close

Вход

Забыли?

вход по аккаунту

?

636

код для вставкиСкачать
INTERNATIONAL JOURNAL OF CLIMATOLOGY
Int. J. Climatol. 19: 863–876 (1999)
TRENDS OF UPPER-AIR CIRCULATION AND WATER VAPOUR
OVER EQUATORIAL SOUTH AMERICA AND ADJACENT OCEANS
SCOTT CURTIS and STEFAN HASTENRATH*
Department of Atmospheric and Oceanic Sciences, Uni6ersity of Wisconsin, Madison, 1225 West Dayton Street, Madison,
WI 53706, USA
Recei6ed 30 April 1998
Re6ised 9 No6ember 1998
Accepted 16 No6ember 1998
ABSTRACT
A novel 40-year upper-air dataset along with satellite measurements have been analyzed to explore the vertical
circulation and water vapour transport and their long-term evolution over equatorial South America and the adjacent
oceans. Lower-tropospheric convergence, mid-tropospheric ascending motion, upper-tropospheric divergence, as well
as precipitable water, are concentrated in the realm of the Pacific and Atlantic Inter-tropical Convergence Zone
(ITCZ) throughout the year and the Amazon basin during austral summer. Water vapour is imported into the
Amazon basin from the equatorial and tropical South Atlantic and during austral summer also from the tropical
North Atlantic. The recognition of long-term developments in the tropical climate system hinges on the homogeneity
of the upper-air dataset. With this qualification, statistically significant increasing trends of lower-tropospheric
convergence, upward motion, upper-tropospheric divergence, convergence of atmospheric water vapour transport,
and precipitable water are found over the Amazon basin throughout the year. In all seasons, the pattern of
atmospheric water vapour transport shows a statistically significant trend towards the development of a clockwise
turning vortex over eastern Brazil, concordant with a centre of falling tendency in 1000 mb height and associated
vortex in the wind field. The trends towards reduced/enhanced water vapour import at the northern/southern side of
the vortex largely offset each other, so that the trend towards greater moisture convergence for the basin as a whole
results from the combination of increasing vapour import through the northern and decreasing export through the
western and southern boundaries of the Amazon watershed. Copyright © 1999 Royal Meteorological Society.
KEY WORDS: equatorial
South America; NCEP–NCAR reanalysis; water vapour; upper-air circulation
1. INTRODUCTION
Long-term evolutions in the global climate system are receiving increased attention in international
endeavours at global change (Houghton et al., 1990) and the implications of the large-scale water budget
are also being recognized (World Meteorological Organization–UNESCO, 1997). In this context, the
equatorial trough zone and the embedded Inter-tropical Convergence Zone (ITCZ) and the three
near-equatorial convection centres merit particular interest because, due to the intense latent heat release,
these limited domains act as hubs of the general circulation. South America contains the Amazon
convection centre and well-developed convergence zones extend from the adjacent Pacific and Atlantic
Oceans into the interior of the continent. The widely publicized deforestation underway in the Amazon
basin raises concern not only with respect to the disturbance of the regional natural environment and a
depletion of the gene pool, but also regarding possible consequences for the atmospheric circulation.
Numerical model experiments (Dickinson and Henderson-Sellers, 1988; Shukla et al., 1990; Nobre et al.,
1991; Walker et al., 1995) inferred substantial basinwide rainfall decrease induced by deforestation. These
* Correspondence to: Department of Atmospheric and Oceanic Sciences, University of Wisconsin, Madison, 1225 West Dayton
Street, Madison, WI 53706, USA.
Contract/grant sponsor: NSF; Contract/grant number: ATM-9732673
CCC 0899–8418/99/080863 – 14$17.50
Copyright © 1999 Royal Meteorological Society
864
S. CURTIS AND S. HASTENRATH
modelling results are not vindicated by satellite measurements of convection and records of rainfall and
river discharge (Richey et al., 1989; Rocha et al., 1989; Chu et al., 1994; Balling and Hughes, 1995; Dias
de Paiva and Clarke, 1995; Hastenrath, 1995; Marengo, 1995; Marengo et al., 1998). A novel upper-air
dataset spanning the recent 40 years, in conjunction with other surface and satellite evidence, invited a
reappraisal of this issue; the spatial continuity of information from the continent to the open oceans being
particularly fortunate. Diagnostically most insightful for the long-term evolution of the hydrometeorological conditions are the large-scale fields of precipitable water, lower-tropospheric convergence, vertical
motion, and upper-tropospheric divergence. An overview of the background circulation is offered in
Section 2, data and methods are described in Section 3, the analyses are presented in Sections 4 and 5, and
a synthesis is provided in Section 6.
2. BACKGROUND
The annual cycle of circulation and climate over the equatorial Americas and adjacent oceans has been
documented in a series of research papers and atlases (Hastenrath and Lamb, 1977, 1978; Virji, 1981;
Chu, 1985; Kousky, 1988; Figueroa and Nobre, 1990; Matsuyama, 1992; Mintz and Serafini, 1992;
Marengo et al., 1994; Rao et al., 1996; Hastenrath, 1997; Matsuyama and Masuda, 1997) and a brief
overview must suffice here. During austral summer, the near-equatorial low pressure trough and
embedded ITCZ are located far south. A separate band of lower-tropospheric convergence extends from
the South Atlantic northwestward into the interior of the continent, the South Atlantic Convergence Zone
(SACZ). A centre of intense convective activity sits over the south-eastern portion of the Amazon basin.
Located to the west of this is an upper-tropospheric topography maximum with divergent outflow, the
‘Bolivian high’. On its poleward side, the Southern Hemispheric Subtropical Westerly Jet is welldeveloped. While the Atlantic ITCZ continues to migrate southward from December–January–February
to around March – April, the centre of most intense convective activity shifts from the south-eastern
portion of the Amazon basin abruptly northwestward after March to attain a location near the
Panama–Colombia border and the adjacent eastern Pacific in May. After February, the Bolivian high
vanishes and the Southern Hemispheric Subtropical Westerly Jet weakens, the Atlantic ITCZ migrates
northward from April to August and begins to shift southward again thereafter. The centre of most
intense convection continues to dwell over the general area of the eastern equatorial Pacific until October,
and then it shifts abruptly southeastward to its December–March position over the south-eastern portion
of the Amazon basin. The peak of rainfall activity varies from November–January in the south-eastern
portion of the Amazon basin to around May –June–July at its northern extremity. Annual rainfall totals
are highest in the westernmost equatorial portion of the basin and here the rains peak around April.
3. DATA AND METHODS
The data sources used in this study include global upper-air analyses and satellite-derived outgoing
longwave radiation (OLR).
The National Centre for Environmental Prediction–National Centre for Atmospheric Research
(NCEP–NCAR) reanalysis (Kalnay et al., 1996; Kousky and Ropelewski, 1997) at a 2.5° latitude–
longitude resolution (Figure 1) was acquired for the years 1958–1997. Data were processed into individual
monthly mean fields. Elements of interest here are the fields of wind, vertical velocity, and humidity.
Information is available for the levels 1000, 925, 850, 700, 600, 500, 400, 300, 250, 200, 150, 100 and 30
mb. The observational input to the NCEP – NCAR exercise in assimilation and modelling became more
plentiful from 1979 onward. Accordingly, both real evolutions in the tropical climate system and
inhomogeneities in input and analysis may contribute to apparent differences between the earlier and later
part of the NCEP – NCAR record.
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
TRENDS OF UPPER-AIR CIRCULATION AND WATER VAPOUR
865
OLR from the scanning radiometer on the NOAA polar-orbiting satellites is available from June 1974
continuously to 1997, except for a gap from March 1978 to January 1979, at a 2.5° latitude–longitude
resolution for nearly the entire globe. A formula proposed by Gadgil et al. (1992) was applied to correct
for bias in OLR due to differences in the equatorial crossing time and uncertainties in scanner calibration
among satellites. In the tropics, low values of OLR are indicative of extensive areas of deep convection.
From the total wind field at selected levels velocity potential and streamfunction were computed as
described in Mancuso (1967), Krishnamurti (1971), and Krishnamurti et al. (1973), with full coverage
from 75°N to 75°S and a grid spacing of 2.5°. An inner subdomain from these near-global fields is used
here. From the fields of velocity potential, maps were then constructed of divergent wind component and
divergence.
Precipitable water and the transport of atmospheric water vapour were calculated as described in Rao
et al. (1996). They showed for this tropical region and the entire annual cycle that the transient eddy
transport is negligible, so that the total transport can be represented by the mean transport.
Precipitable water
PW =
&
1000
q
300
dp
g
(1)
where g is acceleration of gravity, q specific humidity, and the integration using the trapezoidal rule
extends from the 1000 to the 300 mb level.
The vertically integrated transport of atmospheric water vapour
Q=
&
1000
300
qVb
dp
g
(2)
where Vb is the horizontal wind vector. Individual monthly mean values of q and Vb were used here. For
purposes of trend calculations, Q may be separated into its zonal and meridional components as
appropriate.
For purposes of the present study, selected fields were analyzed from the NCEP dataset. Highly
pertinent to trends in the hydrometeorological conditions are the spatial patterns and long-term changes
of precipitable water, transport of atmospheric water vapour, lower-tropospheric convergence, vertical
motion, and upper-tropospheric divergence. Accordingly, the fields analyzed here are precipitable water
and vapour transport in the column from 1000 to 300 mb, divergence and divergent wind component at
850 mb, omega vertical velocity at 500 mb, and divergence and divergent wind component at 200 mb.
Maps are presented for April, the rainy season peak in the western equatorial portion of the Amazon
basin with the largest annual rainfall totals, and for the other cardinal months of January, July and
October. In addition to these map analyses compact index series of precipitable water, divergence of the
transport of atmospheric water vapour, 850 mb divergence, 500 mb vertical motion, and 200 mb
divergence were compiled for the Amazon basin as a whole (Figure 1), with an area of 6.1× 106 km2.
Trends are computed using the least-squares fit equation
y= a+ bx
(3)
Figure 1. Orientation map. Dots indicate gridpoints of NCEP analysis, solid line delineates entire Amazon basin. Tick marks and
letters indicate the N, E, S, and W segments of that boundary
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
866
S. CURTIS AND S. HASTENRATH
Figure 2. Maps of April 1958–1997 long-term mean conditions A – E and trends F – J. A. Mean 200 mb divergence, with isoline
spacing of 2 × 10 − 7 s − 1, and dashed lines indicating negative values; and divergent wind component, with arrows scaled at 1°
latitiude for 2 m s − 1. B. Mean vertical motion at 500 mb, with isoline spacing of 2 ×10 − 4 mb s − 1, dashed lines indicating negative
values or upward motion. C. Mean 850 mb divergence and divergent wind component, with symbols as for panel A. D. Mean
precipitable water with isoline spacing of 5 mm. E. Mean OLR with isoline spacing of 20 W m − 2. F. Trends in 200 mb divergence
with isoline spacing of 2 × 10 − 6 s − 1 per 40 years. G. Trends in 500 mb vertical motion with isoline spacing of 1 × 10 − 4 mb s − 1
per 40 years. H. Trends in 850 mb divergence with isoline spacing of 2× 10 − 6 s − 1 per 40 years. I. Trends in precipitable water with
isoline spacing of 2 mm per 40 years. J. Trends in OLR with isoline spacing of 20 W m − 2 per 40 years. In maps F – J shading
indicates trends significant at 5% level. For maps E and J observation period is 1974 – 1997
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
TRENDS OF UPPER-AIR CIRCULATION AND WATER VAPOUR
867
where b is the slope and a would correspond to the year 1958. The trends were tested for statistical
significance with the null hypothesis b = 0, as described in Spiegel (1975) (p. 289) and Oort et al. (1987).
4. LONG-TERM MEAN CONDITIONS
The 1958–1997 average conditions over the tropical Americas and adjacent oceans are portrayed in Figure
2 for April, the month of largest precipitation in the region of the Amazon basin with the largest annual
rainfall totals. For a more central area of South America and near vicinity, Figure 3 presents maps for
the cardinal months of the year, January, April, July and October. For the yet smaller domain of the Amazon
basin (Figure 1) Table I offers information for the entire annual cycle. The 850 mb level is included in Figure
2 to represent a deep lower-tropospheric layer, in which much of the transport of atmospheric water vapour
is concentrated.
Figure 2E shows for April the Atlantic and Pacific ITCZ and intense convective activity over northern
South America. Precipitable water (Figure 2D) is large in these regions of intense convection. Further,
consistent with the distributions in Figure 2E and D, are the patterns of lower-tropospheric convergence
(Figure 2C), mid-tropospheric vertical motion (Figure 2B), and upper-tropospheric divergence (Figure 2A).
Thus, overall the zones of intense convection and large precipitable water also feature large lowertropospheric inflow, ascending motion, and upper-tropospheric divergent outflow.
The pattern of the transport of atmospheric water vapour is in Figure 3, placed in context with the
lower-tropospheric circulation. Maps for all four cardinal months are meant to illustrate the essentials of
the annual cycle. For January, Figure 3(a) – (c), shows the consistency of the topography and the wind fields
at 1000 mb and the corresponding pattern of the transport of atmospheric water vapour, and Figure 3(d)
portrays the resultant pattern of the divergence of the vapour transport. These fields are further compared
in Figure 3(e)–(h) for April, Figure 3(i) – (l) for July, and Figure 3(m)–(p) for October. The Amazon basin
imports atmospheric water vapour from the equatorial and tropical South Atlantic and in austral summer
also from the tropical North Atlantic.
The spatial patterns for cardinal months portrayed in Figures 2 and 3 are complemented in Table I by
the summary of the annual cycle for the Amazon basin as a whole (Figure 1). Table I part A shows the
most intense lower-tropospheric convergence, ascending motion, upper-tropospheric divergence, and
convective activity, during the austral summer half-year, especially November–March, and the largest
precipitable water from December to April. Particularly large values of convergence of transport of
atmospheric water vapour, lower-tropospheric convergence, upward motion, upper-tropospheric divergence, and convective activity are found from around September to March. Precipitable water is largest
from November through May, or again lagging somewhat behind the vertical motion and convective activity.
Table I part B lists the transport of atmospheric water vapour across the boundaries of the Amazon basin
separately for the north, east, south and west segments (Figure 1), for all months and the year as a whole.
These should be compared with the maps Figure 3(c), (g), (k) and (o). Across the north segment there is
net import throughout the year, largest during austral summer. Of all segments, the east features by far
the greatest net import all year round, it stems from the equatorial and tropical South Atlantic, and reaches
its peak in austral winter. Across the South segment there is small export in all seasons. Consistent with
Table I part B and Figure 3(b), (f), (j) and (n), is the report of Rao et al. (1996) that the Amazon basin
is the main moisture source for central Brazil during the austral summer. In the west segment there is more
substantial export, especially in austral winter. The bottom line of Table I part B gives the resultant water
vapour input into the basin. Dividing these numbers by the total area of the basin of 6.1× 106 km2 and
multiplying by − 1 yields the divergence of the water vapour transport listed in the bottom line of Table
I part A. Table I part B and Figure 3(c), (g), (k) and (o), agree in general magnitude and overall pattern
with the results of Rao et al. (1996). The annual mean of the convergence of water vapour transport in
the bottom line of Table I part A is identical to that presented by Rao et al. (1996).
The evidence on the long-term average conditions summarized in Figures 2 and 3 and Table I serves
as background for the appraisal of long-term evolutions.
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
868
S. CURTIS AND S. HASTENRATH
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
Figure 3. Maps of the 1958–1997 long-term mean conditions for the cardinal months January (a) – (d), April (e)–(h), July (i)–(l), and October (m)–(p). Maps (a), (e), (i) and
(m) in leftmost column present 1000 mb topography, with isoline spacing of 20 m; maps (b), (f), (j) and (n), in second column from left show 1000 mb wind field, with isotach
spacing of 2 m s − 1; maps (c), (g), (k) and (o) in third column from left depict average transport of water vapour, with isoline spacing of 100 kg m − 1 s − 1; maps (d), (h), (l) and
(p) in rightmost column shows divergence of transport of atmospheric water vapour, with isoline spacing of 5 ×10 − 5 kg m − 2 s − 1; dashed lines indicating negative values or
convergence. Dotted line shows boundaries of the Amazon basin
J
F
Part A: Elements as for Figures 5 and 6
200 mb div (10−7 s−1)
+20 +21
500 mb v (10−5 mb s−1)
−32 −33
850 mb div (10−7 s−1)
−14 −14
PW (mm)
48
48
OLR (W m−2)
212
210
div Q (10−6 kg m2 s−1)
−30 −31
M
A
M
+19
−31
−12
48
211
−29
+15
−23
−10
47
221
−20
+10
−13
−6
45
237
−10
Part B: Net water vapour input across boundaries of basin,
N
+30 +27 +22 +14
E
+18 +25 +28 +36
S
−14 −14
−9 −10
W
−15 −19 −22 −28
NESW
+19 +19 +18 +12
J
J
+2
−0
−1
41
251
+3
A
S
O
N
D
Year
−2
+9
+0
38
258
+10
+2
+4
−2
37
255
+5
+9
−10
−7
41
243
−4
+14
−27
−9
45
227
−17
+14
−35
−11
47
218
−29
+15
−32
−12
47
215
−28
+12
−19
−8
44
230
−15
in 10−7 kg m−1 s−1
+7
+1
−2
+40 +41 +40
−10 −15 −15
−30 −30 −29
+6
−2
−6
+1
+40
−16
−27
−3
+6
+42
−19
−26
+2
+12
+34
−19
−17
+10
+19
+25
−14
−13
+18
+28
+19
−14
−15
+17
+14
+32
−14
−23
+9
869
Int. J. Climatol. 19: 863 – 876 (1999)
TRENDS OF UPPER-AIR CIRCULATION AND WATER VAPOUR
Copyright © 1999 Royal Meteorological Society
Table I. Summary of 1958–1997 mean values for the Amazon basin as a whole (see Figure 1)
870
S. CURTIS AND S. HASTENRATH
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
Figure 4. Maps of the 1958–1997 trends for the cardinal months January (a) – (d), April (e) – (h), July (i) – (l), and October (m)–(p). Maps (a), (e), (i) and (m) in leftmost column
present 1000 mb topography, with isoline spacing of 5 m per 40 years; maps (b), (f), (j) and (n) in second column from left show 1000 mb wind field, with isotach spacing of
2 m s − 1 per 40 years; maps (c), (g), (k) and (o) in third column from left depict average transport of water vapour, with isoline spacing of 50 kg m − 1 s − 1 per 40 years; maps
(d), (h), (l) and (p) in rightmost column shows divergence of transport of atmospheric water vapour, with isoline spacing of 5 ×10 − 5 kg m − 2 s − 1 per 40 years; dashed lines
indicating negative values or convergence. Shading denotes trends significant at the 5% level. Dotted line shows boundaries of the Amazon basin
TRENDS OF UPPER-AIR CIRCULATION AND WATER VAPOUR
871
5. TRENDS
The long-term evolutions over the period 1958–1997 are presented in Figure 2F–J for the tropical
Americas and adjacent oceans during April; in Figure 4 for the central domain during the cardinal months
January, April, July and October; in Figures 5 and 6 for the Amazon basin during April and the year as
a whole; and in Table II for the Amazon basin during all months and the year as whole. As discussed in
Section 3, the apparent trends may reflect not only real long-term evolutions in the climate system but
also changes in the observational input to the NCEP–NCAR assimilation and modelling. With this
qualification results have been tested for conventional statistical significance.
Figure 2J shows for April statistically significant increasing convective activity over the tropical North
Pacific, in the realm of the equatorial Atlantic ITCZ, and as well over the northern portion of the
Figure 5. Time series plots for April and Amazon basin as a whole (see Figure 1). A. 200 mb divergence, 10 − 7 s − 1; B. 500 mb
vertical motion, 10 − 4 mb s − 1; C. 850 mb divergence, 10 − 7 s − 1; D. precipitable water, in mm; E. OLR in W m − 2; F. divergence
of water vapour transport, kg m − 1 s − 1. Significance of trend depicted by solid line is given in upper-right corner of diagrams
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
872
S. CURTIS AND S. HASTENRATH
Figure 6. Time series plots for the year as a whole and the entire Amazon basin (Figure 1). A. 200 mb divergence, 10 − 7 s − 1; B.
500 mb vertical motion, 10 − 4 mb s − 1; C. 850 mb divergence, 10 − 7 s − 1; D. precipitable water, in mm; E. OLR, in W m − 2; F.
divergence of water vapour transport, kg m − 1 s − 1. Significance of trend depicted by solid line is given in upper-right corner of
diagrams
Amazon basin. Figure 2I portrays significantly increasing precipitable water in much of the equatorial
zone, markedly also over the Amazon basin. The maps Figure 2F, G and H, depicting the vertical
circulation, feature significant trends broadly concordant with those of OLR and precipitable water,
Figure 2J and I. For the Atlantic sector, comparison with Figure 2C, B and A, reveals the statistically
significant southward shift of the ITCZ noted before (Wagner, 1995). Particularly remarkable is the
evolution over the Amazon basin, where increasing lower-tropospheric convergence, mid-tropospheric
ascending motion, and upper-tropospheric divergence accompany the increasing trends in precipitable
water and convective activity.
On a smaller scale, noteworthy is also the notorious El Niño region of north-western Peru and
Ecuador, for which Figure 2F – I, show mutually consistent trends of increasing upper-tropospheric
divergence, mid-tropospheric ascending motion, lower-tropospheric convergence, and precipitable water.
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
J
F
Part A: Elements as for Figures 5 and 6
200 mb div (10−7 s−1)
+26 +25
500 mb v (10−5 mb s−1)
−31 −29
850 mb div (10−7 s−1)
−10 −11
OLR (W m−2)
−12
−4
PW (mm)
+3
+3
div Q (10−6 kg m2 s−1)
−51 −60
M
A
M
J
J
A
S
O
N
D
Year
+24
−29
−11
−7
+4
−61
+15
−23
−9
−9
+4
−46
+9
−25
−9
−2
+4
−36
+16
−28
−10
−6
+5
−29
+14
−25
−9
+6
+4
−21
+16
−28
−11
+9
+6
−20
+20
−39
−12
+3
+7
−33
+18
−28
−8
−1
+4
−24
+18
−29
−9
−11
+5
−29
+22
−28
−9
−4
+4
−37
+19
−28
−10
−4
+4
−37
+7
+9
+3
−6
+12
+9
−2
+8
+5
+20
+11
−6
+3
+7
+14
+9
−2
+1
+10
+17
+9
−3
−0
+16
+23
+10
−1
+5
+10
+23
Part B: Net water vapour input across boundaries
N
+10
+7 +12
E
−1
+6
−9
S
+4
+8
+9
W
+18 +16 +25
NESW
+31 +37 +37
of basin, in 10−7 kg m−1 s−1
+11 +14 +11
+6
−2
−7
−2
+3
+4
+4
+6
+6
+15 +11
+3
−3
+28 +22 +18 +13
Bold print indicates significance of trend at 5% level according to t-test.
873
Int. J. Climatol. 19: 863 – 876 (1999)
TRENDS OF UPPER-AIR CIRCULATION AND WATER VAPOUR
Copyright © 1999 Royal Meteorological Society
Table II. Summary of 1958–1997 trends in units indicated per 40 years for the Amazon basin as a whole (see Figure 1)
874
S. CURTIS AND S. HASTENRATH
Contrasting with these are the trends for the Peruvian–Bolivian Altiplano, where associations with the
Southern Oscillation tend to run opposite to those for the Ecuador–Peru coast. It may remain open
whether such patterns should be seen in context with increasing frequency of El Niño events.
Figure 4 presents the patterns of trends in lower-tropospheric fields for the cardinal months January,
April, July and October, in an arrangement to be compared with the corresponding maps of long-term
mean conditions in Figure 3. Most noteworthy in all months is the trend toward lower pressure centred
over eastern Brazil and the evolution towards a cyclonic circulation concomitant with this low pressure
center. The trend towards a cyclonic gyre in the transport of atmospheric water vapour is a direct
implication of this evolution in the lower-tropospheric circulation.
For the Amazon basin (Figure 1) detailed time series plots of the long-term development are presented
in Figure 5 for April and in Figure 6 for the year as a whole. There are statistically significant and
mutually consistent trends towards larger lower-tropospheric convergence (Figures 5C and 6C), ascending
motion (Figures 5B and 6B), upper-tropospheric divergence (Figures 5A and 6A), precipitable water
(Figures 5D and 6D), and divergence of moisture transport (Figures 5F and 6F). Compatible with these
evolutions are the decreasing trends in the short series of OLR (Figures 5E and 6E), although these do
not reach statistical significance.
The time series plots Figures 5 and 6 are also interesting regarding the possible effect of more plentiful
observational input to the NCEP – NCAR assimilation and modelling after 1979. Common to the various
time series are the following characteristics; (i) trends are not dominated by contrastingly extreme values
near the beginning and end of the record; (ii) trends are strongest for 1958–1979 and less definite for
1980–1997; (iii) there is no sharp hiatus between the first and second halves of the record. In context, the
apparent trends do not seem to be merely the result of more plentiful observational input after 1979.
Table II for the Amazon basin (Figure 1) places the evidence depicted in Figures 3 and 4 into the larger
context of the annual cycle and the year as a whole. Table II part A shows statistically significant and
mutually consistent trends towards larger convergence of atmospheric water vapour, lower-tropospheric
convergence, ascending motion, upper-tropospheric divergence, and precipitable water, in all months of
the year. Compatible with these evolutions, the shorter series of OLR show prevailingly decreasing trends,
although mostly these do not reach statistical significance.
Table II part B lists the trend of the water vapour transport across the four segments of the basin
boundary indicated in Figure 1. These numbers should be appreciated in context with the maps Figure
4(c), (g), (k) and (o). Across the North segment there is a statistically significant increasing trend of
vapour input throughout the year, largest during austral summer. This is consistent with the long-term
evolution of inter-hemispheric sea surface temperature gradients and the southward shift of the nearequatorial wind confluence in the tropical Atlantic sector reported in earlier work (Wagner, 1995). Little
long-term variation is found for the east segment as a whole, where the opposing trends on the northern
and southern sides of the vortex apparent in Figure 4(c), (g), (k) and (o), largely compensate each other.
The south segment shows throughout the year a weak trend towards smaller moisture export. The west
segment features a statistically significant decreasing trend in export for austral summer and the year as
a whole. The combination of increasing import through the north segment and decreasing export through
the west and south segments results in a statistically significant trend towards increasing convergence of
water vapour transport for the basin as a whole (Table II A, line 6: Table II B, line 5).
The results of Figures 5 and 6 and Table II are complemented by reports of increasing trends in the Rio
Negro water level and various other rivers in the northern portion of the Amazon basin (Marengo et al.,
1998).
6. CONCLUSIONS
In the ongoing research on global change the near-equatorial convergence zones and quasi-permanent
centres of convective activity deserve particular attention because these play a pivotal role in the general
circulation. The present study evaluated the 40-year NCEP dataset to explore the long-term evolution of
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
TRENDS OF UPPER-AIR CIRCULATION AND WATER VAPOUR
875
upper-air circulation, transport of atmospheric water vapour, and precipitable water over equatorial
South America and the contiguous tropical Atlantic and Pacific Oceans. This analysis produced no
evidence for the kind of long-term changes inferred from a host of numerical experiments simulating the
consequences of deforestation in the Amazon basin. Instead, the evolutions over the South American
continent should be appreciated in the context of the large-scale circulation setting. While the observational input to the NCEP – NCAR assimilation and modelling is more plentiful after 1979, trends are
overall strongest over the first two decades of the record.
For the year as a whole and especially during austral summer, the atmosphere over the Amazon basin
exhibits statistically significant and mutually consistent increasing trends in lower-tropospheric convergence, ascending motion, upper-tropospheric divergence, and along with these increasing convergence of
the atmospheric water vapour transport and precipitable water. The increase in precipitable water is seen
as a direct consequence of the enhanced upward motion over the Amazon basin. Indeed, by way of
contrasting example, in the realm of the Pacific ITCZ in April decreasing upward motion is accompanied
by diminishing precipitable water.
The increasing trend in the moisture convergence is consistent with the intensifying upward motion and
also requires appropriate changes in the pattern of the atmospheric water vapour transport. In the
long-term mean, atmospheric water vapour is imported into the Amazon basin from the equatorial and
tropical South Atlantic and in austral summer also from the tropical North Atlantic. Regarding the
evolutions over 40 years in the pattern of atmospheric moisture transport, the most conspicuous feature
is the trend towards a clockwise turning vortex over eastern Brazil, i.e. the segment of the basin boundary
through which, in the long-term mean, the largest share of water vapour is imported from the equatorial
and tropical South Atlantic. This vortex in the moisture transport field coincides with a centre of falling
tendency in 1000 mb height and accompanying vortex in the wind field. However, although conspicuous,
this feature is not relevant: enhanced vapour import on the southern side of the vortex is compensated by
reduced import on its northern side, such that the net effect in the vapour import across the eastern
boundary of the basin is negligible. Instead, the increasing trend in moisture convergence is sustained by
long-term changes in the transport across other portions of the basin boundary where long-term mean
transports are less substantial than in the eastern segment. Thus, import across the northern segment of
the basin boundary shows a statistically significant increasing trend throughout the year and most
markedly in austral summer. This is consistent with acceleration of the North Atlantic tradewinds and
other trends in the circulation of the Atlantic sector reported in earlier work (Wagner, 1995). Contributing
in the same sense are the statistically significant trends towards decreasing water vapour export across the
western and southern segments of the boundary, again most pronounced in austral summer.
Thus, while the equatorial and tropical South Atlantic is the major moisture source region for the
Amazon basin in the long-term mean, the 40-year increasing trend in the convergence of the atmospheric
vapour transport is primarily sustained by increasing moisture import from the tropical North Atlantic,
that accompanies the strengthening of the northeast tradewinds and southward shift of the nearequatorial wind confluence. The increasing upward motion over Amazonia has as a consequence
increasing precipitable water and also diminishing lower-tropospheric outflow downstream across the
western and southern boundaries of the basin. The increasing trend in water vapour convergence into the
Amazon watershed favours an enhancement of precipitation and river discharge, for which the reported
evidence is tenuous.
ACKNOWLEDGEMENTS
This study was supported by NSF Grant ATM-9732673.
REFERENCES
Balling, R.C. and Hughes, W.S. 1995. ‘Comments on ‘‘Detecting climate change concurrent with deforestation in the Amazon basin:
which way has it gone?’’ ’, Bull. Am. Meteorol. Soc., 76, 559.
Chu, P.S. 1985. ‘A contribution to the upper-air climatology of tropical South America’, J. Climatol., 5, 403 – 416.
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
876
S. CURTIS AND S. HASTENRATH
Chu, P.-S., Yu, Z.-P. and Hastenrath, S. 1994. ‘Detecting climate change concurrent with deforestation in the Amazon basin: which
way has it gone?’, Bull. Am. Meteorol. Soc., 75, 579–583.
Dias de Paiva, E.M.C. and Clarke, R.T. 1995. ‘Time trends in rainfall records in Amazonia’, Bull. Am. Meteorol. Soc., 76,
2203 – 2209.
Dickinson, R.E. and Henderson-Sellers, A. 1988. ‘Modeling tropical deforestation: a study of GCM land-surface parameterizations’,
Q. J. R. Meteorol. Soc., 114, 439–462.
Figueroa, S.N. and Nobre, C.A. 1990. ‘Precipitation distribution over central and western tropical South America’, Climanálise, 5,
36 – 44.
Gadgil, S., Guruprasad, A. and Srinivasan, J. 1992. ‘Systematic bias in the NOAA outgoing longwave radiation dataset?’, J. Clim.,
5, 867 – 875.
Hastenrath, S. 1995. Climate Dynamics of the Tropics. Kluwer, Dordrecht, 488 pp.
Hastenrath, S. 1997. ‘Annual cycle of upper air circulation and convective activity over the tropical Americas’, JGR-Atmos.,
102(D4), 4267 – 4274.
Hastenrath, S. and Lamb, P. 1977. Climatic Atlas of the Tropical Atlantic and Eastern Pacific Oceans, University of Wisconsin Press,
Madison, WN, 117 pp.
Hastenrath, S. and Lamb, P. 1978. Heat Budget Atlas of the Tropical Atlantic and Eastern Pacific, University of Wisconsin Press,
Madison, WN, 103 pp.
Houghton, J.T., Jenkins, G.J. and Ephraums, J.J. (eds.) 1990. Climate Change, The IPCC Scientific Assessment. Cambridge
University Press, Cambridge, 364 pp.
Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, M., Saha, S., White, G., Woollen, J., Zhu,
Y., Chelliah, M., Ebisuzaki, W., Higgins, W., Janowiak, J., Mo, K.C., Ropelewski, C., Wang, J., Leetma, A., Reynolds, R., Jenne,
R. and Joseph, D. 1996. ‘The NCEP/NCAR 40-year reanalysis project’, Bull. Am. Meteorol. Soc., 77, 437 – 471.
Kousky, V.E. 1988. ‘Pentad outgoing longwave radiation climatology for the South American sector’, Re6ista. Brasileira Meteorol.,
3, 217 – 231.
Kousky, V.E. and Ropelewski, C.F. 1997. The Tropospheric Seasonally Varying Mean Climate O6er the Western Hemisphere
(1979 – 1995), NCEP/Climate Prediction Center, Atlas No. 3, 135 pp.
Krishnamurti, T.N. 1971. ‘Tropical east–west circulation during the Northern summer’, J. Atmos. Sci., 28, 1342 – 1347.
Krishnamurti, T.N., Kanamitsu, M., Koss, W.J. and Lee, J.D. 1973. ‘Tropical east – west circulation during the Northern winter’,
J. Atmos. Sci., 30, 780–787.
Mancuso, R.L. 1967. ‘A numerical procedure for computing fields of streamfunction and velocity potential’, J. Appl. Meteorol., 6,
994 – 1001.
Marengo, J.A., Miller, J.R., Russell, G.L., Rozenzweig, C.E. and Abramopoulus, F. 1994. ‘Calculations of river runoff in the GISS
GCM: impact of a new land-surface parameterization and runoff routing model on the hydrology of the Amazon river’, Clim.
Dyn., 10, 349 – 361.
Marengo, J.A. 1995. ‘Variations and change in South American streamflow’, Clim. Change, 31, 99 – 117.
Marengo, J.A., Tomasella, J. and Uvo, C.B. 1998. ‘Trends in streamflow and rainfall in tropical South America: Amazonia, Eastern
Brazil and Northwest Peru’, JGR-Atmos., 103(D2), 1775 – 1783.
Matsuyama, H. 1992. ‘The water budget in the Amazon river basin during the FGGE period’, J. Meteor. Soc. Jpn., 70, 1071 – 1083.
Matsuyama, H. and Masuda, K. 1997. ‘Estimates of continental-scale soil wetness and comparison with the soil moisture data of
Mintz and Serafini’, Clim. Dyn., 13, 681–689.
Mintz, Y. and Serafini, Y.V. 1992. ‘A global monthly climatology of the soil moisture and water balance’, Clim. Dyn., 8, 13 – 27.
Nobre, C.A., Sellers, P.J. and Shukla, J. 1991. ‘Amazonian deforestation and regional climate change’, J. Clim., 4, 957 – 988.
Oort, A., Pan, Y.H., Reynolds, R.W. and Ropelewski, C.F. 1987. ‘Historical trends in the surface temperature over the oceans based
on the COADS’, Clim. Dyn., 2, 29–38.
Rao, V.B., Cavalcanti, I.F.A. and Hada, K. 1996. ‘Annual variation of rainfall over Brazil and water vapour characteristics over
South America’, JGR-Atmos., 101(D21), 26539–26551.
Richey, J.E., Nobre, C. and Deser, C. 1989. ‘Amazon river discharge and climate variability: 1903 – 1985’, Science, 246, 101 – 103.
Rocha, H.R., Nobre, C.A. and Barros, M.C. 1989. ‘Variabilidade natural de longo prazo no ciclo hidrológico da Amazônia’,
Climanálise, 3, 36 –43.
Shukla, J., Nobre, C.A. and Sellers, P.J. 1990. ‘Amazonia deforestation and climate change’, Science, 247, 1322 – 1325.
Spiegel, M.R. 1975. Probablility and Statistics. Shaum’s Outline Series, McGraw-Hill, New York, 372 pp.
Virji, H. 1981. ‘A preliminary study of summertime tropospheric circulation patterns over South America estimated from cloud
winds’, Month. Weather Re6., 109, 599–610.
Walker, G.K., Sud, Y.C. and Atlas, R. 1995. ‘Impact of the ongoing Amazonian deforestation on local precipitation: a GCM study’,
Bull. Am. Meteorol. Soc., 76, 346–361.
Wagner, R.G. 1995. ‘Decadal-scale trends in mechanisms controlling meridional sea surface temperature gradients in the tropical
Atlantic’, JGR-Oceans, 101(C7), 16683–16694.
World Meteorological Organization–UNESCO, 1997. Seventh Planning Meeting on World Climate Programme — Water, Final
Report. Koblenz, Germany, May 1997, WCAS P-45, WMO-TD-No. 854, Geneva, Switzerland, 97 pp.
Copyright © 1999 Royal Meteorological Society
Int. J. Climatol. 19: 863 – 876 (1999)
Документ
Категория
Без категории
Просмотров
5
Размер файла
431 Кб
Теги
636
1/--страниц
Пожаловаться на содержимое документа