Forecast Experiments Research Articles

Sensitivity of 24-h Forecast Dryline Position and Structure to Boundary Layer Parameterizations in Convection-Allowing WRF Model Simulations

Abstract Recent NOAA Hazardous Weather Testbed Spring Forecasting Experiments have emphasized the sensitivity of forecast sensible weather fields to how boundary layer processes are represented in the Weather Research and Forecasting (WRF) Model. Thus, since 2010, the Center for Analysis and Prediction of Storms has configured at least three members of their WRF-based Storm-Scale Ensemble Forecast (SSEF) system specifically for examination of sensitivities to parameterizations of turbulent mixing, including the Mellor–Yamada–Janjić (MYJ); quasi-normal scale elimination (QNSE); Asymmetrical Convective Model, version 2 (ACM2); Yonsei University (YSU); and Mellor–Yamada–Nakanishi–Niino (MYNN) schemes (hereafter PBL members). In postexperiment analyses, significant differences in forecast boundary layer structure and evolution have been observed, and for preconvective environments MYNN was found to have a superior depiction of temperature and moisture profiles. This study evaluates the 24-h forecast dryline positions in the SSEF system PBL members during the period April–June 2010–12 and documents sensitivities of the vertical distribution of thermodynamic and kinematic variables in near-dryline environments. Main results include the following. Despite having superior temperature and moisture profiles, as indicated by a previous study, MYNN was one of the worst-performing PBL members, exhibiting large eastward errors in forecast dryline position. During April–June 2010–11, a dry bias in the North American Mesoscale Forecast System (NAM) initial conditions largely contributed to eastward dryline errors in all PBL members. An upgrade to the NAM and assimilation system in October 2011 apparently fixed the dry bias, reducing eastward errors. Large sensitivities of CAPE and low-level shear to the PBL schemes were found, which were largest between 1.0° and 3.0° to the east of drylines. Finally, modifications to YSU to decrease vertical mixing and mitigate its warm and dry bias greatly reduced eastward dryline errors.

Predicting Recessions with Leading Indicators: Model Averaging and Selection over the Business Cycle

Four methods of model selection—equally weighted forecasts, Bayesian model‐averaged forecasts, and two models produced by the machine‐learning algorithm boosting—are applied to the problem of predicting business cycle turning points with a set of common macroeconomic variables. The methods address a fundamental problem faced by forecasters: the most useful model is simple but makes use of all relevant indicators. The results indicate that successful models of recession condition on different economic indicators at different forecast horizons. Predictors that describe real economic activity provide the clearest signal of recession at very short horizons. In contrast, signals from housing and financial markets produce the best forecasts at longer forecast horizons. A real‐time forecast experiment explores the predictability of the 2001 and 2007 recessions. Copyright © 2015 John Wiley & Sons, Ltd.

Impact of Different Satellite Wind Lidar Telescope Configurations on NCEP GFS Forecast Skill in Observing System Simulation Experiments

AbstractThe Global Wind Observing Sounder (GWOS) concept, which has been developed as a hypothetical space-based hybrid wind lidar system by NASA in response to the 2007 National Research Council (NRC) decadal survey, is expected to provide global wind profile observations with high vertical resolution, precision, and accuracy when realized. The assimilation of Doppler wind lidar (DWL) observations anticipated from the GWOS is being conducted as a series of observing system simulation experiments (OSSEs) at the Joint Center for Satellite Data Assimilation (JCSDA). A companion paper (Riishøjgaard et al.) describes the simulation of this lidar wind data and evaluates the impact on global numerical weather prediction (NWP) of the baseline GWOS using a four-telescope configuration to provide independent line-of-sight wind speeds, while this paper sets out to assess the NWP impact of GWOS equipped with alternative paired configurations of telescopes. The National Centers for Environmental Prediction (NCEP) Gridpoint Statistical Interpolation (GSI) analysis system and the Global Forecast System (GFS) were used, at a resolution of T382 with 64 layers, as the assimilation system and the forecast model, respectively, in these lidar OSSEs. A set of 45-day assimilation and forecast experiments from 2 July to 15 August 2005 was set up and executed.In this OSSE study, a control simulation utilizing all of the data types assimilated in the operational GSI/GFS system was compared to three OSSE simulations that added lidar wind data from the different configurations of telescopes (one-, two-, and four-look configurations). First, the root-mean-square error (RMSE) of wind analysis is compared against the nature run. A significant reduction of the stratospheric RMSE of wind analyses is found for all latitudes when lidar wind profiles are used in the assimilation system. The forecast impacts of lidar data on the wind and mass forecasts are also presented. In addition, the anomaly correlations (AC) of geopotential height forecasts at 500 hPa were evaluated to compare the control and different GWOS telescope configuration experiments. The results show that the assimilation of lidar data from the GWOS (one, two, or four looks) can improve the NCEP GFS wind and mass field forecasts. The addition of the simulated lidar wind observations leads to a statistically significant increase in AC scores.

Strategic coordination in forecasting – An experimental study

While reputational herding has been shown to contribute to poor economic forecasting, the underlying behavioral mechanisms have not yet been empirically investigated. We run a forecasting experiment with contradictory incentives for accuracy and coordination, finding subjects’ forecasts to be inaccurate and driven by the coordination motive. Coordination is achieved through the salient, risk-dominant equilibrium, i.e. merely forecasting the current values. Subjects succeeding in coordinating earn significantly more than those striving for accuracy. Our results emphasize that reputational herding should be considered as a driving force for persistently poor prediction accuracy and systematically biased forecasts towards consensus values.

Reconstruction and Forecast Experiments of a Statistical–Dynamical Model of the Western Pacific Subtropical High and East Asian Summer Monsoon Factors

Abstract Abnormal activity of the western Pacific subtropical high (WPSH) may result in extreme weather events in East Asia. However, because the relationship between the WPSH and other components of the East Asian summer monsoon (EASM) system is unknown, it is still difficult to forecast such abnormal activity. The delay-relevant method is used to study 2010 data for abnormal weather and it is concluded that the Indian monsoon latent heat flux, the Somali low-level jet, and the Tibetan high activity index can significantly affect anomalies in the WPSH in the EASM system. By combining genetic algorithms and statistical–dynamical reconstruction theory, a nonlinear statistical–dynamical model of the WPSH and these three influencing factors was objectively reconstructed from actual 2010 data and a dynamically extended forecasting experiment was carried out. To further test the forecasting performance of the reconstructed model, further experiments using data from nine abnormal WPSH years and eight normal WPSH years were performed for comparison. All the results suggest that the forecasts of the subtropical high area index, the Indian monsoon latent heat flux, the Somali low-level jet, and the Tibetan high activity index all have good performance in the short and medium terms (<25 days). Not only is the forecasting trend accurate, but the mean absolute percentage error is ≤9%. This work suggests new areas of research into the association between the WPSH and EASM systems and provides a new method for the prediction of the WPSH area index.

The impact of AIRS atmospheric temperature and moisture profiles on hurricane forecasts: Ike (2008) and Irene (2011)

Atmospheric InfraRed Sounder (AIRS) measurements are a valuable supplement to current observational data, especially over the oceans where conventional data are sparse. In this study, two types of AIRS-retrieved temperature and moisture profiles, the AIRS Science Team product (SciSup) and the single field-of-view (SFOV) research product, were evaluated with European Centre for Medium-Range Weather Forecasts (ECMWF) analysis data over the Atlantic Ocean during Hurricane Ike (2008) and Hurricane Irene (2011). The evaluation results showed that both types of AIRS profiles agreed well with the ECMWF analysis, especially between 200 hPa and 700 hPa. The average standard deviation of both temperature profiles was approximately 1 K under 200 hPa, where the mean AIRS temperature profile from the AIRS SciSup retrievals was slightly colder than that from the AIRS SFOV retrievals. The mean SciSup moisture profile was slightly drier than that from the SFOV in the mid troposphere. A series of data assimilation and forecast experiments was then conducted with the Advanced Research version of the Weather Research and Forecasting (WRF) model and its three-dimensional variational (3DVAR) data assimilation system for hurricanes Ike and Irene. The results showed an improvement in the hurricane track due to the assimilation of AIRS clear-sky temperature profiles in the hurricane environment. In terms of total precipitable water and rainfall forecasts, the hurricane moisture environment was found to be affected by the AIRS sounding assimilation. Meanwhile, improving hurricane intensity forecasts through assimilating AIRS profiles remains a challenge for further study.

Unemployment Rate Forecasts: Evidence from the Baltic States

This paper compares forecasts of unemployment rates in the Baltic states using time-varying parameter models. We are particularly interested in dynamic asymmetries in the unemployment rates and the forecasting performance of different models at different times. We consider each predictor in the time-varying model as either stochastic or deterministic and use a rolling short-term forecast experiment as an out-of-sample test of forecast accuracy. The time span of our data from 2001 to 2014 includes the global financial crisis. We find that the forecasting ability of the models depends on both the forecasting horizon and the moment in time when the forecasts are done. The empirical evidence derived from our investigation suggests that no single model is the best one, but models that include a cyclical component tend to perform better than others. Our findings show that the preferred models differ in the time of increase or decrease in unemployment rates.

Are all movements in food and energy prices transitory? Evidence from India

This paper seeks to answer two questions in the context of the Indian economy. First, are all movements in food and energy prices transitory? Second, is there a significant relationship between permanent and transitory shocks to different components of the aggregate price level? Using monthly price data for India and a multivariate correlated unobserved component model based on Morley et al. ((2003). The Review of Economics and Statistics, 85(2): 235) and Bradley et al. (forthcoming), we find that both food and energy prices have significant permanent components. Further, we find a significant feedback effect between permanent components of food prices and manufacturing good prices. These findings suggest that neglecting food and energy prices may render the core inflation measure based solely on manufacturing prices a biased measure of long run inflation. Finally, using a forecasting experiment, we show that a measure of the trend inflation that explicitly accounts for long run price movements in food and energy prices provides a superior forecast of future inflation when compared with alternative measures of the trend inflation.

Variational Assimilation of Cloud Liquid/Ice Water Path and Its Impact on NWP

AbstractAnalysis of the cloud components in numerical weather prediction models using advanced data assimilation techniques has been a prime topic in recent years. In this research, the variational data assimilation (DA) system for the Weather Research and Forecasting (WRF) Model (WRFDA) is further developed to assimilate satellite cloud products that will produce the cloud liquid water and ice water analysis. Observation operators for the cloud liquid water path and cloud ice water path are developed and incorporated into the WRFDA system. The updated system is tested by assimilating cloud liquid water path and cloud ice water path observations from Global Geostationary Gridded Cloud Products at NASA. To assess the impact of cloud liquid/ice water path data assimilation on short-term regional numerical weather prediction (NWP), 3-hourly cycling data assimilation and forecast experiments with and without the use of the cloud liquid/ice water paths are conducted. It is shown that assimilating cloud liquid/ice water paths increases the accuracy of temperature, humidity, and wind analyses at model levels between 300 and 150 hPa after 5 cycles (15 h). It is also shown that assimilating cloud liquid/ice water paths significantly reduces forecast errors in temperature and wind at model levels between 300 and 150 hPa. The precipitation forecast skills are improved as well. One reason that leads to the improved analysis and forecast is that the 3-hourly rapid update cycle carries over the impact of cloud information from the previous cycles spun up by the WRF Model.

Statistical Prediction of the South China Sea Surface Height Anomaly

Based on the simple ocean data assimilation (SODA) data, this study analyzes and forecasts the monthly sea surface height anomaly (SSHA) averaged over South China Sea (SCS). The approach to perform the analysis is a time series decomposition method, which decomposes monthly SSHAs in SCS to the following three parts: interannual, seasonal, and residual terms. Analysis results demonstrate that the SODA SSHA time series are significantly correlated to the AVISO SSHA time series in SCS. To investigate the predictability of SCS SSHA, an exponential smoothing approach and an autoregressive integrated moving average approach are first used to fit the interannual and residual terms of SCS SSHA while keeping the seasonal part invariant. Then, an array of forecast experiments with the start time spanning from June 1977 to June 2007 is performed based on the prediction model which integrates the above two models and the time-independent seasonal term. Results indicate that the valid forecast time of SCS SSHA of the statistical model is about 7 months, and the predictability of SCS SSHA in Spring and Autumn is stronger than that in Summer and Winter. In addition, the prediction skill of SCS SSHA has remarkable decadal variability, with better phase forecast in 1997–2007.

Predicting Hurricane Intensity and Associated Hazards: A Five-Year Real-Time Forecast Experiment with Assimilation of Airborne Doppler Radar Observations

Abstract Performance in the prediction of hurricane intensity and associated hazards has been evaluated for a newly developed convection-permitting forecast system that uses ensemble data assimilation techniques to ingest high-resolution airborne radar observations from the inner core. This system performed well for three of the ten costliest Atlantic hurricanes: Ike (2008), Irene (2011), and Sandy (2012). Four to five days before these storms made landfall, the system produced good deterministic and probabilistic forecasts of not only track and intensity, but also of the spatial distributions of surface wind and rainfall. Averaged over all 102 applicable cases that have inner-core airborne Doppler radar observations during 2008–2012, the system reduced the day-2-to-day-4 intensity forecast errors by 25%–28% compared to the corresponding National Hurricane Center’s official forecasts (which have seen little or no decrease in intensity forecast errors over the past two decades). Empowered by sufficient computing resources, advances in both deterministic and probabilistic hurricane prediction will enable emergency management officials, the private sector, and the general public to make more informed decisions that minimize the losses of life and property.

Impact of Atmosphere and Land Surface Initial Conditions on Seasonal Forecasts of Global Surface Temperature

Abstract The impact of land surface and atmosphere initialization on the forecast skill of a seasonal prediction system is investigated, and an effort to disentangle the role played by the individual components to the global predictability is done, via a hierarchy of seasonal forecast experiments performed under different initialization strategies. A realistic atmospheric initial state allows an improved equilibrium between the ocean and overlying atmosphere, increasing the model predictive skill in the ocean. In fact, in regions characterized by strong air–sea coupling, the atmosphere initial condition affects forecast skill for several months. In particular, the ENSO region, eastern tropical Atlantic, and North Pacific benefit significantly from the atmosphere initialization. On the mainland, the effect of atmospheric initial conditions is detected in the early phase of the forecast, while the quality of land surface initialization affects forecast skill in the following seasons. Winter forecasts in the high-latitude plains benefit from the snow initialization, while the impact of soil moisture initial state is particularly effective in the Mediterranean region and central Asia. However, the initialization strategy based on the full value technique may not be the best choice for land surface, since soil moisture is a strongly model-dependent variable: in fact, initialization through land surface reanalysis does not systematically guarantee a skill improvement. Nonetheless, using a different initialization strategy for land, as opposed to atmosphere and ocean, may generate inconsistencies. Overall, the introduction of a realistic initialization for land and atmosphere substantially increases skill and accuracy. However, further developments in the procedure for land surface initialization are required for more accurate seasonal forecasts.

Development of an Efficient Regional Four-Dimensional Variational Data Assimilation System for WRF

AbstractThis paper presents the development of a single executable four-dimensional variational data assimilation (4D-Var) system based on the Weather Research and Forecasting (WRF) Model through coupling the variational data assimilation algorithm (WRF-VAR) with the newly developed WRF tangent linear and adjoint model (WRFPLUS). Compared to the predecessor Multiple Program Multiple Data version, the new WRF 4D-Var system achieves major improvements in that all processing cores are able to participate in the computation and all information exchanges between WRF-VAR and WRFPLUS are moved directly from disk to memory. The single executable 4D-Var system demonstrates desirable acceleration and scalability in terms of the computational performance, as demonstrated through a series of benchmarking data assimilation experiments carried out over a continental U.S. domain. To take into account the nonlinear processes with the linearized minimization algorithm and to further decrease the computational cost of the 4D-Var minimization, a multi-incremental minimization that uses multiple horizontal resolutions for the inner loop has been developed. The method calculates the innovations with a high-resolution grid and minimizes the cost function with a lower-resolution grid. The details regarding the transition between the high-resolution outer loop and the low-resolution inner loop are introduced. Performance of the multi-incremental configuration is found to be comparable to that with the full-resolution 4D-Var in terms of 24-h forecast accuracy in the week-long analysis and forecast experiment over the continental U.S. domain. Moreover, the capability of the newly developed multi-incremental 4D-Var system is further demonstrated in the convection-permitting analysis and forecast experiment for Hurricane Sandy (2012), which was hardly computationally feasible with the predecessor WRF 4D-Var system.

Estimating advective near-surface currents from ocean color satellite images

Improved maximum cross correlation (MCC) techniques are used to retrieve ocean surface currents from the sequential ocean color imagery provided by multiple newer generations of satellite sensors on hourly scales in the Yellow Sea and the U.S. East and Gulf coasts. The MCC calculation is validated in a series of Bio-Optical Forecasting (BioCast) experiments with predetermined synthetic velocities, and its products are evaluated by examining the errors and biases with respect to the High Frequency Radar (HFRadar) measurements. The root-mean-square (RMS) errors in our best current products derived from the overlap of satellite sensor swath between the VIIRS sequential orbits are less than 0.17ms−1 in the evaluation area outside of the Chesapeake Bay. The most accurate current products are those derived from the imagery data of Rrs(551), Bb(551) and C(551), while the image sequences of Bb(551) and Zeu_lee are identified as the most suited products for the retrieval of currents because of their best production capacities of valid velocity vectors. Mechanisms between the advective processes and the dynamic changes of bio-optical properties are discussed regarding the performances of various color products on the retrieval of currents. Similarities of velocity distribution in the retrieved vector arrays are collected across different MCC products derived from ocean color datasets that are of different types and derived from different spectral channels of satellite overpasses. The inter-product similarities themselves can be used to characterize the near-surface advection as well and usually have smaller errors than each of the individual MCC currents. Moreover, efforts are also under way to improve the ocean color derived currents by merging several of the MCC products with similarities to increase the total spatial coverage. This study not only seeks the image-derived products best representing the sea surface current structures in coastal areas, but also exploits how these currents can be improved or optimized to support the ocean forecasts.

Sensitivity of the Arctic sea ice concentration forecasts to different atmospheric forcing: a case study

A regional Arctic configuration of the Massachusetts Institute of Technology general circulation model (MITgcm) is used as the coupled ice-ocean model for forecasting sea ice conditions in the Arctic Ocean at the National Marine Environmental Forecasting Center of China (NMEFC), and the numerical weather prediction from the National Center for Environmental Prediction Global Forecast System (NCEP GFS) is used as the atmospheric forcing. To improve the sea ice forecasting, a recently developed Polar Weather Research and Forecasting model (Polar WRF) model prediction is also tested as the atmospheric forcing. Their forecasting performances are evaluated with two different satellite-derived sea ice concentration products as initializations: (1) the Special Sensor Microwave Imager/Sounder (SSMIS) and (2) the Advanced Microwave Scanning Radiometer for EOS (AMSR-E). Three synoptic cases, which represent the typical atmospheric circulations over the Arctic Ocean in summer 2010, are selected to carry out the Arctic sea ice numerical forecasting experiments. The evaluations suggest that the forecasts of sea ice concentrations using the Polar WRF atmospheric forcing show some improvements as compared with that of the NCEP GFS.

Flash flood warnings using the ensemble precipitation forecasting technique: A case study on forecasting floods in Taiwan caused by typhoons

Summary A flash flood is an event that develops rapidly. Given early warnings with sufficient lead time, flood forecasting can help people prepare disaster prevention measures. To provide this early warning, a statistics-based flood forecasting model was developed to evaluate the flooding potential in urban areas using ensemble quantitative precipitation forecasts (the Taiwan Cooperative Precipitation Ensemble Forecast Experiment, TAPEX). The proposed model uses different sources of information, such as (i) the designed capacity of storm sewer systems, (ii) a flood inundation potential database, and (iii) historical flooding observations, to evaluate the potential for flash flooding situations to occur. Using 24-, 48- and 72-h ahead precipitation forecasts from the TAPEX, the proposed model can assess the flooding potential with two levels of risk and at the township scale with a 3-day lead time. The proposed model is applied to Pingtung County, which includes 33 townships and is located in southern Taiwan. A dataset of typhoon storms from 2010 to 2014 was used to evaluate the model performance. The accuracy and threat score for testing events are 0.68 and 0.30, respectively, with a lead time of 24 h. The accuracy and threat score for training events are 0.82 and 0.31, respectively, with a lead time of 24 h. The model performance decreases when the lead time is extended. However, the model demonstrates its potential as a valuable reference to improve emergency responses to alleviate the loss of lives and property due to flooding.

Salinity anomaly as a trigger for ENSO events.

According to the classical theories of ENSO, subsurface anomalies in ocean thermal structure are precursors for ENSO events and their initial specification is essential for skillful ENSO forecast. Although ocean salinity in the tropical Pacific (particularly in the western Pacific warm pool) can vary in response to El Niño events, its effect on ENSO evolution and forecasts of ENSO has been less explored. Here we present evidence that, in addition to the passive response, salinity variability may also play an active role in ENSO evolution, and thus important in forecasting El Niño events. By comparing two forecast experiments in which the interannually variability of salinity in the ocean initial states is either included or excluded, the salinity variability is shown to be essential to correctly forecast the 2007/08 La Niña starting from April 2007. With realistic salinity initial states, the tendency to decay of the subsurface cold condition during the spring and early summer 2007 was interrupted by positive salinity anomalies in the upper central Pacific, which working together with the Bjerknes positive feedback, contributed to the development of the La Niña event. Our study suggests that ENSO forecasts will benefit from more accurate salinity observations with large-scale spatial coverage.

Reconstruction of statistical–dynamical model of the Western Pacific subtropical high and East Asian summer monsoon factors and its forecast experiments

Abnormal Western Pacific subtropical high (WPSH) activities often lead to extreme weather events in East Asia in some years. The relationship between the WPSH and the members of the East Asian summer monsoon (EASM) system is unknown, however. So the forecasting of abnormal WPSH activities is still difficult. Because of adaptive learning and nonlinear superiority of the adaptive-network-based fuzzy inference system (ANFIS), it can be used to analyze and detect the influence and contribution of members of the EASM system on the WPSH anomalies. Based on the abnormal weather data from 2010, using the delay-relevant method and ANFIS, we conclude that the Mascarene cold high index, the Indian monsoon latent heat flux and the Tibetan high activity index can affect WPSH anomalies more significantly than other members of the EASM system. With the combination of genetic algorithms and statistical–dynamical reconstruction theory, a nonlinear statistical–dynamical model of the WPSH and three impact factors are objectively reconstructed from the actual data of 2010, and also, a dynamically extended forecasting experiment is carried out. To further test the forecasting performance of the reconstructed model, more experiments of nine WPSH abnormal years and eight WPSH normal years are also performed for comparison. All the results suggest that the forecasts of the subtropical high area index, the Mascarene cold high index, the Indian monsoon latent heat flux and the Tibetan high activity index all have good performance in the short and medium term (<25 days). Not only is the forecasting trend accurate, but also the root mean square error is no more than 8 %. Our paper not only provides new thinking for research on the association between the WPSH and EASM system, but also provides a new method for the prediction of the WPSH area index.

Unemployment Rate Forecasts in the Baltic States

In this paper we compare forecasts of unemployment rates of the Baltic States using time-varying parameter models. We are particularly interested in dynamic asymmetries in unemployment rates and the forecasting performance of different models in different times. We consider each predictor in time-varying model as stochastic or deterministic and use a rolling short-term forecast experiment as an out-of-sample test of forecast accuracy. The time span of our data 2001-2014 includes global financial crisis. We find that forecasting ability of the models depends on the forecasting horizon as well as on the moment in time for which forecasts are done. The empirical evidence derived from our investigation suggests that there is no single model that is the best one, however models that include cyclical component occur to perform better than others. Our findings show that the preference of the model depends on the time of increase or decrease in unemployment rates.

Corrigendum to &amp;quot;Heavy rainfall episodes over Liguria in autumn 2011: numerical forecasting experiments&amp;quot; published in Nat. Hazards Earth Syst. Sci., 14, 1325–1340, 2014

Corrigendum to &amp;quot;Heavy rainfall episodes over Liguria in autumn 2011: numerical forecasting experiments&amp;quot; published in Nat. Hazards Earth Syst. Sci., 14, 1325–1340, 2014