Skip navigation

RAMMB: Regional and Mesoscale Meteorology Branch logo CIRA: Cooperative Institute for Research in the Atmosphere logo NESDIS: NOAA Environmental Satellite, Data, and Information Service logo

Announcement: Please visit out new web application, SLIDER, for every pixel of real-time GOES-16 and Himawari-8 imagery.

NHC Lightning-based TC Intensity Prediction (Rapid Intensification Index - RII)

Figure 1: 2-hourly Composite Lightning Strikes and Hurricane Ida on 8 November 2009

1) Product Information:

- Who is developing and distributing this product?

This product is being developed by The Cooperative Institute for Research in the Atmosphere (CIRA) in Fort Collins, Colorado, together with the NOAA/NESDIS/STAR RAMM Branch.

- Who is receiving this product, and how?

An experimental version of this product will be produced in real-time based during the 2011 hurricane season. The lightning data are received at CIRA in real-time via a World Wide Lightning Location Network (WWLLN) feed. The RII will be a text product which is distributed from CIRA to NHC via a local ftp server.

- What is the product size?

This is a text product with a file size of approximately 3 KB

2) Product Description:

- Purpose of this product:

Recent studies indicate that rapid changes in lightning density in a tropical cyclone can indicate intensification or weakening of the tropical cyclone, depending on whether the lightning strikes are observed in the outer bands or in the eye wall of the system. Forecasters are already familiar with the operational RII product. Therefore, this experimental RII product will not only provide the NHC forecasters with an additional decision aid for making tropical cyclone intensification/weakening predictions, but will familiarize them with the additional information available from lightning location data.

- Why is this a GOES-R Proving Ground Product?

The RII product demonstrates a new capability that will be possible in the GOES-R era. The RII product is a proxy dataset simulating the future products we will receive from the Geostationary Lightning Mapper (GLM), one of the instruments to be flown on GOES-R. The GLM will detect total strikes intracloud and cloud to ground. For over-land applications the GLM will compliment today’s land based systems that only measure cloud to ground lightning (about 15% of the total). However, over the oceans where land-based lightning information is typically of poor quality or simply unavailable, the GLM presents a new and unique dataset to be exploited, in this case for tropical cyclones.

- How is this product created now?

Since the operational Rapid Intensification Index (RII) described by Kaplan et al (2010) uses a different developmental time period and scales predictors in a different way, the RII used for the Proving Ground was created 1) without lightning information (NLM) and 2) with lightning information (LM)using a developmental dataset that spans 2005-2010. The predictors used in the no-lightning model include those described by Kaplan et al. (2010) RII, but they are scaled differently. In the RII developed for the Proving Ground predictors, including the lightning information, the predictors were scaled using standard statistical normalization (i.e., subtract the mean and divide by the standard deviation). The factors related to RII in Kaplan et al. (2010) and the NLM are provided in the Table below along with the experimental lightning predictors used in the LM (9 & 10).

  1. Previous 12-hour change in intensity [knots]
  2. Scalar measure of vertical wind shear between 850-200 hPa average with 500 km [knots]
  3. 200 hPa divergence averaged within a circle with radius 1000 km
  4. Potential intensity (MPI minus Current Intensity) [knot]
  5. The 850-700 hPa Relative Humidity averaged within 200-800 km of the TC center [%]
  6. Infrared Imagery, % area with pixels <-30 C within 50 to 200km of the TC center [%]
  7. Infrared Imagery, standard deviation of brightness temperature within 50 to 200km [K]
  8. Oceanic heat content along the track (KJ/cm2)
  9. Square root of the Lightning density within 100 km
  10. Square root of the Lightning density in an annulus of 300-400 km

Kaplan, J., M. DeMaria, and JA. Knaff, 2010: A revised tropical cyclone rapid intensification index for the Atlantic and east Pacific basins. Wea. Forecasting, 25, 220-241.

3) Product Examples and Interpretation

A prototype rapid intensity index will be run in near-real time to demonstrate a decision aid using proxy GLM data from various sources. These data are combined in a discriminant analysis algorithm that provides optimal weights of the independent variables to provide a classification of whether or not a tropical cyclone will rapidly intensify in the next 24 hours. This RII algorithm is very similar to an operational guidance product available to NHC. However, the operational algorithm does not include lightning data. The comparison of operational and experimental RII products will provide an estimate of the utility of lightning data for rapid intensity prediction. This product represents an application of one of the GOES-R baseline products. Output from this product is disseminated via a text product hosted on a local ftp server, an example of which is provided below in Table 1. The version to be demonstrated in 2011 will also include the probability of rapid weakening.

                    *   ATLC RAPID INTENSITY INDEX TESTS        *
                    *      GOES DATA AVAILABLE                  *
                    *      OHC  DATA AVAILABLE                  *
                    *  AL04        AL042010  08/03/10  06 UTC   *


   ** 2010 ATLANTIC RI INDEX AL042010 AL04       08/03/10  06 UTC **

 12 HR PERSISTENCE (KT):   5.0 Range:-45.0 to  30.0 Scaled/Wgted Val:  0.7/  1.4
 850-200 MB SHEAR (KT) :   9.5 Range: 26.2 to   3.2 Scaled/Wgted Val:  0.7/  0.9
 D200 (10**7s-1)       :  -2.8 Range:-21.0 to 140.0 Scaled/Wgted Val:  0.1/  0.2
 POT = MPI-VMAX (KT)   : 118.3 Range: 33.5 to 126.5 Scaled/Wgted Val:  0.9/  0.6
 850-700 MB REL HUM (%):  68.2 Range: 56.0 to  85.0 Scaled/Wgted Val:  0.4/  0.3
 % area w/pixels <-30 C:  57.0 Range: 17.0 to 100.0 Scaled/Wgted Val:  0.5/  0.1
 STD DEV OF IR BR TEMP :  19.1 Range: 30.6 to   3.2 Scaled/Wgted Val:  0.4/  0.7
 Heat content (KJ/cm2) :  59.6 Range:  0.0 to 130.0 Scaled/Wgted Val:  0.5/  0.0

 Prob of RI for 25 kt RI threshold=    23% is   1.8 times the sample mean(12.6%)
 Prob of RI for 30 kt RI threshold=    15% is   1.9 times the sample mean( 8.1%)
 Prob of RI for 35 kt RI threshold=    10% is   2.2 times the sample mean( 4.8%)
 Prob of RI for 40 kt RI threshold=     6% is   1.8 times the sample mean( 3.4%)

                    FOR GOES-R PROVING GROUND

AL04       Initial vmax, lat, lon:    35.   13.7   -46.2

 Prob of RI for 30 kt RI threshold=    15%,   no lightning input, exper. algorithm
 Prob of RI for 30 kt RI threshold=    14%, with lightning input, exper. algorithm

Recent Lightning Density History  (Strikes/km2-year)
Date/Time    Inner core (0-100 km)  Rainband (300-400 km)
10 0803 06     0.0                     0.0
10 0803 00     0.0                     0.0
10 0802 18     0.0                     0.0

Sample mean:  21.5                    14.7

 Note: Inner core lightning <  sample mean favors RI
       Rainband lightning   >  sample mean favors RI

Table 1: Example of the experimental RII for Tropical Storm Colin on 8/3/2010 at 6UTC.

Figure 2: This image from 30 June 2010 00UTC depicts lightning time history over 8 hours color-coded in hourly increments to provide information on mesoscale system location and motion.

4) Advantages and Limitations

The RII will provide an additional decision aid tool for rapidly intensifying TCs for NHC forecasters while allowing them to gain experience using lightning location information. In order to provide the RII dataset reliable real-time lightning data need to be received at CIRA.