Past and future changes of moisture transports into the convective areas of
the tropical atmospheric circulation
Matthias Zahn and Richard P. Allan, http://www.nerc-essc.ac.uk/
maz/, m.zahn@reading.ac.uk, r.p.allan@reading.ac.uk
INTRODUCTION
METHOD, step 2
RESULTS (continued)
RESULTS (continued)
large scale atmospheric circulation important
Calculate moisture transport (MT) across
boundary
Changing frequency and extent of the ascending regions.
MT towards tropical land areas
DATA
20˚
studies, which apply time or space averages)
vertically)
applying mean and instantaneous values
V
Pressure level [hPa ]
400
600
0.02
0.04
0.06
0.08
0.1
0.12
0
Shape follows mean
900
920
MT
99%tile experiences
strongest increase
Means higher moisture supply during
extreme precipita- Fig.3: Vertical structure of the
dierence of percentiles of MT.
tion events
200
940
400
960
980
600
1000
0.01
0.011
0
0.012
0.013
0.005
0.014
0.015
10˚
10˚
0˚
0˚
60
−10˚
−10˚
−20˚
−20˚
−30˚
−30˚
0.01
0.015
10˚
0˚
0˚
−10˚
−10˚
−20˚
−20˚
−40˚
−40˚
0˚
−10˚
−10˚
−20˚
−20˚
−30˚
−30˚
22 Jun 2008, 0:00
Fig.1: Examples of a monthly mean and instantaneous ! eld,
blue: upward, red: downward, green line indicates boundary
across which transports are calculated
instantaneous masks exhibit highly irregular
ASCm not physically consistent with instanta-
neous wind and humidity elds
10
15
20
25
30
30
40˚
60˚
80˚
100˚
120˚
140˚
160˚
180˚
−160˚
−140˚
−120˚
−100˚
−80˚
−60˚
−40˚
−20˚
−25
−20
−15
−10
−5
0
5
10
15
20
25
30
0˚
30˚
20˚
10˚
10˚
0˚
0˚
−10˚
−10˚
−20˚
−20˚
−30˚
−30˚
(a) MTin
✭❛✮
❈✷✵ ❆❋❘
✭❜✮
❈✷✶ ❆❋❘
✭❝✮
✁ ❆❋❘
✭❣✮
❈✷✵ ❙❆▼
✭❤✮
❈✷✶ ❙❆▼
✭✐✮
✁ ❙❆▼
60
50
40
Fig.5: Change of percentage of time steps a grid box belongs
to ASC from the instantaneous vertical wind C21 - C20, top:
from ASCi , bottom: ASCm . Red (blue) indicates a box belongs to ASC more (less) frequently. Note the dierent scale of
the colour bar. Right column: Zonal mean percentage a grid
box belongs to ASCi /ASCm . Green denotes C21, red C20.
−8
−6
−4
−2
0
2
4
6
8
−25
−20
10
−15
−10
−5
0
5
10
15
20
25
30
Latitude
ITCZ much more pronounced in ASCm
widening of extent of ASCm in response to
warming
slight or no change of extent of ASCi
Masks of ASCi show a very irregular pattern
Increase for MT into ASC found at lower and
MTi, C20out
MTi, C20in
MTi, C21out
MTi, C21in
MTm, C20out
MTm, C20in
MTm, C21out
MTm, C21in
(b) MTout
out of ASC at mid levels
MT budget increases despite of this counteracting
Increase is most pronounced at high percentiles
of MTi events
Dierent answers from ASCi and ASCm
whether ITCZ extends in response to warming
MT towards land intensify, but no distinct circulation changes were found
Moisture budgets of land areas increase
Budi, C20
Budi, C21
Budm, C20
Budm, C21
References
5
0˚
[ kg *s-1 * m-1 ]
10˚
5
20˚
-2
10˚
-2.8
20˚
-3
20˚
-3.2
30˚
[ kg *s-1 * m-1 ]
30˚
pattern
0
30˚
9
-1.8
0˚
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
100˚ 110˚ 120˚ 130˚ 140˚ 150˚ 160˚ 170˚ 180˚ −170˚ −160˚ −150˚ −140˚ −130˚ −120˚ −110˚ −100˚ −90˚ −80˚ −70˚ −60˚ −50˚ −40˚ −30˚ −20˚ −10˚
11
90˚
10
80˚
9
70˚
8
60˚
7
50˚
[ kg *s-1 * m-1 ]
40˚
12
−30˚
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
Jun 2008
−30˚
Time series of in- and outward transports
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
10˚
10
2069
2072
2075
2078
2081
2084
2087
2090
2093
2096
2099
20˚
8
20˚
7
30˚
30˚
50
40
0.016
6
30˚
2069
2072
2075
2078
2081
2084
2087
2090
2093
2096
2099
0˚
40˚
20˚
0˚
20˚
800
40˚
10˚
−20˚
SUMMARY
99%
97%
95%
90%
75%
50%
25%
10%
5%
3%
1%
-1.6
100˚ 110˚ 120˚ 130˚ 140˚ 150˚ 160˚ 170˚ 180˚ −170˚ −160˚ −150˚ −140˚ −130˚ −120˚ −110˚ −100˚ −90˚ −80˚ −70˚ −60˚ −50˚ −40˚ −30˚ −20˚ −10˚
−40˚
70
MTi ERAint
MTm ERAint
MTm C20
MTm C21
MTi C20
MTi C21
-2.2
90˚
−60˚
20˚
20˚
in accordance with
Hadley Cell
Proles from EH5
and ERAint similar
Lower level inward
and mid level outward MT increase
Fig.2:
Vertical
proles
of
horizontal MT into ASC
with warming
0
-2.4
80˚
−80˚
Latitude
In-/outward pattern
-2.6
70˚
−100˚
−30
2069
2072
2075
2078
2081
2084
2087
2090
2093
2096
2099
60˚
−120˚
30˚
MT [kg *s-1 * m-1 * hPa-1 ]
50˚
−140˚
30˚
−10
1000
-0.005
40˚
−160˚
Changes of percentiles of vertical transports
dicated by subscripts i and m)
results in one ASC/DESC mask per time step
30˚
180˚
−30
MT [kg *s-1 * m-1 * hPa-1]
based on monthly mean and instantaneous ! (in-
20˚
160˚
30
Vertical profiles
0
Define regions of ascending and descending
air motion (ASC and DESC)
10˚
140˚
%
1000
METHOD, step 1
120˚
−30 −25 −20 −15 −10 −5
800
1960-1989: C20, A1B 2070-2100: C21
100˚
%
Pressure level [hPa ]
U
80˚
Frequency of ASCi
linking wind vectors with water content
along all boundary segments (horizontally and
200
!,
and , q, pressure information
0:5 , 30 vertical levels, 6h time intervals
reanalysis (ERAint) 1989 -2008
two IPCC scenarios from ECHAM5-model
60˚
70
RESULTS
Use of high space resolution quasiinstantaneous data (unlike many other such
40˚
Frequency of ASCm
driver of the global water cycle
carries and distributes moisture and energy
thus determines precipitation regionally
how will associated moisture transports change
in a warmer climate?
(c) Budget
Fig.4: Evolution of yearly MT below (a) and above (b) the
reversal level (at which MT = 0 ). (c) time series of the yearly
mean budget. Black lines indicate decadal means.
Zahn,
M.,
and
R.
P.
Allan,
(2011),
Changes
in
water
vapor
transports of the ,ascending
branch
of
the
tropical
circulation,
.
,
(accepted),
Climate
Warming
related
strengthening
of
the
tropical hydrological cycle,
, (submitted), Quantifying present and projected future atmospheric moisture transports onto land,
J. Geophys. Res. 116
Journal of Climate.
Water Resources Re-
search.
Fig.6: MTwall across shorelines of Africa and South America.
Left column: in C20, second column: in C21, right column:
C21-C20. red colours: landward MT, blue: seaward MT.
land- and seaward MT intensify, but rarely
change in sign
intensication of circulation strength and of the
hydrological cycle, but generally no change of
pattern
−150˚
−120˚
−90˚
−30˚
∆MT = 99.3
∆PE = 122.2
75˚
60˚
45˚
−60˚
∆MT = 43.0
∆PE = 50.2
30˚
0˚
30˚
∆MT = 74.6
∆PE = 232.2
∆MT = 53.2
∆PE = 80.2
∆MT = 3.3
∆PE = −3.1
90˚
∆MT = 17.1
∆PE = 27.8
∆MT = 18.3
∆PE = 24.8
∆MT = 44.3
∆PE = 56.1
−15˚
60˚
120˚
150˚
180˚
∆MT = 461.3
∆PE = 249.7
∆MT = −24.5
∆PE = −17.7
∆MT = −179.8
∆PE =
6.9
15˚
0˚
∆MT = −47.7
∆PE = −31.2
∆MT = 69.1
∆PE = 141.8
∆MT = −72.0
∆PE = −36.0
−30˚
−45˚
∆MT = −7.2
∆PE = −0.8
∆MT = −32.1
∆PE = 48.7
−60˚
−75˚
∆MT = 30.3
∆PE = 29.0
Fig.7: Change of moisture budget for land areas from transports and precipitation - evaporation, C21 - C20
−90˚
most land areas get wetter
sign from P-E and MT consistent in most areas
4 times daily MT may be biased by diurnal cycle
(not shown)
ACKNOWLEDGEMENT
Funding by NERC PREPARE project, NE/G015708/1