Search NASA⌕ Search

Engineering topics

Patrick Duran

Publications and source records attributed to Patrick Duran.

22 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Objectively Identifying Transverse Cirrus Bands in Tropical Cyclones using a Convolutional Neural Network

Transverse cirrus bands (TCBs) are bands of upper-level clouds that are regularly seen in mesoscale and synoptic-scale weather systems. In tropical cyclones, their appearance has been subjectively linked to intensification and the diurnal cycle, but these hypothesized relationships have not been rigorously tested due to the difficulty of objectively identifying TCBs in satellite images. This presentation describes a machine learning technique that successfully identifies TCBs in imagery from the GOES-16 Advanced Baseline Imager. The technique uses a convolutional neural network (CNN) that assigns a probability of each pixel in the image being associated with a TCB. Using the CNN, a database of TCBs from 2018 to 2022 was developed for the Atlantic basin. Statistics using this database will be presented, including the relationship between TCBs and deep-layer vertical wind shear, intensity change, and the TC diurnal cycle.

John Mark Mayhall↗
NASA NTRSDate not supplied

Relating Lightning Flash Characteristics to Convective Structure in Tropical Cyclones

For decades, lightning has been used with some success as an indicator of tropical cyclone (TC) intensity change. An increase in inner-core lightning is usually associated with TC intensification, but this is not always the case. Significant lightning outbreaks have been observed in weakening storms, and hurricanes can rapidly intensify in the total absence of inner-core lightning. These conflicting results arise because lightning production depends on convective and microphysical characteristics that can arise during both intensification and weakening phases in TCs. Recent work using flash size and energy metrics provided by the Geostationary Lightning Mapper (GLM) aboard the GOES-R series of satellites shows some promise in distinguishing between lightning outbreaks that correspond to intensification and those that correspond to weakening. If a hurricane is rapidly intensifying in the presence of inner-core lightning, the size and optical energy of the flashes generally increases as intensification proceeds. It is hypothesized that this relationship is governed by changes in the TC-scale circulation that produce enhanced upper-level outflow and large, strongly helical convective updrafts in the inner core. These changes increase the quantity and spatial extent of charged upper-level ice, which allows large intracloud flashes to occur when electric field breakdown occurs. This presentation will present objective diagnostics of convective asymmetry using azimuthal wavenumber decomposition of GOES-R Advanced Baseline Imager brightness temperatures. The degree of symmetry will be related to GLM flash characteristics to better understand how changes in the symmetry of TC inner-core convection relates to flash rate, size, and energy. It will also show how the evolution of inner-core asymmetries relates to the spatial distribution and characteristics of lightning flashes in the inner core.

Patrick Duran↗
NASA NTRSDate not supplied

Objectively Identifying Transverse Cirrus Bands in Tropical Cyclones using a Convolutional Neural Network

Transverse cirrus bands (TCBs) are bands of upper-level clouds regularly seen in mesoscale and synoptic-scale weather systems. In tropical cyclones, their appearance has been subjectively linked to intensification and the diurnal cycle. However, these hypothesized relationships have not been rigorously tested due to the subjective nature of TCBs in satellite images. A machine learning technique that successfully identifies TCBs objectively in imagery from the GOES-16 Advanced Baseline Imager (ABI) has been developed to solve this problem. The technique uses a U-Net convolutional neural network (CNN) that assigns a probability to each pixel in an image based on the likelihood of the pixel being associated with a TCB. Using the U-Net CNN, a database of TCBs from 2019 to 2022 was developed for the Atlantic tropical cyclone basin by defining an appropriate probability threshold that defines the difference between TCB and non-TCB pixels. This threshold is where the Jaccard score, calculated using manually identified TCBs and model identified TCBs, is maximized. Statistics for TCB occurrence will also be presented, including the relationships between TCBs and storm relative motion, shear relative direction, cardinal direction, tropical cyclone intensity, tropical cyclone intensification rates, and time of day.

John Mark Mayhall↗