Reduced Tropical Cyclone Genesis in the Future as Predicted by a Machine Learning Model. Issue 2 (31st January 2022)

• Record Type: Journal Article
• Authors: Qian, QiFeng; Jia, XiaoJing; Lin, Yanluan
• Is Part Of: Earth's future. Volume 10:Issue 2(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-01-31
• Subjects: Environmental sciences -- Periodicals; Environmental sciences; Periodicals; 550
• DOI: 10.1029/2021EF002455
• Languages: English
• ISSNs: 2328-4277

Abstract

Abstract: Due to a lack of observations and limited understanding of the complex mechanisms of tropical cyclone (TC) genesis, the possible TC activity response to future climate change remains controversial. In this work, a machine learning model, called the maximum entropy (MaxEnt) model, is established using various environmental variables. The model performs slightly better than the genesis potential index for historical TC activities based on the spatial correlation coefficient. Using coupled model intercomparison project phase 6 model projections, the MaxEnt model predicts a statistically significant decreasing trend of TC genesis probability under all shared socioeconomic pathway scenarios. In addition, our analysis reveals that TC genesis might have a complex nonlinear relationship with potential intensity, which is different from the positive relationship reported in previous studies and might be the key factor leading to the model predicting reduced TC genesis in the future. Plain Language Summary: Machine learning (ML) models show advantages in capturing the complex nonlinear relationship between predictors and predictands and can save much more computational costs than traditional climate models. The current work uses a maximum entropy (MaxEnt) ML model to predict tropical cyclone (TC) genesis, which shows better skills than some traditional methods. The MaxEnt model predicts a statistically significant decreasing TC genesis trend under all shared socioeconomic pathway scenarios. Compared to other variables, the potential intensity (PI) is the most important factor for the MaxEnt ML model. Furthermore, it reveals that, rather than a simple positive relationship, PI shows a complex nonlinear relationship with TC genesis, which is not noticed by previous studies. Key Points: A constructed various environmental variable‐based machine learning maximum entropy (MaxEnt) model performs well in predicting tropical cyclone genesis The MaxEnt model predicts statistically significant decreasing tropical cyclone (TC) genesis trends in the future under all shared socioeconomic pathway scenarios The most important environmental variable in the MaxEnt model is the potential intensity which shows a nonlinear relationship to TC genesis