PRECIPITATION ANOMALIES IN SOUTHERN SOUTH AMERICA ASSOCIATED WITH A FINER CLASSIFICATION OF EL NIÑO AND LA NIÑA EVENTS - PDF

Description
INTERNATIONAL JOURNAL OF CLIMATOLOGY Int. J. Climatol. 22: (2002) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: /joc.713 PRECIPITATION ANOMALIES IN SOUTHERN SOUTH

Please download to get full document.

View again

of 17
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Speeches

Publish on:

Views: 17 | Pages: 17

Extension: PDF | Download: 0

Share
Transcript
INTERNATIONAL JOURNAL OF CLIMATOLOGY Int. J. Climatol. 22: (2002) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: /joc.713 PRECIPITATION ANOMALIES IN SOUTHERN SOUTH AMERICA ASSOCIATED WITH A FINER CLASSIFICATION OF EL NIÑO AND LA NIÑA EVENTS R.P. KANE* Instituto Nacional de Pesquisas Espaciais INPE, Caixa Postal 515, , São José dos Campos, SP, Brazil Received 9 September 1999 Revised 16 July 2001 Accepted 22 July 2001 ABSTRACT The relationship between rainfall in sub-regions of Uruguay and South Brazil and a finer classification of El Niños (ENs), was studied. ENSOWs were defined as years when an EN existed on the Peruvian coast, the southern oscillation index (SOI = Tahiti minus Darwin pressure) was negative (SO), and Pacific sea-surface temperature (SST) anomalies were positive (W). Further, unambiguous ENSOWs were defined as years when SO and W occurred in the middle of the calendar year, and ambiguous ENSOWs were defined as years when SO and W occurred in the earlier or later part of the calendar year (not in the middle). In contrast with India and some other regions where unambiguous ENSOWs were associated predominantly with droughts, in the case of southern South America the association was with excess rains. Among the ambiguous ENSOWs, some were associated with floods in southern South America, but some had normal or mixed rainfalls (floods in some sub-regions, droughts in others) and a few even had droughts. C events (La Niñas, i.e. no EN, SOI positive, and SST negative) were associated mostly with droughts, but occasionally with floods in southern South America. Many non-events were associated with floods or droughts, indicating that factors other than EN/La Niña could also be important. Copyright 2002 Royal Meteorological Society. KEY WORDS: southern South America; ENSO; rainfall 1. INTRODUCTION Along the coast of Peru Ecuador in South America, there is an ocean current called the Peru or Humboldt current.el Niño (the child, in Spanish) is defined as a warming of this ocean current,so called because it generally develops near Christmas, the birthday of Jesus Christ. Quinn et al. (1978, 1987) determined the occurrence of El Niño events on the basis of the disruption to the fishing industry, hydrological data, sea-surface temperature (SST) and rainfall along and near the Peru Ecuador coast, defining El Niño intensities based on the positive SST anomalies along the coast as: strong, in excess of 3.5 C ; moderate, C; weak, C. El Niños are associated with droughts in many parts of the globe (Ropelewski and Halpert, 1987). However, not all El Niños produce the same effects. Trenberth (1993) refers to different flavours of El Niños. Small changes in Pacific and other SSTs can considerably alter the atmospheric convection distribution. Changes in location, extent, and time of year of SST anomalies result in differences in tropical rainfall. These project onto the changes in tropical heating and, thus, differences in vertical motion, large-scale overturning, and upper tropospheric convergence and divergence, so that there are differences in atmospheric Rossby wave forcing. In addition, the horizontal size of the anomaly affects which scales are forced, and thus the propagation characteristics of the Rossby waves. These different flavours of El Niño southern oscillation (ENSO) events are important. However, these have not been identified so far for every individual El Niño event in the past. Because of several factors involved in the dynamical response of the atmosphere to SST anomalies, and * Correspondence to: R. P. Kane, Instituto Nacional de Pesquisas Espaciasis INPE, Caixa Postal 515, , São José dos Campos, SP, Brazil. Copyright 2002 Royal Meteorological Society 358 R. P. KANE because of SST dependence on the atmospheric surface winds and radiation, observations reveal that there is no simple linear relationship between SST and rainfall or outgoing long-wave radiation (OLR) anomalies. The whole SST pattern, including the gradients and absolute values and their time of occurrence during the year, as well as the proximity of major land masses, are all factors as important as the actual SST anomaly in determining the rainfall anomalies (Trenberth, 1989). As a preliminary study, Kane (1997a,b, 1998) attempted a finer classification of El Niños, in which unambiguous ENSOW (ENSOW-U)-type events (discussed later on) were found to be overwhelmingly associated with droughts in India, southeastern Australia and some other parts of the globe. Thus, some flavour was isolated; but no physical basis for this finer classification was given. Earlier, other workers had attempted to classify El Niño on the basis of SST configurations in the Pacific. For example, Fu et al. (1986) identified two distinct patterns in Pacific SST. In one, the Pacific was warmer east of the dateline, warmer in the Central Pacific, but slightly below normal west of the dateline (e.g. 1957, 1965, 1972, 1982). In another, the Pacific was warmer everywhere (e.g. 1963, 1969). In cases like 1976, there was a mixture of the two. More recently, Ward et al. (1994) classified years according to whether they were wet (excess rains) or dry (droughts) globally (e.g. in Sahel and India), and their average characteristics were studied in terms of SST anomalies in the Pacific. They found that years (which they called Type I) that were associated with a near-global rainfall teleconnection including a tropic-wide oscillation had a strong contrast in SST anomalies between the central/eastern tropical Pacific and western tropical Pacific, leading to a strong perturbation in the longitude of the maximum in the zonal SST profile at 0 10 S in the western Pacific. Ropelewski and Halpert (1987, 1989) identified several regions around the globe as having a coherent ENSO-related response. In addition to the Pacific Ocean basin, there were four regions in Australia, two regions each in North America, South America, the Indian subcontinent, and Africa, and one region in Central America. The regions in South America were northeastern South America (northeast Brazil) and southeastern South America (northeastern Argentina, Uruguay and two stations in southern Brazil). It was shown that, during El Niño years, the rainfall there was higher than average (excess rainfall) from November through to the following February (local summer). Lau and Sheu (1988) and Kiladis and Diaz (1989) reported a similar behaviour for this region. Pisciottano et al. (1994) divided Uruguay into four sub-regions according to differences in precipitation regimes and reported above-average precipitation in El Niño years, from November through to January of the next year, particularly in the northern and western parts, and from March through to July of the following year, in the northern part. For southern Brazil, consisting of the three states of Parana, Santa Catarina and Rio Grande do Sul, Grimm and Feuser (1998) and Grimm et al. (1998) made five subdivisions, whose limits were determined by relief, latitude, and proximity to the Atlantic Ocean, and reported different degrees of ENSO coherence for the different sub-regions. All the sub-regions % consistently had positive anomalies (excess rainfall) during the austral spring of the El Niño year, with a pronounced peak in November, whereas the southeastern part also showed excess rainfall during the austral winter of the following year. For central Chile, Rubin (1955), Quinn and Neal (1983), Aceituno (1988) and Rutland and Fuenzalida (1991) reported excess rainfall in local winter (June August (JJA)) in El Niño years. Quinn and Neal (1983) reported that these El Niño effects were primarily confined to the Chilean subtropics, and that, at times, stations near 40 S register these changes, but to a lesser degree. In this paper, the precipitation anomalies in the subdivisions of Uruguay and southern Brazil, in Argentina, and in central Chile are examined for a finer classification of El Niños described in Kane (1997a,b, 1998) and already used for India, Indonesia and Tasmania (Kane, 1999a,b). The purpose is to identify the regions and years when the relationship with El Niños was strong and, when the relationship was weak, to test whether the failure was due to unpredictability of climate or due to individual features like timing, spatial resolution, etc. of ENSO events. During the year, different locations on the globe have different rainy seasons. If an El Niño starts later than the rainy season, or ends much before the rainy season, no El Niño effects can be expected. 2. DATA For classification of years, El Niño years were obtained from Quinn et al. (1978, 1987). For the SO index (SOI), Wright (1975, 1977, 1984) used atmospheric pressure at a wide range of stations, but the results were SOUTHERN SOUTH AMERICA PRECIPITATION ANOMALIES 359 similar to those of Tahiti minus Darwin pressure difference (Parker, 1983, updated), which is used here. For SST, data used were the equatorial eastern Pacific SST data from Angell (1981, and further personal communication) and central and eastern equatorial Pacific SST data from Wright (1984). (In general, eastern, central and western Pacific SST anomalies are qualitatively similar). Data for SST at Puerto Chicama (Peruvian coast) for 1925 onwards were obtained privately, and those for Pacific SST (El Niño regions 1 + 2, 3, 4, for 1950 onwards) were obtained from the web site of the Climate Prediction Center (CPC) of NOAA, Camp Springs, MD. The locations of the various El Niño regions are: Puerto Chicama (8 S, 80 W); Niño (80 90 W); Niño 3 ( W); Niño 3.5 ( W); Niño 4 (150 W 160 E, through dateline at 180 ). Trenberth (1997) was also consulted. Reynolds SST data (sensed by satellite) were provided by the NOAA CIRES Climate Diagnostics Center, Boulder CO, from their web site (http://www.cdc.noaa.gov.) For rainfall, data for India monsoon rainfall (IMR) were obtained from Parthasarathy et al. (1992) and further information from the India Meteorological Department, for Indonesia from Kripalani and Kulkarni (1997a,b), for Tasmania from Dr Srikanthan (personal communication), for Uruguay from Pisciottano et al. (1994), for southern Brazil from Grimm et al. (1998), for a few stations in northeastern Argentina and Uruguay from Ropelewski and Halpert (1987, 1989), for central east Argentina (30 S, 65 W) from Compagnucci and Vargas (1983), and for Santiago, Chile (34 S, 71 W), from Quinn and Neal (1983) and Rutland and Fuenzalida (1991). Figure 1 shows a map of South Brazil and Uruguay. In South Brazil, the three states, Parana, Santa Catarina and Rio Grande do Sul are divided into five subregions, but region 5 is omitted from analysis (Grimm et al., 1998). In Uruguay, there are four subdivisions (Pisciottano et al., 1994), but data for the whole of Uruguay and its capital Montevideo are also considered, separately. 3. EL NIÑO CHARACTERISTICS AND THEIR IMPLICATIONS FOR RAINFALL 3.1. Finer classification of El Niño events The term ENSO is used nowadays for the general phenomenon of the El Niño southern oscillation. However, for our classification, its components EN and SO are used in their literal sense. Thus, every year was examined to check whether it had an EN (as listed in Quinn et al. (1978, 1987)), and/or an SOI minimum (SO) and/or warm (W) or cold (C) (La Niña) equatorial eastern/central Pacific SST anomalies (data for western Pacific were not available for earlier years, but the entire Pacific generally shows similar SST anomalies qualitatively). For SOI and SST, 12 month running means were used. Several years were ENSOW, i.e. an EN existed near the Peruvian coast, and an SO (SOI minimum) and W (SST maximum) occurred. These were subdivided into two groups: ENSOW-U where an EN existed near the Peruvian coast and the 12 month running means of SO and W were in the middle of the calendar year (May August); ambiguous ENSOW (ENSOW-A) were defined as occasions where an EN existed near the Peruvian coast, but the SO and W were in the early or later part of the calendar year, not in the middle. Besides these, there were other El Niño years of the type ENSO (EN existed, SO existed, but SST was neither warmer W nor colder C, just normal), ENW (EN existed, SO did not exist, but SST was warm W), ENC (EN in the first half of the year, C in the latter half) or just EN (i.e. EN existed only near the Peruvian coast, no SO or W or C). Some other years did not have an EN and were of the types SOW (SO and W existed, but no EN was reported by Quinn et al. (1978, 1987)), SOC (SO in the first half of the year, C in the latter half), SO, W, and C, where the last category C contains all anti-el Niños, i.e. La Niñas. The types ENC or SOC look contradictory. Actually, in these years, EN and/or SO and C did not occur simultaneously. EN or SO occurred in the first part of the year and was followed by a cold Pacific SST, i.e. C (La Niña), event in the latter half. Years not falling into any of these categories were termed non-events. We have this classification ready for all years from 1871 onwards, the period for which a SST index for the eastern and central Pacific is available (Wright, 1984). With this classification, the ENSOW-U-type events showed excellent association with droughts in IMR (June September), as well as for some regional rainfalls in Australia (Kane, 1997a,b). Figure 2 shows the plots of the 12-monthly running averages of Pacific SST (roughly, the Niño 3.5 region) and the SOI for a few selected intervals of three consecutive years, the middle year being an El Niño year. In Figure 2(a), SOI 360 R. P. KANE Figure 1. Map of South Brazil and Uruguay. In South Brazil, the three states of Parana, Santa Catarina and Rio Grande do Sul are divided into five sub-regions. In Uruguay, there are four sub-regions: BON, NW, SW, SE minima SO and SST maxima W (12-monthly running means) are in the middle of the calendar year of the El Niño year (middle part) and these are defined as ENSOW-U years. For such years, the IMR normalized rainfalls had negative deviations. Figure 2(b) shows similar plots for some years of the ENSOW-A type. Here, the SO and W of the El Niño year (middle plot) are not in the middle of the calendar year. For such years, the IMR deviations generally indicated almost normal or sometimes even excess rainfall (see below). During , there were 52 El Niño years. Among these, 16 could be designated as ENSOW-U. These years and the corresponding normalized rainfall anomalies in IMR (given in parentheses) were as follows: 1877( 3.0), 1888( 0.5), 1896( 0.3), 1899( 2.7), 1902( 0.7), 1905( 1.6), 1911( 1.4), 1918( 2.4), 1930( 0.6), 1941( 1.5), 1951( 1.4), 1957( 0.8), 1965( 1.7), 1972( 2.4), 1982( 1.4), 1987( 1.9) all of which (except 1896) had rainfall deficits exceeding 0.5σ. There were 18 years of the type ENSOW-A. These years and the corresponding normalized rainfall anomalies in IMR (given in parentheses) were as follows: 1878(1.5), 1914(0.6), 1919(0.4), 1923( 0.4), 1925( 0.6), 1926(0.6), 1931(0.3), 1940( 0.0), 1948(0.2), 1953(0.8), 1958(0.4), 1963(0.0), 1969( 0.3), 1976(0.0), 1983(1.2), 1992( 0.3), 1997(0.1), 1998(0.1) SOUTHERN SOUTH AMERICA PRECIPITATION ANOMALIES 361 Figure 2. Plots of 12-monthly running means of SST and SOI for selected El Niño events for three consecutive years, where the El Niño event is in the middle year: (a) ENSOW-Us 1905, 1951, 1972, for which SST maxima W and SOI minima SO are in the middle of the calendar year; (b) ENSOW-As 1925, 1958, 1997, for which SST maxima W and SOI minima SO are not in the middle of the calendar year Most of these do not indicate rainfall deficits. Incidentally, some of these (shown bold) are two year events (see below; 1958 from the pair , etc.). It seems that two year events have mostly normal or even excess rainfall in IMR. However, these effects are related to the coincidence or non-coincidence of the monsoon season with the interval when El Niño was operative, as will be shown later. This classification is somewhat subjective, and some cases might have been wrongly classified. For example, in Figure 1(b), the 1925 event had SST maxima and SOI minima in the latter part of the year; but the SST maxima do not occur very late in the year, and 1925, though designated as ENSOW-A, may as well be an ENSOW-U (IMR value was 0.6). In Kane (1997a,b), 1976 was designated as ENSOW-U; but later, it was realized that it was an Ambiguous ENSOW-A. Also, 1888, earlier designated as an SOW, was found to be an El Niño year, an ENSOW-U. There were 18 other types of El Niño year (ENSO, ENW, ENC, EN), most of which did not show severe rainfall deficits in IMR (Kane, 1997a,b, 1998), partly because the El Niño was not intense during the rainy season. In Figure 2(b), the recent very strong El Niño event of is plotted at the bottom and shows SST maxima W and SOI (Tahiti minus Darwin) minima SO in the end part of Hence, in this finer classification, both 1997 and 1998 are ENSOW-A, implying only normal rainfall in India. The observed rainfalls were near normal, as shown above. 362 R. P. KANE Table I. Details of rainfall data, where (0) means the present year and (+) means the next year Location Months Years Type of unit IMR June September Normalized units Indonesia (Indo.) Annual values Normalized units Tasmania, Australia (Tasm.) April March Normalized units Uruguay (Urug.) November(0) January(+) Percent anomalies sub-region I-BON (BON) November(0) January(+) Percent anomalies II-NW (NW) November(0) January(+) Percent anomalies III-SW (SW) November(0) January(+) Percent anomalies IX-SE (SE) November(0) January(+) Percent anomalies Montevideo (Mont.) November(0) January(+) Percent anomalies South Brazil sub-region 1(0) August(0) December(0) Percent anomalies sub-region 2(0) September(0) December(0) Percent anomalies sub-region 3(0) September(0) December(0) Percent anomalies sub-region 4(0) October(0) December(0) Percent anomalies sub-region 2(+) March(+) July(+) Percent anomalies sub-region 3(+) June(+) August(+) Percent anomalies sub-region 4(+) April(+) July(+) Percent anomalies Ropelewski and Halpert (RH) November(0) February(+) Normalized units Central east Argentina (AG) Annual Normalized units Santiago, Chile (ST) Annual Normalized units 3.2. Rainfall response to the finer classification Rainfall deviation units and the periods for which these were available were as given in Table I. For every year for each series, the rainfall deviation was expressed as: +,, positive, negative deviations between 0 and 0.5σ, or 0 to 10% anomalies f, d, mild floods and droughts, between 0.5σ and 1.0σ, or 10 to 20% anomalies F, D, severe floods and droughts, exceeding 1.0σ or 20% anomalies. Table II IV show the rainfall deviations with these symbols for ENSOW-U years, ENSOW-A years and years of other types of El Niño respectively. The following may be noted. (1) The first column in Table II gives the years of 16 (or less) El Niños that were designated as ENSOW-U. S (strong), M (moderate), W (weak) indicate the strengths of these El Niños, and I and II indicate whether these are the first year or second year of double events ( , etc.). The second column shows the rainfall anomalies for the IMR. As can be seen, all these years had negative anomalies, with mostly severe (D) and moderate (d) droughts. Thus, ENSOW-Us have a distinct flavour, suitable for droughts in India. The next column for Indonesia also shows an overwhelming affinity for droughts there (ten out of ten). The next column for Tasmania, Australia, also shows a very good affinity for droughts (14 out of 15). At the bottom, the number of positive (F, f, +) and negative deviations (D, d, ) and the fraction of positive deviations are given. For India and Indonesia, the fraction is zero (0.00), whereas for Tasmania the fraction is very small (0.07). The next six columns give the stati
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x