Transcript Percent Precipitation: Diurnal Cycle DJF 2002-03

Diurnal Cycle of Precipitation
Based on CMORPH
Vernon E. Kousky, John E. Janowiak
and
Robert Joyce
Climate Prediction Center, NOAA
Motivation
• The diurnal cycle is fundamentally important
throughout the Tropics and Subtropics (and in middle
latitudes in summer).
• A better understanding of the diurnal cycle, and its
variability, will lead to better short-range forecasts.
• There is a need to develop a benchmark for evaluating
the diurnal cycle in numerical models.
CMORPH (CPC Morphing technique)
Uses IR data along with passive microwave
data to create global rainfall analyses (60N60S) at high spatial and temporal resolution.
CMORPH uses IR only as a transport vehicle, i.e. IR
data are NOT used to make estimates of rainfall when
passive microwave data are not available.
The underlying assumption is that the error in using IR to
transport precipitation features is less than the error in
using IR to estimate precipitation.
Rationale
• Passive microwave (PMW) data provide much better
rainfall estimates than IR but these data are spatially
and temporally incomplete (instruments on polar
orbiting platforms)
•
IR data are abundant both in space & time (polar &
geostationary platforms)
•
CMORPH combines the superior quality of the
PMW-derived precipitation estimates with the excellent
sampling characteristics of the IR data
Specifics
•
•
•
•
Spatial Grid: 0.0728o lat/lon (8 km at equator)
Temporal Resolution: 30 minutes
Domain: Global (60o N - 60o S)
Period of record: Dec. 2002 – present
For more information about CMORPH:
http://www.cpc.ncep.noaa.gov/products/janowiak/cmorph.html
Methodology
• The seasonal mean precipitation rates, from CMORPH, are
computed for each time interval (1-h or 3-h).
• The mean daily precipitation rates are computed by summing the
1-h rates.
• In the absence of any diurnal variability, the same amount of
rainfall would be expected during each time interval (e.g.,
100/24 = 4.2% for 1-h intervals and 100/8 = 12.5% for 3-h
intervals).
• Brown (green) colors are used to depict times when the observed
% is less (more) than the expected values (precipitation
uniformly distributed throughout the 24-h period).
Global: June-August
Southeast Asia,
India, Malaysia
& Indonesia
Many land areas
experience a
late night/ early
morning
minimum and a
late afternoon/
early evening
maximum in
precipitation.
(Add 5-8 hs for LST)
Land Areas
Southeast Asia,
India, Malaysia
& Indonesia
Many ocean
areas experience
a late night/
early morning
maximum and a
late afternoon/
early evening
minimum in
precipitation.
(Add 5-8 hs for LST)
Ocean Areas
Southeast Asia, India & Indonesia
Animation India
Indonesian region
Animation
Malaysia/Indonesia/Phili
ppines
(5-8 LT)
North &
Central
America:
Land Areas
(18-21 LT)
Many land
areas
experience a
late nightearly morning
minimum and
a late
afternoonearly evening
maximum in
precipitation.
(5-8 LT)
North &
Central
America:
Ocean Areas
(18-21 LT)
Many nearby
ocean areas
experience a
late nightearly morning
maximum and
a late
afternoonearly evening
minimum in
precipitation.
North & Central America: JJA 0304 (1mm/d mask)
South America:
December-February
Comparison: CMORPH and
Gridded Analyses (DJF 0203+0304)
CMORPH
Gridded
CMORPH over estimates precipitation by 40-50% over Brazil.
The bias is being corrected using station observations.
(6-9 LT)
However,
some land
areas
experience a
late nightearly morning
maximum and
a late
afternoonearly evening
minimum in
precipitation.
Land Areas
(18-21 LT)
Many land
areas
experience a
late nightearly morning
minimum and
a late
afternoonearly evening
maximum in
precipitation.
(6-9 LT)
Ocean Areas
(18-21 LT)
Many nearby
ocean areas
experience a
late nightearly morning
maximum and
a late
afternoonearly evening
minimum in
precipitation.
South America: DJF 0203-0405 (1mm/d mask)
Diurnal Cycle
DJF 0203-0405
South America: DJF 02-03+03-04 (1mm/d mask)
DJF - South America (1mm mask)
DJF - South America (1mm mask)
Diurnal Cycle
DJF 02-03
Diurnal Cycle
DJF 02-03 +
03-04
Seasonal
variations of the
Diurnal Cycle of
Precipitation
over the Amazon
Basin
Time-Longitude Diurnal Cycle EQ
DJF 2002-03 +2003-04
Minimum
over the land
~ 12Z.
Maximum
over the
ocean ~ 12Z.
Maximum
over the land
~00Z.
Minimum
over the
ocean ~00Z.
Mean
diurnal
cycle
Repeated
four times
Time-Longitude Diurnal Cycle 10S
DJF 2002-03 +2003-04
Time-Longitude Diurnal Cycle 20S
DJF 2002-03 +2003-04
Time-Longitude Diurnal Cycle 30S
DJF 2002-03 +2003-04
Eastwardpropagating
precipitation
systems
between the
Andes Mts.
and southern
Brazil.
MAM Mean Diurnal Cycle – EQ-5N
Convective
rainfall
systems start
along east
coast on day-1
propagate
westward,
reaching
western
Amazon on
day-3.
West Coast
East Coast
Southeast Brazil
Southeast Brazil (cont.)
Seasonally varying diurnal cycle:
27S, top, and area averaged (2x2
degrees), bottom.
Seasonally varying diurnal cycle
of preciitation (area averaged 2x2
24S
degrees), 24S (top) and 27S
(bottom).
27S
Seasonally varying diurnal cycle
of preciitation (area averaged 2x2
degrees), 2S.
55W
52W
49W
46W
Westward propagating lines occur primarily during February-May.
Conclusions
• CMORPH precipitation analyses are useful in
obtaining a detailed description of the diurnal cycle.
• A pronounced diurnal cycle in precipitation is found
in many areas of South America.
– Daytime peak over high terrain (e.g., Andes Mts.,
Brazilian Planalto).
– Daytime peak near the coast, especially NE South
America, associated with the sea breeze.
– Nocturnal peak in the central La Plata Basin, over
oceanic regions, and over portions of the Amazon Basin
(affected by westward propagating sea breeze-induced
convection).
Future Work
• There is a need to document the regional and
large-scale atmospheric circulation features that
contribute to the formation of westwardpropagating lines over the Amazon Basin.
• Studies should be developed to evaluate the ability
of numerical prediction models to simulate and
predict the occurrence of these westwardpropagating lines.