Interdecadal Changes in the ENSO–Monsoon System
Interdecadal Changes in the ENSO–Monsoon System
C HRISTOPHER T ORRENCE
Advanced Study Program,National Center for Atmospheric Research,Boulder,Colorado
P ETER J.W EBSTER
Program in Atmospheric and Oceanic Sciences,University of Colorado,Boulder,Colorado
(Manuscript received29May1998,in?nal form27November1998)
ABSTRACT
The El Nin?o–Southern Oscillation(ENSO)and Indian monsoon are shown to have undergone signi?cant interdecadal changes in variance and coherency over the last125years.Wavelet analysis is applied to indexes of equatorial Paci?c sea surface temperature(Nin?o3SST),the Southern Oscillation index,and all-India rainfall. Time series of2–7-yr variance indicate intervals of high ENSO–monsoon variance(1875–1920and1960–90) and an interval of low variance(1920–60).The ENSO–monsoon variance also contains a modulation of ENSO–monsoon amplitudes on a12–20-yr timescale.
The annual-cycle(1yr)variance time series of Nin?o3SST and Indian rainfall is negatively correlated with the interannual ENSO signal.The1-yr variance is larger during1935–60,suggesting a negative correlation between annual-cycle variance and ENSO variance on interdecadal timescales.
The method of wavelet coherency is applied to the ENSO and monsoon indexes.The Nin?o3SST and Indian rainfall are found to be highly coherent,especially during intervals of high variance.The Nin?o3SST and Indian rainfall are approximately180?out of phase and show a gradual increase in phase difference versus Fourier period.All of the results are shown to be robust with respect to different datasets and analysis methods.
1.Introduction
Sir Gilbert Walker knew that the Southern Oscillation (SO)and the Indian summer monsoon are intimately linked(Walker and Bliss1937),and he hoped that knowledge of the SO would permit accurate prediction of the monsoon(Normand1953).Since then,numerous studies have shown the in?uence of the Indian monsoon (hereafter simply the‘‘monsoon’’)on tropical circula-tion(Normand1953;Troup1965;Yasunari1990;Ya-sunari and Seki1992)or,conversely,the impact of the El Nin?o–Southern Oscillation(ENSO)on the monsoon (Walker and Bliss1937;Shukla and Paolino1983;Jo-seph et al.1994).
Each year during March–May,the center of tropical convection migrates from the western Paci?c warm pool to the northwest,announcing the arrival of both the Southeast Asian monsoon and the Indian summer mon-soon(Meehl1987).Every few years,an El Nin?o(or warm)event produces a warming of the sea surface temperature(SST)in the central and eastern Paci?c, Corresponding author address:Dr.Christopher Torrence,Ad-vanced Study Program,National Center for Atmospheric Research, P.O.Box3000,Boulder,CO80307-3000.
E-mail:torrence@https://www.wendangku.net/doc/2d12968253.html, accompanied by diminished easterly trade winds and an eastward shift in tropical convection.The opposite La Nin?a(or cold)event,which sometimes follows a warm event,produces an anomalous westward shift in warm SSTs and convection,as well as enhanced easterly trades (Rasmusson and Carpenter1982).
The strength of the monsoon and the occurrence of warm or cold ENSO events depend on the location and magnitude of western Paci?c SSTs and on tropical con-vection(Soman and Slingo1997).A strong monsoon [heavy rains,low sea level pressure(SLP),strong east-erlies]tends to inhibit warm events and favor cold events(Yasunari1990).Conversely,a warm ENSO event(decreased convection and high SLP in the west Paci?c,weak easterlies)tends to suppress the monsoon (Webster1995).No cause or effect is implied(or indeed warranted)in either case.
Since the ENSO–monsoon interaction works in both directions,it is dif?cult to use one to forecast the other. Forecasts of either one are also complicated by the ENSO–monsoon interaction itself,which occurs ulti-mately on intraseasonal timescales through the Madden–Julian oscillation,the timing of monsoon onset,the modulation of monsoon active/break periods,and west-ern Paci?c westerly wind bursts(for a review see Web-ster et al.1998).
?1999American Meteorological Society
Coupled ocean–atmosphere models that include both intraseasonal and interannual processes suggest that the Indian monsoon can have a large impact on ENSO and vice versa(e.g.,Ju and Slingo1995;Wainer and Webster 1996;Meehl1997);yet this knowledge has not been effectively implemented into monsoon or ENSO fore-casting systems.Some of this failure is caused by in-terdecadal changes in both the ENSO–monsoon system and the global ocean–atmosphere background state (Webster and Palmer1997).
Interdecadal changes have been found in tropical rain-fall(Kraus1955;Mooley and Parthasarathy1984), monsoonal atmospheric circulation(Fu and Fletcher 1988;Parthasarathy et al.1991),ENSO frequency and amplitude(Gu and Philander1995;Mak1995;Wang and Wang1996),and ENSO–monsoon relationships(El-liott and Angell1988;Shukla1995).Multidecadal and century-long changes have been found in proxy ENSO records(Cole et al.1993;Knutson et al.1997;Diaz and Markgraf1992).Other studies have examined changes in mean SST(Deser and Blackmon1995;Zhang et al. 1997),especially on the1976–77increase in SSTs in the Paci?c Ocean(Trenberth and Hurrell1994;Wang 1995).
Since ENSO and the monsoon have such a strong teleconnection with global climate anomalies,it is im-portant to examine ENSO and monsoon variability.The interdecadal changes are also important in analyzing the appropriateness and robustness of theories for ENSO and the monsoon,and in determining the expected skill of ENSO–monsoon forecasts.
The purpose of this paper is to give a detailed de-scription of the interdecadal changes in variance and coherency of the ENSO–monsoon system during the last 125yr,along with a measure of their statistical signif-icance.Wavelet analysis is used to isolate the timescales of ENSO–monsoon variability,and signi?cance tests are used to assess the robustness of the results.The method of wavelet coherency is used to link ENSO and monsoon variance.
Section2describes the datasets used.Section3pre-sents the wavelet power spectra and variance time series. Section4uses wavelet coherency to analyze the ENSO–monsoon relationship.Veri?cation with other datasets is given in section5.A summary and discussion are given in section6.The appendix contains a description of wavelet coherency and phase.
2.Data
Three indices will be used as measures of variability in the ENSO–monsoon system:Nin?o3SST,the Southern Oscillation index(SOI),and all-India rainfall.Addi-tional datasets used for veri?cation are described in sec-tion5.a.Nin?o3sea surface temperature
The El Nin?o(or oceanic)component of ENSO is measured here by the Nin?o3SST index,consisting of the area-average SST over the eastern equatorial Paci?c (5?S–5?N,90?–150?W),monthly from January1871to September1998.
The1871–1996data is from the U.K.Meteorological Of?ce(UKMO)Global sea-Ice and Sea Surface Tem-perature(GISST)2.3-b dataset,available on a1??1?resolution grid(courtesy of D.Rowell and J.Arnott, UKMO).As described in Rayner et al.(1996),the GISST2.3b is derived from in situ ship measurements and bias-corrected satellite observations.Data-void re-gions(prominent for pre-1950data)are?lled using em-pirical orthogonal functions(EOFs).The EOFs are de-termined from the1871–1994data using anomalies from the1960–91climatology(Parker et al.1995).The EOF analysis is done on a4??4?resolution grid and consists of a single global EOF mode(to capture any trend)and multiple EOF modes for each ocean basin (Rayner et al.1996).A general discussion of data-qual-ity issues for ship measurements is given in Folland et al.(1984).
The GISST data are used to provide a gridded dataset from which area-average SSTs can be calculated.This approach is thought to be more robust than using the individual ship data,as the original ship reports are not uniformly distributed over the Nin?o regions in either space or time.
The Nin?o3SST index for January1997–September 1998is from the Climate Prediction Center(CPC)op-timally interpolated SST(courtesy of D.Garrett, NOAA).For details of the analysis method see Reynolds and Smith(1994).
b.Southern Oscillation index(SOI)
The Southern Oscillation(or atmospheric)component of ENSO is measured by the SOI(Troup1965),usually de?ned as the anomalous SLP in the eastern Paci?c at Tahiti(17.6?S,149.6?W)minus SLP in the western Pa-ci?c at Darwin,Australia(12.4?S,130.9?E).
Monthly SLP at Tahiti and Darwin from1876to1950 were provided by R.Allan(Commonwealth Scienti?c and Industrial Research Organisation),while1951–Sep-tember1998data are from CPC(courtesy of D.Garrett). The pre-1936SLP at Tahiti are from Ropelewski and Jones(1987),while pre-1882SLP at Darwin are from Allan et al.(1991).Data gaps and inhomogeneities were corrected using linear regression from nearby stations by Allan et al.(1996).
To measure interdecadal changes,the annual cycle is kept within each time series,and the SOI is simply de?ned as the monthly SLP at Tahiti minus the SLP at Darwin(note that the wavelet analysis will separate the residual annual cycle from the ENSO timescales).One possible criticism of this de?nition is the different SLP
variance at Darwin and Tahiti(Trenberth1976).Most of this difference is due to the larger annual cycle at Darwin.The standard deviations(1876–1998)of SLP at Darwin and Tahiti are2.62and1.68mb,respectively; with the monthly mean annual cycle removed,the stan-dard deviations become1.08and1.06mb,respectively. Therefore,changes in the SOI annual cycle will mostly re?ect Darwin,while SOI changes on other timescales will be equally weighted between Darwin and Tahiti.
c.All-India rainfall
Indian rainfall is from the all-India rainfall index, consisting of an area-weighted average of306rain gaug-es distributed across the plains regions of India(Mooley and Parthasarathy1984).Monthly rainfall is available from1871to1994,and the mean(90.9mm)is removed. Note that this is a monthly index and is not just the sum over the monsoon rainy season.
3.Changes in ENSO and monsoon variance
The time series of Nin?o3SST,the SOI,and Indian monsoon rainfall are shown in Fig.1.The SST and SOI have been smoothed to highlight the interannual ENSO component;however,the rest of the analysis uses the unsmoothed time series.Likewise,the Indian rainfall time series shows only the June–September total(in standard deviations from the mean),while the remainder of the paper uses the monthly time series.
The Nin?o3SST and SOI are clearly180?out of phase, as expected for the El Nin?o–Southern Oscillation(Ras-musson and Carpenter1982).Both warm(El Nin?o)and cold(La Nin?a)events occur approximately every2–7 yr,and there is considerable variability in the amplitude of ENSO events(e.g.,the large1997–98El Nin?o)and in the duration of events(e.g.,the prolonged1992–95 warm anomaly).The shading on the monsoon rainfall indicates years associated with El Nin?o events(black bars)or La Nin?a events(gray bars).It is apparent that El Nin?o years are typically associated with de?cient monsoon rainfall,while La Nin?a years are associated with abundant rainfall(Shukla and Paolino1983;Ya-sunari and Seki1992).The relationship is not perfect, suggesting that processes other than ENSO can have an impact on monsoon rainfall(and vice versa).
In Fig.1,one can see evidence of three distinct time intervals:the?rst(1875–1920)has numerous ENSO events and large changes in monsoon rainfall,the second (1920–60)has fewer ENSO events and less monsoon variability,and the third(1960–present)again has more ENSO events and more monsoon variability.These in-terdecadal changes will be examined in the following sections using wavelet analysis.
a.Wavelet analysis method
To isolate the different timescales,the Nin?o3SST,the SOI,and Indian rainfall are analyzed with the wavelet transform.The Morlet wavelet is used,and the trans-form is performed in Fourier space using the method described in Torrence and Compo(1998;hereafter TC98).Further details on wavelet analysis can be found in Foufoula-Georgiou and Kumar(1995).To reduce wraparound effects,each time series is padded with ze-ros.Other wavelet bases,such as the Paul and Mexican hat,were tested and gave the same qualitative results. The Morlet wavelet consists of a complex exponential modulated by a Gaussian,,where t is the
22
i?t/s?t/(2s)
e e
time,s is the wavelet scale,and?
is a nondimensional frequency.For?
0?
6(used here),there are approxi-mately three oscillations within the Gaussian envelope. The wavelet scale s is almost identical to the corre-sponding Fourier period of the complex exponential, and the terms‘‘scale’’and‘‘period’’will be used syn-onymously.
The wavelet power spectrum is de?ned as the absolute value squared of the wavelet transform and gives a mea-sure of the time series variance at each scale(period) and at each time.Note that the wavelet transform con-serves variance,and the time series can be recovered exactly(TC98).To test the signi?cance of peaks in the wavelet power spectrum,a background Fourier spec-trum must be chosen.To test for nonstationary changes in variance,it is most appropriate to choose the global wavelet spectrum(GWS),given by the time average of the wavelet power spectrum(Kestin et al.1998).The wavelet power spectrum is then distributed as chi-square (two degrees of freedom)about the GWS(TC98). The global wavelet spectra are shown in Fig.2.The GWS is equivalent to the Fourier power spectrum smoothed by the Morlet wavelet function in Fourier space(Farge1992).Since the width of the wavelet func-tion is constant in period(the short horizontal lines), the number of smoothed Fourier components decreases with increasing period.This implies a decrease in the degrees of freedom(DOF)with increasing period,and a corresponding increase in the width of the95%con-?dence intervals(the error bars).The decrease in DOF also causes a rise in the1%signi?cance level with pe-riod,despite the choice of a white-noise background (TC98).
The GWS for Nin?o3SST and the SOI show the annual cycle and strong2–8-yr‘‘interannual’’peaks.The In-dian rainfall is dominated by the annual cycle;the rest of the spectrum appears?at(i.e.,white noise),except for a possible peak around2.8yr.The next section will analyze the variation of wavelet power about these mean spectra.
b.Wavelet power spectra
The wavelet power spectra for Nin?o3SST,the SOI, and Indian rainfall are shown in Fig.3.The wavelet power is normalized by dividing each month by the GWS(Fig.2),and thus it measures the deviations from the mean spectrum.The thick black contours indicate
F IG.1.Time series of Nin?o3SST(solid curve,?C),the SOI(dashed curve,mb/2),and the Jun–Sep (JJAS)total monsoon rainfall(bars;mean?853mm,standard deviation??84.2mm).For presentation purposes,the monthly mean annual cycles have been removed,and the SST and SOI have been smoothed by an11-month Lanczos?lter(Trenberth1984).For the monsoon rainfall,black bars indicate El Nin?o year
(0)s and gray bars indicate La Nin?a year(0)s,as de?ned by Kiladis and Diaz
(1989).
F IG.2.The global wavelet power spectra for Nin?o3SST(solid line),the SOI(dashed line)and the all-India rainfall(gray line),as a function of Fourier period.The dotted line is the1%signi?cance level for all three,assuming a white noise process.Before computing variances(?2)and signi?cance levels,the time series were prewhi-tened by removing the monthly mean annual cycle;the annual cycle was then reinstated.The vertical error bars represent the95%con-?dence interval(on a logarithmic scale)for each of the curves(for clarity the bars are only plotted for every power of2).The horizontal lines across each error bar indicate both the baseline of the error bar, and the width of the wavelet function in Fourier space.regions that are signi?cant(at the10%level)above the GWS.
For Nin?o3SST and the SOI(Figs.3a,b),the power is broadly distributed,with peaks in the2–8-yr ENSO band.The10%signi?cance regions indicate that1875–1920and1960–90contain intervals of higher ENSO variance,while1920–60is a time of lower ENSO var-iance.Similar variance changes are found in east Paci?c SST and Darwin SLP(Wang and Wang1996),and in tropical zonal winds(Gu and Philander1995).Here,the use of signi?cance tests allows one to verify that the interdecadal variance changes in ENSO are statistically signi?cant(Torrence and Webster1998).
The wavelet spectrum for all-India rainfall(Fig.3c) is more uniformly distributed about all periods,as can also be seen in the GWS.The1–8-yr band shows in-terdecadal changes similar to Nin?o3SST and the SOI, with a distinct time of low variability from1920to1950. Normand(1953)also noted a decrease in the severity of drought/?ood events after1920.
The tropical biennial oscillation(TBO),or2-yr band, is not especially pronounced,except perhaps in the rain-fall.Since the TBO is thought to occur irregularly (Meehl1997),the Morlet wavelet(which has three os-cillations)is perhaps not the best wavelet function to use.Nevertheless the2-yr variance shows the same de-crease from1920to1960as the2–8-yr band.There is also power in the8–20-yr band during the last50yr,
F IG.3.(a)Wavelet power spectrum(using the Morlet wavelet)of the Nin?o SST.The wavelet power at each period is normalized by the global wavelet spectrum(GWS;Fig.2).The shaded contour levels are at1and2,i.e.,‘‘equal to the GWS’’and‘‘twice the GWS,’’respectively.Cross-hatched regions indicate the‘‘cone-of-in?uence,’’where zero padding has reduced the variance.The thick black contour is the10%signi?cance level above the GWS.(b)Same as(a)but for the SOI.(c)Same as(a)but for all-India rainfall.
which agrees with the results of Mooley and Parthas-arathy(1984).
The wavelet power spectra for Nin?o3SST,the SOI, and all-India rainfall indicate that there is signi?cant(at 10%)nonstationary in variance in the2–8-yr ENSO band.The variance changes also appear to agree on some of the?ner-detail features,especially in the2–4-yr band.These features are more easily seen in the var-iance time series.
c.Time series of2–7-yr variance
To clarify the interdecadal changes in interannual var-iance,Fig.4shows the2–7-yr wavelet variance times series(thick black curves with shading).Each horizontal slice through Fig.3represents a time series of wavelet variance at a particular scale(relative to the GWS).The wavelet power spectra are converted to energy densities by?rst multiplying by the GWS(to remove the nor-malization)and then dividing each time series by its corresponding wavelet scale.The2–7-yr time series of energy density is then summed(TC98).Due to the scale width of the Morlet wavelet,the results are similar for scale averages using2–5yr or3–7yr.The above method is also equivalent to using a Fourier bandpass?lter:the variance is conserved,and the mean amplitude of the variance time series is equal to the total variance within the bandpass?lter(TC98).
The2–7-yr variance time series of Nin?o3SST,SOI, and all-India rainfall(Fig.4)show large changes over the last125yr,with high variance from1875to1920, lower variance from1920to1960,and then high var-iance from1960to1990.The three variance time series are highly correlated with each other,with the Spearman rank correlations given in Table1(the rank correlation is used because the time series are not Gaussian dis-tributed).The signi?cance levels are determined from Student’s t-distribution,with the degrees of freedom es-timated using a decorrelation time of5.2yr,determined from the e-folding time of the2–7-yr scale-averaged wavelet.
In addition to the1875–1920/1920–60/1960–90in-terdecadal change in variance,there is also a12–20-yr ‘‘oscillation’’in variance,which is especially prominent in1875–1920and1960–90(Torrence and Webster 1998).The thin curves in Fig.4are the square of the 2–7-yr bandpass-?ltered time series.The black shading indicates where the?ltered time series was originally positive in Nin?o3SST,or negative in the SOI and Indian rainfall.
For Nin?o3SST and the SOI the12–20-yr variance oscillation corresponds to a12–20-yr amplitude mod-ulation of ENSO events.The ENSO system seems to go through an irregular cycle of increasing amplitude of ENSO events followed by decreasing amplitude.The 12–20-yr cycle is clearly apparent in1875–1920and 1960–90,but less apparent for1920–60.This modu-lation was also found in central Paci?c SST from1949 to1992by Mak(1995).For all-India rainfall the12–20-yr modulation is not as well de?ned,yet there are still intervals of correspondence between ENSO vari-ance and all-India rainfall variance(e.g.,1871–85, 1895–1925,1960–90).The intervals of high ENSO–monsoon correspondence are more apparent in the wavelet coherency(section4).
d.Time series of annual-cycle variance
The annual-cycle variance of Nin?o3SST(relative to the GWS)is given by a slice at1-yr period through Fig. 3a.The time series of1-yr variance for Nin?o3SST and Indian rainfall(not shown)are rank correlated at0.48 (1%signi?cant).When the amplitude of the SST annual cycle in the eastern Paci?c is large,the amplitude of the Indian rainfall annual cycle also tends to be large. The other rank correlations are given in Table1.
F I
G .4.(a)The upper half is the 2–7-yr variance time series for Nin ?o3SST (thick black curve
with gray shading).The thin curve is (Nin ?o3SST)2,where the Nin ?o3SST is bandpass ?ltered 1
2from 2-to 7-yr using a wavelet ?lter.The black shading indicates positive peaks in the ?ltered time series (i.e.,before squaring),while the white are negative peaks.The lower half is the same but for the SOI (with variance increasing downward);the black bars now indicate negative peaks in the ?ltered SOI.(b)The upper half is the same as in (a),while the lower half is for the all-India rainfall;the black bars indicate negative peaks in the ?ltered rainfall.In all plots the hatched bands show where zero padding has reduced the variance.
T ABLE 1.Spearman rank correlation (?100)between wavelet-de-rived time series.Nin ?o3?is the 2–7-yr bandpass-?ltered SST.Italics indicates signi?cant at the 5%level,while bold are signi?cant at the 1%level.The degrees of freedom (DOF)are estimated from the wavelet e -folding time (for each correlation the smaller of the two DOF is used).
2–7-yr variance Nin ?o3
SOI Rain
1-yr variance Nin ?o3
Rain
DOF Nin ?o3(2–7)
SOI (2–7)Rain (2–7)Nin ?o3(1-yr)Rain (1-yr)Nin ?o3?
—
7655?25?43*
—32?04?25*
—?11?07*
—48?29
—?23
22
2122939145
*Zero by construction.
For Nin ?o3SST ,the 1-yr variance is negatively cor-related with the 2–7-yr bandpass ?lter (r ??0.29,5%signi?cant).This agrees with Gu and Philander (1995),who found that during El Nin ?o events the SST annual cycle has a smaller amplitude,while during La Nin ?a the SST annual cycle is larger.This is due to the phase locking of ENSO events to the annual cycle.In the eastern equatorial Paci?c,ENSO events tend to peak in December–February,when the sea surface temperature is normally low.A warm event produces above-normal SSTs during the cooler phase of the annual cycle (weak-ening the annual cycle),while a cold event produces below-normal SSTs (strengthening the annual cycle).Gu and Philander (1995)could ?nd no relationship between interdecadal changes in ENSO and annual-cy-cle variance.Figure 3a suggests,however,that there is
F IG.5.Cumulative variance curves for the1-yr Nin?o3SST variance (thin solid curve),the1-yr Indian rainfall variance(thin dashed curve),and the2–7-yr Nin?o3SST variance(thick curve),using the variance time series from Fig.3.Before calculating,each time series was normalized by dividing by the mean,subtracting1.0,and dividing by the average wavelet scale(either1.0or3.67yr).
higher1-yr power from1935to1960,at the same time as there is lower2–7-yr power.Cole et al.(1993),using proxy rainfall at Tarawa(1?N,172?E),also found in-creased1-yr power from1930to1955and decreased 2–5-yr power from1925to1955.Note in Fig.3b,how-ever,that the higher1-yr power is not seen in the SOI. To clarify the relationship between annual-cycle and ENSO variance,Fig.5shows the cumulative curves for the1-and2–7-yr variance time series.The cumulative value is de?ned as the sum of the time series from an initial time up to a certain time.Before calculating the cumulative variance,the time series was normalized by the mean variance.Upward trends in the cumulative variance indicate intervals of above-normal variance, while upward concavity indicates intervals of increasing variance(Kraus1955).
The cumulative variance indicates that during inter-vals of high ENSO variance(1875–1920and1960–90) the annual-cycle variance is lower than average for both Nin?o3SST and all-India rainfall.Conversely,during the 1920–60interval of low ENSO variance there is higher than average annual-cycle variance.This inverse rela-tionship also appears at the12–20-yr timescale,espe-cially during1875–1910.The apparent phase shift be-tween the1-and2–7-yr cumulative Nin?o3variance on short timescales(prominent from1875to1910)is due to the more rapid variation of the1-yr variance,which causes changes in cumulative variance to appear earlier. To test for signi?cance,a Monte Carlo of10000 white noise time series was used with the above analysis method.None of the peaks in Fig.5is signi?cant(10% level),which could suggest that the ENSO and monsoon time series are stationary in the1-and2–7-yr bands.Nevertheless,it should be noted that the cumulative variance method is very sensitive to the white noise assumption,which is not necessarily appropriate for the annual cycle or ENSO timescales.
4.Wavelet coherency and phase
The Fourier squared coherency is used to identify frequency bands within which two time series are co-varying.Here,the wavelet squared coherency is used to identify both frequency bands and time intervals within which ENSO and the Indian monsoon are cov-arying.The equations for wavelet coherency and phase are given in the appendix.Henceforth,the term‘‘co-herency’’refers to the wavelet squared coherency. The wavelet coherency between Nin?o3SST and all-India rainfall is shown in Fig.6.The5%signi?cance level was determined from a Monte Carlo simulation of 10000sets of two white noise time series with the same length as Nin?o3SST and rainfall.The SST and rainfall show signi?cant coherency in the1-and2–8-yr bands, with low coherency outside of these periods.The in-terdecadal change in ENSO–monsoon variance appears as an interval of low coherency from1920to1960, especially in the2–4-yr band.For1920–60,Trenberth and Shea(1987)found weakened covariance for Darwin versus Pambam,India,while Allan et al.(1996)found weakened spatial correlation patterns between ENSO indexes and SST and SLP.From1950to1990there is a shift in the period of maximum coherency from around 7yr down to around2yr.This shift also appears in the coherency of Nin?o3SST and the SOI(not shown). The usefulness of the wavelet coherency is especially apparent for intervals where both wavelet power spectra (Figs.3a,c)show minimal power,yet there is still high coherency.For example,in1935–45there is a high coherency peak at the2-yr period,which corresponds to the1939–42sequence of El Nin?o/La Nin?a events coupled to weak/strong monsoons(see Fig.1).The peak in1945–55at5yr corresponds to a cycle of cold/warm/ cold Nin?o SST(4yr each)coupled to a cycle of strong (9yr)/weak(2yr)/strong(4yr)monsoons.These results suggest that even during times of low ENSO–monsoon relationship,the two phenomena can still show occa-sional strong interactions.
The wavelet coherency between Nin?o3SST and the SOI(not shown)is greater than0.8for most times in the2–8-yr band,and it exceeds0.95during the high ENSO variance intervals from1875to1910and1960 to1990.In Nin?o3SST–SOI there is low coherency outside of2–8yr,suggesting that independent processes (e.g.,noise)operate at the smallest and largest scales. The vectors in Fig.6indicate the phase difference between Nin?o3SST and all-India rainfall at each time and period(see key).For clarity,only one vector is plotted every2yr,and every other vector is plotted in the period direction(because of the timescale averaging no information is lost).
F IG.6.The wavelet coherency and phase between Nin?o3SST and all-India rainfall.Contours are for wavelet squared coherencies of0.5,0.8,and0.9.The thick black contour is the5%signi?cance level from a Monte Carlo simulation of wavelet coherency between10000sets(two each)of white noise time series.The vectors indicate the phase difference between Nin?o3SST and rainfall(key in the lower right corner).For clarity,vectors are only plotted for coherency?0.5,only one vector is plotted every2yr in time,and every other vector is plotted in period.Cross-hatched regions indicate the cone of in?uence.
In the annual-cycle(1yr)band the Nin?o3SST and rainfall are about120?out of phase.The peak SST in Nin?o3occurs around April,while the peak Indian rain-fall occurs around August,corresponding to a4-month (120?)phase difference.This phase difference appears constant over the1871–1994record.
In the2–8-yr band the Nin?o3SST and rainfall are approximately180?out of phase,indicating that abun-dant rainfall years tend to be associated with cold events, while drought years tend to be associated with warm events.For periods outside2–8yr the phase difference is more random.
The phase difference(rainfall lagging SST)increases from about120?near the2-yr band to about180?near the8-yr band(see especially1875–1910and1930–50). This phase increase is related to the lag between the Indian monsoon season and the typical boreal winter peak of ENSO events.Figure7a shows the regression of November–January(NDJ)Nin?o3SST onto individ-ual months of Nin?o3SST and all-India rainfall for one year before,during,and after the NDJ season.During an El Nin?o(year0)the Indian monsoon tends to be weaker,yet the weak monsoon actually occurs approx-imately4months before the peak Nin?o3SST(depending on the El Nin?o;see Webster et al.1998).Similarly,Fig. 7b shows the regression of June–September all-India rainfall onto Nin?o3SST and all-India rainfall,and it indicates that a strong monsoon(year0)is associated with cold Nin?o3SST,which peaks about four months after the monsoon.Both scenarios produce a phase shift of8months between Nin?o3SST and all-India rainfall. On a biennial timescale this implies that Nin?o3SST leads rainfall by120?(or,equivalently,Nin?o3SST lags by240?).On interannual timescales the4-month sea-sonal lag becomes irrelevant and the phase difference approaches180?out of phase.
Figure7also contains a distinct biennial cycle,es-pecially for rainfall.Weak monsoons tend to be followed the next year by decreased June rainfall(possibly in-dicating delayed onset)yet more rainfall overall.Strong monsoons are followed the next year by increased June rainfall(possibly earlier onset)with less rainfall overall.
5.Veri?cation with other datasets
There are numerous data-quality issues(especially pre-1950)for the GISST dataset,which include changes in measurement methods and inadequate spatial and temporal coverage.In addition,the large-scale patterns of SST and SLP associated with ENSO are continually changing,making it dif?cult to regard a single time series as representative of long-term ENSO activity (Trenberth1976;Trenberth1997).
To check the robustness of the interdecadal changes in ENSO and Indian rainfall,all of the above wavelet analyses were repeated using the following datasets and indexes:
1)Using the GISST data,all four Nin?o regions:
Nin?o1?2(10?S–0?N,80?–90?W),Nin?o3(5?S–5?N,
F IG.7.(a)The linear regression of average Nov–Jan(NDJ;shaded)
Nin?o3SST(1872–1993;122points)onto Nin?o3SST(black curve)
and Indian rainfall(bars)for each month preceeding(year?1),
during(year0),and following(year?1)the NDJ season.To simulate
realistic amplitudes,a regression of?2standard deviations was used.
The monthly mean annual cycle was removed from both time series.
(b)Same as(a),but for the linear regression of total Jun–Sep(JJAS;
shaded)all-India rainfall onto Nin?o3SST and all-India rainfall.
T ABLE2.Spearman rank correlation(?100)between2–7-yr variance time series.Italics indicates signi?cant at the5%level,while bold are signi?cant at the1%level.The signi?cance levels use a5.2-yr e-folding time for the degrees of freedom.
GISST
Nin?o1?2 1871–1998
Nin?o3
1871–1998
Nin?o3.4
1871–1998
Nin?o4
1871–1998
CPC
Nin?o3
1950–98
Wright
S
1881–1986
R
1894–1982
dt
1870–1984
GISST Nin?o3 GISST Nin?o3.4 GISST Nin?o4 CPC Nin?o3 Wright S Wright R Wright dt
SOI 80
74
72
87
73
58
63
58
—
97
87
96
88
83
67
76
—
93
91
88
82
61
74
—
71
85
77
47
64
—
85
82
83
87
—
85
64
70
—
79
85
—
83
90?–150?W),Nin?o3.4(5?S–5?N,120?–170?W),and Nin?o4(5?S–5?N,160?E–150?W);
2)The Nin?o3SST index,from January1950to Sep-
tember1998,using data from the Climate Prediction Center optimally interpolated SST[courtesy of D.
Garrett,NOAA;see Reynolds and Smith(1994)];
and
3)Wright’s(1989)indexes of central Paci?c SST(‘‘S’’;
monthly1881–1986),equatorial Paci?c rainfall (‘‘R’’;monthly1894–1982),and Darwin minus Ta-hiti pressure(‘‘DT’’;seasonal1870–1984);these in-dexes are corrected for inhomogeneities according to Wright(1989).
The results using these datasets are consistent with all of the wavelet results of the previous sections.The wavelet power spectra(e.g.,Fig.3)for all time series show nearly identical patterns.The2–7-yr variance time series(e.g.,Fig.4)contain both the interdecadal changes and the12–20-yr modulation.The rank correlations be-tween the2–7-yr variance time series are highly sig-ni?cant and are given in Table2.
The long-term changes in variance are also robust to other analysis techniques.The1920–60variance min-imum appears when using either a running15-yr vari-ance,or using a sliding15-yr window Fourier power spectrum(not shown).
The interdecadal changes in ENSO–monsoon vari-ance appear to be independent of the datasets used or the analysis method.
6.Summary and discussion
The ENSO and Indian monsoon have undergone con-siderable interdecadal changes in variance and coher-ency over the last125yr.The following results were shown.
R Wavelet power spectra(Fig.3)and variance time se-ries(Fig.4)show interdecadal changes in2–7-yr var-iance,and indicate intervals of high variance(1875–1920and1960–90)and an interval of low variance (1920–60).
R The2–7-yr variance time series contain a12–20-yr oscillation(Fig.4),consisting of a12–20-yr modu-lation of ENSO and monsoon amplitude.
R The annual-cycle variance is negatively correlated with the interannual ENSO signal(Table1),agreeing with Gu and Philander(1995).
R The annual-cycle variance appears to be negatively correlated with ENSO variance on interdecadal time-scales(Fig.5),with higher annual-cycle variance (from1935to1960)during the interval of low ENSO variance.
R The wavelet coherency between Nin?o3SST and all-India rainfall is very high(Fig.6),especially in the 2–8-yr band during intervals of high variance(1875–1920and1960–90).
All of the results are robust with respect to different datasets and indices(Table2),and different analysis methods(running variance and Fourier spectra). Currently,there is no explanation for either the long-term or12–20-yr changes in ENSO–monsoon variance and coherency,although the results suggest three pos-sible explanations.
1)The changes in variance are part of the internal var-
iability of the tropical climate https://www.wendangku.net/doc/2d12968253.html,ing simple dynamical models,Lorenz(1991)found extended intervals,‘‘longer than any built-in timescale,’’of one type of dynamical behavior,followed by long intervals of different behavior.In this context,1875–1920and1960–90exhibit one behavior,while 1920–60exhibits a different https://www.wendangku.net/doc/2d12968253.html,rge changes in ENSO variance have been found in a low-reso-lution coupled ocean–atmosphere GCM,without any changes in extratropical or external forcing(Knutson et al.1997).The Cane–Zebiak model,commonly used for ENSO forecasts,also contains interdecadal changes in ENSO amplitude and frequency,again with no changes in forcing(Zebiak and Cane1991).
2)The changes in variance could be linked to external
changes in the mean background climate or to ex-tratropical changes in variance.Field correlations of 2–7-yr SST variance time series(not shown)indicate possible links on interdecadal timescales to the North Paci?c,although the results are very preliminary.
Connections exist between SSTs in the central equa-torial Paci?c and the North Paci?c(Deser and Black-mon1995;Mak1995),and possibly between the ENSO–monsoon system and the North Atlantic os-cillation(Yasunari and Seki1992).Other studies suggest that changes in Paci?c Ocean climate can in?uence ENSO–monsoon variability(Parthasarathy et al.1991;Trenberth and Hurrell1994;Wang1995;
Wang and Ropelewski1995).
3)A combination of the previous explanations,where
internal variability is modulated in part by external https://www.wendangku.net/doc/2d12968253.html,ing the Cane–Zebiak model coupled to
a simple statistical atmosphere,Kirtman and Schopf
(1998)suggest that interdecadal changes in ENSO are caused by internal variability that manifests itself as changes in the mean tropical climate.Conversely, Gu and Philander(1997)propose an interdecadal oscillation of subsurface temperature anomalies be-tween the extratropics and Tropics,causing changes
in tropical SSTs and a modulation of ENSO vari-ability.
Given the limited data record and the dif?culty of separating the different timescales,it is uncertain which (if any)of the above scenarios is correct.
The increase in annual-cycle variance during times of decreased2–7-yr variance suggest a possible‘‘trans-fer’’of energy between the the annual cycle and ENSO, occurring on interdecadal timescales.The annual cycle during1935–60may have been more ef?cient at redis-tributing heat throughout the Tropics,thereby reducing the amplitude and frequency of ENSO events.These changes in variance may also be related to a shift in monsoonal circulation from‘‘meridional’’to‘‘zonal’’around1940(Fu and Fletcher1988;Parthasarathy et al. 1991),although the relationship and timing are unclear. The interdecadal changes in wavelet coherency be-tween Nin?o3SST and all-India rainfall closely parallel the interdecadal variance changes,suggesting that in-tervals of high variance are associated with clear ENSO–monsoon relationships,and intervals of low var-iance show little ENSO–monsoon relationship.The var-iance–coherency relationship could help in assessing model predictive skill:times of high variance would presumably have higher predictive skill for both the monsoon and ENSO,due to an increase in the usefulness of the ENSO–monsoon connection.Indeed,Walker’s monsoon‘‘foreshadowing’’was successful up until the early1920s when the SOI–monsoon connection became weaker(Normand1953).During1920–60the low co-herency perhaps indicated that other processes(such as Eurasian snow cover)needed to be considered for suc-cessful monsoon forecasts.It is interesting to note that from1921to1950Australian rainfall shows both low persistence and a weakened correlation with the SOI (Simmonds and Hope1997),suggesting that changes in ENSO variance may also affect the predictability of the Australian monsoon.
During the last30years it has become possible to successfully forecast ENSO events up to a year in ad-vance(Webster and Palmer1997).The strong ENSO–monsoon connection suggests that there is some pre-dictability for the coupled ENSO–monsoon system(Ju and Slingo1995).Nevertheless,Shukla(1995)notes that‘‘the relationship between El Nin?o and the Indian monsoon...does not hold during the most recent de-cade.’’As an example,1994was an El Nin?o year,yet the Indian monsoon was unusually strong(Soman and Slingo1997).Similarly,the1997El Nin?o started un-usually early and became very intense,yet the monsoon was close to normal(Webster and Palmer1997).It is possible that the tropical climate system is returning to a period similar to1920–60,with low ENSO–monsoon variability and reduced predictability.
Further work is needed on the cause of interdecadal variance changes in the ENSO–monsoon system,on the relationship between annual-cycle and ENSO variance,
and on the link between variance changes in the Tropics and extratropics.
Acknowledgments.Thanks to https://www.wendangku.net/doc/2d12968253.html,po,C.De-ser,R.Madden,C.Penland,K.Trenberth,and W.Welch.The GISST dataset was kindly provided by D.Rowell,J.Arnott,and the U.K.Hadley Centre Meteorological Of?ce.The Darwin and Tahiti SLP were kindly pro-vided by R.Allan (CSIRO)and D.Garrett (NOAA).This work was funded in part by the NOAA Of?ce of Global Programs Grant NA56GP0230and National Sci-ence Foundation Grants ATM9526030and ATM9525860.The National Center for Atmospheric Research is sponsored by the National Science Foun-dation.
APPENDIX
Wavelet Coherency and Phase
Traditionally,Fourier coherency has been used to identify frequency bands where two time series are re-lated.One would like to develop a wavelet coherency,which could identify both frequency bands and time intervals when the time series were related (Liu 1994).In Fourier analysis,it is necessary to smooth the cross spectrum before calculating coherency (which is oth-erwise identically equal to 1).In previous studies,it was unclear how to smooth the cross-wavelet spectrum to de?ne an appropriate wavelet coherency (Liu 1994).Here,the wavelet coherency is de?ned using smoothing in both time and scale,with the amount of smoothing dependent on both the choice of wavelet and the scale.Given two time series X and Y,with wavelet trans-forms (s )and (s ),where n is the time index and X Y
W W n n s is the scale,the cross-wavelet spectrum is de?ned as
(s )?(s )*(s ),
XY X Y
W W W n n n (A1)
where (*)indicates complex conjugate.The cross-wave-let spectrum accurately decomposes the Fourier co-and quadrature-spectra into timescale space.
The wavelet squared coherency is de?ned as the ab-solute value squared of the smoothed cross-wavelet spectrum,normalized by the smoothed wavelet power spectra,
?1XY
2
|?s W (s )?|n 2R (s )?,
(A2)
n
?1X 2?1Y 2?s |W (s )|??s |W (s )|?n n where ?·?indicates smoothing in both time and scale.Note that in the numerator,both the real and imaginary parts of the cross-wavelet spectrum are smoothed sep-arately before taking the absolute value,while in the denominator it is the wavelet power spectra (after squar-ing)that are smoothed.The factor s ?1is used to convert
to an energy https://www.wendangku.net/doc/2d12968253.html,ing these de?nitions,0?(s )2
R n ?1.Finally,it is noted that because the wavelet trans-form conserves variance (TC98),the wavelet coherency is an accurate representation of the (normalized)co-variance between the two time series.
The wavelet-coherency phase difference is given by
?1XY
?{?s W (s )?}
n ?1
?(s )?tan
.(A3)
n ?1XY
?
?
?{?s W (s )?}
n The smoothed real (?)and imaginary (?)parts should
have already been calculated in (A2).Both (s )and 2
R n ?n (s )are functions of the time index n and the scale s.The smoothing in (A2)and (A3)is done using a weighted running average (or convolution)in both the time and scale directions.It should be noted that just as the Fourier coherency depends on an arbitrary smoothing function,so also does the wavelet coherency.Nevertheless,the width of the Morlet wavelet function in both time and Fourier space provides a ‘‘natural’’width of the smoothing function.The time smoothing uses a ?lter given by the absolute value of the wavelet function at each scale,normalized to have a total weight of unity.For the Morlet wavelet this is just a Gaussian,.The scale smoothing is done using a boxcar 22?t /(2s )e ?lter of width ?j 0,the scale-decorrelation length.For the Morlet wavelet,?j 0?0.60(TC98).Using different ?lter widths and shapes produces either smoother (and larger)coherency or noisier (smaller)coherency yet still gives the same qualitative results.The above ?lters are the best ‘‘compromise,’’as they provide the minimal amount of smoothing necessary to include two inde-pendent points in both the time and scale dimensions.
REFERENCES
Allan,R.J.,N.Nicholls,P .D.Jones,and I.J.Butterworth,1991:A
further extension of the Tahiti–Darwin SOI,early ENSO events,and Darwin pressure.J.Climate,4,743–749.
,J.A.Lindesay,and D.E.Parker,1996:El Nin ?o Southern Oscillation and Climatic Variability.CSIRO,408pp.
Cole,J.E.,R.G.Fairbanks,and G.T.Shen,1993:Recent variability
in the Southern Oscillation:Isotopic results from a Tarawa atoll coral.Science,260,1790–1793.
Deser,C.,and M.L.Blackmon,1995:On the relationship between
tropical and North Paci?c sea surface temperature variations.J.Climate,8,1677–1680.
Diaz,H.F .,and V.Markgraf,Eds.,1992:El Nin ?o:Historical and
Paleoclimatic Aspects of the Southern Oscillation.Cambridge University Press,476pp.
Elliott,W.P .,and J.K.Angell,1988:Evidence for changes in South-ern Oscillation relationships during the last 100years.J.Climate,1,729–737.
Farge,M.,1992:Wavelet transforms and their applications to tur-bulence.Annu.Rev.Fluid Mech.,24,395–457.
Folland,C.K.,D.E.Parker,and F .E.Kates,1984:Worldwide marine
temperature ?uctuations 1856–1981.Nature,310,670–673.Foufoula-Georgiou,E.,and P .Kumar,Eds.,1995:Wavelets in Geo-physics.Academic Press,373pp.
Fu,C.-B.,and J.Fletcher,1988:Large signals of climatic variation
over the ocean in the Asian monsoon region.Adv.Atmos.Sci.,5,389–404.
Gu,D.,and S.G.H.Philander,1995:Secular changes of annual and
interannual variability in the Tropics during the past century.J.Climate,8,864–876.,and ,1997:Interdecadal climate ?uctuations that depend on exchanges between the Tropics and extratropics.Science,275,805–807.
Joseph,P .V.,J.K.Eischeid,and R.J.Pyle,1994:Interannual var-
iability of the onset of the Indian summmer monsoon and its association with atmospheric features,El Nin?o,and sea surface temperature anomalies.J.Climate,7,81–105.
Ju,J.,and J.Slingo,1995:The Asian summer monsoon and ENSO.
Quart.J.Roy.Meteor.Soc.,121,1133–1168.
Kestin,T.A.,D.J.Karoly,J.-I.Yano,and N.A.Rayner,1998:Time–frequency variability of ENSO and stochastic simulations.J.
Climate,11,2258–2272.
Kiladis,G.N.,and H.F.Diaz,1989:Global climatic anomalies as-sociated with extremes in the Southern Oscillation.J.Climate, 2,1069–1090.
Kirtman,B.P.,and P.S.Schopf,1998:Decadal variability in ENSO predictability and prediction.J.Climate,11,2804–2822. Knutson,T.R.,S.Manabe,and D.Gu,1997:Simulated ENSO in a global coupled ocean–atmosphere model:Multidecadal ampli-
tude modulation and CO
2sensitivity.J.Climate,10,138–161.
Kraus,E.B.,1955:Secular changes of tropical rainfall regimes.
Quart.J.Roy.Meteor.Soc.,81,198–210.
Liu,P.C.,1994:Wavelet spectrum analysis and ocean wind waves.
Wavelets in Geophysics,E.Foufoula-Georgiou and P.Kumar, Eds.,Academic Press,151–166.
Lorenz,E.N.,1991:Chaos,spontaneous climatic variations and de-tection of the greenhouse effect.Greenhouse Gas-Induced Cli-matic Change:A Critical Appraisal of Simulations and Obser-vations,M.E.Schlesinger,Ed.,Elsevier,457–469.
Mak,M.,1995:Orthogonal wavelet analysis:Interannual variability in the sea surface temperature.Bull.Amer.Meteor.Soc.,76, 2179–2186.
Meehl,G.A.,1987:The annual cycle and interannual variability in the tropical Paci?c and Indian Ocean regions.Mon.Wea.Rev., 115,27–50.
,1997:The South Asian monsoon and the tropospheric biennial oscillation.J.Climate,10,1921–1943.
Mooley,D.A.,and B.Parthasarathy,1984:Fluctuations in all-India summer monsoon rainfall during1871–1978.Climate Change, 6,287–301.
Normand,C.,1953:Monsoon seasonal forecasting.Quart.J.Roy.
Meteor.Soc.,79,463–473.
Parker,D.E.,M.Jackson,and E.B.Horton,1995:The GISST2.2 sea surface temperature and sea-ice climatology.Hadley Centre for Climate Prediction and Research Tech.Rep.CRTN63,35 pp.[Available from Hadley Centre for Climate Prediction and Research,U.K.Meteorological Of?ce,London Road,Bracknell, Berkshire RG122SY,United Kingdom.]
Parthasarathy,B.,K.Rupa Kumar,and A.A.Munot,1991:Evidence of secular variations in Indian monsoon rainfall–circulation re-lationships.J.Climate,4,927–938.
Rasmusson,E.M.,and T.H.Carpenter,1982:Variations in tropical sea surface temperature and surface wind?elds associated with the Southern Oscillation/El Nin?o.Mon.Wea.Rev.,110,354–384.
Rayner,N.A.,E.B.Horton,D.E.Parker,C.K.Folland,and R.B.
Hackett,1996:Version2.2of the Global Sea-Ice and Sea Surface Temperature Data Set,1903–1994.Hadley Centre for Climate Prediction and Research Tech.Rep.CRTN74,35pp.[Available from Hadley Centre for Climate Prediction and Research,U.K.
Meteorological Of?ce,London Road,Bracknell,Berkshire RG122SY,United Kingdom.]
Reynolds,R.W.,and T.M.Smith,1994:Improved global sea surface temperature analyses using optimum interpolation.J.Climate, 7,929–948.
Ropelewski,C.F.,and P.D.Jones,1987:An extension of the Tahiti–Darwin Southern Oscillation index.Mon.Wea.Rev.,115,2161–2165.Shukla,J.,1995:Predictability of the tropical atmosphere,the tropical oceans and TOGA.Proc.Int.Sci.Conf.on the Tropical Ocean Global Atmosphere(TOGA)Programme,WCRP-91,WMO/TD 717,Vol.2,Melbourne,Australia,WCRP,725–730.
,and D.A.Paolino,1983:The Southern Oscillation and long-range forecasting of the summer monsoon rainfall over India.
Mon.Wea.Rev.,111,1830–1837.
Simmonds,I.,and P.Hope,1997:Persistence characteristics of Aus-tralian rainfall anomalies.Int.J.Climatol.,17,597–613. Soman,M.K.,and J.Slingo,1997:Sensitivity of the Asian summer monsoon to aspects of sea surface temperature anomalies in the tropical Paci?c Ocean.Quart.J.Roy.Meteor.Soc.,123,309–336.
Torrence,C.,and https://www.wendangku.net/doc/2d12968253.html,po,1998:A practical guide to wavelet analysis.Bull.Amer.Meteor.Soc.,79,61–78.
,and P.J.Webster,1998:The annual cycle of persistence in the El Nin?o–Southern Oscillation.Quart.J.Roy.Meteor.Soc.,124, 1985–2004.
Trenberth,K.E.,1976:Spatial and temporal variations of the South-ern Oscillation.Quart.J.Roy.Meteor.Soc.,102,639–653.
,1984:Signal versus noise in the Southern Oscillation.Mon.
Wea.Rev.,112,326–332.
,1997:The de?nition of El Nin?o.Bull.Amer.Meteor.Soc.,78, 2771–2777.
,and D.J.Shea,1987:On the evolution of the Southern Os-cillation.Mon.Wea.Rev.,115,3078–3096.
,and J.W.Hurrell,1994:Decadal atmosphere–ocean variations in the Paci?c.Climate Dyn.,9,303–319.
Troup,A.J.,1965:The‘‘southern oscillation.’’Quart.J.Roy.Meteor.
Soc.,91,490–506.
Wainer,I.,and P.J.,Webster,1996:Monsoon/El Nin?o–Southern Os-cillation relationships in a simple coupled ocean–atmosphere model.J.Geophys.Res.,101,25599–25614.
Walker,G.T.,and E.W.Bliss,1937:World weather VI.Mem.Roy.
Meteor.Soc.,4,119–139.
Wang,B.,1995:Interdecadal changes in El Nin?o onset in the last four decades.J.Climate,8,267–285.
,and Y.Wang,1996:Temporal structure of the Southern Os-cillation as revealed by wave-form and wavelet analysis.J.Cli-mate,9,1586–1598.
Wang,X.L.,and C.F.Ropelewski,1995:An assessment of ENSO-scale secular variability.J.Climate,8,1584–1599. Webster,P.J.,1995:The annual cycle and the predictability of the tropical coupled ocean–atmosphere system.Meteor.Atmos.
Phys.,56,33–55.
,and T.N.Palmer,1997:The past and future of El Nin?o.Nature, 390,562–564
,V.Magan?a,T.N.Palmer,J.Shukla,R.A.Tomas,M.Yanai, and T.Yasunari,1998:Monsoons:Processes,predictability and the prospects for prediction.J.Geophys.Res.,103,14451–14510.
Wright,P.B.1989:Homogenized long-period Southern Oscillation indices.Int.J.Climatol.,9,33–54.
Yasunari,T.,1990:Impact of Indian monsoon on the coupled at-mosphere/ocean system in the tropical Paci?c.Meteor.Atmos.
Phys.,44,29–41.
,and Y.Seki,1992:Role of the Asian monsoon on the inter-annual variability of the global climate system.J.Meteor.Soc.
Japan,70,177–189.
Zebiak,S.E.,and M.A.Cane,1991:Natural climate variability in
a coupled model.Greenhouse Gas-Induced Climatic Change:A
Critical Appraisal of Simulations and Observations,M. E.
Schlesinger,Ed.,Elsevier,457–469.
Zhang,Y.,J.M.Wallace,and D.S.Battisti,1997:ENSO-like in-terdecadal variability:1900–93.J.Climate,10,1004–1020.
云南马关扣哈锡多金属矿矿床地质特征
云南马关扣哈锡多金属矿矿床地质特征 锡多金属矿矿床产于马关都龙老君山花岗岩岩体中近东西向断裂破碎带中及过渡相含斑二云母花岗岩节理裂隙中。由多期多成因成矿热液、多种物源、多期次区域变质、热液作用下运移、活化、聚集和沉淀形成矿体。 标签：高温热液石英脉型锡、钨矿床内接触带断裂破碎带云南马关扣哈 滇东南地区横跨特提斯—喜马拉雅和滨太平洋两大构造域，以哀牢山为界，西部为青藏滇板块，东部为扬子板块。老君山锡多金属矿区位于扬子板块次级构造单元：屏边—西筹山弧区西南缘，即扬子板块与青藏滇板块、哀牢山地体碰撞带东侧，属华南锡钨多金属成矿带，特殊的大地构造位置，造就了区内丰富的矿产资源。 1矿区地质 1.1地层 矿区内地层发育不全，南岭锡钨成矿带西部，属华南锡钨多金属成矿带，扣哈锡多金属矿区位于老君山环形构造西北侧。区域上主要出露地层为寒武系（∈），为一套厚数千米的地槽型复理石矿源层建造，并具有不同程度的变质。已知的锡、钨、铅、锌等绝大多数均赋存于寒武系地层的一定层位中。 1.2构造 矿区紧邻马关—都龙断裂北侧，区内构造极为简单，主要为发育于花岗岩体内部的节理裂隙。受区域主干断裂构造的影响，花岗岩体内节理裂隙配套构造表现为近东西向的张性结构面。所有已知矿体均产于近东西向或近南北向的节理裂隙中。文山-麻栗坡断裂、南温河断裂、马关-都龙断裂对区内构造发展、矿产分布具有明显的控制作用（图1）。 文山-麻栗坡断裂：是本区规模最大的一条超壳断裂，具剪切压扭性质，沿北西方向延伸。该断裂自晚古生代至中生代具有继承性活动，控制了东西两个构造单元内的地层、构造和岩浆岩的分布，断裂两侧岩相及矿产均有明显差异[1]。 马关-都龙断裂：位于老君山花岗岩体西侧，总体走向北西，平面上呈一弧形，断裂带宽50～20m， 沿断裂带及附近有花岗斑岩脉和长英岩脉分布，并具热液矿化蚀变。断裂旁侧次级构造和有利含矿层中分布着一系列重要的锡多金属矿床，如铜街、曼家寨、花石头、老寨等[2]，反映该断裂为区内重要的导矿构造。 1.3岩浆岩
华南钨和锡大规模成矿作用的差异及其原因初探_华仁民
2010年2月Feb.,2010矿床地质 M IN ERAL D EPOSI T S 第29卷第1期 Vol.29No.1 文章编号:0258-7106(2010)01-0009-15 华南钨和锡大规模成矿作用的差异及其原因初探X 华仁民,李光来,张文兰,胡东泉,陈培荣,陈卫锋,王旭东 (内生金属矿床成矿机制研究国家重点实验室,南京大学地球科学与工程学院,江苏南京210093) 摘要钨和锡都是中国的优势矿产资源,主要分布在华南尤其是南岭及其邻近地区,关系密切、相互共生,但两者之间仍然存在着很明显的差异。在空间分布上表现为东钨西锡:南岭东段,钨矿密集产出;中段,钨锡并重,但锡矿化明显增强;西段则为大规模锡成矿作用。在成矿时代上,钨和锡都以160~150M a为成矿高峰期,但锡还有雪峰期、印支期成矿作用,以及燕山晚期的又一个成矿高峰期。锡的成矿作用及与其相关的花岗岩类显示出与地幔物质有较为密切的关系,因此,锡的大规模成矿作用或发生在有地幔物质参与的/十-杭带0附近,或发生在地壳强烈拉张的燕山晚期。钨与锡在元素地球化学性质上的差异,以及华南东、西两部分在构造背景、沉积物特征和岩浆活动等方面的差异,是造成华南钨与锡大规模成矿作用差异的根本原因。 关键词地质学;钨;锡;大规模成矿作用;差异;华南 中图分类号:P618.67;P618.44文献标志码:A A tentative discussion on differences between large-scale tungsten and tin mineralizations in South China H UA RenMin,LI GuangLai,ZHANG WenLan,H U DongQuan, CH EN PeiRong,CH EN WeiFeng and WANG XuDong (State K ey Labor ator y for M iner al Deposit Research,School of Earth Science and T echnolog y, N anjing U niversity,Nanjing210093,Jiangsu,China) Abstract Both tungsten and tin ar e the most prevailing mineral resources in https://www.wendangku.net/doc/2d12968253.html,rg e-scale tung sten and tin mineralizations too k place mostly in South China,especially in the N anling M ountains and the neighboring areas.T he two elements ar e v er y closely r elated to and accompanied wit h each other and always occur together in most or e deposits.Ho wever,differences do ex ist between tungsten and t in miner alizations in many aspects.Spat ial distribution shows that the eastern sector of the N anling M ountains is character ized by strong and dense W mineralizat ion,w hile Sn mineralization becomes stronger westwards.China's tw o lar gest Sn or e deposits are dis-tr ibuted in the western margin of the South China fold system.Although both tungsten and tin mineralizations have the same highest or e-forming epoch of160~150M a,the t in miner alization is spanned fr om t he late Xuefeng Period(~800M a)to L ate Yanshanian M ov ement(80M a).T he L ate Yanshanian Mov ement is in fact another highlight of tin mineralization.T he tin-bearing g ranite and t in miner alization show much closer relationship with the participatio n of mantle derived mat erials than tungsten mineralizatio n. T herefore,the larg e-scale tin mineralization took place either near the so-called Sh-i Hang Zone where long-term deep faulting w as ac-tiv ated and A-ty pe granites w er e emplaced or during t he L ate Yanshanian per iod when the South China continent was in a strong ten-sional geodynamic environment.T he pr incipal factors responsible for these differ ences may include the div ersity of geochemical proper-ties betw een the two elements and the differences in tectonic backgr ounds,str atigraphic char acteristics and magmatic activities be-tw een the eastern part and the western par t of Sout h China. X本文得到国家重点基础研究发展计划(2007CB411404)及国家自然科学基金重点项目(40730423)的资助 第一作者简介华仁民,男,1946年生,教授,主要从事矿床学研究与教学。Email:huarenmin@https://www.wendangku.net/doc/2d12968253.html, 收稿日期2009-10-12;改回日期2009-12-01。许德焕编辑。
浙江平阳龙尾钨锡铍多金属矿地质特征及成因探讨
浙江平阳龙尾钨锡铍多金属矿地质特征及成因探讨 浙江省平阳县龙尾钨锡铍多金属矿床位于浙东-闽东沿海火山喷发带内，处于山门火山洼地的北东侧边缘地带。该破火山内环状构造发育，且被花岗斑岩和辉绿岩等脉岩充填。通过综合分析地质、物化探、岩浆岩等条件，从而认为本矿床属于外接触带石英脉型钨锡铍多金属矿。 标签：龙尾钨锡铍多金属矿破火山五层楼 1前言 浙江省平阳龙尾钨锡铍多金属矿位于浙江省东南部飞云江南岸平阳县与瑞安市接壤地带，矿床主要发育于平阳山门沉积型火山洼地东部的内部边缘，跟燕山期火山活动有关。围绕该破火山边缘及中心分布有铜、铅锌多金属、钨锡、铌钽、银、金、铀、黄铁矿、明矾石及高岭土等多个矿床，具有较好的成矿背景前景。目前矿区正在进行普查地质工作，本文旨在通过综合分析矿区的地质特征，总结找矿标志，以期对矿床成因认识和找矿工作有所裨益。 2区域地质背景 矿区位于华南褶皱系浙东南褶皱带温州—临海拗陷带南段之泰顺—青田拗断带的中部，位于北东向温州—镇海大断裂与泰顺—黄岩大断裂之间，松阳—平阳大断裂从矿区北东部通过，山门沉积型火山洼地东部的内部边缘（图1），区内岩浆侵入作用频繁，火山、断裂构造极为发育，属于矾山—山门—南田明矾石、叶蜡石、萤石、伊利石、多金属成矿远景区的东部，区域成矿地质条件有利。 3矿区地质特征 矿区出露地层主要为上侏罗统一套火山碎屑岩和下白垩统河湖相沉积岩组成。其中上侏罗统主要有高坞组、西山头组、茶湾组和九里坪组火山岩系列，其主要岩性由流纹质晶屑熔结凝灰岩、流纹质晶屑玻屑熔结凝灰岩、粉砂质泥岩、沉凝灰岩等火山碎屑岩夹河湖相沉积岩组成，产状微斜内倾，厚度>2000m；下白垩统主要由馆头组、朝川组沿山门盆地中心呈北东向展布，为内陆湖相沉积的一套紫红色砾岩、砂砾岩和青灰色泥岩、粉砂岩等，厚度>2700m。高坞组、西山头组、茶湾组和九里坪组主要分布于山门复活破火山盆地边缘。 矿区内构造主要由断裂构造和火山机构组成。其中北西向平阳—松阳大断裂西起衢州之北，经松阳、平阳经过该区延入东海。该构造主要形成于燕山中晚期，白垩纪后期活动较为强烈，与成矿关系密切。自矿区北西延至南东，贯彻整个矿区，断裂宽数米，倾向北东，倾角75～80°，断面呈舒缓波状，具擦痕，倾伏角15°，倾伏方向南东，显示左行。构造带中见构造角砾岩及糜棱岩化，并具硅化、黄铁矿化及局部绿泥石化等蚀变。矿区位于双尖山破火山的北部边缘地带，本地区火山构造显示了线状区域构造与环状火山构造的复合关系。该破火山内环状构
东坡_黄沙坪钨锡多金属矿成矿规律浅析
中华民居2012年3月 前言 东坡—黄沙坪地区位于南岭多金属成矿带中段北缘，大地构造位置为炎陵—郴州—蓝山北东向深断裂与郴州——邵阳北西向断裂带的交接部位，在主要成矿带与次级褶皱断裂带的相交部位分布着东坡、芙蓉、黄沙坪等多金属矿田。区内矿产资源丰富，以钨、锡、钼、铋、铅、锌为主，除少数矿床外，其主要矿床成因类型均属岩浆热液型矿床，总体上看它们在成矿空间和时间上有着密切的联系，它们均是岩浆活动、地质构造及地层岩性综合控制的产物。 1．矿床成矿系列 经区内的基本构造格局和深部构造，岩浆岩的分布及其成因分类，地层岩性组合，矿化组合及分带特征和有成因联系的若干矿床在空间分布上的规律性等的研究，将本区内生金属矿床划分两个成矿系列，即与酸性中浅成花岗岩类有关的钨锡多金属矿成矿系列和与中酸性深源浅成花岗闪长岩类有关的铅锌多金属矿床成矿系列。 1.1 与酸性、中浅成花岗岩类有关的钨锡多金属矿成矿系列 本系列矿床主要出现于一相对隆起带，经受了加里东至燕山期的构造运动，褶皱断裂发育，岩浆活动繁且规模较大，深部推断有多个隐伏岩浆岩带，地表出露的岩基、岩株体多为壳源型（Ⅰ类）花岗岩，侵入时代为加里东至燕山期，其中以燕山早期最为发育，具有多期多阶段的特点。区域地球物理、地球化学特征为一区域性航磁正值高磁区和区域性重力低值区、以W、Sn、Pb、Zn、Nb、Ta、Be等地球化学高背景区。已知有东坡、芙蓉等钨锡多金属矿田（床）。 对成矿起主导作用的为燕山早期多期次、多阶段的花岗岩岩浆活动，地表出露的岩体是深部大岩基有关的高侵位岩体，是岩浆不断演化的产物，具有丰富的物质基础，并且随着岩浆的分异演化使成矿金属元素组分得以聚集，从岩浆中分异出的挥发组分和含矿溶液具有较高的压力梯度，并向岩浆侵入前缘汇集，在构造断裂的诱导下向所开拓的空间运移，于造当的有利空间和岩性条件下，含矿气水热液经过充填、交代、改造又叠加等矿方式，在内、外接触带及其附近地区形成成因上具有同源演化的一系列不同矿种和类型的矿床，构成一个与酸性中浅成花岗岩类有关的岩浆演化成矿系列。 1.2 与中酸性、深源浅成花岗闪长岩类有关的铜铅锌多金属矿成矿 系列 本系列矿床主要出现于本区西部，郴州一桂阳断裂的北西侧，为一相对拗陷带，岩浆活动规模较小，地表出露多属同熔型（Ⅱ类）的花岗闪长斑岩、花岗斑岩等小侵入体，时代为燕山早、晚期。地球物理、地球化学特征为一区域性重力高和负值平稳磁场区，区域化探异常以Pb、Zn、Ag、As异常为主，已知有宝山、大坊等铜铅锌金银多金属矿床。 矿床的形成主要为与基底断裂有成因联系的同熔型花岗岩浆为成矿提供了丰富的物质来源，盖层的紧闭线型褶皱及其伴生产出的逆冲断裂为成矿热液的运移提供了通道、为矿化富集提供了有利的储矿空间，在有利的岩性组合和构造部份，含矿气水热液经过交代、充填等成矿作用，形成不同的矿化组合和矿床类型，并且有明显的分带性。 本区一些沉积改造岩浆热液叠加型矿床，虽具有叠加成矿系列的特点，但主要为岩浆活动使金属元素迁移富集而成，如玛瑙山铁锰多金属矿等，故未单独列为一个系列。 2．找矿标志 2.1 地质标志 2.1.1 岩浆岩标志 ①燕山期复式岩体是形成区内多金属矿的首要条件，燕山早期晚阶段和燕山晚期早阶段的花岗岩与钨锡矿形成有着不可分割的联系，而铅锌矿的形成与燕山晚期的花岗岩关系密切。当大岩体中有小岩体分布时对成矿更有利。 ②北东向展布的花岗斑岩与铅锌矿在空间上紧密伴生。 ③岩体侵位较高的突起部位及其附近是寻找蚀变岩体型（云英岩型）及裂隙充填型锡铅锌矿的重要部位。 2.1.2 构造标志 ①南北向断裂是本区主要的导岩、导矿构造，钨锡铅锌多金属矿矿床往往发育在南北向与北东向两组断裂构造交汇的部位及其附近次级构造中。 ②北北东向和北北西向断裂构造，往往是主要容矿构造，是寻找含锡石英脉、云英岩脉以及锡石硫化物及铅锌矿的有利部位。 ③在花岗岩体与中、上泥盆统及二叠系栖霞组碳酸盐岩的接触带，捕虏体及岩体顶面凹陷，侧洼或缓倾斜面，是厚大矽卡岩型矿体的产出部位。 ④在岩体附近的中、上泥盆统碳酸盐岩中，节理、裂隙、层理、劈理等小构造特别发育的地段，有利于热液的渗滤交代，有利于成矿。 2.1.3 地层标志 ①与岩体接触的中、上泥盆统及石炭系中上统及二叠系下统栖霞组的碳酸盐岩是寻找矽卡岩型和网脉大理岩型锡多金属矿的重要部位。 ②震旦系浅变质砂岩、板岩是裂隙充填型锡石—硫化物矿床、含钨锡石英脉矿床及裂隙充填型的铅锌矿的主要控矿层位。 2.2 物化探异常标志 1、物化探综合异常区往往是重要的矿化集中区。 2、重力负异常是寻找矽卡岩矿床的间接找矿标志；磁异常的存在与否是判别矽卡岩型多金属矿的典型标志；各种明显的电法异常是寻找硫化物多金属矿的重要标志。 3、W、Sn、Mo、Bi、Pb、Zn化探异常与磁法、电法异常重叠时出现的综合异常，找矿效果好。 2.3 矿化蚀变标志 2.1.1 矿化标志 ①区内矿化水平及垂直分带较明显，由岩体向外依次为W、Sn、Mo、Bi —Sn、Cu、Pb、Zn —Pb、Zn — Hg、Sb矿化，可以借以某一矿物组合进行预测另一种矿化的大致地段。 ②地表（或者浅部）含锡硫化物脉可直接预测深部蚀变岩体型及矽卡岩型锡矿床。 ③地表网脉大理岩很可能预示着深部矽卡岩及云英岩的存在。 ④在花岗岩与碳酸盐岩接触带附近的褐红色、黑色疏松土状物，或有铁帽存是寻找矽卡岩型锡矿的标志。 2.1.2 蚀变标志 ①矽卡岩化是直接找矿标志。当矽卡岩中有云英岩脉穿插，则下面有可能找到云英岩型矿体。 ②云英岩是找云英岩型锡矿的直接标志，蚀变越强，则矿化越强。 ③钠长石化、萤石化是寻找蚀变岩体型锡矿的重要标志。 ④硅化、绿泥石化及绢云母化组合是寻找充填型铅锌银矿的间接标志。 3．矿床分布规律 3.1 矿床的时间分布规律 东坡—黄沙坪地区的绝大多数与有色金属成矿有关的岩体多属燕山早期第二、第三阶段侵入定位的。区内成矿是多期多阶段的，经历了三期五阶段，即岩浆期、矽卡岩期和石英硫化物期，成矿阶段有早矽卡岩阶段、晚矽卡岩阶段、氧化物阶段、早硫化物阶段和晚硫化物阶段。 总的看来，与钨锡成矿有关的岩体，形成时代相对较早；而与铅锌成矿有关的岩体，形成时代相对较晚。通过一批典型的成矿岩体研究发现，不少同期多阶段多期次形成的复式岩体，如千里山、骑田岭等岩体，在其周围的某一部分自岩体向外，可以出现有钨锡铅锌等矿化分带现象（如柿竹园—野鸡尾矿区）。 一般地说，燕山早期第一次侵入活动在本区形成了较大范围的W、Sn、Mo、Bi、Pb、Zn矿化，但较弱；第二次侵入活动进一步扩大了前次矿化范围和强度，出现较强的W、Sn、Mo、Bi、Pb、Zn矿化；第三次侵入活动挥发分较前两次更加充分，进一步强化了锡的矿化，并出现大量的铍矿化。故本区锡铅锌矿化与燕山期第二次、第三次岩浆侵入活动有关，主要成矿期为燕山早期中、晚阶段。另外，有时也可出现不同侵入阶段产生的矿化，在同一空间相互叠加而造成分带不明显的现象。 东坡-黄沙坪钨锡多金属矿成矿规律浅析 陈 杰 许东明 张怡军 （湖南省湘南地质勘察院） 摘 要：通过分析东坡—黄沙坪地区各类钨锡铅锌多金属矿床的对比分析，总结出在大地构造背景下的特殊成矿物质来源和特殊的成矿作用，探索燕山期花岗岩型钨锡多金属矿床的形成机理，并建立全区的成矿模式。 关键词：钨锡矿；分布规律；成矿模式 ?224?