A dynamical‐statistical forecast model for the annual frequency of western Pacific tropical cyclones based on the NCEP Climate Forecast System version 2

Abstract

A dynamical‐statistical forecast model for the annual tropical cyclones over the western North Pacific is developed based on the empirical relationship between the actual annual number of tropical cyclones (ANTCs) and the dynamical predictions of large‐scale variables by the Climate Forecast System version 2 of the National Centers for Environmental Prediction (NCEP). On interannual time scales, the ANTCs are significantly and negatively correlated with the July–October tropical North Atlantic sea surface temperature, tropical western Pacific vertical zonal wind shear (WPVZWS), and subtropical Pacific geopotential height at 500 hPa (HGT500). They are also positively correlated with the zonal wind at 850 hPa over the tropical Pacific Ocean. Skillful forecasts of the above four potential predictors are made with the 24‐member ensemble predictions by the NCEP model. The two‐predictor model with the HGT500 and the WPVZWS shows the most skillful hindcasts at 0–2 month leads assessed by the leave‐one‐out cross validation for the ANTCs over the 31 year record between 1982 and 2012. The corresponding correlation coefficients and the root‐mean‐square errors (RMSEs) between the observed and hindcast ANTCs are in the ranges from 0.73 to 0.79 and from 3.11 to 2.75, respectively. Observed ANTCs during El Niño–Southern Oscillation events are generally well captured with RMSEs ranging from 3.12 to 3.04 during El Niño years and from 3.62 to 2.44 during La Niña years. The forecast skill of the model for the past 10 years (2003–2012) is competitive with the current forecast schemes. The forecast model initialized in March, May, and June 2013 suggests an inactive season for 2013, with about 22 tropical cyclones.

Statistical‐dynamical seasonal forecast of western North Pacific and East Asia landfalling tropical cyclones using the high‐resolution GFDL FLOR coupled model

“…[] (Figure ). Two regions for SST related to cluster 2 are located in the tropical Pacific (La Niña‐like pattern) [ Wang and Chan , ; Camargo and Sobel , ] and North Atlantic [ Li et al ., ; Huo et al ., ; Yu et al ., ]. The SST region related to cluster 3 is located in the North Atlantic with the spatial pattern similar to the Atlantic tripole mode [ Deser and Timlin , ], although the correlations in the tropical Atlantic are not statistically significant.…”

“…The SST region related to cluster 3 is located in the North Atlantic with the spatial pattern similar to the Atlantic tripole mode [ Deser and Timlin , ], although the correlations in the tropical Atlantic are not statistically significant. The association between the Atlantic tripole mode and WNP TC activity could be supported by previous studies [ Li et al ., ; Huo et al ., ; Yu et al ., ]. The Atlantic Meridional Mode strongly modulates WNP TC activity in both observations and a suite of FLOR‐FA experiments; negative (positive) AMM phases are conducive to more (less) WNP TCs [ Zhang et al ., c].The correlation between Atlantic tripole mode index and Atlantic Meridional Mode index is significant at 0.01 level (value of 0.75) for the period 1961–2013 in the observations.…”

“…It is noted that the spatial correlation maps of SST and Z500 with WNP TC frequency are nearly opposite to one another for clusters 4 and 5 (Figure ), reflecting different phases of the climate oscillation (e.g., ENSO and AMM). The SST, Z500, and VWS [ Chan et al ., ; Li et al ., ] domains of cluster 6 are typical of El Niño years [ Wang and Chan , ; Camargo and Sobel , ], concurrent with SST regions in the Indian ocean [e.g., Wang et al ., ; Zhan et al , ] and North Atlantic [ Li et al ., ; Huo et al ., ; Yu et al ., ]. TC geneses in cluster 6 (Figure ) are located southeast of those in cluster 4, consistent with the southeast‐northwest pattern of TC genesis during El Niño/La Niña years [ Wang and Chan , ; Camargo and Sobel , ].…”

“…To complement statistical models, state‐of‐the‐art climate models, including atmospheric general circulation models (AGCMs) and coupled GCMs (CGCMs) have shown encouraging skill in seasonal forecasting of TC frequency [e.g., Vitart et al ., ; Smith et al ., ; Zhao et al ., ; Chen and Lin , ; Vecchi et al ., ; Camp et al ., ], and landfall over East Asia [e.g., Sun and Ahn , ; Au‐Yeung and Chan , ; Huang and Chan , ; Tan et al ., ]. Over the years, hybrid models combining dynamic and statistical models, have proven highly skillful in the prediction of TCs in the North Atlantic and the WNP [e.g., Klotzbach , ; Elsner et al ., ; Vecchi et al ., 2010; LaRow , ; Villarini and Vecchi , ; Li et al ., ; Vecchi et al ., ; Kim et al ., ]. Huang and Chan [2012] reported that the dynamically downscaled forecasts of TC landfall over East Asia outperform the original CFS2 prediction with a correlation of 0.27 and a RMSE of 3.72, while that for TC frequency is 0.55 with a leave‐one‐out cross‐validation.…”

“…Sun and Ahn [] reported that the coupled climate model (PNU‐CGCM) exhibited virtually no forecasting ability in predicting 6–9 month and 3–5‐month leads for landfalling TCs over East Asia. Based on statistical relationships between TC frequency/landfall and physical processes tied to dynamical models, the so‐called hybrid dynamical‐statistical models have been capable of predicting seasonal TC activity over the North Atlantic and the WNP [e.g., Wang et al ., ; Kim and Webster , ; Vecchi et al ., 2010; Sun and Chen , ; Li et al ., ; Villarini and Vecchi , ; Vecchi et al ., ; Kim et al ., ; Vecchi and Villarini et al , ]. Hybrid models take advantage of both statistical associations derived from the observations and dynamic models, and environmental variables simulated by dynamic models to improve the skill of the prediction.…”