Browse

You are looking at 101 - 110 of 9,661 items for :

  • Journal of Applied Meteorology and Climatology x
  • Refine by Access: All Content x
Clear All
Christina Kumler-Bonfanti, Jebb Stewart, David Hall, and Mark Govett

Abstract

Extracting valuable information from large sets of diverse meteorological data is a time-intensive process. Machine-learning methods can help to improve both speed and accuracy of this process. Specifically, deep-learning image-segmentation models using the U-Net structure perform faster and can identify areas that are missed by more restrictive approaches, such as expert hand-labeling and a priori heuristic methods. This paper discusses four different state-of-the-art U-Net models designed for detection of tropical and extratropical cyclone regions of interest (ROI) from two separate input sources: total precipitable water output from the Global Forecast System (GFS) model and water vapor radiance images from the Geostationary Operational Environmental Satellite (GOES). These models are referred to as International Best Track Archive for Climate Stewardship (IBTrACS)-GFS, Heuristic-GFS, IBTrACS-GOES, and Heuristic-GOES. All four U-Nets are fast information extraction tools and perform with an ROI detection accuracy ranging from 80% to 99%. These are additionally evaluated with the Dice and Tversky intersection-over-union (IoU) metrics, having Dice coefficient scores ranging from 0.51 to 0.76 and Tversky coefficients ranging from 0.56 to 0.74. The extratropical cyclone U-Net model performed 3 times as fast as the comparable heuristic model used to detect the same ROI. The U-Nets were specifically selected for their capabilities in detecting cyclone ROI beyond the scope of the training labels. These machine-learning models identified more ambiguous and active ROI missed by the heuristic model and hand-labeling methods that are commonly used in generating real-time weather alerts, having a potentially direct impact on public safety.

Restricted access
R. Bassett, P. J. Young, G. S. Blair, F. Samreen, and W. Simm

Abstract

Lagos, Nigeria, is rapidly urbanizing and is one of the fastest-growing cities in the world, with a population that is increasing at almost 500 000 people per year. Yet the impacts on Lagos’s local climate via its urban heat island (UHI) have not been well explored. Considering that the tropics already have year-round high temperatures and humidity, small changes are very likely to tip these regions over heat-health thresholds. Using a well-established model, but with an extended investigation of uncertainty, we explore the impact of Lagos’s recent urbanization on its UHI. Following a multiphysics evaluation, our simulations, against the background of an unusually warm period in February 2016 (during which temperatures regularly exceeded 36°C), show a 0.44°C ensemble-time-mean increase in nighttime UHI intensity between 1984 and 2016. The true scale of the impact is seen spatially as the area over which ensemble-time-mean UHIs exceed 1°C was found to increase steeply from 254 km2 in 1984 to 1572 km2 in 2016. The rate of warming within Lagos will undoubtedly have a high impact because of the size of the population (12+ million) already at risk from excess heat. Significant warming and modifications to atmospheric boundary layer heights are also found in rural areas downwind, directly caused by the city. However, there is limited long-term climate monitoring in Lagos or many similarly expanding cities, particularly in the tropics. As such, our modeling can only be an indication of this impact of urbanization, and we highlight the urgent need to deploy instrumentation.

Open access
Federico Flores, Andrés Arriagada, Nicolás Donoso, Andrés Martínez, Aldo Viscarra, Mark Falvey, and Rainer Schmitz

Abstract

In desert environments, intense radiative cooling of the surface during the night leads to rapid cooling of the adjacent air, resulting in a strong temperature inversion conducive to cold-air-pool formation. In this work observations are analyzed to investigate the structure of a nocturnal cold-air pool inside a semiclosed basin located near Sierra Gorda in the Atacama Desert in Chile and its effect on dust dispersion in the area. The measurement campaign was conducted over a 5-day period (14–19 August) in 2017 and included ceilometer data, vertical profiles of temperature, a grid of fixed ground stations, and mobile temperature sensors. We focus our attention on the conditions during periods of high levels of dust pollution, in order to understand the atmospheric conditions that contribute to these episodes. The analysis of the available data confirms the development of an intense nocturnal cold-air pool, which is reflected in a strong nocturnal potential temperature inversion (18 K in 150 m) and a 30°C diurnal temperature range. A comparison of the vertical distribution of dust and temperature shows that the capping inversion controls the location of the dust cloud. As a consequence, the highest dust concentrations were observed inside the cold pool, below the capping inversion, proving that within the basin the dust is confined to the layer where its source is located.

Restricted access
Sarah M. Borg, Steven M. Cavallo, and David D. Turner

Abstract

Tropopause polar vortices (TPVs) are long-lived, coherent vortices that are based on the dynamic tropopause and characterized by potential vorticity anomalies. TPVs exist primarily in the Arctic, with potential impacts ranging from surface cyclone generation and Rossby wave interactions to dynamic changes in sea ice. Previous analyses have focused on model output indicating the importance of clear-sky and cloud-top radiative cooling in the maintenance and evolution of TPVs, but no studies have focused on local observations to confirm or deny these results. This study uses cloud and atmospheric state observations from Summit Station, Greenland, combined with single-column experiments using the Rapid Radiative Transfer Model to investigate the effects of clear-sky, ice-only, and all-sky radiative cooling on TPV intensification. The ground-based observing system combined with temperature and humidity profiles from the European Centre for Medium-Range Weather Forecasts’s fifth major global reanalysis dataset, which assimilates the twice-daily soundings launched at Summit, provides novel details of local characteristics of TPVs. Longwave radiative contributions to TPV diabatic intensity changes are analyzed with these resources, starting with a case study focusing on observed cloud properties and associated radiative effects, followed by a composite study used to evaluate observed results alongside previously simulated results. Stronger versus weaker vertical gradients in anomalous clear-sky radiative heating rates, contributing to Ertel potential vorticity changes, are associated with strengthening versus weakening TPVs. Results show that clouds are sometimes influential in the intensification of a TPV, and composite results share many similarities to modeling studies in terms of atmospheric state and radiative structure.

Restricted access
David Kristovich
Open access
David Kristovich
Open access
Brent Knutson, Wenbo Tang, and Pak Wai Chan

Abstract

The operational light detection and ranging (lidar) data from the Hong Kong International Airport (HKIA) in China are assimilated in the six-nest, high-resolution Weather Research and Forecasting (WRF) Model. The existing radar data assimilation schemes in the WRF data assimilation (WRFDA) package have been adapted to accommodate the high temporal frequency and spatial resolution of the lidar observations. The weather data are then used to produce Lagrangian coherent structures to detect atmospheric hazards for flights. The coherent structures obtained from the various datasets are contrasted against flight data measured on aircraft. It is found that both WRF and WRFDA produce coherent structures that are more distinguishable than those obtained from two-dimensional retrieval, which may improve the detection of true wind shear hazards.

Restricted access
Andrew D. Magee and Anthony S. Kiem

Abstract

Catastrophic impacts associated with tropical cyclone (TC) activity mean that the accurate and timely provision of TC outlooks are important to people, places, and numerous sectors in Australia and beyond. In this study, we apply a Poisson regression statistical framework to predict TC counts in the Australian region (AR; 5°–40°S, 90°–160°E) and its four subregions. We test 10 unique covariate models, each using different representations of the influence of El Niño–Southern Oscillation (ENSO), Indian Ocean dipole (IOD), and southern annular mode (SAM) and use an automated covariate selection algorithm to select the optimum combination of predictors. The performance of preseason TC count outlooks generated between April and October for the AR TC season (November–April) and in-season TC count outlooks generated between November and January for the remaining AR TC season are tested. Results demonstrate that skillful TC count outlooks can be generated in April (i.e., 7 months prior to the start of the AR TC season), with Pearson correlation coefficient values between r = 0.59 and 0.78 and covariates explaining between 35% and 60% of the variance in TC counts. The dependence of models on indices representing Indian Ocean sea surface temperature highlights the importance of the Indian Ocean for TC occurrence in this region. Importantly, generating rolling monthly preseason and in-season outlooks for the AR TC season enables the continuous refinement of expected TC counts in a given season.

Open access
Domingo Muñoz-Esparza, Robert D. Sharman, and Wiebke Deierling

Abstract

We explore the use of machine learning (ML) techniques, namely, regression trees (RT), for the purpose of aviation turbulence forecasting at upper levels [20–45 kft (~6–14 km) in altitude]. In particular, we develop a series of RT-based algorithms that include random forests (RF) and gradient-boosted regression trees (GBRT) methods. Numerical weather prediction model prognostic variables and derived turbulence diagnostics based on 6-h forecasts from the 3-km High-Resolution Rapid Refresh model are used as features to train these data-driven models. Training and evaluation are based on turbulence estimates of eddy dissipation rate (EDR) obtained from automated in situ aircraft reports. Our baseline RF model, consisting of 100 trees with 30 layers of maximum depth, significantly reduces forecast errors for EDR < 0.1 m2/3 s−1 (which corresponds roughly to null and light turbulence) when compared with a simple regression model, increasing the probability of detection and in turn reducing the number of false alarms. Model complexity reduction via GBRT and feature-relevance analyses is performed, indicating that considerable execution speedups can be achieved while maintaining the model’s predictive skill. Overall, the ML models exhibit enhanced performance in discriminating the EDR forecast among the light, moderate, and severe turbulence categories. In addition, these artificial intelligence techniques significantly simplify the generation of new NWP and grid-spacing specific turbulence forecast products.

Open access
Natalie P. Thomas, Michael G. Bosilovich, Allison B. Marquardt Collow, Randal D. Koster, Siegfried D. Schubert, Amin Dezfuli, and Sarith P. Mahanama

Abstract

Heat waves are extreme climate events that have the potential to cause immense stress on human health, agriculture, and energy systems, so understanding the processes leading to their onset is crucial. There is no single accepted definition for heat waves, but they are generally described as a sustained amount of time over which temperature exceeds a local threshold. Multiple different temperature variables are potentially relevant, because high values of both daily maximum and minimum temperatures can be detrimental to human health. In this study, we focus explicitly on the different mechanisms associated with summertime heat waves manifested during daytime hours versus nighttime hours over the contiguous United States. Heat waves are examined using the National Aeronautics and Space Administration Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2). Over 1980–2018, the increase in the number of heat-wave days per summer was generally stronger for nighttime heat-wave days than for daytime heat-wave days, with localized regions of significant positive trends. Processes linked with daytime and nighttime heat waves are identified through composite analysis of precipitation, soil moisture, clouds, humidity, and fluxes of heat and moisture. Daytime heat waves are associated with dry conditions, reduced cloud cover, and increased sensible heating. Mechanisms leading to nighttime heat waves differ regionally across the United States, but they are typically associated with increased clouds, humidity, and/or low-level temperature advection. In the midwestern United States, enhanced moisture is transported from the Gulf of Mexico during nighttime heat waves.

Restricted access