The Forecast Skills and Predictability Sources of Marine Heatwaves in the NUIST-CFS1.0 Hindcasts

Using monthly observations and ensemble hindcasts of the Nanjing University of Information Science and Technology Climate Forecast System(NUIST-CFS1.0) for the period 1983–2020, this study investigates the forecast skill of marine heatwaves(MHWs) over the globe and the predictability sources of the MHWs over the tropical oceans. The MHW forecasts are demonstrated to be skillful on seasonal-annual time scales, particularly in tropical oceans. The forecast skill of the MHWs over the tropical Pacific Ocean(TPO) remains high at lead times of 1–24 months, indicating a forecast better than random chance for up to two years. The forecast skill is subject to the spring predictability barrier of El Nino-Southern Oscillation(ENSO). The forecast skills for the MHWs over the tropical Indian Ocean(TIO), tropical Atlantic Ocean(TAO), and tropical Northwest Pacific(NWP) are lower than that in the TPO. A reliable forecast at lead times of up to two years is shown over the TIO, while a shorter reliable forecast window(less than 17 months) occurs for the TAO and NWP.Additionally, the forecast skills for the TIO, TAO, and NWP are seasonally dependent. Higher skills for the TIO and TAO appear in boreal spring, while a greater skill for the NWP emerges in late summer-early autumn. Further analyses suggest that ENSO serves as a critical source of predictability for MHWs over the TIO and TAO in spring and MHWs over the NWP in summer.

Improvement of wind field hindcasts for tropical cyclones

This paper presents a study on the improvement of wind field hindcasts for two typical tropical cyclones, i.e., Fanapi and Meranti, which occurred in 2010. The performance of the three existing models for the hindcasting of cyclone wind fields is first examined, and then two modification methods are proposed to improve the hindcasted results. The first one is the superposition method, which superposes the wind field calculated from the parametric cyclone model on that obtained from the cross-calibrated multi-platform （CCMP） reanalysis data. The radius used for the superposition is based on an analysis of the minimum difference between the two wind fields. The other one is the direct modification method, which directly modifies the CCMP reanalysis data according to the ratio of the measured maximum wind speed to the reanalyzed value as well as the distance from the cyclone center. Using these two methods, the problem of underestimation of strong winds in reanalysis data can be overcome. Both methods show considerable improvements in the hindcasting of tropical cyclone wind fields, compared with the cyclone wind model and the reanalysis data.

Ensemble Hindcasts of ENSO Events over the Past 120 Years Using a Large Number of Ensembles

Based on an intermediate coupled model （ICM）, a probabilistic ensemble prediction system （EPS） has been developed. The ensemble Kalman filter （EnKF） data assimilation approach is used for generating the initial ensemble conditions, and a linear, first-order Markov-Chain SST anomaly error model is embedded into the EPS to provide model-error perturbations. In this study, we perform ENSO retrospective forecasts over the 120 year period 1886-2005 using the EPS with 100 ensemble members and with initial conditions obtained by only assimilating historic SST anomaly observations. By examining the retrospective ensemble forecasts and available observations, the verification results show that the skill of the ensemble mean of the EPS is greater than that of a single deterministic forecast using the same ICM, with a distinct improvement of both the correlation and root mean square （RMS） error between the ensemble-mean hindcast and the deterministic scheme over the 12-month prediction period. The RMS error of the ensemble mean is almost 0.2℃ smaller than that of the deterministic forecast at a lead time of 12 months. The probabilistic skill of the EPS is also high with the predicted ensemble following the SST observations well, and the areas under the relative operating characteristic （ROC） curves for three different ENSO states （warm events, cold events, and neutral events） are all above 0.55 out to 12 months lead time. However, both deterministic and probabilistic prediction skills of the EPS show an interdecadal variation. For the deterministic skill, there is high skill in the late 19th century and in the middle-late 20th century （which includes some artificial skill due to the model training period）, and low skill during the period from 1906 to 1961. For probabilistic skill, for the three different ENSO states, there is still a similar interdecadal variation of ENSO probabilistic predictability during the period 1886~2005. There is high skill in the late 19th century from 1886 to 1905, and a decline to a minimum of skill around 1910-50s, beyond which skill rebounds and increases with time until the 2000s.

Sub-seasonal to Seasonal Hindcasts of Stratospheric Sudden Warming by BCC_CSM1.1(m):A Comparison with ECMWF

This study focuses on model predictive skill with respect to stratospheric sudden warming(SSW) events by comparing the hindcast results of BCC_CSM1.1(m) with those of the ECMWF's model under the sub-seasonal to seasonal prediction project of the World Weather Research Program and World Climate Research Program. When the hindcasts are initiated less than two weeks before SSW onset, BCC_CSM and ECMWF show comparable predictive skill in terms of the temporal evolution of the stratospheric circumpolar westerlies and polar temperature up to 30 days after SSW onset. However, with earlier hindcast initialization, the predictive skill of BCC_CSM gradually decreases, and the reproduced maximum circulation anomalies in the hindcasts initiated four weeks before SSW onset replicate only 10% of the circulation anomaly intensities in observations. The earliest successful prediction of the breakdown of the stratospheric polar vortex accompanying SSW onset for BCC_CSM(ECMWF) is the hindcast initiated two(three) weeks earlier. The predictive skills of both models during SSW winters are always higher than that during non-SSW winters, in relation to the successfully captured tropospheric precursors and the associated upward propagation of planetary waves by the model initializations. To narrow the gap in SSW predictive skill between BCC_CSM and ECMWF, ensemble forecasts and error corrections are performed with BCC_CSM. The SSW predictive skill in the ensemble hindcasts and the error corrections are improved compared with the previous control forecasts.

Wave hindcast under tropical cyclone conditions in the South China Sea:sensitivity to wind fields

Reliable wave information is critical for marine engineering.Numerical wave models are useful tools to obtain wave information with continuous spatiotemporal distributions.However,the accuracy of model results highly depends on the quality of wind forcing.In this study,we utilize observations from five buoys deployed in the northern South China Sea from August to September 2017.Notably,these buoys successfully recorded wind field and wave information during the passage of five tropical cyclones of different intensities without sustaining any damage.Based on these unique observations,we evaluated the quality of four widely used wind products,namely CFSv2,ERA5,CCMP,and ERAI.Our analysis showed that in the northern South China Sea,ERA5 performed best compared to buoy observations,especially in terms of maximum wind speed values at 10 m height(U10),extreme U10 occurrence time,and overall statistical indicators.CFSv2 tended to overestimate non-extreme U10 values.CCMP showed favorable statistical performance at only three of the five buoys,but underestimated extreme U10 values at all buoys.ERAI had the worst performance under both normal and tropical cyclone conditions.In terms of wave hindcast accuracy,ERA5 outperformed the other reanalysis products,with CFSv2 and CCMP following closely.ERAI showed poor performance especially in the upper significant wave heights.Furthermore,we found that the wave hindcasts did not improve with increasing spatiotemporal resolution,with spatial resolution up to 0.5°.These findings would help in improving wave hindcasts under extreme conditions.

Multi-Year Simulations and Experimental Seasonal Predictions for Rainy Seasons in China by Using a Nested Regional Climate Model (RegCM_NCC) Part Ⅱ:The Experimental Seasonal Prediction

A nested regional climate model has been experimentally used in the seasonal prediction at the China National Climate Center （NCC） since 2001. The NCC/IAP （Institute of Atmospheric Physics） T63 coupled GCM （CGCM） provides the boundary and initial conditions for driving the regional climate model （RegCM_NCC）. The latter has a 60-km horizontal resolution and improved physical parameterization schemes including the mass flux cumulus parameterization scheme, the turbulent kinetic energy closure scheme （TKE） and an improved land process model （LPM）. The large-scale terrain features such as the Tibetan Plateau are included in the larger domain to produce the topographic forcing on the rain-producing systems. A sensitivity study of the East Asian climate with regard to the above physical processes has been presented in the first part of the present paper. This is the second part, as a continuation of Part Ⅰ. In order to verify the performance of the nested regional climate model, a ten-year simulation driven by NCEP reanalysis datasets has been made to explore the performance of the East Asian climate simulation and to identify the model＇s systematic errors. At the same time, comparative simulation experiments for 5 years between the RegCM2 and RegCM_NCC have been done to further understand their differences in simulation performance. Also, a ten-year hindcast （1991-2000） for summer （June-August）, the rainy season in China, has been undertaken. The preliminary results have shown that the RegCM_NCC is capable of predicting the major seasonal rain belts. The best predicted regions with high anomaly correlation coefficient （ACC） are located in the eastern part of West China, in Northeast China and in North China, where the CGCM has maximum prediction skill as well. This fact may reflect the importance of the largescale forcing. One significant improvement of the prediction derived from RegCM_NCC is the increase of ACC in the Yangtze River valley where the CGCM has a very low, even a negative, ACC. The reason behind this improvement is likely to be related to the more realistic representation of the large-scale terrain features of the Tibetan Plateau. Presumably, many rain-producing systems may be generated over or near the Tibetan Plateau and may then move eastward along the Yangtze River basin steered by upper-level westerly airflow, thus leading to enhancement of rainfalls in the mid and lower basins of the Yangtze River. The real-time experimental predictions for summer in 2001, 2002, 2003 and 2004 by using this nested RegCM-NCC were made. The results are basically reasonable compared with the observations.

Improved ENSO Forecasts by Assimilating Sea Surface Temperature Observations into an Intermediate Coupled Model

A simple method for initializing intermediate coupled models （ICMs） using only sea surface temperature （SST） anomaly data is comprehensively tested in two sets of hindcasts with a new ICM. In the initialization scheme, both the magnitude of the nudging parameter and the duration of the assimilation are considered, and initial conditions for both atmosphere and ocean are generated by running the coupled model with SST anomalies nudged to the observations. A comparison with the observations indicates that the scheme can generate realistic thermal fields and surface dynamic fields in the equatorial Pacific through hindcast experiments. An ideal experiment is performed to get the optimal nudging parameters which include the nudging intensity and nudging time length. Twelve-month-long hindcast experiments are performed with the model over the period 1984-2003 and the period 1997-2003. Compared with the original prediction results, the model prediction skills are significantly improved by the nudging method especially beyond a 6-month lead time during the two different periods. Potential problems and further improvements are discussed regarding the new coupled assimilation system.

Predictability of the East Asian Winter Monsoon Interannual Variability as Indicated by the DEMETER CGCMS

The interannual variability of East Asian winter monsoon （EAWM） circulation from the Development of a European Multi-Model Ensemble （MME） System for Seasonal to Inter-Annual Prediction （DEMETER） hindcasts was evaluated against observation reanalysis data. We evaluated the DEMETER coupled general circulation models （CGCMs）＇ retrospective prediction of the typical EAWM and its associated atmospheric circulation. Results show that the EAWM can be reasonably predicted with statistically significant accuracy, yet the major bias of the hindcast models is the underestimation of the related anomalies. The temporal correlation coefficient （TCC） of the MME-produced EAWM index, defined as the first EOF mode of 850- hPa air temperature within the EAWM domain （20^-60~N, 90^-150~E）, was 0.595. This coefficient was higher than those of the corresponding individual models （range： 0.39-0.51） for the period 1969 2001; this result indicates the advantage of the super-ensemble approach. This study also showed that the ensemble models can reasonably reproduce the major modes and their interannual variabilities for sea level pressure, geopotential height, surface air temperature, and wind fields in Eurasia. Therefore, the prediction of EAWM interannual variability is feasible using multimodel ensemble systems and that they may also reveal the associated mechanisms of the EAWM interannual variability.

Evaluating the Accuracy of ERA5 Wave Reanalysis in the Water Around China

Wave parameters, such as wave height and wave period, are important for human activities, such as navigation, ocean engineering and sediment transport, etc. In this study, wave data from six buoys around Chinese waters, are used to assess the quality of wave height and wave period in the ERA5 reanalysis of the European Centre for Medium-Range Weather Forecasts. Annual hourly data with temporal resolution are used. The difference between the significant wave height(SWH) of ERA 5 and that of the buoy varies from-0.35 m to 0.30 m for the three shallow locations;for the three deep locations, the variation ranges from-0.09 m to 0.09 m. The ERA5 SWH data show positive biases, indicating an overall overestimation for all locations, except for E2 and S1 where underestimation is observed. During the tropical cyclone period, a large(about 32%) underestimation of the maximum SWH in the ERA5 data is observed. Hence, the ERA5 SWH data cannot be used for design applications without site-specific validation. The difference between the annual wave period from ERA5 and the mean wave period from the buoys varies from-1.31 s to 0.4 s. Inter-comparisons suggest that the ERA5 dataset is consistent with the annual mean SWH. However, for the average period, the performance is not good, and half of the correlation coefficients in the four points are less 50%. Overall, the deep water area simulation effect is better than that in the shallow water.

The spring prediction barrier in ENSO hindcast experiments using the FGOALS-g model

The Flexible Global Ocean-Atmosphere-Land System Model-gamil (FGOALS-g) was used to study the spring prediction barrier (SPB) in an ensemble system. This coupled model was developed and maintained at the State Key Laboratory of Atmospheric Sciences and Geophysical Fluid Dynamics (LASG). There are two steps in our hindcast experiments. The first is to integrate the coupled model continuously with sea surface temperature (SST) nudging, from 1971 to 2006. The second is to carry out a series of one-year hindcasts without SST nudging, by adopting initial values from the first step on January 1 st , April 1st , July 1st , and October 1st , from 1982 to 2005. We generate 10 ensemble members for a particular start date (1st ) by choosing different atmospheric and land conditions around the hindcast start date (1st through 10th ). To estimate the predicted SST, two methods are used: (1) Anomaly Correlation Coefficient and its rate of decrease; and (2) Talagrand distribution and its standard deviation. Results show that FGOALS-g offers a reliable ensemble system with realistic initial atmospheric and oceanic conditions, and high anomaly correlation (>0.5) within 6 month lead time. Further, the ensemble approach is effective, in that the anomaly correlation of ensemble mean is much higher than that of most individual ensemble members. The SPB exists in the FGOALS-g ensemble system, as shown by anomaly correlation and equal likelihood. Nevertheless, the role of the ensemble mean in reducing the SPB of ENSO prediction is significant. The rate of decrease of the ensemble mean is smaller than the largest deviations by 0.04-0.14. At the same time, the ensemble system "equal likelihood" declines during spring. An ensemble mean helps give a correct prediction direction, departing from largely-deviated ensemble members.

A hindcast of the Bohai Bay oil spill during June to August 2011

An operational three-dimensional oil spill model is developed by the National Marine Environmental Forecasting Center（NMEFC）, State Oceanic Administration, China, and the model has been running for 9 a. On June 4 and 17,2011, oil is spilled into the sea water from two separate oil platforms in the Bohai Bay, i.e., Platforms B and C of Penglai 19-3 oilfield. The spill causes pollution of thousands of square kilometres of sea area. The NMEFC’s oil spill model is employed to study the Penglai 19-3 oil-spill pollution during June to August 2011. The wind final analysis data of the NMEFC, which is based on a weather research and forecasting（WRF） model, are analyzed and corrected by comparing with the observation data. A corrected current filed is obtained by forcing the princeton ocean model（POM） with the corrected wind field. With the above marine environmental field forcing the oil spill model, the oil mass balance and oil distribution can be produced. The simulation is validated against the observation, and it is concluded that the oil spill model of the NMEFC is able to commendably simulate the oil spill distribution. Thus the NMEFC’s oil spill model can provide a tool in an environmental impact assessment after the event.

Interannual Variability and Scenarios Projection of Sea Ice in Bohai Sea Part Ⅰ: Variation Characteristics and Interannual Hindcast

The Bohai Sea is one of the southernmost areas for sea ice formation in the northern hemisphere.Sea ice disasters in this body of water severely affect marine activities and the safety of coastal residents.In this study,we analyze the variation characteristics of the sea ice in the Bohai Sea and establish an annual regression model based on predictable mode analysis method.The results show the following:1)From 1970 to 2018,the average ice grade is(2.6±0.8),with a maximum of 4.5 and a minimum of 1.0.Liaodong Bay(LDB)has the heaviest ice conditions in the Bohai Sea,followed by Bohai Bay(BHB)and Laizhou Bay(LZB).Interannual variation is obvious in all three bays,but the linear decreasing trend is significant only in BHB.2)Three modes are obtained from empirical orthogonal function analysis,namely,single polarity mode with the same sign of anomaly in all of the three bays and strong interannual variability(82.0%),the north–south dipole mode with BHB and LZB showing an opposite sign of anomalies to that in LDB and strong decadal variations(14.5%),and a linear trend mode(3.5%).Critical factors are analyzed and regression equations are established for all the principal components,and then an annual hindcast model is established by synthesizing the results of the three modes.This model provides an annual spatial prediction of the sea ice in the Bohai Sea for the first time,and meets the demand of operational sea ice forecasting.

Wind-wave hindcast in the Yellow Sea and the Bohai Sea from the year 1988 to 2002

We performed long-term wind-wave hindcast in the Yellow Sea and the Bohai Sea from the year 1988 to 2002, and then analyzed the regional wave climate. Comparisons between model results and satellite data are generally consistent on monthly mean significant wave height. Then we discuss the temporal and spatial characteristics of the climatological monthly mean significant wave heights and mean wave periods. The climatologically spatial patterns are observed as increasing from northwest to southeast and from offshore to deep-water area for both significant wave height and mean wave period, and the patterns are highly related to the wind forcing and local topography. Seasonal variations of wave parameters are also significant. Furthermore, we compute the extreme values of wind and significant wave height using statistical methods. Results reveal the spatial patterns of N-year return significant wave height in the Yellow Sea and the Bohai Sea, and we discuss the relationship between extreme values of significant wave height and wind forcing.

Preliminary assessment on the hindcast skill of the Arctic Oscillation with decadal experiment by the BCC_CSM1.1 climate model

The prediction skill of Arctic Oscillation (AO) in the decadal experiments with the Beijing Climate Center Climate System Model version 1.1 (BCC_CSM1.1) is assessed. As compared with the observations and historical experiments, the contribution of initialization for climate model to predict the seasonal scale AO and its interannual variations is estimated. Results show that the spatial correlation coefficient of AO mode simulated by the decadal experiment is higher than that in the historical experiment. The two groups of experiments reasonably reproduce the characteristics that AO indices are the strongest in winter and the weakest in summer. Compared with historical experiments, the correlation coefficient of the monthly and winter AO indices are higher in the decadal experiments. In particular, the correlation coefficient of monthly AO index between decadal hindcast and observation reached 0.1 significant level. Furthermore, the periodicity of the monthly and spring AO indices are achieved only in the decadal experiments. Therefore, the initial state of model is initialized by using sea temperature data may help to improve the prediction skill of AO in the decadal prediction experiments to some extent.

Potential Predictability of Sea Surface Temperature in a Coupled Ocean-Atmosphere GCM

Using the Flexible Global Ocean-Atmosphere-Land System model （FGOALS） version g1.11, a group of seasonal hindcasting experiments were carried out. In order to investigate the potential predictability of sea surface temperature （SST）, singular value decomposition （SVD） analyses were applied to extract dominant coupled modes between observed and predicated SST from the hindcasting experiments in this study. The fields discussed are sea surface temperature anomalies over the tropical Pacific basin （20~0S-20~0N, 120~0E- 80~0W）, respectively starting in four seasons from 1982 to 2005. On the basis of SVD analysis, the simulated pattern was replaced with the corresponding observed pattern to reconstruct SST anomaly fields to improve the ability of the simulation. The predictive skill, anomaly correlation coefficients （ACC）, after systematic error correction using the first five modes was regarded as potential predictability. Results showed that： 1） the statistical postprocessing approach was effective for systematic error correction; 2） model error sources mainly arose from mode 2 extracted from the SVD analysis-that is, during the transition phase of ENSO, the model encountered the spring predictability barrier; and 3） potential predictability （upper limits of predictability） could be high over most of the tropical Pacific basin, including the tropical western Pacific and an extra 10-degrees region of the mid and eastern Pacific.

Interannual variability of dimethylsulfide in the Yellow Sea

The 22-year(1998-2019)surface seawater dimethylsulfi de(DMS)concentrations in the Yellow Sea(YS)were hindcasted based on satellite sea surface temperature(SST)and chlorophyll-a(Chl-a)data using a generalized additive mixed model(GAMM).A continuous monthly dataset of DMS concentration in the YS was obtained after using the data interpolation empirical orthogonal function(DINEOF)to reconstruct missing information in the dataset.Then,the interannual DMS variability in the YS was analyzed.The results indicated that the monthly climatological DMS concentration in the YS was 3.61 nmol/L.DMS concentrations in the spring and summer were signifi cantly higher than those in the autumn and winter.DMS concentrations were highest in coastal YS waters and lowest primarily in off shore YS waters.Interannual DMS variability between 1998 and 2019 was subdivided into two inverse phases:with the exception of the central YS,DMS increased before the turning point and decreased after.The turning point in interannual DMS variation was earlier in the inshore YS as compared to the central YS.Spectrum analysis identifi ed some signifi cant patterns of interannual variation in the DMS anomaly in the YS.Chl a appeared to be the main factor infl uencing interannual trends in DMS in the YS.Interannual DMS variability was under the joint control of Chl a and SST.However,short-term interannual DMS variation(2-3 years)was primarily related to SST,while longer term interannual DMS variation(6-8 years)was signifi cantly correlated with Chl a and SST.

ENSO Hindcast Experiments Using a Coupled GCM

A group of seasonal hindcast experiments are conducted using a coupled model known as the Flexible Global Ocean-Atmosphere-Land System Modelgamil1.11 (FGOALS-g1.11) developed at the State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics (LASG).Two steps are included in our ElNi o-Southern Oscillation (ENSO) hindcast experiments.The first step is to integrate the coupled GCM with the Sea Surface Temperature (SST) strongly nudged towards the observation from 1971 to 2006.The second step is to remove the SST nudging term.The authors carried out a one-year hindcast by adopting the initial values from SST nudging experiments from the first step on January 1st,April 1st,July 1st,and October 1st from 1982 to 2005.In the SST nudging experiment,the model can reproduce the observed equatorial thermocline anomalies and zonal wind stress anomalies in the Pacific,which demonstrates that the SST nudging approach can provide realistic atmospheric and oceanic initial conditions for seasonal prediction experiments.The model also demonstrates a high Anomaly Correlation Coefficient (ACC) score for SST in most of the tropical Pacific,Atlantic Ocean,and some Indian Ocean regions with a 3-month lead.Compared with the persistence ACC score,this model shows much higher ACC scores for the Ni o-3.4 index for a 9-month lead.

Wave Hindcast for the Neighbouring Seas of Korea Based on Loosely Coupled Wave-Tide-Surge Model

A hindcast simulation of 75 typhoons and winter monsoons which affected the coastal areas of Korean Peninsula is performed by use of a third generation ocean wave prediction model, WAM-cycle 4 model, loosely coupled with a com-bined tide and surge model. Typhoon wind fields are derived from the planetary marine boundary layer model for effective neutral winds embedding the vortical storm wind from the parameterized Rankin vortex type model in the limited areas of the overall modeled region. The hindcasted results illustrate that significant wave heights (SWH) considering the wave-tide-surge coupled process are significantly different from the results via the decoupled case especially in the region of the estuaries of the Changjiang Estuary, The Hangzhou Bay, and the southwestern tip of Korean Peninsula. This extensive model simulation is the first attempt to investigate the strong wave-tide-surge interaction for the shallow depth area along the coasts of the Yellow Sea and the East China Sea Continental shelf.

Relationship between coral growth rate and sea surface temperature in the northern part of South China Sea during the past 100 a

To research the relationship between coral growth rate and sea surface temperature ( Tss), 5 cores of living Porites lutea were collected from the Xisha Islands and the southern Hainan Island waters and measured. The results of the study show that there is an obviously positive correlation between the coral growth rates and the Tss records from the northern part of South China Sea. The annual growth rates of the five samples of Porites lutea during the past 100 a are in the range of 7-15 mm/a, and their mean value is 11 mm/a. The correlation coefficients between the coral growth rates and the Tss records from the waters during 1961-1993 are in the range of 0. 77-0.89. As a result, a thermometer of the coral growth rate is established. A hindcasting Tss in the waters from 1993 to 1961 has been obtained with an error of about 0.12-0.17℃ . Based upon the calculated result, the rising rate of Tss in the northern part of South China Sea during the past 100 a is 0. 20℃ , which is higher than that of the air temperature in China (0.09℃/100 a), but lower than that of the global temperature and that of Tss in the tropical western Pacific Ocean.

Prediction model for spring dust weather frequency in North China

It is of great social and scientific importance and also very difficult to make reliable prediction for dust weather frequency (DWF) in North China. In this paper, the correlation between spring DWF in Beijing and Tianjin observation stations, taken as examples in North China, and seasonally averaged surface air temperature, precipitation, Arctic Oscillation, Antarctic Oscillation, South Oscillation, near surface meridional wind and Eurasian westerly index is respectively calculated so as to construct a prediction model for spring DWF in North China by using these climatic factors. Two prediction models, i.e. model-I and model-II, are then set up respectively based on observed climate data and the 32-year (1970 -2001) extra-seasonal hindcast experiment data as reproduced by the nine-level Atmospheric General Circulation Model developed at the Institute of Atmospheric Physics (IAP9L-AGCM). It is indicated that the correlation coefficient between the observed and predicted DWF reaches 0.933 in the model-I, suggesting a high prediction skill one season ahead. The corresponding value is high up to 0.948 for the subsequent model-II, which involves synchronous spring climate data reproduced by the IAP9L-AGCM relative to the model-I. The model-II can not only make more precise prediction but also can bring forward the lead time of real-time prediction from the model-I’s one season to half year. At last, the real-time predictability of the two models is evaluated. It follows that both the models display high prediction skill for both the interannual variation and linear trend of spring DWF in North China, and each is also featured by different advantages. As for the model-II, the prediction skill is much higher than that of original approach by use of the IAP9L-AGCM alone. Therefore, the prediction idea put forward here should be popularized in other regions in China where dust weather occurs frequently.