasts are compared to springtime radiosonde observations upstream from deep convection to gain a better understanding of the thermodynamic characteristics of these PBL schemes in this regime. The morning PBLs are all too cool and dry despite having little bias in PBL depth (except for YSU). In the evening, the local schemes produce shallower PBLs that are often too shallow and too moist compared to nonlocal schemes. However, MYNN is nearly unbiased in PBL depth, moisture, and potential temperature, which is comparable to the background North American Mesoscale model (NAM) forecasts. This result gives confidence in the use of the MYNN scheme in convection-allowing configurations of WRF-ARW to alleviate the typical cool, moist bias of the MYJ scheme in convective boundary layers upstream from convection. The morning cool and dry biases lead to an underprediction of mixed-layer CAPE (MLCAPE) and an overprediction of mixed-layer convective inhibition (MLCIN) at that time in all schemes. MLCAPE and MLCIN forecasts improve in the evening, with MYJ, QNSE, and MYNN having small mean errors, but ACM2 and YSU having a somewhat low bias. Strong observed capping inversions tend to be associated with an underprediction of MLCIN in the evening, as the model profiles are too smooth. MLCAPE tends to be overpredicted (underpredicted) by MYJ and QNSE (MYNN, ACM2, and YSU) when the observed MLCAPE is relatively small (large).

Use of Multiple Verification Methods to Evaluate Forecasts of Convection from Hot- and Cold-Start Convection-Allowing Models

Abstract This study uses both traditional and newer verification methods to evaluate two 4-km grid-spacing Weather Research and Forecasting Model (WRF) forecasts: a “cold start” forecast that uses the 12-km North American Mesoscale Model (NAM) analysis and forecast cycle to derive the initial and boundary conditions (C0) and a “hot start” forecast that adds radar data into the initial conditions using a three-dimensional variational data assimilation (3DVAR)/cloud analysis technique (CN). These forecasts were evaluated as part of 2009 and 2010 NOAA Hazardous Weather Test Bed (HWT) Spring Forecasting Experiments. The Spring Forecasting Experiment participants noted that the skill of CN’s explicit forecasts of convection estimated by some traditional objective metrics often seemed large compared to the subjectively determined skill. The Gilbert skill score (GSS) reveals CN scores higher than C0 at lower thresholds likely due to CN having higher-frequency biases than C0, but the difference is negligible at higher thresholds, where CN’s and C0’s frequency biases are similar. This suggests that if traditional skill scores are used to quantify convective forecasts, then higher (>35 dBZ) reflectivity thresholds should be used to be consistent with expert’s subjective assessments of the lack of forecast skill for individual convective cells. The spatial verification methods show that both CN and C0 generally have little to no skill at scales <8–12Δx starting at forecast hour 1, but CN has more skill at larger spatial scales (40–320 km) than C0 for the majority of the forecasting period. This indicates that the hot start provides little to no benefit for forecasts of convective cells, but that it has some benefit for larger mesoscale precipitation systems.

Stratosphere‐to‐troposphere transport revealed by ground‐based lidar and ozonesonde at a midlatitude site

This paper presents ozone structures measured by a ground‐based ozone lidar and ozonesonde at Huntsville, Alabama, on 27–29 April 2010 originating from a stratosphere‐to‐troposphere transport event associated with a cutoff cyclone and tropopause fold. In this case, the tropopause reached 6 km and the stratospheric intrusion resulted in a 2‐km thick elevated ozone layer with values between 70 and 85 ppbv descending from the ∼306‐K to 298‐K isentropic surface at a rate of ∼5 km day−1. The potential temperature was provided by a collocated microwave profiling radiometer. We examine the corresponding meteorological fields and potential vorticity (PV) structures derived from the analysis data from the North American Mesoscale model. The 2‐PVU (PV unit) surface, defined as the dynamic tropopause, is able to capture the variations of the ozone tropopause estimated from the ozonesonde and lidar measurements. The estimated ozone/PV ratio, from the measured ozone and model derived PV, for the mixing layer between the troposphere and stratosphere is ∼41 ppbv/PVU with an uncertainty of ∼33%. Within two days, the estimated mass of ozone irreversibly transported from the stratospheric into the troposphere is between 0.07 Tg (0.9 × 1033 molecules) and 0.11 Tg (1.3 × 1033molecules) with an estimated uncertainty of 59%. Tropospheric ozone exhibited enormous variability due to the complicated mixing processes. Low ozone and large variability were observed in the mid‐troposphere after the stratospheric intrusion due to the westerly advection including the transition from a cyclonic system to an anticyclonic system. This study using high temporal and vertical‐resolution measurements suggests that, in this case, stratospheric air quickly lost its stratospheric characteristics once it is irreversibly mixed down into the troposphere.

Improving Numerical Weather Predictions of Summertime Precipitation over the Southeastern United States through a High-Resolution Initialization of the Surface State

Abstract It is hypothesized that high-resolution, accurate representations of surface properties such as soil moisture and sea surface temperature are necessary to improve simulations of summertime pulse-type convective precipitation in high-resolution models. This paper presents model verification results of a case study period from June to August 2008 over the southeastern United States using the Weather Research and Forecasting numerical weather prediction model. Experimental simulations initialized with high-resolution land surface fields from the National Aeronautics and Space Administration’s (NASA) Land Information System (LIS) and sea surface temperatures (SSTs) derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) are compared to a set of control simulations initialized with interpolated fields from the National Centers for Environmental Prediction’s (NCEP) 12-km North American Mesoscale model. The LIS land surface and MODIS SSTs provide a more detailed surface initialization at a resolution comparable to the 4-km model grid spacing. Soil moisture from the LIS spinup run is shown to respond better to the extreme rainfall of Tropical Storm Fay in August 2008 over the Florida peninsula. The LIS has slightly lower errors and higher anomaly correlations in the top soil layer but exhibits a stronger dry bias in the root zone. The model sensitivity to the alternative surface initial conditions is examined for a sample case, showing that the LIS–MODIS data substantially impact surface and boundary layer properties. The Developmental Testbed Center’s Meteorological Evaluation Tools package is employed to produce verification statistics, including traditional gridded precipitation verification and output statistics from the Method for Object-Based Diagnostic Evaluation (MODE) tool. The LIS–MODIS initialization is found to produce small improvements in the skill scores of 1-h accumulated precipitation during the forecast hours of the peak diurnal convective cycle. Because there is very little union in time and space between the forecast and observed precipitation systems, results from the MODE object verification are examined to relax the stringency of traditional gridpoint precipitation verification. The MODE results indicate that the LIS–MODIS-initialized model runs increase the 10 mm h−1 matched object areas (“hits”) while simultaneously decreasing the unmatched object areas (“misses” plus “false alarms”) during most of the peak convective forecast hours, with statistically significant improvements of up to 5%. Simulated 1-h precipitation objects in the LIS–MODIS runs more closely resemble the observed objects, particularly at higher accumulation thresholds. Despite the small improvements, however, the overall low verification scores indicate that much uncertainty still exists in simulating the processes responsible for airmass-type convective precipitation systems in convection-allowing models.

Assessment of Forecasts during Persistent Valley Cold Pools in the Bonneville Basin by the North American Mesoscale Model

Abstract North American Mesoscale Model (NAM) forecasts of low-level temperature and dewpoint during persistent valley cold pools in the Bonneville Basin of Utah are assessed. Stations near the east sidewall have a daytime cold and nighttime warm bias. This is due to a poor representation of the steep slopes on this side of the basin. Basin stations where the terrain is better represented by the model have a distinct warm, moist bias at night. Stations in snow-covered areas have a cold bias for both day and night. Biases are not dependent on forecast lead or validation time. Several potential causes for the various errors are considered in a series of sensitivity experiments. An experiment with 4-km grid spacing, which better resolves the gradient of the slopes on the east side of the basin, yields smaller errors along the east corridor of the basin. The NAM assumes all soil water freezes at a temperature of 273 K. This is likely not representative of the freezing temperature in the salt flats in the western part of the basin, since salt reduces the freezing point of water. An experiment testing this hypothesis shows that reducing the freezing point of soil water in the salt flats leads to an average error reduction between 1.5 and 4 K, depending on the station and time of day. Using a planetary boundary layer scheme that has greater mixing alleviates the cold bias over snow somewhat, but the exact source of this bias could not be determined.

A Retrospective Evaluation of the Storm Surge Produced by Hurricane Gustav (2008): Forecast and Hindcast Results

Abstract The evolution and convergence of modeled storm surge were examined using a high-resolution implementation of the Advanced Circulation Coastal Ocean and Storm Surge (ADCIRC) model for Hurricane Gustav (2008). The storm surge forecasts were forced using an asymmetric gradient wind model (AWM), directly coupled to ADCIRC at every time step and at every grid node. A total of 20 forecast advisories and best-track data from the National Hurricane Center (NHC) were used as input parameters into the wind model. Differences in maximum surge elevations were evaluated for ensembles comprised of the final 20, 15, 10, and 5 forecast advisories plus the best track. For this particular storm, the final 10–12 forecast advisories, encompassing the last 2.5–3 days of the storm’s lifetime, give a reasonable estimate of the final storm surge and inundation. The results provide a detailed perspective of the variability in the storm surge due to variability in the meteorological forecast and how this changes as the storm approaches landfall. This finding is closely tied to the consistency and accuracy of the NHC storm track forecasts and the predicted landfall location and, therefore, cannot be generalized to all storms in all locations. Nevertheless, this first attempt to translate variability in forecast meteorology into storm surge variability provides useful insights for guiding the potential use of storm surge models for forecast purposes. Model skill was also evaluated for Hurricane Gustav by comparing observed water levels with hindcast modeled water levels forced by river flow, tides, and several sources of wind data. The AWM (which ingested best-track information from NHC) generated winds that were slightly higher than those from NOAA’s Hurricane Research Division (HRD) H*Wind analyses and substantially greater than the North American Mesoscale (NAM) model. Surge obtained using the AWM more closely matched the observed water levels than that computed using H*Wind; however, this may be due to the neglect of the contribution of wave setup to the surge, especially in exposed areas. Several geographically distinct storm surge response regimes, some characterized by multisurge pulses, were identified and described.

Neighborhood-Based Verification of Precipitation Forecasts from Convection-Allowing NCAR WRF Model Simulations and the Operational NAM

Abstract Since 2003 the National Center for Atmospheric Research (NCAR) has been running various experimental convection-allowing configurations of the Weather Research and Forecasting Model (WRF) for domains covering a large portion of the central United States during the warm season (April–July). In this study, the skill of 3-hourly accumulated precipitation forecasts from a large sample of these convection-allowing simulations conducted during 2004–05 and 2007–08 is compared to that from operational North American Mesoscale (NAM) model forecasts using a neighborhood-based equitable threat score (ETS). Separate analyses were conducted for simulations run before and after the implementation in 2007 of positive-definite (PD) moisture transport for the NCAR-WRF simulations. The neighborhood-based ETS (denoted 〈ETS〉r) relaxes the criteria for “hits” (i.e., correct forecasts) by considering grid points within a specified radius r. It is shown that 〈ETS〉r is more useful than the traditional ETS because 〈ETS〉r can be used to diagnose differences in precipitation forecast skill between different models as a function of spatial scale, whereas the traditional ETS only considers the spatial scale of the verification grid. It was found that differences in 〈ETS〉r between NCAR-WRF and NAM generally increased with increasing r, with NCAR-WRF having higher scores. Examining time series of 〈ETS〉r for r = 100 and r = 0 km (which simply reduces to the “traditional” ETS), statistically significant differences between NCAR-WRF and NAM were found at many forecast lead times for 〈ETS〉100 but only a few times for 〈ETS〉0. Larger and more statistically significant differences occurred with the 2007–08 cases relative to the 2004–05 cases. Because of differences in model configurations and dominant large-scale weather regimes, a more controlled experiment would have been needed to diagnose the reason for the larger differences that occurred with the 2007–08 cases. Finally, a compositing technique was used to diagnose the differences in the spatial distribution of the forecasts. This technique implied westward displacement errors for NAM model forecasts in both sets of cases and in NCAR-WRF model forecasts for the 2007–08 cases. Generally, the results are encouraging because they imply that advantages in convection-allowing relative to convection-parameterizing simulations noted in recent studies are reflected in an objective neighborhood-based metric.

An empirically derived emission algorithm for wind‐blown dust

A wind‐blown dust emission algorithm was developed by matching the frequency of high–aerosol optical depth (AOD) events derived from the MODIS Deep Blue algorithm with the frequency of friction velocities derived from National Centers for Environmental Prediction's North American Mesoscale model. The threshold friction velocity is defined as the velocity that has the same frequency of as the 0.75 AOD. The AODs are converted to an emission flux that is used to compute the linear regression slope of the flux to the friction velocity. The slope represents the potential of a particular land surface to produce airborne dust and, in combination with the friction velocity, is used as a predictor for wind‐blown dust emissions. Calculations for a test period of June and July 2007 showed the model prediction to capture the major measured plume events in timing and magnitude, although peak events tended to be overpredicted and many of the near‐background level ambient concentrations were underpredicted. Most of the airborne dust loadings are attributed to locations with relatively low threshold friction velocities (<45 cm s−1), although these locations only composed of 9% of the total number of source locations. There was some evidence that the duration of wind‐blown dust plume events was comparable to the 3 day sampling frequency of the IMPROVE monitoring network. Higher temporal frequency AIRNow observations at Phoenix showed a surprisingly good fit with the magnitude of the model‐predicted peak concentrations.

NAM Model Forecasts of Warm-Season Quasi-Stationary Frontal Environments in the Central United States

Abstract Using a composite procedure, North American Mesoscale Model (NAM) forecast and observed environments associated with zonally oriented, quasi-stationary surface fronts for 64 cases during July–August 2006–08 were examined for a large region encompassing the central United States. NAM adequately simulated the general synoptic features associated with the frontal environments (e.g., patterns in the low-level wind fields) as well as the positions of the fronts. However, kinematic fields important to frontogenesis such as horizontal deformation and convergence were overpredicted. Surface-based convective available potential energy (CAPE) and precipitable water were also overpredicted, which was likely related to the overprediction of the kinematic fields through convergence of water vapor flux. In addition, a spurious coherence between forecast deformation and precipitation was found using spatial correlation coefficients. Composite precipitation forecasts featured a broad area of rainfall stretched parallel to the composite front, whereas the composite observed precipitation covered a smaller area and had a WNW–ESE orientation relative to the front, consistent with mesoscale convective systems (MCSs) propagating at a slight right angle relative to the thermal gradient. Thus, deficiencies in the NAM precipitation forecasts may at least partially result from the inability to depict MCSs properly. It was observed that errors in the precipitation forecasts appeared to lag those of the kinematic fields, and so it seems likely that deficiencies in the precipitation forecasts are related to the overprediction of the kinematic fields such as deformation. However, no attempts were made to establish whether the overpredicted kinematic fields actually contributed to the errors in the precipitation forecasts or whether the overpredicted kinematic fields were simply an artifact of the precipitation errors. Regardless of the relationship between such errors, recognition of typical warm-season environments associated with these errors should be useful to operational forecasters.

Evaluation of WRF Model Output for Severe Weather Forecasting from the 2008 NOAA Hazardous Weather Testbed Spring Experiment

Abstract This study assesses forecasts of the preconvective and near-storm environments from the convection-allowing models run for the 2008 National Oceanic and Atmospheric Administration (NOAA) Hazardous Weather Testbed (HWT) spring experiment. Evaluating the performance of convection-allowing models (CAMs) is important for encouraging their appropriate use and development for both research and operations. Systematic errors in the CAM forecasts included a cold bias in mean 2-m and 850-hPa temperatures over most of the United States and smaller than observed vertical wind shear and 850-hPa moisture over the high plains. The placement of airmass boundaries was similar in forecasts from the CAMs and the operational North American Mesoscale (NAM) model that provided the initial and boundary conditions. This correspondence contributed to similar characteristics for spatial and temporal mean error patterns. However, substantial errors were found in the CAM forecasts away from airmass boundaries. The result is that the deterministic CAMs do not predict the environment as well as the NAM. It is suggested that parameterized processes used at convection-allowing grid lengths, particularly in the boundary layer, may be contributing to these errors. It is also shown that mean forecasts from an ensemble of CAMs were substantially more accurate than forecasts from deterministic CAMs. If the improvement seen in the CAM forecasts when going from a deterministic framework to an ensemble framework is comparable to improvements in mesoscale model forecasts when going from a deterministic to an ensemble framework, then an ensemble of mesoscale model forecasts could predict the environment even better than an ensemble of CAMs. Therefore, it is suggested that the combination of mesoscale (convection parameterizing) and CAM configurations is an appropriate avenue to explore for optimizing the use of limited computer resources for severe weather forecasting applications.

Observations of Gulf of Tehuantepec Gap Wind Events from QuikSCAT: An Updated Event Climatology and Operational Model Evaluation

Abstract A climatology of gale- and storm-force gap wind events in the Gulf of Tehuantepec is constructed for the first time using 10 yr of ocean surface vector wind data from the SeaWinds scatterometer on board NASA’s Quick Scatterometer (QuikSCAT) satellite. These wind events are among the most severe that occur within the National Hurricane Center’s (NHC) area of marine forecasting responsibility outside of tropical cyclones. The 10-yr climatology indicates that on average 11.9 gale-force events and 6.4 storm-force events occur in the Gulf of Tehuantepec each cold season. About 84% of these events occur between November and March, with the largest number of gale-force events occurring in December. Storm-force events are most frequent in January. Operational numerical weather prediction model forecasts of these events from the NCEP Global Forecast System (GFS) and North American Mesoscale (NAM) models were evaluated during the 2006/07 cold season. Results show that neither model is able to consistently forecast storm-force Tehuantepec wind events; however, the models do have some ability to forecast gale-force events. The NAM model showed a significant increase in probability of detection over the GFS, possibly due to increased horizontal and vertical resolutions as well as differences in boundary layer mixing and surface flux schemes. Finally, the prospects of observing these gap wind events in the post-QuikSCAT era will be discussed.

Comparison of Aura MLS Water Vapor Measurements with GFS and NAM Analyses in the Upper Troposphere–Lower Stratosphere

Abstract Water vapor mixing ratios in the upper troposphere and lower stratosphere measured by the Aura Microwave Limb Sounder (MLS) version 2.2 instrument have been compared with Global Forecast System (GFS) analyses at five levels within the 300–100-hPa layer and North American Mesoscale (NAM) model analyses at six levels within the 300–50-hPa layer over the two years of 2005 and 2006 at four analysis times (e.g., 0000, 0600, 1200, and 1800 UTC). Probability density functions of the vapor mixing ratios suggest that both analyses are often moister than Aura MLS values, but NAM model analyses agree somewhat better with Aura MLS measurements than GFS model analyses over the same North American domain at the five common levels. Examining five subsets of the global GFS domain, the GFS model analysis is moister than Aura MLS estimates everywhere but at 150 and 100 hPa in all regions outside of the tropics. NAM model analysis water vapor mixing ratios exceeded the Aura MLS values at all levels from 250 to 150 hPa in all four seasons of both years and some seasons at 100 and 50 hPa. Moist biases in winter and spring of both years were similar at all levels, but these moist biases in summer and fall were smaller in 2005 than in 2006 at all levels. These differences may be due to the change in the NAM from using the Eta Model to using the Weather Research and Forecasting model (WRF) in June 2006.

Evaluations of NAM Forecasts on Midtropospheric Perturbation-Induced Convective Storms over the U.S. Northern Plains

Abstract In the U.S. northern plains, summer progressive convective storms that occur in weakly forced environments are often coupled with short-wave perturbations that are embedded in the midlevel northwesterly flow. These midtropospheric perturbations (MPs) are capable of inducing propagating convection that contributes to a majority of the rainfall over the northern plains during July and August. There is a possibility that the difficulties of numerical weather prediction models in forecasting summer convective rainfall over the northern plains are partly attributed to their deficiency in forecasting MPs. The present study tests this possibility through examining operational forecasts by the North American Mesoscale (NAM) model during the summers of 2005 and 2006. Forecasted MPs exhibit slower propagation speeds and weaker relative vorticity than the observations leading to systematic position errors. Underpredicted vorticity magnitudes weaken horizontal vorticity advection that influences the vorticity tendency throughout the MP life cycle and, in turn, slows the propagation speed of MPs. Moreover, biases of weak ambient flow speed and vortex stretching contribute to the magnitude and propagation speed errors of MPs. Skill scores of precipitation forecasts associated with MPs are low, but can be considerably improved after removing the MP position error that displaces the rainfall pattern. The NAM also tends to underpredict precipitation amounts. A modified water vapor budget analysis reveals that the NAM insufficiently generates atmospheric humidity over the central United States. The shortage of moisture in the forecast reduces the water vapor flux convergence that is part of the precipitation process. The precipitation bias may feed back to affect the MP growth through the bias in heating, thus further slowing the perturbation.

Verification of Extratropical Cyclones within the NCEP Operational Models. Part II: The Short-Range Ensemble Forecast System

Abstract This paper verifies the strengths and positions of extratropical cyclones around North America and the adjacent oceans within the Short Range Ensemble Forecast (SREF) system at the National Centers for Environmental Prediction (NCEP) during the 2004–07 cool seasons (October–March). The SREF mean for cyclone position and central pressure has a smaller error than the various subgroups within SREF and the operational North American Mesoscale (NAM) model in many regions on average, but not the operational Global Forecast System (GFS) for many forecast times. Inclusion of six additional Weather Research and Forecasting (WRF) model members into SREF during the 2006–07 cool season did not improve the SREF mean predictions. The SREF has slightly more probabilistic skill over the eastern United States and western Atlantic than the western portions of the domain for cyclone central pressure. The SREF also has slightly greater probabilistic skill than the combined GFS and NAM for central pressure, which is significant at the 90% level for many regions and thresholds. The SREF probabilities are fairly reliable, although the SREF is overconfident at higher probabilities in all regions. The inclusion of WRF did not improve the SREF probabilistic skill. Over the eastern Pacific, eastern Canada, and western Atlantic, the SREF is overdispersed on average, especially early in the forecast, while across the central and eastern United States the SREF is underdispersed later in the forecast. There are relatively large biases in cyclone central pressure within each SREF subgroup. As a result, the best-member diagrams reveal that the SREF members are not equally accurate for the cyclone central pressure and displacement. Two cases are presented to illustrate examples of SREF developing large errors early in the forecast for cyclones over the eastern United States.

Next-Day Convection-Allowing WRF Model Guidance: A Second Look at 2-km versus 4-km Grid Spacing

Abstract During the 2007 NOAA Hazardous Weather Testbed (HWT) Spring Experiment, the Center for Analysis and Prediction of Storms (CAPS) at the University of Oklahoma produced convection-allowing forecasts from a single deterministic 2-km model and a 10-member 4-km-resolution ensemble. In this study, the 2-km deterministic output was compared with forecasts from the 4-km ensemble control member. Other than the difference in horizontal resolution, the two sets of forecasts featured identical Advanced Research Weather Research and Forecasting model (ARW-WRF) configurations, including vertical resolution, forecast domain, initial and lateral boundary conditions, and physical parameterizations. Therefore, forecast disparities were attributed solely to differences in horizontal grid spacing. This study is a follow-up to similar work that was based on results from the 2005 Spring Experiment. Unlike the 2005 experiment, however, model configurations were more rigorously controlled in the present study, providing a more robust dataset and a cleaner isolation of the dependence on horizontal resolution. Additionally, in this study, the 2- and 4-km outputs were compared with 12-km forecasts from the North American Mesoscale (NAM) model. Model forecasts were analyzed using objective verification of mean hourly precipitation and visual comparison of individual events, primarily during the 21- to 33-h forecast period to examine the utility of the models as next-day guidance. On average, both the 2- and 4-km model forecasts showed substantial improvement over the 12-km NAM. However, although the 2-km forecasts produced more-detailed structures on the smallest resolvable scales, the patterns of convective initiation, evolution, and organization were remarkably similar to the 4-km output. Moreover, on average, metrics such as equitable threat score, frequency bias, and fractions skill score revealed no statistical improvement of the 2-km forecasts compared to the 4-km forecasts. These results, based on the 2007 dataset, corroborate previous findings, suggesting that decreasing horizontal grid spacing from 4 to 2 km provides little added value as next-day guidance for severe convective storm and heavy rain forecasters in the United States.