Modeling western North Pacific tropical cyclones associated with ENSO

September 7, 2016
ENSO and tropical cyclones in the Pacific
Composites of tropical cyclone track density anomaly for eastern Pacific El Niño (1982, 1986, 1991, 1997, 1nd 2006; left), central Pacific El Niño (1987, 1994, 2002, 2004, and 2009; central), and La Niña (1984, 1985, 1988, 1995, 1998, 1999, 2000 and 2007; right) during 1982–2009 from observations (top) and multi-model ensemble mean (bottom). The white contours represent the 95% significance level estimated by the Monte Carlo test. 

The western North Pacific is the most active basin of the global oceans with the highest annual number of tropical cyclones. It is well known that the El Niño–Southern Oscillation (ENSO) strongly affects the interannual variability of these tropical cyclones. Previous observational studies show that western North Pacific tropical cyclone activity is above normal during El Niño years and below normal during La Niña. It is also found that there are differences in the tropical cyclone activity between eastern Pacific El Niño and central Pacific El Niño. In assessing the feasibility of dynamical seasonal forecast of tropical cyclones, it would be relevant to know whether state-of-art global climate models can simulate the western North Pacific tropical cyclone response to different types of ENSO.

With a primary focus on modeling studies of tropical cyclones, the US CLIVAR Hurricane Working Group initiated a set of simulations with five high-resolution atmospheric global climate modelsforced with the observed time-varying sea surface temperature from 1982 to 2009. A recent paper by Han et al. evaluates the performance of the five individual models, as well as a multi-model ensemble mean, in simulating the interannual variability of western North Pacific tropical cyclones associated with ENSO. 

What they found is that for the annual tropical cyclone counts, the multi-model ensemble mean has the closest long-term mean value and the smallest root-mean-square error compared to the observations. Most models can reproduce the interannual variability of these cyclones well, with stronger activity during El Nino and weaker activity during La Niña. However, none of the models capture the distinctions between eastern Pacific and central Pacific El Niño as found in observations. The incapability of the models in distinguishing the differences in tropical cyclone activity between the two types of El Nino is likely due to the model biases in response to the shift of tropical heating associated with eastern Pacific and central Pacific El Niño.

Written by 
Hui Wang and Arun Kumar, NOAA Climate Prediction Center

Rongqing Han1, Hui Wang2, Zeng-Zhen Hu2, Arun Kumar2, Weijing Li1, Lindsey N. Long2,3, Jae-Kyung E. Schemm2, Peitao Peng2, Wanqiu Wang2, Dong Si1, Xiaolong Jia1, Ming Zhao4, Gabriel A. Vecchi4, Timothy E. LaRow5, Young-Kwon Lim6,7, Siegfried D. Schubert7, Suzana J. Camargo8, Naomi Henderson8, Jeffrey A. Jonas9, and Kevin J. E. Walsh10

1National Climate Center, China Meteorological Administration, Beijing, China

2NOAA/NWS/NCEP/Climate Prediction Center

3Innovim

4NOAA/Geophysical Fluid Dynamics Laboratory

5Florida State University

6Goddard Earth Sciences Technology and Research, I. M. Systems Group

7Global Modeling and Assimilation Office, NASA Goddard Space Flight Center

8Lamont-Doherty Earth Observatory, Columbia University

9Columbia University and NASA Goddard Institute for Space Studies

10University of Melbourne, Victoria, Australia