Skip Navigation Links 
NOAA logo - Click to go to the NOAA homepage National Weather Service   NWS logo - Click to go to the NWS homepage
The Weather Prediction Center



Follow the Weather Prediction Center on Facebook Follow the Weather Prediction Center on Twitter
NCEP Quarterly Newsletter
WPC Home
Analyses and Forecasts
   National High & Low
   WPC Discussions
   Surface Analysis
   Days ½-2½ CONUS
   Days 3-7 CONUS
   Days 4-8 Alaska
   Flood Outlook
   Winter Weather
   Storm Summaries
   Heat Index
   Tropical Products
   Daily Weather Map
   GIS Products
Current Watches/

Satellite and Radar Imagery
  GOES-East Satellite
  GOES-West Satellite
  National Radar
Product Archive
WPC Verification
   Medium Range
   Model Diagnostics
   Event Reviews
   Winter Weather
International Desks
Development and Training
WPC Overview
   About the WPC
   WPC History
   Other Sites
Meteorological Calculators
Contact Us
   About Our Site is the U.S. Government's official web portal to all federal, state, and local government web resources and services.
Excessive Rainfall Discussion
(Caution: Version displayed is not the latest version. - Issued 2027Z Sep 19, 2019)
Version Selection
Versions back from latest:  0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19   20   
Abbreviations and acronyms used in this product
Geographic Boundaries -  Map 1: Color  Black/White       Map 2: Color  Black/White

Excessive Rainfall Discussion
NWS Weather Prediction Center College Park MD
427 PM EDT Thu Sep 19 2019

Day 1
Valid 1740Z Thu Sep 19 2019 - 12Z Fri Sep 20 2019



...Eastern Texas and Extreme Western Louisiana...
1600 UTC Update: Made minor tweaks to the ERO outlooks along the
Upper TX Coast and into eastern TX, based largely on the current
mesoanalysis and observational trends. Pulled the western
periphery of the Moderate Risk area a bit farther to the west,
aligning with the orientation and projected track of the dominant
inflow band south of TD Imelda. This band, currently tracking
slowly southward through the Houston metro region, has a history
of producing hourly rainfall rates of 3-4+ inches. In looking at
the 12Z HREF neighborhood probabilities -- probabilities of
exceeding 3"/hour are highest (above 50%) through 17-18Z, then
diminish to below 40% by 20-21Z.


Previous discussion below...

Significant and life-threatening flash flooding is ongoing early
this morning across far southeast TX within a nearly stationary
band of intense rainfall. This band is focusing well to the south
of the center of Imelda, within a low level convergence
axis...with persistent southerly flow into this axis supplying a
continuous influx of instability. Unfortunately this band is not
expected to move much through the morning hours. As Imelda lifts
slowly northward the background convergent flow will try to shift
northward as well. However, the convection will also be attempting
to push southward into the strong inflow and better instability
pool. Thus the situation favors a persistence of the band into the
morning hours without much movement. After 12z we may begin to see
a slow southward shift of the band into the Houston metro area,
but this is uncertain. Hopefully by later this morning into the
afternoon hours we will begin to at least see a
broadening/weakening of the heavy rain band. This scenario is
depicted by the high res guidance and fits the typical diurnal
trend with a general weakening/broadening of the low level inflow.
Thus, while not a sure bet, appears more likely than not that the
heavy rainfall threat becomes less organized this afternoon into
the evening hours. What happens tonight is uncertain...but at
least some threat we see another increase in the flash flood
threat overnight as low level inflow increases and tightens with
the diurnal cycle. By this time the circulation of Imelda will be
quite uncertain if there will be enough there to allow
for an intense banding scenario to evolve. There is certainly some
chance of this occurring somewhere over east TX tonight, so will
need to monitor trends. While the activity overnight should not be
able to match what we are seeing this morning over far southeast
TX, it still has the potential to bring a significant flash flood
threat and worth monitoring.

A HIGH risk will be carried into the new day 1 ERO for far
southeast TX. A dangerous and life-threatening situation is
ongoing and will continue into the morning hours across this
region. A moderate risk extends northward across east TX, to
account for rainfall closer to the center of Imelda and any
potential redevelopment tonight. We were able to trim back some on
the northern extent of the Slight and Moderate risk areas. Overall
it appears that rainfall near the center of Imelda will not be as
significant as some of the earlier guidance suggested. Instead the
significant rainfall will tend to remain focused south of the
center within slow moving intense bands within the convergent flow
south of the system.

Convection ongoing at 12z may pose an isolated flash flood threat
into the morning hours across portions of eastern NE into
IA...although the threat should generally be on a downward trend
through the morning hours. Redevelopment is likely this afternoon
from the TX Panhandle into western KS/NE along the dry line. This
activity may stay tied to the low level boundary long enough to
pose a localized flash flood threat. Do note at least some
mid/upper level connection to the tropical Pacific in the layered
PW fields...which may help increase rainfall efficiency.
Additional development appears likely over NM...and this may be
able to persist through the evening hours as it moves towards west
TX and encounters increasing southeasterly inflow. This may
continue the isolated flash flood potential over these areas into
the overnight.


Day 2
Valid 12Z Fri Sep 20 2019 - 12Z Sat Sep 21 2019


...Northeast Texas / Southeast Oklahoma / Southwest Arkansas...
In the 2030Z update we expanded the coverage of Slight Risk here,
The risk is conditional, betting that nocturnal convergence into a
remnant mid level circulation will be sufficient to spark a new
area of heavy rainfall. There are a few lesser developed solutions
from the mesoscale models, namely the 12Z NAM CONUS Nest and
WRF-NMMB. More often than not, however, mid level vorticity
centers that originated as part of a tropical cyclone are
efficient at generating these nocturnal rainfall maxima, more like
that seen in the 12Z WRF-ARW and ARW-2. By Thursday afternoon the
mid level circulation was evident over the Texas Piney Woods,
while convergence was supporting a band of deep convection closer
to the coast of southeast Texas. This daytime convection should
gradually loosen or propagate westward, leaving room for deep
layer convergence to increase into the mid level center late
tonight as it lifts into northeast Texas. Once initiated,
southerly and southwesterly inflow of 20-plus knots should
maintain a convective cluster or two, with a fairly well portioned
balance between northward directed mean layer winds and southward
directed propagation vectors - or in other words, slow movement.
This balance could yield isolated very heavy rainfall from late
Thursday night through the first half of the day on Friday. One or
more clusters could occur as the mid level center begins to shear
and continues to lift northward, and models have some heavy rain
signal extending into the Ozarks in northwest Arkansas, hence the
need to expand the Slight Risk. Fortunately, antecedent conditions
are dry here, but tropical rain rates may overwhelm creeks and
streams. Precipitable Water values around 2.00 inches for
mid-September are well above the 90th percentile, and will
approach daily records at LZK and SHV.

We also expanded Marginal Risk westward to capture a separate
convective environment where more traditional high plains storms
are likely to form in the increasingly difluent upper flow regime
ahead of a deep western U.S. trough. Rain rates 1 to 2 inches an
hour are possible with supercell and multicell modes from the
Texas panhandle into western Oklahoma. Flash flooding will be
dependent on local cell mergers / brief training, and antecedent
wetness is not especially pronounced.

...Northern Plains...
Here in the 2030Z update we maintained a broad Slight Risk area
affecting the central and eastern Dakotas and western Minnesota
Friday / Friday night. The odds that this event would need to see
an upgrade to Moderate Risk do not appear to have trended in any
direction. The PW anomalies are impressive, and there will be a
good deal of instability. Convective mode, and widespread
overturning / cold pool production, however, along with
displacement of the heaviest QPF from the lowest flash flood
guidance values, suggests Slight Risk best captures the storyline.
We did expand Marginal Risk a little farther west to catch an
extension of weak instability toward the evolving mid level low
center expected to come out of the mountains in Wyoming and
Montana early Saturday.

An anomalously deep upper trough and associated surface front will
pull toward the northern Plains Friday night. An increasingly
coupled upper jet structure is forecast as one deep layer
shortwave lifts into central Canada during the evenign, followed
quickly by a secondary jet max rotating through the base of the
trough. The initial wave is well timed to promote vigorous
thunderstorm growth during peak heating and beyond, into the low
level jet cycle Friday evening.

Flow from 925-500mb will become southerly, driving tremendous
moist advection northward, noted by large 850mb moisture transport
vectors and PWATs surging to 1.75+", more than 3 standard
deviations above the climatological mean. As the 850mb wind
reaches 40-50 kts immediately ahead of the cold front, the
accompanying WAA pushes MUCape to 3000 J/kg. Although storm
motions are forecast to be rapid to the north as evidenced by
0-6km mean wind vectors of 25-35kts, deep layer unidirectional
shear parallel to the length of the cold front suggests training
of echoes with heavy rainfall is likely, particularly across North
Dakota with the lead wave during the evening. Somewhat lesser
coverage is expected in South Dakota, but upscale growth is
possible throughout the region owing to favorable inflow and deep
layer ascent, resulting in the broad Slight Risk area.


Day 3
Valid 12Z Sat Sep 21 2019 - 12Z Sun Sep 22 2019


...Northern Kansas to Northern Illinois...
In the 2030Z update, and along with input from WFOs Topeka and
Pleasant Hill, we upgraded to a Moderate Risk of excessive
rainfall from northeast KS into northern MO for rainfall expected
Saturday night. No doubt, it is difficult to pinpoint a mesoscale
heavy rain area this far out in time, and the bulls-eye may move a
little bit between now and Saturday. The increased lead time was
deemed worth it, though, to get the message out well before the
weekend. There are a number of synoptically evident heavy rainfall
ingredients checking a lot of boxes over this area. In simplest
terms, a deep western U.S. trough will pull out into the Plains,
running into rich moisture that is enhanced by high level blow-off
from two tropical storms in the eastern Pacific. Model QPFs are
not overwhelming in magnitude, but they are relatively consistent
in placement, and both the UKMET and ECMWF have produced greater
than 5 inch spot totals on consecutive runs. Most likely, 5-inch
amounts or greater would focus along the CAPE gradient which the
NAM predicts over the Moderate Risk area. This matches exactly to
the QPF placement in the 00Z ECMWF and 12Z UKMET. Especially of
concern is the persistence of strong 850-mb moisture flux which
sets up Saturday night to the rear of a 110-knot upper jet streak.
Despite what should be rapid storm motions to the east on 30+ kts
cloud layer winds, 0-6km mean flow parallel to the frontal
boundary suggests that training is likely. Rainfall rates should
exceed 1"/hr, and may approach 2"/hr the latter half of Day 3 as
warm cloud depths surge above 13kft to increase rainfall
efficiency. Wherever outflow becomes established as a barrier to
the moist southwesterly low level flow, some local extreme totals
pushing closer to 10 inches are not out of the question, although
with no mesoscale guidance that goes that far out in time - this
is a little bit of conjecture based on experience.

Rainfall will also be quite persistent downstream through parts of
MO / IA / IL, covered by a Slight Risk. Rain rates and event
totals should decline steadily with eastward extent, as
instability becomes exhausted in favor of the ongoing activity
back to the west. Some pronounced training could occur over
northeast Missouri and adjacent IA/IL, however, especially early
in the event Saturday afternoon. Based on the latest large scale
guidance we would expect the back edge of heavier rain should
begin to shift eastward early in what would be the Day 4 period,

...North Dakota...
The NAM indicates continued generation of a few hundred J/kg
MUCAPE over western North Dakota Saturday morning within the
elevated frontal zone / wrap-around moisture associated with an
ejecting, tight vort max. Given 2-week precipitation running
6-times normal, and the anticipated widespread 0.50 to 1.50 inches
occurring through Friday night, this environment looks to warrant
a Marginal Risk that will linger into the Day 3 period, daytime on
Saturday. Rain rates briefly up to one inch with convective
enhancement - per the 12Z NAM CONUS Nest - could exacerbate any
ongoing flooding.


Day 1 threat area:
Day 2 threat area:
Day 3 threat area: