QSCAT High-Resolution Hurricane Analysis
1. Hurricane Frances (September 2004), standard QSCAT and UHR images
Here we present high-res figures for hurricane Frances (September 2004. Included are standard QSCAT (R2) at 25x25km
resolution, and David Long's (BYU) ultra high-resolution (UHR) winds at 2.5x2.5km resolution.
Click on the images for enlarged versions of plots.
Fig. 1.1 (a) BYU hi-res wind speed, Frances, rev 27096, and (b) BYU wind vectors and standard QSCAT (red)
with NOAA best track hurricane center (blue cross).
Fig. 1.2 (a) Standard QSCAT wind retrievals (rain-flagged WVC in red), and (b) Axis-symmetric hurrricane wind field from
conventional UHR winds (all WVC included).
Fig. 1.3 (a) Residual 1: conv UHR winds minus axis-symmetric average, and
(b) 0.1° x 0.1° average Residual 1.
2. BYU UHR images, from simultaneous wind/rain (SWR) and experimental hurricane retrievals (MAP)
The following figures have been prepared by David Long and Brent Williams at BYU. MAP data represent the
"Maximum Aposteriori Probability" wind retrievals.
Fig. 2.1 Slices of the mean wind speeds orthogonal to the mean flow for large and small storms with
various magnitudes of the mean flow. The corresponding slices of the simplified hurricane model are also plotted.
Data are derived from all Atlantic hurricanes.
Fig. 2.2 Density plots of the H*WIND wind speed versus the three wind retrieval schemes for the same
observation of hurricane Isabel: (a) conventional ultra-high resolution (UHR),
(b) MAP estimation, and (c) MAP ambiguity selection.
Fig. 2.3 Hurricane Floyd (1999) examples. The wind vector fields are down-sampled by 10 for plotting.
The black dots represent the eye center reported by the new method and the white x's represent the conventional
high resolution eye center based on the curl of the vector field:
(a) conventional ultra-high resolution (UHR) wind field,
(b) field-wise MAP estimation of the wind field,
and (c) result of the model-based MAP ambiguity selection.
Fig. 2.4 Hurricane Frances, rev 27096, from BYU:
(a) top-left: standard L2B QSCAT, (b) top-right: conventional UHR winds, (c) bottom-left: MAP estimation, and
(d) MAP ambiguty selections.
Fig. 2.5 Hurricane Frances, rev 27096, Simultaneous Wind / Rain (SWR) from BYU:
(a) top: SWR winds,
(b) bottom-left: SWR rain, and (c) SWR regimes (nothing=wind dominated,
red=rain dominated, green=wind&rain).
Fig. 2.6 Hurricane Katrina, rev 32244, from BYU:
(a) top-left: standard L2B QSCAT, (b) top-right: conventional UHR winds, (c) bottom-left: MAP estimation, and
(d) MAP ambiguty selections.
Fig. 2.7 Hurricane Katrina, rev 32244, Simultaneous Wind / Rain (SWR) from BYU:
(a) top-left: conventional UHR winds, (b) top-right: SWR winds,
(c) bottom-left: SWR rain, and (d) SWR regimes (nothing=wind dominated,
red=rain dominated, green=wind&rain).
3. Frances, MAP images
The following plots show the MAP hurricane model fit wind speed fields. Also marked is the
interpolated NOAA best track hurricane center (yellow cross).
Fig. 3.1 Hurricane Frances, rev 27096, MAP hurricane model fit speed field:
(a) standard MAP, (b) hand selected hurricane eye center ("handeye" MAP), and
(c) the latest file from 8/23/07.
In the next figures, the latest MAP files (from 8/23/07) are presented.
Fig. 3.2 (a) MAP wind speed, Frances, rev 27096, and (b) MAP wind vectors and standard QSCAT (red)
with NOAA best track hurricane center (blue cross).
Fig. 3.3 Frances MAP (a) axis-symmetric average, and (b) Residual 1: winds minus axis-symmetric average.
Fig. 3.4 Frances MAP (a) 0.1° x 0.1° average Res 1, and (b) Residual 2 = Residual 1 - av Res 1.
Fig. 3.5 Frances Windstress Curl (left side) and Divergence (right side) of MAP
(a) total vectors, (b) axis-symmetric average, (c) Res 1, (d) av Res 1, and (e) Res 2.
4. Frances, Rain Effect on MAP images
In the following figures, MAP images for Frances are examined with respect to rain effect. The figures are based
on MAP analysis of rev 27096 with the new "rain regime" rain flag (8/23/07). Also shown is the simultaneous wind/rain retrieved
(SWR) "rain_rate_integrated" (in km mm/hr).
Fig. 4.1 MAP Res1 wind vectors, Frances, rev 27096, (a) colored with integrated rain rate, and
(b) Res 1 "rain regime" vectors (green = wind & rain backscatter of same order, red = rain dominated).
The impact of rain-flagging on windstress curl and divergence calculations is shown below. Valid averages
of tau are only calculated when there are at least 4 wind retrievals per 0.1° x 0.1° bin. Curl and divergence are only
calculated when all 4 adjacent tau values are available for computing gradients.
Fig. 4.2 MAP av Res1, Frances rev 27096, including all rain-flagged data, (a) curl , and (b) divergence.
Fig. 4.3 MAP av Res1, Frances rev 27096, excluding rain-flagged data for (a) curl, and
(b) divergence; Top row: excluding only "red" regime-flagged retrievals, and Bottom row:
excluding "red" and "green" retrievals.
5. Katrina, MAP images
The following plots show the MAP hurricane retrievals for Katrina rev 32244, from BYU (8/23/07).
Fig. 5.1 (a) MAP wind speed, Katrina, rev 32244, and (b) MAP wind vectors and standard QSCAT (red)
with NOAA best track hurricane center (blue cross).
Fig. 5.2 Katrina MAP (a) axis-symmetric average, and (b) Residual 1: winds minus axis-symmetric average.
Fig. 5.3 Katrina MAP (a) 0.1° x 0.1° average Res 1, and (b) Residual 2 = Residual 1 - av Res 1.
Fig. 5.4 Katrina Windstress Curl (left side) and Divergence (right side) of MAP
(a) total vectors, (b) axis-symmetric average, (c) Res 1, and (d) av Res 1.
6. Katrina, Rain Effect on MAP images
In the following figures, MAP images for Katrina are examined with respect to rain effect. The figures are based
on MAP analysis of rev 32244 with the new "rain regime" rain flag (8/23/07). Also shown is the simultaneous wind/rain retrieved
(SWR) "rain_rate_integrated" (in km mm/hr).
Fig. 6.1 MAP Res1 wind vectors, Katrina, rev 32244, (a) colored with integrated rain rate, and
(b) Res 1 "rain regime" vectors (green = wind & rain backscatter of same order, red = rain dominated).
The impact of rain-flagging on windstress curl and divergence calculations is shown below. Valid averages
of tau are only calculated when there are at least 4 wind retrievals per 0.1° x 0.1° bin. Curl and divergence are only
calculated when all 4 adjacent tau values are available for computing gradients.
Fig. 6.2 MAP av Res1, Katrina rev 32244, including all rain-flagged data, (a) curl , and (b) divergence.
Fig. 6.3 MAP av Res1, Katrina rev 32244, excluding rain-flagged data for (a) curl, and
(b) divergence; Top row: excluding only "red" regime-flagged retrievals, and Bottom row:
excluding "red" and "green" retrievals.
7. Katrina, WRF images
The following images are from WRF simulation for Katrina, courtesy of Svetla Veleva at JPL.
Fig. 7.1 (a) WRF wind speed, Katrina, on Aug 28, 2005 11:30, during rev 32244, and (b) WRF Res 1 wind vectors,
with hurricane center determined from minimum in WRF speed map (yellow and blue crosses).
Fig. 7.2 WRF Katrina (a) 0.1° x 0.1° average Res 1, and (b) Residual 2 = Residual 1 - av Res 1.
Fig. 7.3 WRF Katrina Windstress Curl (left side) and Divergence (right side)
(a) total vectors, (b) axis-symmetric average, (c) Res 1, and (d) av Res 1.
last modified on August 30, 2007
Go
Back
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)