Automated eTRaP Information

The eTRaP is a simple ensemble whose members are the 6-hourly totals from the single-orbit TRaPs (hereafter, simply TRaP). This ensemble approach allows for the generation of probabilistic forecasts of rainfall in addition to deterministic rainfall totals similar to what is currently provided by the TRaP product.

Each eTRaP is made up of forecasts using observations from potentially several microwave sensors-currently AMSU, TRMM, SSMI and AMSRE-initialized at several observation times, and possibly using several different track forecasts. The diversity among the ensemble members helps to reduce the large (unknown) errors associated with a single-sensor, single-track TRaP. The large number of perturbations leads to ensembles with many members, allowing probability forecasts to be issued with good precision and reliability.

The TRaPs are weighted according to their latency and relative accuracy based on validation conducted for US landfalling tropical cyclones in 2004, 2005, 2007 and 2008. When weights were computed based on equitable threat score, the sensor weights were all equal (all set to 1) whereas the latency weights were assigned linearly decreasing values of 1, 0.85, 0.70, and 0.55.

The requisite data needed to create an eTRaP include a forecast and at least two TRaP members. Often, a large number of 6h TRaPs are available for generating an ensemble, such that more than a thousand ensemble members are possible. Some culling procedures are invoked to keep the ensemble to a manageable size. If more than one TRaP is issued from a given satellite overpass then only the latest TRaP is included in the ensemble. An exception is when the TRaPs are based on forecasts from different operational centers, in which case both are retained because the different track forecasts give useful information on track uncertainty. If, after this step is taken, the number of potential ensemble members exceeds 200, the permutations are randomly culled to reduce the number to 200 in order to speed computing time.

An eTRaP will be produced, centered on the synoptic hours (e.g., 00Z, 06Z, 12Z, 18Z), from single-orbit TRaP segments with start times up to 3 hours after the synoptic hour. Therefore, a 00Z eTRaP would be available to customers around 0315Z. Similarly, the 06Z ensemble would be available at 0915Z; the 12Z at 1515Z and the 18Z at 2115Z. Users may notice delays when more than 2 tropical cyclones are active simultaneously anywhere on the globe. The eTRaP will consist of deterministic and probabilistic rain forecasts for each of four 6h time periods (e.g., 00-06h, 06-12h, 12-18h, 18-24h) as well as the 24 hour cumulative time period.

To summarize:

00Z eTRaP issued ~0315Z
0-6hr Rain Amount and POPs are valid 00Z-06Z
6-12hr Rain Amount and POPs are valid 06Z-12Z
12-18hr Rain Amount and POPs are valid 12Z-18Z
18-24hr Rain Amount and POPs are valid 18Z-00Z
24hr Total Rain Amount and POPs are valid 00Z-00Z

06Z eTRaP issued ~0915Z
0-6hr Rain Amount and POPs are valid 06Z-12Z
6-12hr Rain Amount and POPs are valid 12Z-18Z
12-18hr Rain Amount and POPs are valid 18Z-00Z
18-24hr Rain Amount and POPs are valid 00Z-06Z
24hr Total Rain Amount and POPs are valid 06Z-06Z

12Z eTRaP issued ~1515Z
0-6hr Rain Amount and POPs are valid 12Z-18Z
6-12hr Rain Amount and POPs are valid 18Z-00Z
12-18hr Rain Amount and POPs are valid 00Z-06Z
18-24hr Rain Amount and POPs are valid 06Z-12Z
24hr Total Rain Amount and POPs are valid 12Z-12Z

18Z eTRaP issued ~2115Z
0-6hr Rain Amount and POPs are valid 18Z-00Z
6-12hr Rain Amount and POPs are valid 00Z-06Z
12-18hr Rain Amount and POPs are valid 06Z-12Z
18-24hr Rain Amount and POPs are valid 12Z-18Z
24hr Total Rain Amount and POPs are valid 18Z-18Z

The deterministic rain forecast provided in the eTRaP is the probability-matched ensemble mean (PMQPF) which has the same relative spatial distribution of rain as the ensemble mean, but the intensity distribution is transformed using probability (histogram) matching to have the same intensity distribution as the full ensemble. The purpose of this transformation is to remove some of the excess light rain caused by the averaging process, and to restore the heavy rain accumulations that may have been lost during averaging. Probabilistic forecasts at four rain thresholds will be provided for each time period, including the 24-hour cumulative period. The thresholds chosen for computing probabilistic forecasts are 25, 50, 75 and 100 mm for 6h forecast periods, and 50, 100, 150 and 200 mm for the 24h forecast period.

The end result is 25 files per eTRaP: 5 time periods of 0-6hr, 6-12h, 12-18h, 18-24h, 24h total x 5 eTRaP output files consisting of four probabilistic forecasts and the PMQPF. This can be visualized easily by viewing the experimental web page located at https://www.ospo.noaa.gov/products/ocean/tropical/etrap.html
The term "No Recent TRaP (<2)" is displayed when the requisite data for creating the eTRaP product- at least two TRaP members-are not available.

Due to differences in the convention used by JTWC and the RSMCs to number tropical cyclones, the web page may at times contain separate links for what is, in fact, the same system. For example, RSMC Reunion may be warning on Tropical Depression 12 at the same time JTWC is carrying it as 15S. Once the RSMC has named a tropical cyclone and that name is reflected in the forecast bulletins of JTWC, the eTRaPs generated by both agencies will appear on the storm page bearing the RSMC's name for the system.

eTRaPs are archived by storm and year. In the storm archive page there are three possible reasons an eTRaP was not created:

• The storm had not yet formed or was recently given a proper name by the RSMC. This appears as "Not Formed or Renamed" on the archive page.
• The storm has dissipated or been renamed in which case "Renamed or Ended" appears on the archive page.
• The requisite data for creating the eTRaP product. at least two TRaP members.are not available in which case "No Recent TRaP (<2)" appears on the archive page.

In addition to the NOAA/NESDIS web page, the eTRaP output will be available to users via anonymous ftp for five days in three formats: McIDAS area file, text file and gif file. McIDAS users can easily display the data while users of the text format can develop the means to view the data on their own systems. The eTRaP text file format is very similar to that of the TRaP text file format to minimize disruption to customers. Sample text files, McIDAS area files and gif files can be accessed through the following site: http://satepsanone.nesdis.noaa.gov/pub/TRAP/ETRAP/example/ and current eTRaP files can be found at: http://satepsanone.nesdis.noaa.gov/pub/TRAP/ETRAP/

More detailed information can be found in the document Explanation of Digital eTRaP Formats located at etrap-expl.html

Questions or feedback may be directed to Michael Turk