Prediction of Tropical Cyclone Genesis from Mesoscale Convective Systems Using Machine Learning

Tropical cyclone (TC) genesis is a problem of great significance in climate and weather research. Although various environmental conditions necessary for TC genesis have been recognized for a long time, prediction of TC genesis remains a challenge due to complex and stochastic processes involved during TC genesis. Different from traditional statistical and dynamical modeling of TC genesis, in this study, a machine learning framework is developed to determine whether a mesoscale convective system (MCS) would evolve into a tropical cyclone. The machine learning models 1) are built upon a number of essential environmental predictors associated with MCSs/TCs, 2) predict whether MCSs can become TCs at different lead times, and 3) provide information about the relative importance of each predictor, which can be conducive to discovering new aspects of TC genesis. The results indicate that the machine learning classifier, AdaBoost, is able to achieve a 97.2% F1-score accuracy in predicting TC genesis over the entire tropics at a 6-h lead time using a comprehensive set of environmental predictors. A robust performance can still be attained when the lead time is extended to 12, 24, and 48 h, and when this machine learning classifier is separately applied to the North Atlantic Ocean and the western North Pacific Ocean. In contrast, the conventional approach based on the genesis potential index can have no more than an 80% F1-score accuracy. Furthermore, the machine learning classifier suggests that the low-level vorticity and genesis potential index are the most important predictors to TC genesis, which is consistent with previous discoveries.

Investigate the relationship between Storm Formation and Tropical Cyclone Genesis Potential Index in the Vietnam East Sea

In this paper, the relationship between Tropical Cyclone (TC) Genesis Potential Index (GPI) and the number of TC (NTC) associated with ENSO over the Vietnam East Sea (VES) was investigated. Observed TC data of the Regional Specialized Meteorological Center (RSMC) Tokyo Typhoon Center and ERA Interim reanalysis data for the period 1985-2015 were used. The results show a good agreement between GPI and NTC over the VES with the correlation coefficient is 0.84. There were more TCs formed over the VES during La Nina years and less TCs during El Nino years. There were positive anomalies of GPI, environmental factors (relative humidity, sea surface temperature, absolute vorticity, potential intensity)over the region where the highest densityof TCs genesis locatedduring La Nina years while there were negative anomalies found during El Nino years. Relative humidity has the largest contribution to the positive difference GPI between La Nina years and El Nino years, the less contribution comes from the potential intensity, absolute vorticity, and wind shear. Keywords: GPI, Tropical Cyclone Genesis, ENSO, Vietnam East Sea. References: [1] K.A. Emanuel, D.S. Nolan, Tropical cyclone activity and global climate, Reprints, 26th Conference on hurricane and Tropical Meteorology, American meteorological Society: Miami, (2004) 240–241.[2] D.S. Nolan, E.D. Rappin, K.A. Emanuel., Tropical cyclogenesis sensitivity to environmental parameters in radiative-convective equilibrium, Quarterly Journal of the Royal Meteorological Society. 133 (2007) 2085–2107.[3] S.J. Camargo, K.A. Emanuel, A.H. Sobel, Use of the Genesis Potential Index to Diagnose ENSO effected on Tropical Cyclone Genesis, American Meteorological Society.20 (2007) 4819-4834[4] C.L. Bruyere, G.J. Holland, E. Towler, Investigating the Used of a Genesis Potential Index for Tropical Cyclones in the North Atlatic Basin, American Meteorological Society..25 (2012) 8611-8626[5] Song Yuan, Wang Lei, Lei Xiaoyan and Wang Xidong, Tropical cyclone genesis potential index over western north Pacific simulated by CMIP5 models, (2015).[6] Lei Wang, Diagnostic of the ENSO modulation of Tropical cyclogenesis over the southern South China Sea using a genesis potential index, Acta Oceanol. Sin., Vol. 31, No. 5 (2012) 54-68.[7] Xin Kieu-Thi, Hang Vu-Thanh, Truong Nguyen-Minh, Duc Le, Linh Nguyen-Manh, Izuru Takayabu, Hidetaka Sasaki, Akio Kito, Rainfall and tropical cyclone activity over Vietnam simulated and projected by the Non-Hydrostatic Regional Climate Model – NHRCM, Journal of the Meteorological Society of Japan. 94A (2016) 135-150.[8] https://www.jma.go.jp/jma/jma-eng/jma-center/ rsmc-hp-pub-eg/trackarchives.html[9] Trần Quang Đức, Xu thế biến động của một số đặc trưng ENSO, Tạp chí Khoa học Đại học Quốc gia Hà Nội, Khoa học Tự nhiên và Công nghệ. 1S (2011) 29-36.[10] https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php[11] E. Palmen, Formation and development of tropical cyclones, Proceedings of tropical cyclone Symposium, Brisbane, Australian Bur. Meteorol., Melbourne, (1956) 213-231[12] M. DeMaria, The effect of vertical wind shear on tropical cyclone intensity change, Jounal of Atmospheric Sicences. 53 (1996) 2076-2087.[13] S.J. Camargo, Diagnosis of the MJO modulation of Tropical cyclogenesis using an empirical index. American Meteorological Society. 66 (2009) 3061-3074.[14] S.J. Camargo, A.H. Sobel, Anthony G. Barnston, K.A. Emanuel, Tropical cyclone genesis potential index in climate models, Tellus A: Dynamic Meteorology and Ocenaography. 59:4 (2007) 428-443. doi: 10.1111/j.1600-0870.2007. 00238.

An Intraseasonal Genesis Potential Index for Tropical Cyclones during Northern Hemisphere Summer

An intraseasonal genesis potential index (ISGPI) for Northern Hemisphere (NH) summer is proposed to quantify the anomalous tropical cyclone genesis (TCG) frequency induced by boreal summer intraseasonal oscillation (BSISO). The most important factor controlling NH summer TCG is found as 500-hPa vertical motion (ω500) caused by the prominent northward shift of large-scale circulation anomalies during BSISO evolution. The ω500 with two secondary factors (850-hPa relative vorticity weighted by the Coriolis parameter and vertical shear of zonal winds) played an effective role globally and for each individual basin in the northern oceans. The relative contributions of these factors to TCG have minor differences by basins except for the western North Atlantic (NAT), where low-level vorticity becomes the most significant contributor. In the eastern NAT, the BSISO has little control of TCG because weak convective BSISO and dominant 10–30-day circulation signal did not match the overall BSISO life cycle. The ISGPI is shown to reproduce realistic intraseasonal variability of TCG, but the performance is phase-dependent. The ISGPI shows the highest fidelity when BSISO convective anomalies have the largest amplitude in the western North Pacific and the lowest when they are located over the north Indian Ocean and eastern North Pacific. Along the NH major TCG zone, the TCG probability changes from a dry to a wet phase by a large factor ranging from 3 to 12 depending on the basins. The new ISGPI for NH summer can simulate more realistic impact of BSISO on TC genesis compared to canonical GPI derived by climatology.