Search

As part of a project aimed at exploring the use of next-generation high-performance computing technologies for numerical weather prediction, we have ported two physics modules from the Common Community Physics Package (CCPP) to Graphics Processing Unit (GPU) and obtained accelerations of up to 10× relative to a comparable multi-core CPU. The physics parameterizations accelerated in this work are the aerosol-aware Thompson microphysics (TH) scheme and the Grell–Freitas (GF) cumulus convection scheme. Microphysics schemes are among the most time-consuming physics parameterizations, second to only radiative process schemes, and our results show better acceleration for the TH scheme than the GF scheme. Multi-GPU implementations of the schemes show acceptable weak scaling in a single node with 8 GPUs, and perfect weak scaling on multiple nodes using one GPU per node. The lack of inter-node communication for column physics parameterizations contributes to their scalability, however, physics parameterizations are run along with dynamics, so the overall multi-GPU performance is often governed by the latter. In the context of optimizing CCPP physics modules, our observations underscore that the extensive use of automatic arrays within inner subroutines hampers GPU performance due to serialized memory allocations. We have used the OpenACC directive programming language for this work because it allows for easy porting of large amounts of code and makes code maintenance more manageable compared to low-level languages like CUDA and OpenCL. [ABSTRACT FROM AUTHOR]

Estimates of soil moisture from two National Oceanic and Atmospheric Administration (NOAA) models are compared to in situ observations. The estimates are from a high-resolution atmospheric model with a land surface model [High-Resolution Rapid Refresh (HRRR) model] and a hydrologic model from the NOAA Climate Prediction Center (CPC). Both models produce wetter soils in dry regions and drier soils in wet regions, as compared to the in situ observations. These soil moisture differences occur at most soil depths but are larger at the deeper depths below the surface (100 cm). Comparisons of soil moisture variability are also assessed as a function of soil moisture regime. Both models have lower standard deviations as compared to the in situ observations for all soil moisture regimes. The HRRR model's soil moisture is better correlated with in situ observations for drier soils as compared to wetter soils—a trend that was not present in the CPC model comparisons. In terms of seasonality, soil moisture comparisons vary depending on the metric, time of year, and soil moisture regime. Therefore, consideration of both the seasonality and soil moisture regime is needed to accurately determine model biases. These NOAA soil moisture estimates are used for a variety of forecasting and societal applications, and understanding their differences provides important context for their applications and can lead to model improvements. Significance Statement: Soil moisture is an essential variable coupling the land surface to the atmosphere. Accurate estimates of soil moisture are important for forecasting near-surface temperature and moisture, predicting where clouds will form, and assessing drought and fire risks. There are multiple estimates of soil moisture available, and in this study, we compare soil moisture estimates from two different National Oceanic and Atmospheric Administration (NOAA) models to in situ observations. These comparisons include both soil moisture amount and variability and are conducted at several soil depths, in different soil moisture regimes, and for different seasons and years. This comprehensive assessment allows for an accurate assessment of biases within these models that would be missed when conducting analyses more broadly. [ABSTRACT FROM AUTHOR]

Numerical models of the atmosphere are based on the best theory available. Understandably, the theoretical assessment of errors induced by the use of such models is confounding. Without clear theoretical guidance, the experimental separation of the model-induced part of the total forecast error is also challenging. In this study, the forecast error and ensemble perturbation variances were decomposed. Smaller- and larger-scale components, separated as a function of the lead time, were independent. They were associated with features with completely vs. only partially lost skill, respectively. For their phenomenological description, the larger-scale variance was further decomposed orthogonally into positional and structural components. An analysis of the various components revealed that chaotically amplifying initial perturbation and error predominantly led to positional differences in forecasts, while structural differences were interpreted as an indicator of the model-induced error. Model-induced errors were found to be relatively small. These results confirmed earlier assumptions and limited empirical evidence that numerical models of the atmosphere may be near perfect on the scales they well resolve. [ABSTRACT FROM AUTHOR]

The upcoming photometric surveys, such as the Rubin Observatory's Legacy Survey of Space and Time (LSST) will monitor unprecedented number of active galactic nuclei (AGN) in a decade‐long campaign. Motivated by the science goals of LSST, which includes the harnessing of broadband light curves of AGN for photometric reverberation mapping (PhotoRM), we implement the existing formalism to estimate the lagged response of the emission line flux to the continuum variability using only multi‐band photometric light curves. We test the PhotoRM method on a set of 19 artificial light curves simulated using a stochastic model based on the damped random walk process. These light curves are sampled using different observing strategies, including the two proposed by the LSST, in order to compare the accuracy of time‐lag retrieval based on different observing cadences. In addition, we apply the same procedure for time‐lag retrieval to the observed photometric light curves of NGC 4395, and compare our results to the existing literature. [ABSTRACT FROM AUTHOR]

The consequences of complex disturbed environments in the vicinity of a supermassive black hole are not well represented by standard statistical models of optical variability in active galactic nuclei (AGN). Thus, developing new methodologies for investigating and modeling AGN light curves is crucial. Conditional Neural Processes (CNPs) are nonlinear function models that forecast stochastic time series based on a finite amount of known data without the use of any additional parameters or prior knowledge (kernels). We provide a CNP algorithm that is specifically designed for simulating AGN light curves. It was trained using data from the All‐Sky Automated Survey for Supernovae, which included 153 AGN. We present CNP modeling performance for a subsample of five AGNs with distinctive difficult‐to‐model properties. The performance of CNP in predicting temporal flux fluctuation was assessed using a minimizing loss function, and the results demonstrated the algorithm's usefulness. Our preliminary parallelization experiments show that CNP can efficiently handle large amounts of data. These results imply that CNP can be more effective than standard tools in modeling large volumes of AGN data (as anticipated from time‐domain surveys such as the Vera C. Rubin Observatory's Legacy Survey of Space and Time). [ABSTRACT FROM AUTHOR]

Motivated by upcoming photometric and spectroscopic surveys [Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST), Manuakea Spectroscopic Explorer], we design the statistical proxies to measure the cadence effects on active galactic nuclei (AGNs) variability-observables [time-lags, periodicity, and structure-function (SF)]. We constructed a multiple-regression model to statistically identify the cadence-formal error pattern knowing AGN time-lags and periodicity from different surveys. We defined the simple metric for the SF's properties, accounting for the 'observed' SF's deviation relative to those obtained from the homogenously sampled light curves. We tested the regression models on different observing strategies: the optical data set of long light curves of eight AGN with peculiarities and the artificial data sets based on several idealized and LSST-like cadences. The SFs metric is assessed on synthetic data sets. The regression models (for both data types) predict similar cadences for time-lags and oscillation detection, whereas for light curves with low variability (⁠|${\sim}10{{\ \rm per\ cent}}$|⁠), cadences for oscillation detection differ. For higher variability (⁠|${\sim}20{{\ \rm per\ cent}}$|⁠), predicted cadences are larger than for |$F_{var}\sim 10{{\ \rm per\ cent}}$|⁠. The predicted cadences are decreasing with redshift. SFs with dense and homogenous cadences are more likely to behave similarly. SFs with oscillatory signals are sensitive to the cadences, possibly impacting LSST-like operation strategy. The proposed proxies can help to select spectroscopic and photometric-surveys cadence strategies, and they will be tested further in larger samples of objects. [ABSTRACT FROM AUTHOR]

The High-Resolution Rapid Refresh Ensemble (HRRRE) is a 36-member ensemble analysis system with 9 forecast members that utilizes the Advanced Research version of the Weather Research and Forecasting (ARW-WRF) dynamic core and the physics suite from the operational Rapid Refresh/High-Resolution Rapid Refresh deterministic modeling system. A goal of HRRRE development is a system with sufficient spread among members, comparable in magnitude to the random error in the ensemble mean, to represent the range of possible future atmospheric states.HRRRE member diversity has traditionally been obtained by perturbing the initial and lateral boundary conditions of each member, but recent development has focused on implementing stochastic approaches in HRRRE to generate additional spread. These techniques were tested in retrospective experiments and in the May 2019 Hazardous Weather Testbed Spring Experiment (HWT-SE). Results show a 6%-25% increase in the ensemble spread in 2-m temperature, 2-m mixing ratio, and 10-m wind speed when stochastic parameter perturbations are used in HRRRE (HRRRE-SPP). Case studies from HWT-SE demonstrate that HRRRE-SPP performed similar to or better than the operational High-Resolution Ensemble Forecast system, version 2 (HREFv2), and the nonstochastic HRRRE. However, subjective evaluations provided by HWTSE forecasters indicated that overall, HRRRE-SPP predicted lower probabilities of severe weather (using updraft helicity as a proxy) compared to HREFv2. A statistical analysis of the performance of HRRRE-SPP and HREFv2 from the 2019 summer convective season supports this claim, but also demonstrates that the two systems have similar reliability for prediction of severe weather using updraft helicity. [ABSTRACT FROM AUTHOR]

U.S. National Weather Service (NWS) forecasters assess and communicate hazardous weather risks, including the likelihood of a threat and its impacts. Convection-allowing model (CAM) ensembles offer potential to aid forecasting by depicting atmospheric outcomes, including associated uncertainties, at the refined space and time scales at which hazardous weather often occurs. Little is known, however, about what CAM ensemble information is needed to inform forecasting decisions. To address this knowledge gap, participant observations and semi structured interviews were conducted with NWS forecasters from national centers and local weather forecast offices. Data were collected about forecasters’ roles and their forecasting processes, uses of model guidance and verification information, interpretations of prototype CAM ensemble products, and needs for information from CAM ensembles. Results revealed forecasters’ needs for specific types of CAM ensemble guidance, including a product that combines deterministic and probabilistic output from the ensemble as well as a product that provides map-based guidance about timing of hazardous weather threats. Forecasters also expressed a general need for guidance to help them provide impact-based decision support services. Finally, forecasters conveyed needs for objective model verification information to augment their subjective assessments and for training about using CAM ensemble guidance for operational forecasting. The research was conducted as part of an interdisciplinary research effort that integrated elicitation of forecasters’ CAM ensemble needs with model development efforts, with the aim of illustrating a robust approach for creating information for forecasters that is truly useful and usable. [ABSTRACT FROM AUTHOR]

A stochastically perturbed parameterization (SPP) approach that spatially and temporally perturbs parameters and variables in the Mellor–Yamada–Nakanishi–Niino planetary boundary layer scheme (PBL) and introduces initialization perturbations to soil moisture in the Rapid Update Cycle land surface model was developed within the High-Resolution Rapid Refresh convection-allowing ensemble. This work is a follow-up study to a work performed using the Rapid Refresh (RAP)-based ensemble. In the present study, the SPP approach was used to target the performance of precipitation and low-level variables (e.g., 2-m temperature and dewpoint, and 10-m wind). The stochastic kinetic energy backscatter scheme and the stochastic perturbation of physics tendencies scheme were combined with the SPP approach and applied to the PBL to target upper-level variable performance (e.g., improved skill and reliability). The three stochastic experiments (SPP applied to PBL only, SPP applied to PBL combined with SKEB and SPPT, and stochastically perturbed soil moisture initial conditions) were compared to a mixed-physics ensemble. The results showed a positive impact from initial condition soil moisture perturbations on precipitation forecasts; however, it resulted in an increase in 2-m dewpoint RMSE. The experiment with perturbed parameters within the PBL showed an improvement in low-level wind forecasts for some verification metrics. The experiment that combined the three stochastic approaches together exhibited improved RMSE and spread for upper-level variables. Our study demonstrated that, by using the SPP approach, forecasts of specific variables can be improved. Also, the results showed that using a single-physics suite ensemble with stochastic methods is potentially an attractive alternative to using multiphysics for convection allowing ensembles. [ABSTRACT FROM AUTHOR]

Traditional ensemble probabilities are computed based on the number of members that exceed a threshold at a given point divided by the total number of members. This approach has been employed for many years in coarse-resolution models. However, convection-permitting ensembles of less than ~20 members are generally underdispersive, and spatial displacement at the gridpoint scale is often large. These issues have motivated the development of spatial filtering and neighborhood postprocessing methods, such as fractional coverage and neighborhood maximum value, which address this spatial uncertainty. Two different fractional coverage approaches for the generation of gridpoint probabilities were evaluated. The first method expands the traditional point probability calculation to cover a 100-km radius around a given point. The second method applies the idea that a uniform radius is not appropriate when there is strong agreement between members. In such cases, the traditional fractional coverage approach can reduce the probabilities for these potentially well-handled events. Therefore, a variable radius approach has been developed based upon ensemble agreement scale similarity criteria. In this method, the radius size ranges from 10 km for member forecasts that are in good agreement (e.g., lake-effect snow, orographic precipitation, very short-term forecasts, etc.) to 100 km when the members are more dissimilar. Results from the application of this adaptive technique for the calculation of point probabilities for precipitation forecasts are presented based upon several months of objective verification and subjective feedback from the 2017 Flash Flood and Intense Rainfall Experiment. [ABSTRACT FROM AUTHOR]

A stochastic parameter perturbation (SPP) scheme consisting of spatially and temporally varying perturbations of uncertain parameters in the Grell-Freitas convective scheme and the Mellor-Yamada-Nakanishi- Niino planetary boundary scheme was developed within the Rapid Refresh ensemble system based on the Weather Research and Forecasting Model. Alone the stochastic parameter perturbations generate insufficient spread to be an alternative to the operational configuration that utilizes combinations of multiple parameterization schemes. However, when combined with other stochastic parameterization schemes, such as the stochastic kinetic energy backscatter (SKEB) scheme or the stochastic perturbation of physics tendencies (SPPT) scheme, the stochastic ensemble system has comparable forecast performance. An additional analysis quantifies the added value of combining SPP and SPPT over an ensemble that uses SPPT only, which is generally beneficial, especially for surface variables. The ensemble combining all three stochastic methods consistently produces the best spread-skill ratio and generally outperforms the multiphysics ensemble. The results of this study indicate that using a single-physics suite ensemble together with stochastic methods is an attractive alternative to multiphysics ensembles and should be considered in the design of future high-resolution regional and global ensembles. [ABSTRACT FROM AUTHOR]

The accurate and timely depiction of the state of the atmosphere on multiple scales is critical to enhance forecaster situational awareness and to initialize very short-range numerical forecasts in support of nowcasting activities. The Local Analysis and Prediction System (LAPS) of the Earth System Research Laboratory (ESRL)/Global Systems Division (GSD) is a numerical data assimilation and forecast system designed to serve such very finescale applications. LAPS is used operationally by more than 20 national and international agencies, including the NWS, where it has been operational in the Advanced Weather Interactive Processing System (AWIPS) since 1995. Using computationally efficient and scientifically advanced methods such as a multigrid technique that adds observational information on progressively finer scales in successive iterations, GSD recently introduced a new, variational version of LAPS (vLAPS). Surface and 3D analyses generated by vLAPS were tested in the Hazardous Weather Testbed (HWT) to gauge their utility in both situational awareness and nowcasting applications. On a number of occasions, forecasters found that the vLAPS analyses and ensuing very short-range forecasts provided useful guidance for the development of severe weather events, including tornadic storms, while in some other cases the guidance was less sufficient. [ABSTRACT FROM AUTHOR]

The weather research and forecasting (WRF) model and the maximum likelihood ensemble filter (MLEF) data assimilation approach are used to examine the potential impact of observations from the future Geostationary Operational Environmental Satellite, generation R (GOES-R) on improving our knowledge about clouds. Synthetic radiances are assimilated from the 10.35 μm channel of the GOES-R advanced baseline imager (ABI) employing a ‘non-identical twins’ experimental setup. The experimental results are examined for an extratropical cyclone named Kyrill that produced unusually strong winds, widespread damage and fatalities in Western Europe in January 2007. The data assimilation problem is especially challenging for this case, as there is a large error in the model-simulated radiances resulting from incorrect cloud location. Although this problem is difficult to eliminate, data assimilation results indicate the potential of GOES-R data to significantly reduce these errors. [ABSTRACT FROM AUTHOR]

In the current study, a technique that offers a way to evaluate ensemble forecast uncertainties produced either by initial conditions or different model versions, or both, is presented. The technique consists of first diagnosing the performance of the forecast ensemble and then optimizing the ensemble forecast using results of the diagnosis. The technique is based on the explicit evaluation of probabilities that are associated with the Gaussian stochastic representation of the weather analysis and forecast. It combines an ensemble technique for evaluating the analysis error covariance and the standard Monte Carlo approach for computing samples from a known Gaussian distribution. The technique was demonstrated in a tutorial manner on two relatively simple examples to illustrate the impact of ensemble characteristics including ensemble size, various observation strategies, and configurations including different model versions and varying initial conditions. In addition, the authors assessed improvements in the consensus forecasts gained by optimal weighting of the ensemble members based on time-varying, prior-probabilistic skill measures. The results with different observation configurations indicate that, as observations become denser, there is a need for larger-sized ensembles and/or more accuracy among individual members for the ensemble forecast to exhibit prediction skill. The main conclusions relative to ensembles built up with different physics configurations were, first, that almost all members typically exhibited some skill at some point in the model run, suggesting that all should be retained to acquire the best consensus forecast; and, second, that the normalized probability metric can be used to determine what sets of weights or physics configurations are performing best. A comparison of forecasts derived from a simple ensemble mean to forecasts from a mean developed from variably weighting the ensemble members based on prior performance by the probabilistic measure showed that the latter had substantially reduced mean absolute error. The study also indicates that a weighting scheme that utilized more prior cycles showed additional reduction in forecast error. [ABSTRACT FROM AUTHOR]

To assist in optimizing a mixed-physics ensemble for warm season mesoscale convective system rainfall forecasting, the impact of various physical schemes as well as their interactions on rainfall when different initializations were used has been investigated. For this purpose, high-resolution Weather Research and Forecasting (WRF) model simulations of eight International H2O Project events were performed. For each case, three different treatments of convection, three different microphysical schemes, and two different planetary boundary layer (PBL) schemes were used. All cases were initialized with both Local Analyses and Prediction System (LAPS) “hot” start analyses and 40-km Eta Model analyses. To evaluate the impacts of the variation of two different physical schemes and their interaction on the simulated rainfall under the two different initial conditions, the factor separation method was used. The sensitivity to the use of various physical schemes and their interactions was found to be dependent on the initialization dataset. Runs initialized with Eta analyses appeared to be influenced by the use of the Betts–Miller–Janjić scheme in that model’s assimilation system, which tended to reduce the WRF’s sensitivity to changes in the microphysical scheme compared with that present when LAPS analyses were used for initialization. In addition, differences in initialized thermodynamics resulted in changes in sensitivity to PBL and convective schemes. With both initialization datasets, the greatest sensitivity to the simulated rain rate was due to changes in the convective scheme. However, for rain volume, substantial sensitivity was present due to changes in both the physical parameterizations and the initial datasets. [ABSTRACT FROM AUTHOR]

In recent years, a mixed-physics ensemble approach has been investigated as a method to better predict mesoscale convective system (MCS) rainfall. For both mixed-physics ensemble design and interpretation, knowledge of the general impact of various physical schemes and their interactions on warm season MCS rainfall forecasts would be useful. Adopting the newly emerging Weather Research and Forecasting (WRF) model for this purpose would further emphasize such benefits. To pursue this goal, a matrix of 18 WRF model configurations, created using different physical scheme combinations, was run with 12-km grid spacing for eight International H2O Project (IHOP) MCS cases. For each case, three different treatments of convection, three different microphysical schemes, and two different planetary boundary layer schemes were used. Sensitivity to physics changes was determined using the correspondence ratio and the squared correlation coefficient. The factor separation method was also used to quantify in detail the impacts of the variation of two different physical schemes and their interaction on the simulated rainfall. Skill score measures averaged over all eight cases for all 18 configurations indicated that no one configuration was obviously best at all times and thresholds. The greatest variability in forecasts was found to come from changes in the choice of convective scheme, although notable impacts also occurred from changes in the microphysics and planetary boundary layer (PBL) schemes. Specifically, changes in convective treatment notably impacted the forecast of system average rain rate, while forecasts of total domain rain volume were influenced by choices of microphysics and convective treatment. The impact of interactions (synergy) of different physical schemes, although occasionally of comparable magnitude to the impacts from changing one scheme alone (compared to a control run), varied greatly among cases and over time, and was typically not statistically significant. [ABSTRACT FROM AUTHOR]

Warm season convective system rainfall forecasts remain a particularly difficult forecast challenge. For these events, it is possible that ensemble forecasts would provide helpful information unavailable in a single deterministic forecast. In this study, an intense derecho event accompanied by a well-organized band of heavy rainfall is used to show that for some situations, the predictability of rainfall even within a 12–24-h period is so low that a wide range of simulations using different models, different physical parameterizations, and different initial conditions all fail to provide even a small signal that the event will occur. The failure of a wide range of models and parameterizations to depict the event might suggest inadequate representation of the initial conditions. However, a range of different initial conditions also failed to lead to a well-simulated event, suggesting that some events are unlikely to be predictable with the current observational network, and ensemble guidance for such cases may provide limited additional information useful to a forecaster. [ABSTRACT FROM AUTHOR]

The large-scale forcing associated with 20 mesoscale convective system (MCS) events has been evaluated to determine how the magnitude of that forcing influences the rainfall forecasts made with a 10-km grid spacing version of the Eta Model. Different convective parameterizations and initialization modifications were used to simulate these Upper Midwest events. Cases were simulated using both the Betts–Miller–Janjić (BMJ) and the Kain–Fritsch (KF) convective parameterizations, and three different techniques were used to improve the initialization of mesoscale features important to later MCS evolution. These techniques included a cold pool initialization, vertical assimilation of surface mesoscale observations, and an adjustment to initialized relative humidity based on radar echo coverage. As an additional aspect in this work, a morphology analysis of the 20 MCSs was included. Results suggest that the model using both schemes performs better when net large-scale forcing is strong, which typically is the case when a cold front moves across the domain. When net forcing is weak, which is often the case in midsummer situations north of a warm or stationary front, both versions of the model perform poorly. Runs with the BMJ scheme seem to be more affected by the magnitude of surface frontogenesis than the KF runs. Runs with the KF scheme are more sensitive to the CAPE amount than the BMJ runs. A fairly well-defined split in morphology was observed, with squall lines having trailing stratiform regions likely in scenarios associated with higher equitable threat scores (ETSs) and nonlinear convective clusters strongly dominating the more poorly forecast weakly forced events. [ABSTRACT FROM AUTHOR]