Identification of tropical cyclones via deep convolutional neural network based on satellite cloud images

Tropical Cyclones (TCs) are one of the most destructive natural disasters. For the prevention and mitigation of TC-induced disasters, real-time monitoring and prediction of TCs is essential. At present, satellite cloud images (SCIs) are utilized widely as a basic data source for such studies. Although great achievements have been made in this field, lack of concerns on the identification of TC fingerprint from SCIs have become a potential issue, since it is a prerequisite step for follow-up analyses. This paper presents a methodology which identifies TC fingerprint via Deep Convolutional Neural Network (DCNN) techniques based on SCIs of more than 200 TCs over the Northwest Pacific basin. Two DCNN models have been proposed and validated, which are able to identify the TCs from not only single-TC featured SCIs but also multi-TCs featured SCIs. Results show that both models can reach 96 % of identification accuracy. As the TC intensity strengthens, the accuracy becomes better. To explore how these models work, heat maps are further extracted and analyzed. Results show that all the fingerprint features are focused on clouds during the testing process. For the majority of TC images, the cloud features in TC’s main parts, i.e., eye, eyewall and primary rainbands, are most emphasized, reflecting a consistent pattern with the subjective method.

Topographic Rossby Waves at Two Different Periods in the Northwest Pacific Basin

To clarify characteristics and mechanisms of mesoscale variability in the deep ocean, we conducted a two-dimensional observation with a 3 × 3 grid mooring array around site R (30°N, 147°E) during 2014–16. We analyze the obtained velocity data together with past mooring observation data in the northwest Pacific basin and outputs of an ocean general circulation model (OGCM). In our two-dimensional mooring observations, the variability of zonal and meridional velocities at a depth of 4000 m was prominent at periods of 174 and 58 days, respectively. The variability at periods of 174 and 58 days propagated to the northwest and west-southwest, respectively, as a single plane wave. The variability at the period of 58 days was considered to be topographic Rossby waves (TRWs) under stratification originated in the Kuroshio Extension region north of site R, as demonstrated by our previous study. At the period of 174 days, zonal and meridional wavenumbers estimated from the phase lag for zonal velocities also satisfied the dispersion relation of TRWs under stratification. Backward ray tracing from site R indicated that energy of TRWs propagated from the eastern slope of the Shatsky Rise to site R almost along f/H contours, where f is the Coriolis parameter and H is water depth. The orientation of major axis of variance ellipses at periods of 174 days and longer, obtained from the past mooring observations and the OGCM outputs, tended to be parallel to f/H contours, being consistent with the direction of energy propagation of TRWs.