NCEP Climate Forecast System Version Research Articles

Effects of Arctic sea-ice concentration on turbulent surface fluxes in four atmospheric reanalyses

Abstract. A prerequisite for understanding the local, regional, and hemispherical impacts of Arctic sea-ice decline on the atmosphere is to quantify the effects of sea-ice concentration (SIC) on the turbulent surface fluxes of sensible and latent heat in the Arctic. We analyse these effects utilising four global atmospheric reanalyses, ERA5, JRA-55, MERRA-2, and NCEP/CFSR (including both the NCEP Climate Forecast System Reanalysis (CFSR) and the NCEP Climate Forecast System Version 2 (CFSv2)), and evaluate their uncertainties arising from inter-reanalysis differences in SIC and in the sensitivity of the turbulent surface fluxes to SIC. The magnitude of the differences in SIC is up to 0.15 but typically around 0.05 in most of the Arctic over all four seasons. Orthogonal-distance regression and ordinary-least-squares regression analyses indicate that the greatest sensitivity of both the latent and the sensible heat flux to SIC occurs in the cold season, November to April. For these months, using daily means of data, the average sensitivity is 400 W m−2 for the latent heat flux and over 800 W m−2 for the sensible heat flux per unit of SIC (change in SIC from 0 to 1), with differences between reanalyses that are as large as 300 W m−2 for the latent heat flux and 600 W m−2 for the sensible heat flux per unit of SIC. The sensitivity is highest for the NCEP/CFSR reanalysis. Comparing the periods 1980–2000 and 2001–2021, we find that the effect of SIC on turbulent surface fluxes has weakened owing to the increasing surface temperature of sea ice and sea-ice decline. The results also indicate signs of a decadal-scale improvement in the mutual agreement between reanalyses. The effect of SIC on turbulent surface fluxes arises mostly via the effect of SIC on atmosphere–surface differences in temperature and specific humidity, whereas the effect of SIC on wind speed (via surface roughness and atmospheric-boundary-layer stratification) partly cancels out in the turbulent surface fluxes, as the wind speed increases the magnitudes of both upward and downward fluxes.

Observational characteristics and causes of extreme heavy rainfall in typhoon Duksuri (2305)

Super Typhoon Duksuri (2305), made landfall in Fujian Province, causing significant impacts from July 27th to July 31st. The typhoon resulted in heavy to torrential rainfall in multiple provinces, particularly in southern Taiwan, eastern Fujian, and the Beijing-Tianjin-Hebei region. Daily accumulated precipitation reached a maximum of 200 to 220 millimeters. Using ERA5 reanalysis data and NCEP Climate Forecast System Version 2 reanalysis data, this article conducted an observational analysis of Duksuri. The results indicate that the interaction between subtropical high-pressure systems, typhoons, and the strong southeast and east winds created favorable conditions for the prolonged duration of Duksuri and extreme precipitation. The extreme rainfall in southern Taiwan and eastern Fujian was primarily caused by the abundant moisture associated with the typhoon itself. In contrast, the extreme rainfall in the Beijing-Tianjin-Hebei region was primarily due to the northward movement of the remnants of the typhoon, hindered by the high-pressure systems, resulting in longer and concentrated precipitation. Additionally, both regions are located on the windward slopes of mountainous areas, which significantly enhance the rainfall due to orographic lifting.

Evaluation of Multiweek Tropical Cyclone Forecasts in the Philippines

Abstract In the pursuit of providing tropical cyclone (TC) forecasts beyond the conventional time scales covered by weather forecasting in the Philippines, this study has examined the multiweek (i.e., from week 1 to week 4) TC forecast skill in the country. TC forecasts derived from three ensemble models, namely, the NCEP Climate Forecast System version 2 (CFSv2), the European Centre for Medium-Range Weather Forecasts Ensemble Prediction System (ECMWF), and the NCEP Global Ensemble Forecast System version 12 (GEFSv12) from 6 October 2020 to 31 October 2021 were verified. Results revealed that the ECMWF model is consistently the most skillful in multiweek TC prediction over the domain bounded by 110°–155°E and 0°–27°N in the western North Pacific. The ECMWF obtained hit rates ranging from 0.25 to 0.31, low false alarm rates of 0–0.33, and the highest equitable threat scores among the models. In contrast to this, the GEFSv12 and CFSv2 models had varying skills, with the former performing better in the first two weeks and the latter in longer lead times. It is further revealed that the three models generally underestimate the observed number of storms, storm days, and accumulated cyclone energy. Moreover, the study shows that the forecast TC tracks have a significant (p < 0.05) positional bias toward the right of observed tracks beyond week 1, and that they tend to propagate slower than observations especially in week 1 and week 2. These findings contribute to better understanding the strengths and limitations of these ensemble models useful for eventual provision of multiweek TC forecasts in the Philippines.

Improving the CFSv2 prediction of the Indian Ocean Dipole based on a physical‐empirical model and a deep‐learning approach

AbstractDespite significant advances in seasonal climate forecasts, the reliability of both dynamical and empirical models for the Indian Ocean Dipole (IOD) prediction is still limited to a lead time of one season or less. In this study, the skill of the NCEP Climate Forecast System version 2 (CFSv2) for the IOD prediction during the period 1982–2014 is evaluated. The results indicate that the model performance for the IOD prediction is the worst in spring among the four seasons, which is manifested in the fact that a skilful prediction of spring IOD event is limited to a lead time of only about 1–2 months. To improve the forecast of spring IOD events, a physical‐empirical (PE) model and a convolutional neural network (CNN) model are established in the present study. The IOD in April–May–June (AMJ) is taken as the predictand, and the CFSv2‐predicted sea surface height (SSH) in AMJ and the observed Laptev sea ice in the preceding December are used as the two predictors. The original CFSv2‐predicted IOD time series has an insignificant correlation with the observed IOD time series with a temporal correlation coefficient (TCC) of 0.03; the PE model (CNN model) can largely improve the IOD prediction with a TCC of 0.74 (0.77) between the PE‐model‐predicted (CNN‐model‐predicted) IOD and the observed IOD during AMJ. Thus, the PE model and the CNN model developed in the present study can be applied to improve the IOD predictions from numerical models in the future.

Sub-seasonal Prediction of the South China Sea Summer Monsoon Onset in the NCEP Climate Forecast System Version 2

Sub-seasonal Prediction of the South China Sea Summer Monsoon Onset in the NCEP Climate Forecast System Version 2

Indian Monsoon Teleconnections and the Impact of Correcting Tropical Diabatic Heating

Abstract The Indian summer monsoon is partly modulated by persistent remote forcing from the tropical Indo-Pacific, evident in the dominant observed teleconnection patterns, namely, El Niño–Southern Oscillation (ENSO) and the equatorial Indian Ocean Oscillation (EQUINOO). In the atmosphere, these teleconnections are presumably driven by diabatic heating, primarily associated with the release of latent heat in condensation with rainfall. However, in coupled atmosphere–ocean models, biases result in large systematic errors in tropical heating. This study seeks to understand the extent that teleconnections are forced by tropical heating and whether or not correcting tropical heating biases improves monsoon prediction skill. We examine a series of reforecasts made with the NCEP Climate Forecast System version 2 in which the “added heating” technique is applied to largely remove tropical heating biases. We isolate the ENSO and EQUINOO signals and examine the ability to reproduce and predict these teleconnections in the model run with and without tropical heating correction. Improving ENSO and EQUINOO-related heating does result in increased prediction skill in monsoon circulation teleconnection patterns. Prediction of other relevant tropical and subtropical circulation indices is improved; however, the impact on the Indian monsoon as a whole is limited. EQUINOO exhibits large internal variability in the model, and despite imposing realistic EQUINOO heating, the monsoon circulation is relatively insensitive in the model. This suggests that either the EQUINOO teleconnection in nature does not emerge as a forced response to tropical heating, and/or the model is unable to reproduce the relationship due to separate deficiencies. Significance Statement India receives over 80% of its annual rainfall during the summer in association with the monsoon. A strong socioeconomic dependence on agriculture makes India sensitive to year-to-year variations in monsoon rainfall, thus predicting and understanding such variations is of great value. Coincident changes in tropical atmospheric heating (and cooling) may be more predictable and presumably impact the monsoon; however, causality has yet to be demonstrated and quantified, particularly for the tropical Indian Ocean. This motivates our modeling study to diagnose the role of tropical heating for the Indian monsoon and whether or not correcting heating errors improves monsoon prediction.

Land Surface Temperature Retrieval Using High-Resolution Vertical Profiles Simulated by WRF Model

This work gives a first insight into the potential of the Weather Research and Forecasting (WRF) model to provide high-resolution vertical profiles for land surface temperature (LST) retrieval from thermal infrared (TIR) remote sensing. WRF numerical simulations were conducted to downscale NCEP Climate Forecast System Version 2 (CFSv2) reanalysis profiles, using two nested grids with horizontal resolutions of 12 km (G12) and 3 km (G03). We investigated the utility of these profiles for the atmospheric correction of TIR data and LST estimation, using the moderate resolution atmospheric transmission (MODTRAN) model and the Landsat 8 TIRS10 band. The accuracy evaluation was performed using 27 clear-sky cases over a radiosonde station in Southern Brazil. We included in the comparative analysis NASA’s Atmospheric Correction Parameter Calculator (ACPC) web-tool and profiles obtained directly from the NCEP CFSv2 reanalysis. The atmospheric parameters from ACPC, followed by those from CFSv2, were in better agreement with parameters calculated using in situ radiosondes. When applied into the radiative transfer equation (RTE) to retrieve LST, the best results (RMSE) were, in descending order: CFSv2 (0.55 K), ACPC (0.56 K), WRF G12 (0.79 K), and WRF G03 (0.82 K). Our findings suggest that there is no special need to increase the horizontal resolution of reanalysis profiles aiming at RTE-based LST retrieval. However, the WRF results were still satisfactory and promising, encouraging further assessments. We endorse the use of the well-known ACPC and recommend the NCEP CFSv2 profiles for TIR atmospheric correction and LST single-channel retrieval.

Evaluation of Mean State in NCEP Climate Forecast System (Version 2) Simulation Using a Stochastic Multicloud Model Calibrated With DYNAMO RADAR Data

AbstractStochastic parameterizations are continuously providing promising simulations of unresolved atmospheric processes for global climate models (GCMs). One of the stochastic multi‐cloud model (SMCM) features is to mimic the life cycle of the three most common cloud types (congestus, deep, and stratiform) in tropical convective systems. To better represent organized convection in the Climate Forecast System version 2 (CFSv2), the SMCM parameterization is adopted in CFSv2 (SMCM‐CTRL) in lieu of the pre‐existing revised simplified Arakawa–Schubert (RSAS) cumulus scheme and has shown essential improvements in different large‐scale features of tropical convection. But the sensitivity of the SMCM parameterization from the observations is yet to be ascertained. Radar data during the Dynamics of the Madden‐Julian Oscillation (DYNAMO) field campaign is used to tune the SMCM in the present manuscript. The DYNAMO radar observations have been used to calibrate the SMCM using a Bayesian inference procedure to generate key time scale parameters for the transition probabilities of the underlying Markov chains of the SMCM as implemented in CFS (hereafter SMCM‐DYNAMO). SMCM‐DYNAMO improves many aspects of the mean state climate compared to RSAS, and SMCM‐CTRL. Significant improvement is noted in the rainfall probability distribution function over the global tropics. The global distribution of different types of clouds, particularly low‐level clouds, is also improved. The convective and large‐scale rainfall simulations are investigated in detail.

Reanalysis profile downscaling with WRF model and sensitivity to PBL parameterization schemes over a subtropical station

Atmospheric profiles are essential in weather prediction, atmospheric research, and satellite data calibration. In this study, we assess the use of the Weather Research and Forecasting (WRF) model to downscaling NCEP Climate Forecast System Version 2 (CFSv2) reanalysis atmospheric profiles. Moreover, a sensitivity analysis was conducted to Yonsei University (YSU) and Mellor–Yamada–Janjic (MYJ) Planetary Boundary Layer (PBL) parameterization schemes. WRF simulations were carried out using two nested grids with horizontal resolutions of 12 and 3 km. Simulated profiles were evaluated against radiosonde observations in Southern Brazil. Results indicate that WRF can skillfully simulate vertical profiles of water vapor mixing ratio (q) and potential temperature (θ). The entire profile comparison yielded high correlation coefficients (mostly higher than 0.9), low moist and cold average biases, and overall RMSE (MAE) of 0.84 (0.44) g/Kg and 2.30 (1.61) K for q and θ, respectively. In the lower troposphere (0–3 km), errors are larger for q and lower for θ. Statistically, there is no significant difference in the retrieved profiles on increasing the horizontal resolution. The setting choice is a compromise between spatial detail needs and computational cost. Furthermore, the overall results did not indicate any particular PBL scheme that stands out in all case days. The performance depends on the meteorological conditions that prevail in the different cases. The WRF model and even data directly from NCEP CFSv2 reanalysis are useful to represent the vertical structure of the atmosphere when local radiosondes are not available.

A modelling framework for a better understanding of the tropically-forced component of the Indian monsoon variability

An experimental technique is introduced for controlling atmospheric diabatic heating in a fully-coupled climate model (NCEP Climate Forecast System version 2), with the motivation of better understanding the role of tropical forcing in modulating the Indian summer monsoon and contributing to global seasonal prediction skill. The ‘added heating’ mechanistic approach has the benefit of not interfering with any internal model feedbacks. The approach is applied in an iterative fashion to correct bias in the three-dimensional tropical heating over the Indo-Pacific, defined as the difference between the model ensemble mean heating and that estimated from ERA-Interim. The June through September seasonal mean, trend and parabolic fit in diabatic heating are corrected for each year separately over a 20-year set of seasonal re-forecasts, resulting in a 60–90% reduction in the mean-squared error of Indo-Pacific heating from an initial set of control re-forecasts, two-thirds of which is associated with the climatological bias. This results in higher skill in the climatological mean and inter-annual variability of local low-level winds and the underlying sea surface temperature (SST). Improvements are most significant over the equatorial Indian Ocean with a removal of the climatological low-level westerly bias and SST gradient bias along with substantially more skill in inter-annual variability. Although there is very little improvement in the model rainfall over India, there is improvement in the low-level circulation. The lack of significant improvement in rainfall prediction may be partly related to the representation of convection and/or the coarse resolution grid of the model.

Reanalysis Profile Downscaling with WRF Model and Sensitivity to PBL Parameterization Schemes Over a Subtropical Station

Atmospheric profiles are essential in weather prediction, atmospheric research, and satellite data calibration. In this study, we assess the use of the Weather Research and Forecasting (WRF) model to downscaling NCEP Climate Forecast System Version 2 (CFSv2) reanalysis atmospheric profiles. Moreover, a sensitivity analysis was conducted to Yonsei University (YSU) and Mellor–Yamada–Janjic (MYJ) Planetary Boundary Layer (PBL) parameterization schemes. WRF simulations were carried out using two nested grids with horizontal resolutions of 12 and 3 km. Simulated profiles were evaluated against radiosonde observations in Southern Brazil. Results indicate that WRF can skillfully simulate vertical profiles of water vapor mixing ratio ( q ) and potential temperature ( θ ). The entire profile comparison yielded high correlation coefficients (mostly higher than 0.9), low moist and cold average biases, and overall RMSE (MAE) of 0.84 (0.44) g/Kg and 2.30 (1.61) K for  q and θ , respectively. In the lower troposphere (0-3 km), errors are larger for q and lower for θ . Statistically, there is no significant difference in the retrieved profiles on increasing the horizontal resolution. The setting choice is a compromise between spatial detail needs and computational cost. Furthermore, the overall results did not indicate any preferred PBL scheme that outperforms in all case days. The performance depends on the meteorological conditions that prevail in the different cases. The WRF model and even data directly from NCEP CFSv2 reanalysis are useful to represent the vertical structure of the atmosphere when local radiosondes are not available.

Cooperative effects of tropical Pacific and Atlantic SST forcing in southern China winter precipitation variability

This study investigates the variability of winter precipitation over southern China (SC) and the impacts of sea surface temperature (SST) forcing in the tropical Pacific and Atlantic Oceans. The SC winter precipitation displays two major spatial modes: a same-sign mode with the maximum center located in southeastern China, and a north–south dipole mode with opposite variations between the lower reach of the Yangtze River and the southeastern coast of China. Both modes are associated with the variation of the western North Pacific (WNP) anticyclone. The same-sign mode is associated with canonical El Nino through the Pacific-East Asian teleconnection in the lower troposphere. The north–south dipole mode is subject to the impact of the western Pacific warm pool associated with the central Pacific El Nino and equatorial Atlantic forcing. The influence of equatorial Atlantic warming is through an upper-tropospheric zonal wave pattern from the tropical Atlantic to the mid-latitudes of East Asia. Both the tropical Pacific and Atlantic oceanic forcing induce a north–south dipole pattern of winter precipitation anomalies over SC by affecting the northward extension of the WNP anticyclone. Moreover, the observed statistical relationship between the dipole mode and the equatorial Atlantic SST anomalies is present in hindcasts/forecasts of the NCEP Climate Forecast System version 2 (CFSv2), confirming a potential source of the predictability of the north–south dipole-like variation of the winter precipitation in SC.

The most predictable patterns and prediction skills of subseasonal prediction of rainfall over the Indo-Pacific regions by the NCEP Climate Forecast System

The most predictable patterns and prediction skills of subseasonal rainfall prediction for the Indo-Pacific regions are investigated using the daily hindcast product of the NCEP Climate Forecast System version 2. A maximum signal-to-noise empirical orthogonal function analysis indicates that on 30–60-day time scale the most predictable patterns exhibit a zonal dipole over the tropical Indian Ocean and the surrounding oceanic areas of the Maritime Continent in boreal winter and spring, associated with the eastward propagation of the Madden–Julian Oscillation. Differently, the most predictable patterns in boreal summer and autumn present a meridional dipole between a tropical belt (from the Arabian Sea, the Bay of Bengal, and the South China Sea to the western Pacific) and an equatorial belt (from the equatorial Indian Ocean to the equatorial Maritime Continent), related to the northward propagation of intraseasonal oscillation. Both the 30–60-day and 10–20-day patterns can be predicted at lead times around one half of their respective longest life cycles, with worst skills in summer. For the winter–spring 30–60-day patterns, the atmospheric drive on oceans plays an important role, and the corresponding winds at both upper and lower levels are consistent with the Gill response, suggesting the importance of the evaporation-wind feedback mechanism. During summer–autumn, cold sea surface temperature anomalies cause a change in land-sea thermal contrast, which is important for monsoon variations.

Analysis of the typhoon wave distribution simulated in WAVEWATCH- III model in the context of Kuroshio and wind-induced current

To investigate the relationship between surface currents and wave distributions in typhoons, we took the Typhoon Talim in 2017 as a case, and found that the track of the typhoon winds up to 50 m/s was almost consistent with the Kuroshio track, particularly from September 13 to 16, 2017. The surface current data, derived from the NCEP Climate Forecast System Version 2 (CFSv2) from the National Center of Atmospheric Research (NCAR), revealed that the speed of the wind-induced current exceeded that of the Kuroshio in the region with the maximum wind speed. In this study, was utilized a third-generation numeric wave model, WAVEWATCH-III (the latest version 5.16), developed by the National Oceanic and Atmospheric Administration (NOAA), to simulate the wave fields of Typhoon Talim using the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis wind data in 0.125°×0.125° grid as the forcing field. We found that the root-mean-square error (RMSE) of the significant wave height (SWH) was 0.34 m when validated against measurements from altimeter Jason-2. In addition, we discovered that the SWH had a similar tendency to the change in the surface current speed that was approximately 0.5 m/s at the beginning of Typhoon Talim. However, the relationship became weak as the surface current speed was below 0.2 m/s. Our findings show that the distribution of typhoon waves is resulted from the interaction of surface current and the wind-sea portion of the wave system, since the distribution pattern of wind-sea is consistent with the surface current, and there is a weak relationship between surface current and swell.

Seasonal Climate Prediction Models for the Number of Landfalling Tropical Cyclones in China

Two prediction models are developed to predict the number of landfalling tropical cyclones (LTCs) in China during June–August (JJA). One is a statistical model using preceding predictors from the observation, and the other is a hybrid model using both the aforementioned preceding predictors and concurrent summer large-scale environmental conditions from the NCEP Climate Forecast System version 2 (CFSv2). (1) For the statistical model, the year-to-year increment method is adopted to analyze the predictors and their physical processes, and the JJA number of LTCs in China is then predicted by using the previous boreal summer sea surface temperature (SST) in Southwest Indonesia, preceding October South Australia sea level pressure, and winter SST in the Sea of Japan. The temporal correlation coefficient between the observed and predicted number of LTCs during 1983–2017 is 0.63. (2) For the hybrid prediction model, the prediction skill of CFSv2 initiated each month from February to May in capturing the relationships between summer environmental conditions (denoted by seven potential factors: three steering factors and four gene-sis factors) and the JJA number of LTCs is firstly evaluated. For the 2- and 1-month leads, CFSv2 has successfully reproduced these relationships. For the 4-, 3-, and 2-month leads, the predictor of geopotential height at 500 hPa over the western North Pacific (WNP) shows the worst forecasting skill among these factors. In general, the summer relative vorticity at 850 hPa over the WNP is a modest predictor, with stable and good forecasting skills at all lead times.

LOW-FREQUENCY CIRCULATION AND EXTENDED-RANGE FORECAST IN CONNECTION WITH AUTUMN EXTREME HEAVY RAINFALL PROCESS IN HAINAN

Observed daily precipitation data from the National Meteorological Observatory in Hainan province and daily data from the National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis-2 dataset from 1981 to 2014 are used to analyze the relationship between Hainan extreme heavy rainfall processes in autumn (referred to as EHRPs) and 10–C30 d low-frequency circulation. Based on the key low-frequency signals and the NCEP Climate Forecast System Version 2 (CFSv2) model forecasting products, a dynamical-statistical method is established for the extended-range forecast of EHRPs. The results suggest that EHRPs have a close relationship with the 10–C30 d low-frequency oscillation of 850 hPa zonal wind over Hainan Island and to its north, and that they basically occur during the trough phase of the low-frequency oscillation of zonal wind. The latitudinal propagation of the low-frequency wave train in the middle-high latitudes and the meridional propagation of the low-frequency wave train along the coast of East Asia contribute to the ‘north high (cold), south low (warm)’ pattern near Hainan Island, which results in the zonal wind over Hainan Island and to its north reaching its trough, consequently leading to EHRPs. Considering the link between low-frequency circulation and EHRPs, a low-frequency wave train index (LWTI) is defined and adopted to forecast EHRPs by using NCEP CFSv2 forecasting products. EHRPs are predicted to occur during peak phases of LWTI with value larger than 1 for three or more consecutive forecast days. Hindcast experiments for EHRPs in 2015–C2016 indicate that EHRPs can be predicted 8–C24 d in advance, with an average period of validity of 16.7 d.

A conceptual prediction model for seasonal drought processes using atmospheric and oceanic standardized anomalies: application to regional drought processes in China

Abstract. Reliable drought prediction is fundamental for water resource managers to develop and implement drought mitigation measures. Considering that drought development is closely related to the spatial–temporal evolution of large-scale circulation patterns, we developed a conceptual prediction model of seasonal drought processes based on atmospheric and oceanic standardized anomalies (SAs). Empirical orthogonal function (EOF) analysis is first applied to drought-related SAs at 200 and 500 hPa geopotential height (HGT) and sea surface temperature (SST). Subsequently, SA-based predictors are built based on the spatial pattern of the first EOF modes. This drought prediction model is essentially the synchronous statistical relationship between 90-day-accumulated atmospheric–oceanic SA-based predictors and SPI3 (3-month standardized precipitation index), calibrated using a simple stepwise regression method. Predictor computation is based on forecast atmospheric–oceanic products retrieved from the NCEP Climate Forecast System Version 2 (CFSv2), indicating the lead time of the model depends on that of CFSv2. The model can make seamless drought predictions for operational use after a year-to-year calibration. Model application to four recent severe regional drought processes in China indicates its good performance in predicting seasonal drought development, despite its weakness in predicting drought severity. Overall, the model can be a worthy reference for seasonal water resource management in China.

Extended-range prediction of South Atlantic convergence zone rainfall with calibrated CFSv2 reforecast

During the 2010–2011 wet season in Brazil, widespread landslides triggered by heavy rainfall killed hundreds of people and displaced nearly 35,000. The extreme precipitation was associated with the formation of the South Atlantic convergence zone (SACZ). Even though the physical mechanisms behind the formation and persistence of subtropical convergence zones are still unclear, we demonstrate that early predictions of heavy rainfall in the SACZ region are possible. Precipitation rate hindcasts from the NCEP Climate Forecast System version 2 are calibrated with the aid of a gridded precipitation dataset. When the calibration was applied to the 2010–2011 events, the hindcasts were able to depict both active and break phases of the SACZ with up to 2 weeks in advance during a period of relatively weak intraseasonal variability associated with the Madden–Julian Oscillation (MJO).

Subseasonal Dynamical Prediction of East Asian Cold Surges

Abstract Predictability of East Asian cold surges is studied using daily data from the hindcasts of 45-day integrations by the NCEP Climate Forecast System version 2 (CFSv2). Prediction skills of the CFSv2 in forecasting cold surges, their annual variation, and their physical links to large-scale atmospheric circulation patterns are examined. Results show that the climatological characteristics of the East Asian winter monsoon can be reasonably reproduced by the CFSv2. The model can well capture the frequency, intensity, and location of cold surges at a lead time of about two weeks. Obviously, fewer-than-observed cold surge days are found in the predictions when the lead time is above 14 days. The spatiotemporal evolutions of high-, mid-, and low-level circulation patterns during cold surge occurrences are all accurately indicated in the CFSv2 prediction. Except for precipitation, the other variables associated with cold surges, such as geopotential height, wind, sea level pressure, and surface air temperature, exhibit higher skills. The lead time of skillful prediction of precipitation is limited to around 1 week, with systematic wet biases over the South China Sea, the Philippine Islands, and the northwest Pacific, but dry biases over India, the Indo-China Peninsula, and most high-latitude regions. Wave train–like patterns of geopotential height and wind differ distinguishably when cold surges occur in northern and southern regions (using 35°N as the dividing line), and the CFSv2 gives a consistent prediction to these anomalous patterns. A weaker-than-observed Siberian high and weaker northerly winds over eastern China are found in the predictions especially at longer lead times.

Seasonal prediction and predictability of Eurasian spring snow water equivalent in NCEP Climate Forecast System version 2 reforecasts

The spring snow water equivalent (SWE) over Eurasia plays an important role in East Asian and Indian monsoon rainfall. This study evaluates the seasonal prediction capability of NCEP Climate Forecast System version 2 (CFSv2) retrospective forecasts (1983–2010) for the Eurasian spring SWE. The results demonstrate that CFSv2 is able to represent the climatological distribution of the observed Eurasian spring SWE with a lead time of 1–3 months, with the maximum SWE occurring over western Siberia and Northeastern Europe. For a longer lead time, the SWE is exaggerated in CFSv2 because the start of snow ablation in CFSv2 lags behind that of the observation, and the simulated snowmelt rate is less than that in the observation. Generally, CFSv2 can simulate the interannual variations of the Eurasian spring SWE 1–5 months ahead of time but with an exaggerated magnitude. Additionally, the downtrend in CFSv2 is also overestimated. Because the initial conditions (ICs) are related to the corresponding simulation results significantly, the robust interannual variability and the downtrend in the ICs are most likely the causes for these biases. Moreover, CFSv2 exhibits a high potential predictability for the Eurasian spring SWE, especially the spring SWE over Siberia, with a lead time of 1–5 months. For forecasts with lead times longer than 5 months, the model predictability gradually decreases mainly due to the rapid decrease in the model signal.