1. Introduction
The 1976/77 climate transition is a global coupled atmosphere–ocean system variation that links the Pacific sea surface temperature (SST) variability and climate parameters almost all over the world (Namias 1978; Houghton et al. 2001; Solomon et al. 2007). Some authors, such as Mantua et al. (1997) and Mantua and Hare (2002) among others, related it to the Pacific decadal oscillation (PDO). The PDO is a low-frequency (20–30 yr) oceanic oscillation of the North Pacific SSTs, defined by Mantua et al. (1997) as the first empirical orthogonal function (EOF) of the monthly SSTs from the Hadley SST dataset, to the north of 20°N. The PDO was devised to reflect the interdecadal variability of the North Pacific in a simple index with substantial expression through the tropical and South Pacific (Dettinger et al. 2001). During the “cold” PDO phase, “colder”-than-normal SSTs are favored in the eastern and equatorial Pacific offshore North America and values above normal in the northwestern and western sectors of the North Pacific. During the “warm” PDO phase the eastern North Pacific warms and the eastern sector cools. The last cold PDO phase ranges from 1947 to 1976, and afterward, in the austral 1976/77 summer, it changed to a warm phase that apparently ended in 1999, although a new cold phase of the interdecadal oscillation in not yet fairly established (Huang et al. 2005).
The shift from the cold to the warm PDO phase during the 1970s was climatically evidenced as a significant change in the statistical population of more than 40 environmental variables along the Pacific coasts and in the Americas between the periods 1968–75 and 1977–84 (Ebbesmeyer et al. 1991). This makes the 1970s PDO shift, or even the 1970s El Niño–like variability shift, a special issue that deserves to be called the 1976/77 climate transition. Globally, the signal of change is evidenced as a predominance of climate conditions similar to those observed during El Niño events [the positive phase of the El Niño–Southern Oscillation (ENSO)], for which the central-equatorial Pacific SST variability is between 2 and 6 yr, approximately. After the 1976/77 summer, not only is the frequency of the El Niño events higher than the La Niña ones, but some El Niño signatures on climate are the stronger and longer lasting (McPhaden and Zhang 2002).
For North America, the 1976/77 climate transition dynamics implies a deepening of the Aleutian low that drives equatorward the jet stream path over the United States with concomitant changes in the dry/wet patterns in Mexico and southern United States (Trenberth 1990; Trenberth and Hurrel 1994). In South America (SA) the 1976/77 climate transition leads to an enhancement of the conditions imposed on the basic flow during the positive ENSO phase, that is, El Niño–like situations, at tropical and subtropical latitudes (Nogués-Paegle et al. 2002; Marengo 2004).
The basic SA low-level atmospheric circulation features established in the summer season (Fig. 1) consist of the quasi-stationary subtropical South Pacific anticyclone (SPA) to the west over the Pacific Ocean, the quasi-stationary subtropical South Atlantic anticyclone (SAA) to the east in the Atlantic, the “Chaco” continental orographic–dynamic–thermal low (ChL; Seluchi et al. 2003), the easterlies over tropical–equatorial Atlantic and Amazonia, and the midlatitude westerlies as a result of anticyclone/cyclone synoptic-scale activity. The easterlies are naturally deflected by the Andes orography toward the southeast and south into subtropical latitudes producing a north–south low-level flow (LLF) that links the Amazonian forests with the SA subtropical plains. During El Niño events, subsidence is increased over Amazonia and the Atlantic Ocean, around 5°N, because of an anomalous upper-troposphere east–west circulation (an anomaly in the Walker circulation) generated by increased tropical convection in the equatorial Pacific (Lenters and Cook 1995, 1999; Nogués-Paegle et al. 2002; Vera et al. 2006). The subsidence is associated with an intensification of the low-level easterlies in north Brazil and increased moisture budget and precipitation in Amazonia. The greater easterlies favor enhanced LLF southward into SA, with increased precipitation at tropical and subtropical latitudes and weakening of the activity of the South Atlantic Convergence Zone (SACZ; Carvalho et al. 2002, 2004). In other words, the anomalous mass distribution associated with the positive ENSO phase is based on a pressure decrease in the South Pacific from subtropical–tropical latitudes toward central SA and a pressure enhancement in the Atlantic basin toward northwestern Africa (Nogués-Paegle et al. 2002). As a result, both the summer zonal pressure gradient, which is established between the SAA and the ChL, and the concomitant meridional LLF can be reinforced or enhanced. Marengo (2004) shows that after the 1976/77 summer enhanced El Niño conditions (enhanced easterlies and precipitation) are observed over Amazonia and equatorial and the tropical North and South Atlantic. Hence, a further enhanced LLF with associated moisture advection from tropical to subtropical lands could be expected to occur after the 1976/77 climate transition. This LLF variability has not yet been confirmed at multidecadal scales; nonetheless, several studies reveal increased precipitation within the entire SA subtropical portion (Barros et al. 1996; Agosta et al. 1999; Minetti et al. 2003; Liebmann et al. 2004; Boulanger et al. 2005; Haylock et al. 2006).
Therefore, our aim is to analyze the effects of the 1976/77 climate transition during the austral summer upon the low-level atmospheric circulation in southern SA, south of 15°S. Changes in the basic atmospheric circulation features such as the SPA, the SAA, and the westerlies will be examined using principal component analysis (PCA) and inhomogeneity tests. The analysis will be complemented by examining climate changes observed within southern South America in the 1970s. Particularly, the summer precipitation variability in central-western Argentina (CWA; approximately at 28°–37°S and 65–70°W) underwent a significant shift toward lower frequencies in the 1976/77 summer, producing a prolonged wet event during the last decades of the twentieth century (Compagnucci et al. 2002). A comparative analysis between the atmospheric circulation associated with CWA summer extreme precipitation and the circulation field changes is presented.
2. Data and methodology
Monthly and 1200 UTC daily atmospheric fields of geopotential height (GH; in m), specific humidity (Q; in g kg−1), and vector wind (V; in m s−1) from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis I data products on a 2.5° latitude–longitude resolution are used for the atmospheric circulation analysis. The reanalysis is available from 1949; however, because of the scarcity of data over the Southern Hemisphere for assimilation prior to the International Geophysics Year, only data after 1958 are considered in this study (Kistler et al. 2001). Furthermore, while it is true that Bengtsson et al. (2004) have argued that the changes resulting from the inclusion of satellite observations, beginning in 1979, in the reanalysis processing can affect trends, this does not represent a problem since spatial patterns and spatial means are considered in this study, particularly in the subtropics and midlatitudes. As noted subsequently, the NCEP–NCAR data products analysis is compared with direct observational products, which further precludes the influence or presence of spurious behavior in the detected atmospheric circulation patterns. In turn, since the 1976/77 climate transition toward “warmer” conditions in the equatorial central Pacific SSTs may be reversing since early 2000s (Huang et al. 2005), the atmospheric circulation analysis period will only be extended until the summer of 1997/98, encompassing a period of 40 yr. The turning point of the ENSO-like SSTs’ multidecadal variability lies just in between.
Positive (negative) values of the percentage index anomaly denote wet (dry) summers. In the present study the index is computed for the period 1958/59–2003/04 using the 1958/59–1997/98 baseline.
Monthly precipitation data from the University of Delaware (Udel dataset) on a 0.5° latitude–longitude resolution is used to compare the precipitation fields in areas beyond CWA. The Udel dataset is built through the interpolation of land-observed precipitation on an equally spaced grid, available for the period 1953–99 (Udel version 2 is available online at www.cdc.noaa.gov/cgi-bin/db_search/DBSearch.p/?Dataset=Univ.+Of+Delaware+Precipitation+And+Air+Temp.&Variable=Precipitation&group=1&submit=Search). Agosta and Compagnucci (2006) show that the spatial patterns obtained using the Udel dataset and the meteorological stations within CWA are comparable.
The effects of the austral summer 1976/77 climate transition on the atmospheric circulation are first detected using Yamamoto’s statistic test for interannual time series inhomogeneity (Yamamoto et al. 1986). The formulation of the test is as follows.
Here Cpb and Cpa are the confidence limits of probability p for Mb and Ma, respectively. It is reasonable to conclude that the discontinuity of time means, or the inhomogeneity in time series, can be detected with p% confidence at the reference years when the value of Jt,p is over 1.0. Usually Jt,p is greater than 1.0 during the several years around the exact year of discontinuity. If that is the case, then the year of maximum Jt,p is assumed as the discontinuity year tr, as suggested in Yamamoto et al. (1986).
Thus, the test permits the detection of the occurrence of an inhomogeneity (a “jump” or a leap) in time series, determining the year tr, when Jt,1-α > 1.0 is maximum, at 1-α interval of confidence. The year tr is defined as the exact reference year for the inhomogeneity in the time series. In our analysis Yamamoto’s test is applied to the tropospheric summer GH fields centered on the 1970s over southern South America. The statistics are estimated for different size samples from N = 7 to N = 12 to verify the stability. The analysis of the low–midtroposphere GH summer fields over South America during the 1970s inhomogeneity shows areas of Jt,1-α > 1.0 that maximize in the 1976/77 summer. Then, this year results are the exact reference year according to the test. Note that in this region the method applied does not detect jumps in the 1979 timeframe that would suggest the presence of spurious behavior from reanalysis assimilation issues, as previously noted.
We will show the maps corresponding to the exact reference year at N = 9. The areas of discontinuity are shaded in gray at a confidence level α = 0.05. The inhomogeneity test is complemented by estimating the summer GH difference field between the climatology means in the period 1958/59–1975/76 (CL1) and 1979/80–1997/98 (CL2), that is, before and after the global climate transition, each period having 18 yr for comparison. The time when the transition or jump actually occurs is not considered in any of the 18-yr averages to ensure that the system is in a quasi steady state, that is, so that no actual transition patterns contaminate the samples. In consequence, the “old” and “new” circulation patterns can be more clearly discerned. The difference between the means is performed as CL2 − CL1 and their significance is tested using the Student’s t test for differences in means at α = 0.05, assuming unequal variances between the samples (Moser and Stevens 1992).
In the PCA, raw 850-hPa GH daily fields are used as the input data matrix (referred to as T mode, according to Preisendorfer 1988) on the domain 15°–70°S and 110°–30°W. The similitude matrix is the correlation between variables (daily fields). For the analysis, summer is defined from 1 October to 31 March, for two distinct periods: 1958/59–1976/77 and 1978/79–1997/98. The PCA technique helps find spatial patterns (PC scores) that typify the atmospheric circulation, which can then be climatically identified with regional synoptic regimes. According to North et al. (1982), the resulting PCs are degenerated when eigenvalues are close to one another. Richman (1986) and Compagnucci and Richman (2008) suggest applied rotation to avoid the problem. Despite the unrotated eigenvalues in the current study are well separated, the Varimax rotation is applied to the input data matrix to prevent potential degeneration. Yet the Varimax rotation allows us to isolate individual modes of variation and to better interpret the physics of the real fields embedded in it (Richman 1986; Compagnucci and Richman 2008). The analysis is complemented performing 30-daily field compositions of 850-hPa V, 850-hPa Q, and 850- and 500-hPa GH for those days with the highest PC loadings (correlation value time series) associated with each PC score (spatial pattern). The differences found in the PC scores and associated explained variances for each analyzed period can help determine essential features before and after the 1976/77 climate transition. Furthermore, the summer GH and precipitation fields associated with extreme dry and wet years in the CWA region are composited and compared to the PC spatial patterns found. The criterion to determine extreme summers in CWA is that of the first and third quartiles for the P-index distribution, which are displayed in Table 2. As already noted, such a comparison is useful to validate the fields obtained through the analysis of NCEP–NCAR data products.
3. Results
a. The low-level atmospheric circulation climatology before and after the 1976/77 summer over southern South America
The significant changes between the summer GH field climatology CL1 (1958/59–1976/77) and CL2 (1979/80–1997/98) are detected using the two-tailed Student’s t test for differences in means at α = 0.05, assuming unequal variances between the samples. The difference between the means is performed as CL2 − CL1. In Fig. 2, the shading in the left panels shows the areas where the difference anomalies are significant according to the Student’s t test. The right panels show the results after applying Yamamoto’s statistics on the summer GH fields during the period 1970–1981 (see section 2), which revealed that the 1976/77 summer is the exact reference year in some areas in the vicinity of southern SA.
The low-tropospheric (850 hPa) GH difference field shows a significant anticyclonic low-level tropospheric anomalous circulation over subtropical latitudes in the continent and at midlatitudes over the southwestern South Atlantic (Fig. 2a, left column). The midtropospheric (500 hPa) significant GH anomalies give anticyclonic anomalous circulation in the continent centered at midlatitudes (Fig. 2b, left column). The tropical-to-midlatitude significant areas of the GH anomalies in the lower troposphere are noticeably similar in shape and location to those of the significant Yamamoto’s statistics values at the exact reference year 1976/77 [max(J0.95,1976/77 ≥ 1.0) Fig. 2, right column]. Recalling that Yamamoto’s statistics show the areas of significant jumps in the interannual time series, the similarity in shape and location of these features is indicative that the significant GH differences between the climatology CL2 and CL1 are associated with a statistical “jump” or sudden change in the population of the tropospheric circulation around the 1976/77 summer. Furthermore, the subtropical–midlatitude anticyclonic anomalies denoted by the difference field between CL2 and CL1 point to lesser midlatitude cyclonic activity and/or a southward displacement of the SAA over southern SA after the 1976/77 summer.
In contrast, at latitudes south of 60°S, the low-to-midtroposphere GH differences are negative (cyclonic), yet they are not detected by Yamamoto’s statistics. For this reason these anomalies are suspected not to be related to a jump, as the climate transition of 1976/77 can be, but ought to be viewed as a progressive process.
There are previous results that hypothesize that such a process could be a tendency or trend in one of the principal modes of variability of the SH atmospheric circulation. Primarily, van Loon et al. (1993) encountered a sudden change by the late 1970s that consists of an intensification of a midlatitude hemispherical wavenumber 3 with a decrease in pressure over Antarctica using the Australian database in the period 1972–89. Later, Hurrel and van Loon (1994) showed changes in the annual cycle of surface winds and pressure at mid- and high latitudes, due to a weakening of the Semiannual Oscillation (SO), by comparing data from the 1970s to the 1980s. Afterward, a principal mode of oscillation in the SH atmospheric circulation was identified at mid–high latitudes that is detected both from reanalysis information (Thompson and Wallace 2000) and surface station data (Marshall 2003). This mode shows a significant tendency in the last decades (Marshall 2003; among others). The spatial pattern associated with this mode is essentially a zonally asymmetric, equivalent barotropic annular structure that has synchronically opposing-sign anomalies between the Antarctic and the midlatitudes. It is known as the southern annular mode (SAM; Limpasuvan and Hartmann 1999). A positive (negative) SAM phase is associated with low (high) pressures over Antarctica and higher (lower) pressures at midlatitudes, near 45°S. This tendency was first detected by Kidson (1999), and Gong and Wang (1999) using NCEP–NCAR reanalysis data products. Thompson and Solomon (2002) showed that it is mainly a significant summer/autumn signal, using exclusively radiosonde data over Antarctica. More recently, Marshall (2003) confirmed the tendency toward a positive SAM phase from the 1970s using an unbiased empirical index (the normalized monthly pressure around 45° and 60°S from station data). In addition, he identified Hurrel and van Loon’s (1994) SO weakening with the SAM positive tendency.
In consequence, the high-latitude negative GH differences between CL2 − CL1, which are not evidenced as a jump by Yamamoto’s statistics, could probably be the natural expression of the observed tendency toward a positive SAM. In fact, the hemispherical CL2 − CL1 field shows negative anomalies around Antarctica and positive anomalies at midlatitudes (figures not shown). In consequence, the high-latitude GH difference anomalies are unrelated to the climate transition 1976/77.
Conversely, some meteorological variables can support the significant and positive GH difference anomalies between periods before and after 1976/77 that are detected as a “jump” at southern SA mid-to-subtropical latitudes. Barros and Scasso (1994) found a positive tendency in pressure over Patagonia, around 45°S, linked to a southward displacement of the westerlies in the area from 1970s. Likewise, Camilloni (1999) suggested a southward displacement of the SAA (about 5°) to explain the observed variations in the urban heat island effect over Buenos Aires (35°S, 59°W). Besides, Barros et al. (2000a) suggested that upper-tropospheric circulation systems over subtropical Argentina may have been displaced southward in order to account for the positive trend in precipitation observed since 1970s. Furthermore, Carril et al. (1997) showed that the Mendoza River shifted its streamflow tendency from negative to positive in 1977. Waylen et al. (2000) found that the change in the Mendoza River streamflow (33°S approximately) is also present in other Argentine Andean rivers around 1976, and suggested the return to the flow conditions previous to 1935. Furthermore, results from Minetti and Vargas (1998), Minetti et al. (2003), and Rusticucci and Penalba (2003) show negative precipitation trends in southwestern SA (southern Chile and Patagonia) for the last three decades. The latter suggest that the trends can be related to decrease in frontal activity at midlatitudes, as observed in the latitude bands 30°–40°S in Australia (Allan and Haylock 1993). Haylock et al. (2006), using canonical correlation analysis on precipitation and SSTs, indicate that the negative trends in precipitation over southern SA could be explained by a reduction in cyclonic activity in the area. In this sense, using rotated EOFs, Vera (2003) found a significant change by 1976, as if a “jump” had taken place in the score series of the third principal pattern, associated with Southern Hemisphere midlatitudes storm-track activity over southeastern Pacific.
Notwithstanding, the mid-1970s changes in the tropospheric circulation over SA are not constrained to mid- and subtropical latitudes alone. According to Chen et al. (2001), from the mid-1970s there is an increase in low-level moisture convergence within the Amazonia basin that is consistent with positive trends in the observed precipitation in northeastern Brazil (Hastenrath and Greischar 1993; Haylock et al. 2006), Amazonia (Depaiva and Clarke 1995; Chen et al. 2001, 2003), and southern Brazil (Barros et al. 2000b; Haylock et al. 2006). The moisture convergence increase is explained by an enhancement of the equatorial easterlies over Amazonia, transferring moisture from the Atlantic to the basin (Marengo 2004). The easterlies encountering the Andes are deflected poleward toward southern and southeastern SA tropical and subtropical continental areas through two main circulation features that work at intraseasonal-to-interannual scales: the SA low-level jet and the SACZ (Saulo et al. 2000; Carvalho et al. 2004). This is consistent with the positive trends observed since the mid-1970s in precipitation observed in eastern subtropical Argentina (Barros et al. 2000a) as well as in the streamflow river levels of the Paraná-Del Plata hydrological basin (Boulanger et al. 2005) and the Mar Chiquita Lake level (Piovano et al. 2004).
In particular, the summer precipitation variability in CWA (the west of subtropical Argentina) shows the occurrence of a prolonged wet event from 1973 to the early 2000s, lasting around 30 yr (Fig. 3). In this period, 20 out of 30 summers are above the regional average. The change means an increase in regional precipitation of over 20%. A similar prolonged wet event has never been recorded since 1900, and, in fact, it interrupted a significant period of alternating wet–dry sequences, each phase lasting roughly 9 yr until the summer of 1976/77 (Agosta et al. 1999; Compagnucci et al. 2002; Agosta and Compagnucci 2006). Furthermore, the case of the Mar Chiquita Lake level rise illustrates the situation in subtropical central Argentina where the change is also detected as a significant “jump” in 1977 (Piovano et al. 2004).
As can be seen, the long-term climate variations detected primarily in precipitation over southern South America are not isolated: they reveal changes in the atmospheric circulation during the 1970s. Likewise, some climate variables point to the austral 1976/77 summer as a decisive moment in their records. As a matter of fact, the current hypothesis of regional climate change over SA during summer suggests an enhanced low-level meridional circulation over southern SA, from tropical to subtropical latitudes, associated with the easterlies deflected by the Andes and a strengthening of the SSA over the continent. This could be due to an anomalous mass distribution between the Pacific and the Atlantic as a consequence of the global climate transition in 1976/77, which may resemble El Niño event conditions at longer time scales, as indicated by several authors (i.e., Nogués-Paegle et al. 2002; McPhaden and Zhang 2002; Kayano and Andreoli 2007). The results so far presented further explore this important issue by comparing the climatology before and after the 1976/77 summer. The next section will show the specific manner in which the summer low-level atmospheric circulation has changed after the 1976/77 climate transition over southern SA.
b. Changes in the summer low-level atmospheric circulation features
To classify the summer low-level daily atmospheric circulation over southern SA into principal patterns, the Varimax-rotated PCA is performed for 850-hPa daily GH fields on the domain 15°–70°S and 110°–30°W for the period 1958/59–1976/77 (PCA pre-1977) and 1978/79–1997/98 (PCA post-1977). Following Craddock and Flood’s (1969) criterion to determine the number of PCs to be retained for rotation, the log-eigenvalue’s (LEV) versus the eigenvector number (PC order) diagram, where the random data are a slow slope tendency (Farmer 1971), shows a linear adjustment up to the fifth PC for both the unrotated PCA pre-1977 and post-1977 (Fig. 4). Hence, considering a rotation using up to the fifth PC, the total variance, estimated with respect to the total variance in each period and accounted for by each PC score, is close to 96% explained variance for each analysis. The 850- and 500-hPa GH, 850-hPa vector wind (V), and 850-hPa specific humidity anomalies (Q) daily fields are composited for the 30 highest PC loadings associated with each PC score (by this composition criterion, the PC loadings result to be always over 0.8). The spatial patterns given by each rotated PCj score (j = 1, . . . , 5) for the PCA pre-1977 and PCA post-1977 are shown in Fig. 5 (left column and right column, respectively). Between both periods there are changes in the explained variance of some PCs that alter the order of the corresponding spatial models. All the PCs appear in the direct mode, but in case of the PC5 both modes, direct and inverse, must be considered. However, the explained variances associated with each mode are low (less than 1%).
The pre-1977 PC1 shows a spatial pattern (score) whose shape is similar to that of the post-1977 PC1, which is called model A [Fig. 5, (M–A)]. Each PC1 explains 37.39% and 36.68% of each total variance, respectively. The model A shows the systems displaced slightly westward in the period following 1977. This effect suggests a strengthening of the SAA over the continent during this season. The composites of 850- and 500-GH daily fields for model A (Figs. 6a,b) show cyclonic systems at high latitudes with a trough axis extending to lower latitudes over the Pacific and a ridge axis from the SSA to mid- and high latitudes in the lower troposphere. The GH values are higher over the continent in the period post-1977 (Fig. 6a, right column). The midtroposphere composite daily GH field is similar to those of lower troposphere for both periods. There appear trough systems at mid- and high latitudes in the South Pacific and ridge systems in the South Atlantic basin whose influence extends over subtropical Argentina (Figs. 6a,b). The previous composite features are the consequence of a low-level airmass flow from the South Atlantic from the east-northeast as shown in Fig. 6c. It is associated with positive percentage specific humidity anomalies (%Q) over subtropical central-west Argentina and Patagonia that are slightly higher in the second period (Fig. 6d). This is consistent with westward displacement of the systems shown by model A after 1977. In midtroposphere the trough ridge eastward configuration results in generalized neutral conditions for airmass lift in the continent according to the quasigeostrophic omega equation (Holton 1992), even more since the ridge systems over the Atlantic are closer to the continent than the Pacific trough systems.
The pre-1977 PC2 shows a similar spatial pattern to that of post-1977 PC3, therefore they are named model B [Fig. 5 (M–B)]. They explain 35.00% and 17.83% of the total variances, respectively. The model describes cyclonic systems south of Patagonia, which are deeper and westward displaced in the period post-1977 together with both semipermanent anticyclones shifted south. The lower-troposphere composite GH daily field verifies the westward shift of the cyclonic systems after 1977, highlighting a stronger gradient over Patagonia (Fig. 7). In midtroposphere, the trough systems extend northwest to southeast in the latitude band 30°–60°S around 90°–75°W. The tropospheric circulation in the continent configuration gives westerly winds at midlatitudes, northerly winds at subtropical and tropical latitudes in eastern SA and northwesterly winds at tropical latitudes in the LLF area. The midlatitude westerlies are stronger in the second period while the LLF appears to be weaker. After 1977, both the SPA and the SAA appear located at their climatic position. In both periods, negative %Q anomalies are observed mainly in subtropical central-west Argentina and partly in Patagonia, for which reason precipitation is expected to be inhibited under this thermodynamic condition. Positive %Q anomalies are observed in Paraguay, northeastern Argentina, and southern Brazil (Fig. 7d). The fact that model B explains half of the total variance during the period after 1977 in comparison to the previous period, implies a reduced amount of cyclonic activity at midlatitudes that can also be associated with lower precipitation. This is consistent with the negative trends in precipitation over Patagonia (Rusticucci and Penalba 2003; Haylock et al. 2006; among others) and the climatic decrease in the cyclonic activity from the mid-1970s suggested by some authors (Key and Chang 1999; Simmonds and Keay 2000; Pezza and Ambrizzi 2003) and as discussed above. Indeed, the phenomenon is strongly related to the positive GH anomalies at midlatitudes encountered for the difference field between the climatology CL2 (after 1977) minus CL1 (before 1977) analyzed in the previous section, which are closely related to the 1976/77 climate transition following Yamamoto’s test. Also, the midlatitude anticyclonic anomalies associated with the tendency toward a positive phase of the SAM (Marshall 2003) could be contributing to the lower variance explained by this model. Furthermore, since the mid-1970s, the CWA region has undergone a prolonged wet event when compared to the whole twentieth century, with the 1976/77 summer as the turning point in its low-frequency precipitation variability, as can be seen in Fig. 3. Then, the lower cyclonic activity, which is related with reduced moisture in subtropical central-west areas, could account for the less frequent occurrence of CWA dry summers during recent decades (Agosta and Compagnucci 2006).
The composite 850 hPa GH anomalies for extreme dry summers in CWA during the periods before and after the 1976/77 summer, are shown in Fig. 8a, left and right panels, respectively. The extreme summers are defined using the first- and third-quartile criterion for the distribution of the empirical precipitation index, devised from meteorological stations in CWA, displayed in Table 2. Note the similar shape between the 850 GH summer anomalies and model B, in the period before 1977. Afterward, the 850 GH anomalies during CWA extreme dry summers seem to be a combination of model B and model D (see Fig. 6 right column, M–B and M–D). In other words, model B depicts quite well the low atmospheric circulation anomalies observed during extreme dry summer occurrences in CWA. Therefore, the lower midlatitude cyclonic activity after 1976/77 is one of the reasons by which there have been fewer dry events since.
The third model, model C, corresponds to the PC3 before 1977, with 15% of the explained variance, and to the PC2 after 1977, with twice the explained variance, 31.15% [Fig. 5 (M–C) ]. Its broad pattern shows a westerly flow perturbed by anticyclonic activity at high latitudes near the continent. This model suggests an important change, after 1976/77. After 1976/77, with respect to the previous state, the SAA oriental flank extends over the continent contributing to the connection between subtropical latitudes and the Atlantic, together with a westward shift of the southeastern Pacific anticyclonic center and its ridge over Patagonia. The lower-troposphere daily composite field before 1976/77 shows anticyclonic activity over Patagonia in connection with the SPA (Figs. 9a–b, left column) and its associated midtroposphere ridge structure to the west, and a subtropical trough structure right west of the Andes. An anticyclonic mass flow gyre is observed in Patagonia and the southwestern Atlantic, giving a southeasterly wind component in northern Patagonia, and a slight easterly wind component in subtropical Argentina. Northerly winds are observed over Brazil and the adjacent coasts. Strong westerlies, at higher latitudes, and the typical anticyclonic gyre occur in the subtropical Pacific (Fig. 9c, left column). The highest specific humidity anomalies are observed in tropical and subtropical latitudes with two cores, one in the northeastern Argentina, Uruguay, and southern Brazil and the other in CWA and northern Patagonia (Fig. 9d, left column). The net low-level atmospheric circulation before 1976/77 yields moisture advection toward subtropical latitudes from the midlatitude South Atlantic and toward tropical latitudes from the northern equatorial region of Brazil. The midtroposphere subtropical wave structure in the Pacific produces dynamic conditions for upward motion in the subtropical continent because of the differential vorticity advection according to the quasigeostrophic theory. Both features favor precipitation in the continent, north of 40°S. For the period after 1976/77, the daily composite fields also outline the differential feature of model C, highlighting substantial changes (Figs. 9a–d, right column). The 850 GH daily compositions show the SAA expansion over the continent at tropical and subtropical latitudes, owing to the higher GH values observed there with respect to the previous period. Also, a westward withdrawal of the SPA and associated ridge is apparent because of a trough system at midlatitudes in Patagonia and the southwestern South Atlantic that is connected to a cyclonic core at higher latitudes (Figs. 9a–b, right column). In the midtroposphere, trough systems appear in subtropical latitudes of the South Pacific and higher latitudes of the South Atlantic, while ridge systems are observed at higher latitudes over the Pacific. The joint feature gives a diffluent-to-confluent flow (Fig. 9b, right column). The winds blow with a northerly component in the areas north of 35°S, while weak-to-strong westerlies are observed to the south. The positive %Q values are observed in subtropical latitudes of Argentina, Paraguay, Uruguay, and southern Brazil (Fig. 8d, right column). Comparing model C [Figs. 5 (M–C) and 9a] to the 850 GH composite anomalies for wet extreme summers in CWA (Fig. 8b), it is evident that the model reasonably resembles these events, even after taking into account the differential feature between the periods before and after 1976/77. The differential feature for the second period in the CWA summer compositions consists of anticyclonic anomalies over northeastern tropical latitudes, extending from the strengthening of the SAA western flank toward the southwestern tip of the continent. Previously, the positive anomalies were located at midlatitudes over Patagonia and the southwestern Atlantic (Fig. 7b). The precipitation compositions for extreme wet summer in CWA using the Udel dataset is shown in Fig. 10, also showing changes after 1977. Note that before 1977, during CWA extreme wet summers, the positive precipitation anomalies were restricted to the west of central Argentina and northern Patagonia. Instead, after 1977, the positive precipitation anomalies extended toward large continental areas at subtropical latitudes. This confirms that the differential feature in the low-level atmospheric circulation anomaly compositions for CWA extreme wet summers is not casual but a clear climate shift or change within the southern cone of South America, which even affects the way different climate regions are related to each other (Agosta and Compagnucci 2006).
Furthermore, since model C seems to reflect the atmospheric circulation feature associated with extreme wet summers in CWA, the fact that its explained variance doubles for the period after 1976/77 can account for the higher amount of wet events recorded in CWA after the mid-1970s. The double variance explained by model C after the 1976/77 climate transition implies an enhanced occurrence of meridional mass flow from tropical to subtropical latitudes in the southern plains of SA together with enhanced moisture amounts in the area and enhanced dynamical conditions favorable to upward motions, all of which contribute to increased precipitation. In consequence, the trends and jumps observed in hydrological and precipitation series in the subtropical latitudes of SA can be due to the change in the amount of explained variance associated with this atmospheric synoptic circulation regime.
The remaining two PC scores do not show substantial differences in shape and explained variances between the PCA pre- and post-1977. The pre-1977 PC4 and post-1977 PC4 explain around 11% and 12% of the variance, respectively. Therefore they appear to be neutral as to changes observed in the CWA precipitation. The PC4 model D spatial pattern shows a northwest–southeast-oriented ridge axis near southwestern SA and two cyclonic centers at mid–high latitudes in the South Pacific and South Atlantic (Fig. 4). Model D can be associated with a weakening of the southern flank of the SAA and the strengthening of the southern flank of the SPA, which affect Patagonia, together with a weakening of the westerlies in southern Patagonia and higher latitudes. It seems to be part of a spatial wavenumber 3 or higher. The troposphere 850- and 500-hPa GH daily compositions show a midlatitude wave with a ridge centered near 75°W. The associated mass flow highlights a frontal zone between 35° and 40°S over the continent (figures not shown). The model corresponds to negative specific humidity anomalies in central and east Argentina and the southwestern South Atlantic basin (figures not shown).
The pre-1977 PC5 and post-1977 PC5 each explain less than 1.7% of the total variance, and occur preferentially in the direct mode, each mode explaining less than 1% of the variance (Fig. 4). The spatial pattern is related to strengthening (weakening) of the SAA and weakening (strengthening) of the SPA in the direct (inverse) mode. Indeed, the model E has low participation in the atmospheric condition that produced precipitation in CWA. Therefore, the daily field compositions are not discussed.
4. Conclusions
The austral 1976/77 summer climate transition impact on atmospheric circulation variability has been studied in detail for North America, yet few studies have focused the consequences over South America (Trenberth 1995; Zhang et al. 1997; Garreaud and Battisti 1999; Huang et al. 2005). A vast number of studies do refer to it as a possible source of change or trend enhancement for precipitation, river streamflow, lake level, and even temperature in many SA regions (Robertson and Mechoso 1998; Chen et al. 2001; Compagnucci et al. 2002; Piovano et al. 2004; Chen et al. 2003; Liebmann et al. 2004; Haylock et al. 2006; among others). Marengo (2004) points out a change in the easterlies over the Atlantic and Amazonia, by comparing the periods before and after the 1976/77 summer, and associated it with enhanced moisture and precipitation there. In addition, there exists the possibility of an enhanced southward LLF over SA subtropical latitudes in order to account for precipitation increase in subtropical Argentina and southeastern Brazil (Vera et al. 2006 and references therein). Yet all these studies, except Marengo (2004), have not directly linked such processes with the climate transition or studied the change in its own right. Therefore, in this work the aim has been to specify the manner in which the atmospheric circulation over southern South America and adjacent areas, from 15° to 70°S, is altered after the climate transition 1976/77.
It is found that the midlatitude cyclonic activity is reduced near southern South America after the 1976/77 summer. This is highlighted by the change in the amount of variance explained by the model B in the rotated PCA, which is related to midlatitude cyclone systems and low moisture values in subtropical Argentina [Figs. 5 (M–B) and 7]. Before 1977, model B accounts for an important amount of the total variance, that is, 35%, while afterward this drops to approximately 18%. The reduced cyclonic activity can explain the fewer occurrences of dry events in the summer precipitation over the CWA region and, possibly, in other subtropical regions [Figs. 5 (M–B) and 7]. Furthermore, this reduction is intimately linked to the 1976/77 climate transition via a possible southward influence of the SAA [Figs. 5 (M–C) and 9], and it appears to be reinforced by the tendency toward a positive phase of the SAM at higher latitudes in recent decades (Marshall 2003). A reduction in Southern Hemisphere cyclonic activity over the last decades has been previously pointed (Haylock et al. 2006 and references therein).
Another relevant feature encountered is the expansion of the SAA over the continent at tropical and subtropical latitudes after the 1976/77 summer, altering the atmospheric circulation over subtropical Argentina, Uruguay, and southern Brazil. There appears to be a wind rotation that alters the extent of the moisture accretion in the subtropical latitudes, north of 40°S; that is, being less southwestward and more confined to southeastern South America. The effect is disclosed by model C for which, after 1977, the midlatitude anticyclonic gyre over Patagonia and the southwestern South Atlantic, connected with the southern flank of the SPA, disappears and the south and western flanks of the SAA strengthen, resulting in a reduced easterly flow and a more northerly flow over subtropical latitudes (Figs. 5 and 9). Also, the model C explained variance changes from 15% to 31%. The increase in the amount of explained variance is indicative of the new atmospheric circulation associated with model C, which takes a more prominent role in the synoptic regime after 1977. The possible dynamical mechanism that operates on the SAA can be the mass redistribution through an anomalously enhanced Walker circulation established between the Pacific and the Atlantic after the transition 1976/77, as suggested by many authors (Lenters and Cook 1995, 1999; Nogués-Paegle et al. 2002; Kayano and Andreoli 2007).
Therefore, the combination of the observed changes associated with models B and C can account for the increase in precipitation and water budget over southern SA, in particular at subtropical and midlatitudes. Specifically, the CWA prolonged wet summer sequence, recorded since the mid-1970s, during which 20 out of 30 summers have record precipitations above the regional mean (Agosta and Compagnucci 2006), can owe its existence to the reduction in cyclonic activity, which is associated with general dry conditions in CWA and subtropical Argentina (the model B), as well as to the enhancement of the atmospheric circulation that favors the occurrence of wet conditions in CWA (the model C) at daily scales. Summing up, midlatitude cyclonic activity implies a westerly flow across the high Andes that results in dry advection in CWA and hence dry conditions in the area, as described in model B from Fig. 7, both before and after 1976/77. After 1977, reduction in cyclonic activity at midlatitudes together with a stronger northerly flow, which brings in higher humidity levels from lower latitudes (as given by the differential feature in model C from Fig. 9), can account for the wetter conditions in CWA in recent decades.
The current analysis has not detected substantial changes in intensity of the LLF connecting the equatorial–tropical latitudes with the subtropical latitudes in SA from the NCEP–NCAR reanalysis dataset. However, the large dynamic atmospheric circulation features, such as the expansion of the SAA and the decreased midlatitude cyclonic activity, seem to point to a consistent and coherent moisture transport mechanism from lower to higher latitudes with enhanced occurrence after the 1976/77 climate transition.
Finally, according to Compagnucci et al. (2002), the CWA lower-frequency variability shows a significant quasi-18-yr cycle from the beginning of the twentieth century until the mid-1970s. Furthermore, the CWA variability at annual scales is statistically independent from the ENSO events in summer. What is more, CWA summer variability cannot be directly influenced by the multidecadal Pacific variability, which has significantly lower frequencies of about 30–40 yr; that is, such low frequency is not observed in CWA variability, in agreement with Compagnucci et al. (2002). However, the CWA summer variability has indeed undergone a change during the 1970s in phase with the PDO or El Niño–like phase changes at that time. Thus, to a certain extent, the change in the basic atmospheric circulation features over SSA during the 1976/77 climate transition could be rather unique, and it should not be solely attributed to the PDO or the El Niño–like variability. Other possible factors involved in the change could be nonlinear internal variability of the climate system, as well as enhanced GHG forcing and/or solar variability response (Solomon et al. 2007). Nonetheless, all these aspects need further model research to be fully understood.
Acknowledgments
This research was funded by grants of the University of Buenos Aires, UBA EXT 002, UBA EXT 095, and of the National (Argentine) Research and Technique Council (CONICET), PIP 5006 and PIP 5276. We express our gratitude also to Carmelite Order for all their help.
REFERENCES
Agosta, E. A., and R. H. Compagnucci, 2006: Atmospheric conditions during wet and dry summer extremes in central-west Argentina. Proc. Eighth Int. Conf. on Southern Hemisphere Meteorology and Oceanography, Foz do Iguaçu, Brazil, Amer. Meteor. Soc. and the Brazilian Institute for Space Research (INPE), 1439–1452.
Agosta, E. A., R. H. Compagnucci, and M. W. Vargas, 1999: Cambios en el régimen interannual de la precipitación en la región Centro-Oste de Argentina (Changes in the interannual precipitation regime over Central-West Argentina). Meteorologica, 24 , 63–84.
Allan, R. J., and M. R. Haylock, 1993: Circulation features associated with the winter rainfall decrease in southwestern Australia. J. Climate, 6 , 1356–1367.
Barros, V. R., and L. M. Scasso, 1994: Surface pressure and temperature anomalies in Argentina in connection with the Southern Oscillation. Atmósfera, 94 , 1159–1171.
Barros, V. R., M. Castañeda, and M. Doyle, 1996: Recent precipitation trends in southern South America to the east of the Andes: An indication of a mode of climatic variability. Greenhouse Gas Emission under Developing Countries Point of View, L. P. Rosa and M. A. Santos, Eds., COPPE, 41–67.
Barros, V. R., M. E. Casteñeda, and M. Doyle, 2000a: Recent precipitation trends in southern South America east of the Andes: An indication of climatic variability. Southern Hemisphere Paleo and Neo-climates, P. P. Smolka and W. Volkheimer, Eds., Springer-Verlag, 187–206.
Barros, V. R., M. Gonzalez, B. Liebmann, and I. Camilloni, 2000b: Influence of the South Atlantic Convergence Zone and South Atlantic sea surface temperature on interannual summer rainfall variability in southeastern South America. Theor. Appl. Meteor., 67 , 123–133.
Bengtsson, L., S. Hagemann, and K. I. Hodges, 2004: Can climate trends be calculated from reanalysis data? J. Geophys. Res., 109 .D11111, doi:10.1029/2004JD004536.
Boer, G. J., K. Hamilton, and W. Zhu, 2005: Climate sensitivity and climate change under strong forcing. Climate Dyn., 24 , 685–700.
Boulanger, J. P., J. Leloup, O. Penalba, M. Rusticucci, F. Lafon, and M. Vargas, 2005: Observed precipitation in the Paraná-Plata hydrological basin: Long term trends, extreme conditions and ENSO teleconnections. Climate Dyn., 24 , 393–413.
Camilloni, I., 1999: Temporal variability of the Buenos Aires’ urban heat island intensity. Proc. Int. Conf. on Urban Climatology ICUC’99, Sydney, Australia, International Society of Biometeorology, World Meteorological Organisation, and the Australian Bureau of Meteorology.
Carril, A. F., M. E. Doyle, V. R. Barros, and M. N. Núñez, 1997: Impacts of climate change on the oases of the Argentinean cordillera. Climate Res., 9 , 121–129.
Carvalho, L. M. V., C. Jones, and B. Liebmann, 2002: Extreme precipitation events in southeastern South America and large-scale convective patterns in the South Atlantic convergence zone. J. Climate, 15 , 2377–2394.
Carvalho, L. M. V., C. Jones, and B. Liebmann, 2004: The South Atlantic Convergence Zone: Intensity, form, persistence, and relationships with intraseasonal to interannual activity and extreme rainfall. J. Climate, 17 , 88–108.
Chen, T. C., J. H. Yoon, K. J. St Croix, and E. S. Takle, 2001: Suppressing impacts of the Amazonian deforestation by the global circulation change. Bull. Amer. Meteor. Soc., 82 , 2209–2216.
Chen, T. C., E. S. Takle, J. H. Yoon, K. J. St. Crox, and P. Hsieh, 2003: Impacts on tropical South American rainfall due to changes in global circulation. Proc. Seventh Int. Conf. on Southern Hemisphere Meteorology and Oceanography, Wellington, New Zealand, Amer. Meteor. Soc., 92–93.
Compagnucci, R. H., and M. Richman, 2008: Can principal component analysis provide atmospheric circulation or teleconnection patterns? Int. J. Climatol., 28 , 703–726. doi:10.1002/joc.1574.
Compagnucci, R. H., E. A. Agosta, and M. W. Vargas, 2002: Climatic change and quasi-oscillations in central-west Argentina summer precipitation: Main features and coherent behaviour with southern African region. Climate Dyn., 18 , 421–435.
Craddock, J. M., and C. R. Flood, 1969: Eigenvectors for representing the 500 mb geopotential surface over the Northern Hemisphere. Quart. J. Roy. Meteor. Soc., 95 , 576–593.
Depaiva, E. M. C. D., and R. T. Clarke, 1995: Time trends in rainfall records in Amazonia. Bull. Amer. Meteor. Soc., 76 , 2203–2209.
Dettinger, M. D., D. S. Battisti, R. D. Garreaud, G. J. McCabe, and C. M. Bitz, 2001: Interhemispheric effects of interannual and decadal ENSO-like climate variations on the Americas. Interhemispheric Climate Linkages: Present and Past Climates in the Americas and their Societal Effects, V. Markgraf, Ed., Academic Press, 1–16.
Ebbesmeyer, C. C., D. R. Cayan, D. R. McLain, F. H. Nichols, D. H. Peterson, and T. Redmond, 1991: 1976 Step in the Pacific Climate: Forty environmental changes between 1968–1975 and 1977–1984. California Department of Water Resources, Interagency Ecological Studies Program Tech. Rep. 26, 115–126.
Farmer, S. A., 1971: An investigation into the results of principal component analysis of data derived from random numbers. Statistician, 20 , 63–72.
Garreaud, R. D., and D. S. Battisti, 1999: Interannual ENSO and interdecadal ENSO-like variability in the Southern Hemisphere. J. Climate, 12 , 2113–2123.
Gong, D., and S. Wang, 1999: Definition of Antarctic Oscillation Index. Geophys. Res. Lett., 26 , 459–462.
Gu, D. F., and S. G. H. Philander, 1997: Interdecadal climate fluctuations that depend on exchanges between the tropics and extratropics. Science, 275 , 805–807.
Hastenrath, S., and L. Greischar, 1993: Further work on the prediction of northeast Brazil rainfall anomalies. J. Climate, 6 , 743–758.
Haylock, M. R., and Coauthors, 2006: Trends in total and extreme South American rainfall in 1960–2000 and links with sea surface temperature. J. Climate, 19 , 1490–1512.
Holton, J., 1992: An Introduction to Dynamic Meteorology. 3rd ed. Academic Press, 511 pp.
Houghton, J. T., Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden, X. Dai, K. Maskell, and C. A. Johnson, 2001: Climate Change 2001: The Scientific Basis. Cambridge University Press, 881 pp.
Huang, H-P., R. Seager, and Y. Kushnir, 2005: The 1976/77 transition in precipitation over the Americas and the influence of tropical sea surface temperature. Climate Dyn., 24 , 721–740.
Hurrell, J. W., and H. van Loon, 1994: A modulation of the atmospheric annual cycle in the Southern Hemisphere. Tellus, 46A , 325–338.
Kayano, M. T., and R. V. Andreoli, 2007: Relations of South American summer rainfall interannual variations with the Pacific Decadal Oscillation. Int. J. Climatol., 27 , 531–540.
Key, J. R., and A. C. K. Chang, 1999: Multidecadal global and regional trends in 1000 mb and 500 mb cyclone frequencies. Geophys. Res. Lett., 26 , 2053–2056.
Kidson, J. W., 1999: Principal modes of Southern Hemisphere low frequency variability obtained from NCEP–NCAR reanalyses. J. Climate, 12 , 2808–2830.
Kistler, R., and Coauthors, 2001: The NCEP–NCAR 50-Year Reanalysis: Monthly means CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82 , 247–267.
Lenters, J. D., and K. H. Cook, 1995: Simulation and diagnosis of the regional summertime precipitation climatology of South America. J. Climate, 8 , 2988–3005.
Lenters, J. D., and K. H. Cook, 1999: Summertime precipitation variability over South America: Role of the large-scale circulation. Mon. Wea. Rev., 127 , 409–431.
Liebmann, B., and Coauthors, 2004: An observed trend in central South American precipitation. J. Climate, 17 , 4357–4367.
Limpasuvan, V., and D. L. Hartmann, 1999: Eddies and the annular modes of climate variability. Geophys. Res. Lett., 26 , 3133–3136.
Mantua, N. J., and S. R. Hare, 2002: The Pacific Decadal Oscillation. J. Oceanogr., 58 , 35–44.
Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace, and R. C. Francis, 1997: A Pacific interdecadal climate oscillation with impacts on salmon production. Bull. Amer. Meteor. Soc., 78 , 1069–1079.
Marengo, J. A., 2004: Interdecadal variability and trends of rainfall across the Amazon basin. Theor. Appl. Climatol., 78 , 79–96.
Marshall, G. J., 2003: Trends in the southern annular mode from observations and reanalyses. J. Climate, 16 , 4134–4143.
McPhaden, M. J., and D. Zhang, 2002: Slowdown of the meridional overturning circulation in the upper Pacific Ocean. Nature, 415 , 603–608.
Minetti, J. L., and W. M. Vargas, 1998: Trends and jumps in the annual precipitation in South America, south of 15°S. Atmósfera, 11 , 205–223.
Minetti, J. L., W. M. Vargas, A. G. Poblete, L. R. Acuna, and G. Casagrande, 2003: Non-linear trends and low frequency oscillations in annual precipitation over Argentina and Chile, 1931–1999. Atmósfera, 16 , 119–135.
Moser, B. K., and G. R. Stevens, 1992: Homogeneity of variance in the two-sample means test. Amer. Stat., 46 , 19–21.
Namias, J., 1978: Multiple causes of the North American abnormal winter 1976–1977. Mon. Wea. Rev., 106 , 279–295.
Nogués-Paegle, J., and Coauthors, 2002: Progress in pan American CLIVAR research: Understanding the South American Monsoon. Meteorologica, 27 , 1–30.
North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110 , 699–706.
Pezza, A. B., and T. Ambrizzi, 2003: Variability of Southern Hemisphere cyclone and anticyclone behavior: Further analysis. J. Climate, 16 , 1075–1083.
Piovano, E. L., D. Ariztegui, S. M. Bernasconi, and J. A. McKenzie, 2004: Stable isotopic record of hydrological changes in subtropical Laguna Mar Chiquita (Argentina) over last 230 years. Holocene, 14 , 525–535.
Preisendorfer, R. W., 1988: Principal Component Analysis in Meteorology and Oceanography. Elsevier, 425 pp.
Richman, M. B., 1986: Rotation of principal components. Int. J. Climatol., 6 , 293–335.
Robertson, A., and C. Mechoso, 1998: Interannual and decadal cycles in river flows of Southeastern South America. J. Climate, 11 , 2570–2581.
Rusticucci, M., and O. Penalba, 2003: Interdecadal changes in the precipitation seasonal cycle over southern South America and their relationship with surface temperature. Climate Res., 16 , 1–15.
Saulo, A. C., M. Nicolini, and S. C. Chou, 2000: Model characterization of South American low-level flow during the 1997–1998 spring–summer season. Climate Dyn., 16 , 867–881.
Seager, R., N. Harnik, W. A. Robinson, Y. Kushnir, M. Ting, H. P. Huang, and J. Velez, 2005: Mechanisms of ENSO-forcing of hemispherically symmetric precipitation variability. Quart. J. Roy. Meteor. Soc., 131 , 1501–1527.
Seluchi, M. E., A. C. Saulo, M. Nicolini, and P. Satyamurty, 2003: The Northwestern Argentinean Low: A study of two typical events. Mon. Wea. Rev., 131 , 2361–2378.
Simmonds, I., and K. Keay, 2000: Variability of Southern Hemisphere extratropical cyclone behavior, 1958–97. J. Climate, 13 , 550–561.
Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Miller, Eds. 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.
Thompson, D. W. J., and J. M. Wallace, 2000: Annular modes on the extratropical circulation. Part I: Month-to-month variability. J. Climate, 13 , 1000–1016.
Thompson, D. W. J., and S. Solomon, 2002: Interpretation of recent Southern Hemisphere climate change. Science, 296 , 895–899.
Trenberth, K. E., 1990: Recent observed interdecadal climate changes in the Northern Hemisphere. Bull. Amer. Meteor. Soc., 71 , 988–993.
Trenberth, K. E., 1995: Atmospheric circulation climate changes. Climatic Change, 31 , 427–453.
Trenberth, K. E., and J. W. Hurrell, 1994: Decadal atmosphere–ocean variations in the Pacific. Climate Dyn., 9 , 303–319.
van Loon, H., J. W. Kidson, and A. B. Mullan, 1993: Decadal variation of the annual cycle in the Australian dataset. J. Climate, 6 , 1227–1231.
Vera, C., 2003: Interannual and interdecadal variability of atmospheric synoptic-scale activity in the Southern Hemisphere. J. Geophys. Res., 108 .8077, doi:10.1029/2000JC000406.
Vera, C., and Coauthors, 2006: Toward a unified view of the American monsoon systems. J. Climate, 19 , 4997–5000.
Waylen, P., R. Compagnucci, and R. M. Caffera, 2000: Interannual and interdecadal variability in streamflow from the Argentine Andes. Phys. Geogr., 21 , 452–465.
Yamamoto, R., Y. Iwashima, and T. Sanga, 1986: An analysis of climatic jump. J. Meteor. Soc. Japan, 64 , 273–280.
Zhang, Y., J. M. Wallace, and D. Battisti, 1997: ENSO-like interdecadal variability: 1900–93. J. Climate, 10 , 1004–1020.
Basic low-level atmospheric circulation features in the austral summer season. The contours are the seasonally averaged 850-hPa GH, and the arrows are the 850-hPa vector wind, from the NCEP–NCAR reanalysis I. SPA: subtropical South Pacific anticyclone. SAA: subtropical South Atlantic anticyclone. ChL: “Chaco” continental low. LLF: north-to-south low-level flow. Right corner map: political division of the area.
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
(left) Difference between the long-term seasonal GH average for the period 1958/59–1976/77 (CL1) and for 1979/80–1997/98 (CL2) estimated as CL2 minus CL1 for (a) 850 and (b) 500 hPa. (right) Yamamoto’s statistics applied to the seasonal GH at (a) 850 and (b) 500 hPa at the exact reference year 1976/77 (see text). Shaded areas: significant values at α = 0.05.
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
Percentage precipitation index anomaly for CWA estimated on the period 1959–2004 as P − 100% (vertical bars). Horizontal thick line: first quartile k1 = −19.7% and third quartile k3 = +18.5% from the percentage index anomaly distribution. In the period 1973–2003, 20 summers are above normal and 11 are below normal.
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
LEV and the eigenvector number (PC order) diagram for the unrotated PCA before 1977 and after 1977.
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
Varimax-rotated PC scores (spatial patterns) obtained by the rotated PCA applied to 850-hPa daily geopotential fields from 1 Oct to 31 Mar (summer season), for the period before 1977 (pre-1977) and after 1977 (post-1977). Percentage values: variance explained by the PC; the sign “+” indicates the direct mode and “−” the inverse mode. M-A: model A; M-B: model B; M-C: model C; M-D: model D; and M-E: model E.
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
Composite daily fields for the first highest 30 PC loadings associated with model A (M-A) obtained by rotated PCA (R-PCA) (left column) before 1977 (pre-1977) and (right column) after 1977 (post-1977), for (a) the 850-hPa GH (in m), (b) 500-hPa GH (in m), (c) 850-hPa vector wind (V-wind in m s−1), and (d) 850-hPa specific humidity anomaly (%Q, percentage daily anomaly estimated from each climatic baseline).
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
Same as Fig. 6 but for model B (M-B).
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
Summer 850-hPa GH composition for extreme (a) dry and (b) wet summers in CWA according to the precipitation index anomaly displayed in Table 2. The shaded areas are the significant values at α = 0.05.
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
Same as Fig. 6 but for model C (M-C).
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
Percentage precipitation composition for extreme wet summers in CWA (left) before 1977 and (right) after 1977. Rectangle: schematic boundary containing the CWA region.
Citation: Journal of Climate 21, 17; 10.1175/2008JCLI2137.1
List of meteorological stations used to devise the precipitation index P (see text) over the CWA region located at approximately 28°–37°S and 65°–70°W.
Dry and wet summer extremes according to the quartile criterion for the P-index distribution (Fig. 3), k1 = −19.7% and k3 = +18.5%. ΔP: percentage precipitation index anomaly (P − 100%). The year indicated is of the end of the summer season.
