Search Results

You are looking at 1 - 6 of 6 items for :

  • Author or Editor: Robert T. DeMaria x
  • Refine by Access: All Content x
Clear All Modify Search
John A. Knaff
and
Robert T. DeMaria

Abstract

The development of an infrared (IR; specifically near 11 μm) eye probability forecast scheme for tropical cyclones is described. The scheme was developed from an eye detection algorithm that used a linear discriminant analysis technique to determine the probability of an eye existing in any given IR image given information about the storm center, motion, and latitude. Logistic regression is used for the model development and predictors were selected from routine information about the current storm (e.g., current intensity), forecast environmental factors (e.g., wind shear, oceanic heat content), and patterns/information (e.g., convective organization, tropical cyclone size) extracted from the current IR image. Forecasts were created for 6-, 12-, 18-, 24-, and 36-h forecast leads. Forecasts were developed using eye existence probabilities from North Atlantic tropical cyclone cases (1996–2014) and a combined North Atlantic and North Pacific (i.e., Northern Hemisphere) sample. The performance of North Atlantic–based forecasts, tested using independent eastern Pacific tropical cyclone cases (1996–2014), shows that the forecasts are skillful versus persistence at 12–36 h, and skillful versus climatology at 6–36 h. Examining the reliability and calibration of those forecasts shows that calibration and reliability of the forecasts is good for 6–18 h, but forecasts become a little overconfident at longer lead times. The forecasts also appear unbiased. The small differences between the Atlantic and Northern Hemisphere formulations are discussed. Finally, and remarkably, there are indications that smaller TCs are more prone to form eye features in all of the TC areas examined.

Full access
Mark DeMaria
,
Robert T. DeMaria
,
John A. Knaff
, and
Debra Molenar

Abstract

A large sample of Atlantic and eastern North Pacific tropical cyclone cases (2005–10) is used to investigate the relationships between lightning activity and intensity changes for storms over water. The lightning data are obtained from the ground-based World Wide Lightning Location Network (WWLLN). The results generally confirm those from previous studies: the average lightning density (strikes per unit area and time) decreases with radius from the storm center; tropical storms tend to have more lightning than hurricanes; intensifying storms tend to have greater lightning density than weakening cyclones; and the lightning density for individual cyclones is very episodic. Results also show that Atlantic tropical cyclones tend to have greater lightning density than east Pacific storms. The largest lightning density values are associated with sheared cyclones that do not intensify very much. The results also show that when the lightning density is compared with intensity change in the subsequent 24 h, Atlantic cyclones that rapidly weaken have a larger inner-core (0–100 km) lightning density than those that rapidly intensify. Thus, large inner-core lightning outbreaks are sometimes a signal that an intensification period is coming to an end. Conversely, the lightning density in the rainband regions (200–300 km) is higher for those cyclones that rapidly intensified in the following 24 h in both the Atlantic and east Pacific. When lightning density parameters are used as input to a discriminant analysis technique, results show that lightning information has the potential to improve the short-term prediction of tropical cyclone rapid intensity changes.

Full access
Mark DeMaria
,
John A. Knaff
,
Richard Knabb
,
Chris Lauer
,
Charles R. Sampson
, and
Robert T. DeMaria

Abstract

The National Hurricane Center (NHC) Hurricane Probability Program (HPP) was implemented in 1983 to estimate the probability that the center of a tropical cyclone would pass within 60 n mi of a set of specified points out to 72 h. Other than periodic updates of the probability distributions, the HPP remained unchanged through 2005. Beginning in 2006, the HPP products were replaced by those from a new program that estimates probabilities of winds of at least 34, 50, and 64 kt, and incorporates uncertainties in the track, intensity, and wind structure forecasts. This paper describes the new probability model and a verification of the operational forecasts from the 2006–07 seasons.

The new probabilities extend to 120 h for all tropical cyclones in the Atlantic and eastern, central, and western North Pacific to 100°E. Because of the interdependence of the track, intensity, and structure forecasts, a Monte Carlo method is used to generate 1000 realizations by randomly sampling from the operational forecast center track and intensity forecast error distributions from the past 5 yr. The extents of the 34-, 50-, and 64-kt winds for the realizations are obtained from a simple wind radii model and its underlying error distributions.

Verification results show that the new probability model is relatively unbiased and skillful as measured by the Brier skill score, where the skill baseline is the deterministic forecast from the operational centers converted to a binary probabilistic forecast. The model probabilities are also well calibrated and have high confidence based on reliability diagrams.

Full access
John A. Knaff
,
Scott P. Longmore
,
Robert T. DeMaria
, and
Debra A. Molenar

Abstract

A new and improved method for estimating tropical-cyclone (TC) flight-level winds using globally and routinely available TC information and infrared (IR) satellite imagery is presented. The developmental dataset is composed of aircraft reconnaissance (1995–2012) that has been analyzed to a 1 km × 10° polar grid that extends outward 165 km from the TC center. The additional use of an azimuthally average tangential wind at 500 km, based on global model analyses, allows the estimation of winds at larger radii. Analyses are rotated to a direction-relative framework, normalized by dividing the wind field by the observed maximum, and then decomposed into azimuthal wavenumbers in terms of amplitudes and phases. Using a single-field principal component method, the amplitudes and phases of the wind field are then statistically related to principal components of motion-relative IR images and factors related to the climatological radius of maximum winds. The IR principal components allow the wind field to be related to the radial and azimuthal variability of the wind field. Results show that this method, when provided with the storm location, the estimated TC intensity, the TC motion vector, and a single IR image, is able to estimate the azimuthal wavenumber 0 and 1 components of the wind field. The resulting wind field reconstruction significantly improves on the method currently used for satellite-based operational TC wind field estimates. This application has several potential uses that are discussed within.

Full access
Mark DeMaria
,
John A. Knaff
,
Michael J. Brennan
,
Daniel Brown
,
Richard D. Knabb
,
Robert T. DeMaria
,
Andrea Schumacher
,
Christopher A. Lauer
,
David P. Roberts
,
Charles R. Sampson
,
Pablo Santos
,
David Sharp
, and
Katherine A. Winters

Abstract

The National Hurricane Center Hurricane Probability Program, which estimated the probability of a tropical cyclone passing within a specific distance of a selected set of coastal stations, was replaced by the more general Tropical Cyclone Surface Wind Speed Probabilities in 2006. A Monte Carlo (MC) method is used to estimate the probabilities of 34-, 50-, and 64-kt (1 kt = 0.51 m s−1) winds at multiple time periods through 120 h. Versions of the MC model are available for the Atlantic, the combined eastern and central North Pacific, and the western North Pacific. This paper presents a verification of the operational runs of the MC model for the period 2008–11 and describes model improvements since 2007. The most significant change occurred in 2010 with the inclusion of a method to take into account the uncertainty of the track forecasts on a case-by-case basis, which is estimated from the spread of a dynamical model ensemble and other parameters. The previous version represented the track uncertainty from the error distributions from the previous 5 yr of forecasts from the operational centers, with no case-to-case variability. Results show the MC model provides robust estimates of the wind speed probabilities using a number of standard verification metrics, and that the inclusion of the case-by-case measure of track uncertainty improved the probability estimates. Beginning in 2008, an older operational wind speed probability table product was modified to include information from the MC model. This development and a verification of the new version of the table are described.

Full access
Galina Chirokova
,
John A. Knaff
,
Michael J. Brennan
,
Robert T. DeMaria
,
Monica Bozeman
,
Stephanie N. Stevenson
,
John L. Beven
,
Eric S. Blake
,
Alan Brammer
,
James W. Darlow
,
Mark DeMaria
,
Steven D. Miller
,
Christopher J. Slocum
,
Debra Molenar
, and
Donald W. Hillger

Abstract

Visible satellite imagery is widely used by operational weather forecast centers for tropical and extratropical cyclone analysis and marine forecasting. The absence of visible imagery at night can significantly degrade forecast capabilities, such as determining tropical cyclone center locations or tracking warm-topped convective clusters. This paper documents ProxyVis imagery, an infrared-based proxy for daytime visible imagery developed to address the lack of visible satellite imagery at night and the limitations of existing nighttime visible options. ProxyVis was trained on the VIIRS day/night band imagery at times close to the full moon using VIIRS IR channels with closely matching GOES-16/17/18, Himawari-8/9, and Meteosat-9/10/11 channels. The final operational product applies the ProxyVis algorithms to geostationary satellite data and combines daytime visible and nighttime ProxyVis data to create full-disk animated GeoProxyVis imagery. The simple versions of the ProxyVis algorithm enable its generation from earlier GOES and Meteosat satellite imagery. ProxyVis offers significant improvement over existing operational products for tracking nighttime oceanic low-level clouds. Further, it is qualitatively similar to visible imagery for a wide range of backgrounds and synoptic conditions and phenomena, enabling forecasters to use it without special training. ProxyVis was first introduced to National Hurricane Center (NHC) operations in 2018 and was found to be extremely useful by forecasters becoming part of their standard operational satellite product suite in 2019. Currently, ProxyVis implemented for GOES-16/18, Himawari-9, and Meteosat-9/10/11 is being used in operational settings and evaluated for transition to operations at multiple NWS offices and the Joint Typhoon Warning Center.

Significance Statement

This paper describes ProxyVis imagery, a new method for combining infrared channels to qualitatively mimic daytime visible imagery at nighttime. ProxyVis demonstrates that a simple linear regression can combine just a few commonly available infrared channels to develop a nighttime proxy for visible imagery that significantly improves a forecaster’s ability to track low-level oceanic clouds and circulation features at night, works for all current geostationary satellites, and is useful across a wide range of backgrounds and meteorological scenarios. Animated ProxyVis geostationary imagery has been operational at the National Hurricane Center since 2019 and is also currently being transitioned to operations at other NWS offices and the Joint Typhoon Warning Center.

Open access