Radio Constraints on the Identifications and Redshifts of Submm Galaxies

a r X i v :a s t r o -p h /9907083v 3 20 A u g 1999

R ECEIVED :J ULY 01,1999;A CCEPTED A UGUST 20,1999

Preprint typeset using L A T E X style emulateapj

RADIO CONSTRAINTS ON THE IDENTIFICATIONS AND REDSHIFTS OF SUBMM GALAXIES

I AN S MAIL ,1,2R.J.I VISON ,3,4F.N.O WEN ,5A.W.B LAIN 6&J.-P.K NEIB 7

1)Department of Physics,University of Durham,South Road,Durham DH13LE,UK

3)Department of Physics &Astronomy,University College London,Gower Street,London WC1E 6BT,UK

5)NRAO,P.O.Box 0,1003Lopezville Road,Socorro,NM 878016)Cavendish Laboratory,Madingley Road,Cambridge CB3OHE,UK

7)Observatoire Midi-Pyrénées,CNRS-UMR5572,14Avenue E.Belin,31400Toulouse,France

Received:July 01,1999;Accepted August 20,1999

ABSTRACT

We present radio maps from the Very Large Array (VLA)for 16sources detected in a sub-millimeter (submm)survey of the distant Universe.Our deep VLA 1.4-GHz maps allow us to identify radio counterparts or place

stringent limits (<～

20μJy in the source plane)on the radio ?ux of the submm sources.We compare the spectral indices of our sources between 850μm and 1.4GHz to empirical and theoretical models for distant starburst galaxies and active galactic nuclei (AGN)as a function of redshift.In this way we can derive redshift limits for the submm sources,even in the absence of an optical or near-infrared counterpart.We conclude that the submm population brighter than ～1mJy has a median redshift of at least ～2,more probably ～2.5–3,with almost all galaxies at z ?1.This estimate is a strong lower limit as both misidenti?cation of the radio counterparts and non-thermal emission from an AGN will bias our redshift estimates to lower values.The high median redshift means that the submm population,if predominately powered by starbursts,contributes a substantial fraction of the total star formation density at high redshifts.A comparison of the spectral index limits with spectroscopic redshifts for proposed optical counterparts to individual submm galaxies suggests that half of the submm sources

remain unidenti?ed and thus their counterparts must be fainter than I >～

24.Subject headings:cosmology:observations —galaxies:evolution —galaxies:formation —infrared:galaxies

—radio:galaxies

1.INTRODUCTION

Faint submm sources are likely to be highly obscured star-burst galaxies and AGN,within which optical/UV radiation from massive stars or an active nucleus is absorbed by dust and reradiated in the far-infrared.The dust emission peaks at λ～100μm and thus long-wavelength observations of dis-tant dusty galaxies can bene?t as this peak is redshifted into their window.At 850μm,this increase balances the geomet-rical dimming at higher redshifts,resulting in a constant ?ux density out to z ～5–10and the opportunity to select very high redshift galaxies.

A number of deep submm surveys have been published (Smail,Ivison &Blain 1997;Barger et al.1998,1999b;Hughes et al.1998;Blain et al.1999a;Eales et al.1999)providing counts which are in good agreement at 850-μm ?ux densities above 2mJy (the confusion limit of the blank-?eld surveys).The surface density of submm galaxies reaches ～2per sq.ar-cmin by 1mJy (Blain et al.1999a).If these galaxies lie at z >～

1,then they have bolometric luminosities of >～

1012L ⊙and they are the distant analogs of the local ultraluminous infrared galaxy (ULIRG)population.However,the observed surface density of submm galaxies is several orders of magnitude greater than that expected from the local ULIRG population (Smail et al.1997)indicating very substantial evolution of these systems in the dis-tant Universe.The integrated emission from this population can account for the bulk of the extragalactic background detected at 850μm by COBE (e.g.Fixsen et al.1998),and hence con?rms these galaxies as an important source of radiation in the Uni-verse (Blain et al.1999a,1999b).

To identify the era of obscured emission in the Universe,whether from AGN or starbursts,we have to measure the red-shifts of a complete sample of submm galaxies.Several groups have attempted this (Hughes et al.1998;Barger et al.1999a;Lilly et al.1999).Hughes et al.(1998)concluded that the bulk of the population is at z ～2–4,based on photometric redshift limits for the probable counterparts of ?ve submm sources in the Hubble Deep Field (HDF ,c.f.Richards 1999and Downes et al.1999).Barger et al.(1999a)undertook a spectroscopic survey of the same submm sample analysed here and concluded that the median redshift was ～1.5–2,with the bulk of the population having z ～1–3.Finally,Lilly et al.(1999)used archival spectroscopy and broad-band photometry of submm sources from the Eales et al.(1999)survey to claim that the population spans z =0.1–3,with a third at z <1.The differ-ences between these studies are signi?cant and important for our understanding of the nature of submm galaxies.

It is very dif?cult to achieve high completeness in optical spectroscopic surveys of submm galaxies (e.g.Barger et al.1999a)due to the very different behaviour of the K correc-tions for distant galaxies between submm and optical pass-bands.However,even a crude estimate of the median redshift of a complete sample of submm galaxies would provide a pow-erful insight into the relative dominance of obscured and unob-scured emission at different epochs (Blain et al.1999b).

In a recent paper,Carilli &Yun (1999,CY)demonstrated that using the spectral index between the submm (850μm)and radio (1.4GHz)wavebands,α8501.4=0.42log 10(S 850/S 1.4),it was possible to obtain crude redshift limits for distant dusty galax-

2Royal Society University Research Fellow.4PPARC

Advanced Fellow.

1

2Radio Observations of Submm

Galaxies

Fig.1.15×15arcsec VLA maps of the16submm sources for which we have radio observations.These are ordered as in Table1and the maps span a range in sensitivity and resolutions (§2.1).The panels are centered on the nominal submm position in each case and we show the typical error boxes for the sources.We mark the radio counterpart by a+where they are

identi?ed.The panels have north top and east left and correspond to>～100h?1

50kpc at z>1.The contour levels are at apparent1.4-GHz?ux densities of3,5,10,20,50,100×the map noise

listed in§2.1(except for SMM J21536+1741and SMM J14010+0252where they are100×these values). ies,irrespective of the nature of the emission mechanism,AGN

or starburst.CY employed a number of theoretical and em-

pirical spectral energy distributions(SEDs)to investigate the

range inα850

1.4for different assumed SEDs and showed that these

models adequately described the small sample of high-redshift galaxies for which both radio and submm observations were available.As pointed out by Blain(1999,B99),if we adopt lower dust temperatures for the submm population than are seen in the local sources used in CY’s models,then the allowed range

of redshifts is slightly lower for a given value ofα850

1.4.Never-

theless,the modest scatter between the models in CY suggests that this technique can provide useful limits on the redshifts of submm galaxies in the absence of an optical counterpart.

In this paper we apply the CY analysis to deep radio obser-vations of a complete sample of submm galaxies selected from the SCUBA Cluster Lens Survey(Smail et al.1998).Our aim is to constrain the redshift distribution of this population and in the process test the optical identi?cations and spectroscopic redshifts from Smail et al.(1998)and Barger et al.(1999a).We present the observations and their analysis in§2,discuss our results in§3and give our main conclusions in§4.

2.OBSERV ATIONS,REDUCTION AND ANALYSIS

The850-μm maps on which our survey is based were ob-tained using the long-wavelength array of the Sub-millimeter Common-User Bolometer Array(SCUBA,Holland et al.1999) on the James Clerk Maxwell Telescope(JCMT)8.The details of the observations,their reduction and analysis are given in Smail et al.(1997,1998)and Ivison et al.(1998).Each?eld covers an area of5.2arcmin2with a typical1σsensitivity of～1.7mJy, giving a total survey area of0.01deg2.The median ampli?ca-tion by the cluster lenses for background sources detected in

our?elds is expected to be2.5+5.0

?1.5(Blain et al.1999a;Barger

et al.1999a),and so we have effectively surveyed an area of about15arcmin2in the source plane to an equivalent1σsensi-tivity of0.7mJy.The follow-up of these submm sources also bene?ts from the achromatic ampli?cation,which boosts the apparent brightness of counterparts in all other wavebands.

2.1.Radio Observations

All the radio maps used in this work were obtained with the VLA9at1.4GHz in A or B con?guration,giving effec-tive resolutions of1.5′′and5′′,respectively.More details of the reduction and analysis of these maps are given in the fol-lowing references(we list the VLA con?guration and1σmap noise for each cluster):Morrison et al.(1999)for Cl0024+16 (B/15μJy),A370(B/10μJy)and Cl0939+47(B/9μJy);Ivison et al.(1999)for A1835(B/16μJy)and Ivison et al.(2000)for MS0440+02(A/15μJy)and Cl2244?02(A/17μJy).No deep radio map is available for A2390,although shallower observa-tions were used to study the submm/radio spectral index of the central cluster galaxy(Edge et al.1999).

Radio counterparts were searched for around the nominal po-sitions of the submm sources based on the SCUBA astrome-try and2σerror-boxes of6′′or8′′diameter depending upon whether the submm source was respectively a4σor3σde-tection(Fig.1).The size of these error-boxes includes both the systematic and random errors in the source positions and they have been con?rmed as realistic using two sources with CO interferometric observations(Frayer et al.1998,1999). To assign radio?uxes to the individual submm sources we have adopted the conservative approach of identifying the ra-dio counterpart as the brightest radio source within the submm error-box(Fig.1).The behaviour of the radio-submm spec-tral index is such that a brighter radio counterpart leads to a lower redshift estimate.Thus by selecting the brightest avail-able radio source,we should obtain a lower bound on the red-shift of the submm source.Apparent radio?uxes or3σlimits are listed in Table1.Note that a faint radio source lies just outside the error-box of SMM J00265+1710,and so we adopt the radio?ux of that source as a lower limit in this case.For SMM J02399?0134,the radio counterpart is confused due to emission from a nearby bright cluster elliptical(Fig.1),and so we only quote an approximate1.4-GHz?ux for this galaxy.

2.2.Radio-Submm Spectral Indices

The850-μm and1.4-GHz?uxes or limits for16of the sources in Smail et al.(1998),for which we have radio obser-vations,are listed in Table1in order of their apparent submm ?uxes,along with their proposed spectroscopic redshifts,z spec, from Barger et al.(1999a).Where a radio counterpart is iden-ti?ed the spectroscopic redshift of the closest optical candidate is listed in the table.The errors onα850

1.4

are calculated assum-ing the1σ?ux uncertainties in each band(§2.1and Smail et al. 1998).For non-detections at1.4GHz we use the3σ?ux limit of the relevant radio map.The two central cluster galaxies in our sample are not included in our analysis.

Using theα850

1.4

values or limits,the redshift ranges,zα,are

8The JCMT is operated by the Joint Astronomy Centre on behalf of the United Kingdom Particle Physics and Astronomy Research Council,the Netherlands Organisation for Scienti?c Research,and the National Research Council of Canada.

9The VLA is run by NRAO and is operated by Associated Universities Inc.,under a cooperative agreement with the National Science Foundation.

Smail et al.3

T ABLE1

Spectral Properties of the Submm Sample

Submm S850S1.4α850

1.4zαz d spec Comments

Source(mJy)(μJy)Reliable Uncertain

R EFERENCES.—a)Ivison et al.(1998)–b)Smail et al.(1999)–c)Ivison et al.(1999)–d)Barger et al.(1999a)–e)Soucail et al.(1999)–f)Smail et al.(1998)—g) Edge et al.(1999).

derived from the extremes of the predictions from the four CY

models(two empirical SEDs representing Arp220and M82,

and two models with dust temperatures of T d～50–60K and

emissivities ofβ=1.0orβ=1.5)and a further model from

B99,withβ=1.5and a T d=30K to illustrate the minimum

possible redshift assuming a very low T d.For a model SED

at z>～1,a variation ofδβ=+0.2orδT d=?10K results in an

change ofδα8501.4<～+0.1,equivalent to an uncertainty in the de-

rived redshift ofδz/z～0.2.

There are three caveats to bear in mind when usingα850

1.4to

estimate redshifts for distant galaxies.First,most of the dis-tant galaxies which CY used to compare with their model pre-dictions show some signs of AGN activity.If these AGN also contribute to the1.4-GHz non-thermal emission of the galaxy

they will lower the observedα850

1.4values(as some obviously do

in Fig.1a of CY).This will mean that any radio-quiet submm sources could lie at the high end of the predictedα8501.4range at each epoch.Secondly,the CY and B99models we use as-sume effective dust temperatures for the galaxies,T d≥30K.If the dust in distant obscured galaxies is much cooler then this,it will again shift the predicted redshifts systematically lower(see B99).Finally,as mentioned in CY,the effects of inverse Comp-ton scattering of radio photons off the microwave background may reduce the radio luminosities of star-forming galaxies at

z>～3and hence increaseα850

1.4for the most distant galaxies.

Nevertheless,the relatively good agreement shown in CY of the spectral indices of distant galaxies with the models is an

important con?rmation thatα850

1.4indices can be used to derive

robust lower limits to the redshifts of submm sources without reliable spectroscopic identi?cations.

3.RESULTS AND DISCUSSION

We show in Fig.2three cumulative redshift distributions for the population representing extreme interpretations of the zαlimits from the spectral index models.We see that even mak-ing the most conservative assumptions about the likely redshifts from theα850

1.4

indices we still predict a median redshift for the submm population above an intrinsic850-μm?ux of1mJy of ～2,and more likely closer to z～2.5–3.

Comparing the cumulative redshift distribution to that de-rived from the(incomplete)spectroscopic study of this sample (Barger et al.1999a)we see broad similarities.However,com-parison of redshifts for individual sources from the two stud-ies(Table1),while showing good agreement for those submm sources with reliable identi?cations(e.g.Ivison et al.1998, 1999)also indicates that the majority of the uncertain spec-troscopic IDs are likely to be incorrect.Barger et al.(1999a) obtained spectroscopy of most of the possible optical counter-parts within each submm error-box.We can therefore state that the true submm sources must be fainter than the faintest spec-troscopic target.Including the two optical blank-?eld sources already known(Smail et al.1998),we conclude that approx-imately half of the submm population are therefore currently unidenti?ed.These submm sources have no radio counterparts and are too faint for optical spectroscopy,I>～24,their identi?-cation will thus be very dif?cult.

Our median redshift is compatible with the results of Hughes et al.(1998)and CY for the?ve submm sources in the HDF based on analyses of their SEDs and radio-submm indices.The only other submm survey for which spectroscopic redshift in-formation has been published is by Lilly et al.(1999)for the Eales et al.(1999)sample—they?nd～2—similar to our median redshift.However,Lilly et al.(1999)claim that a third of the submm population lies at z<1;in contrast,we?nd no galaxies in our?eld sample at z<1.This apparent contrad-ication may result simply from the small sizes of the current samples or might indicate that foreground bright optical galax-ies are lensing the distant submm sources detected in the?eld surveys(see Blain,M?ller&Maller1999;Hughes et al.1998). The detection rate of radio counterparts to the submm sources is higher for the intrinsically brighter sources.All the submm sources with observed?uxes above～10mJy(intrinsic

4Radio Observations of Submm Galaxies ?uxes of>～4mJy)have radio counterparts,while the majority

of the fainter sources do not(this is consistent with CY’s re-

sults in the HDF).The detections and astrometry of the fainter

sources are suf?ciently reliable that this result is not due to spu-

rious detections(see Ivison et al.1999).Making the conser-

vative assumption of placing all of the non-detections at their

lower bounds onα8501.4,we?nd that this distribution is consis-

tent with being drawn from theα850

1.4distribution of the brighter

sources with a probability of P～0.4.However,simply com-

paring theα850

1.4indices for the two subsamples we see that half

of the radio-detected bright submm sources haveα850

1.4values

lower than the lowest limit on the undetected sources,the like-lihood that this occurs by chance is only P～3×10?4suggest-

ing that there may be real differences between theα850

1.4val-

ues for the two subsamples.Several factors could cause this, most simply the apparently fainter submm sources may be at higher redshifts(a correlation which could naturally exist in a low density Universe).Alternatively,intrinsic differences in the spectral indices of fainter sources would occur if they contain a lower fraction of radio-loud AGN or have typically cooler dust temperatures(or higher emissivities).Both effects are plausi-ble given what we know about the correlations of AGN frac-tions,dust temperature and emissivity with luminosity in local ULIRG samples(Sanders&Mirabel1996).Further detailed observations of both distant and local ULIRGs are needed to distinguish between these

possibilities.

Fig.2.The cumulative redshift distribution for the full submm sample.We have used the spectroscopic redshifts of those sources thought to be reliable(Table1)and combined

these with the probable redshift ranges of the remaining sources derived from theirα850

1.4 indices or limits.The solid line shows the cumulative distribution if we assume the mini-mum redshift distribution which is obtained if all sources are assumed to lie at their lower zαlimit given in Table1(the dashed line is the equivalent analysis but restricted to just the CY models).The effect of non-thermal radio emission,which drives down theα850

1.4

in-dices,means that this is a very conservative assumption if some fraction of the population harbor radio-loud AGN.The dot-dashed line assumes a?at probability distribution for the sources within their zαranges and a maximum redshift of z=6for those sources where

we only have a lower limit onα850

1.4

.Finally,the dotted line is the cumulative redshift dis-tribution from Barger et al.(1999a)with two of the source identi?cations corrected as in Smail et al.(1999)and all blank-?eld/ERO candidates placed at z=4.

The relatively high median redshift we?nd for the submm population,～2–3,indicates that their equivalent star for-mation density at these epochs is around0.5M⊙yr?1Mpc?3 (Blain et al.1999a),roughly three times that seen in UV-selected samples(Steidel et al.1999).Emission from dust heated by obscured AGN will reduce this estimate,but it is dif?-cult not to conclude that the submm galaxies contain a substan-tial fraction of the star formation in the high redshift Universe.

4.CONCLUSIONS

?We present radio maps of16galaxies selected in a deep submm survey.We combine submm and radio?uxes(or limits) to determine the radio-submm spectral indices of these galaxies and interpret these using model predictions to derive the red-shifts for a complete sample of faint submm galaxies.

?We?nd a median redshift～2for the submm popu-lation down to S850～1mJy under conservative assumptions, and～2.5–3for more reasonable assumptions.Median redshifts below?2are only possible if the bulk of the emission is coming from dust at T d?30K(compared to the 40–50K typically seen in well-studied,distant submm galax-ies,or their low-redshift analogs:ULIRGs).As a result we ?nd no evidence for a signi?cant low-redshift,z<1,tail in our distribution in contrast to Lilly et al.(1999).

?We compare the individual redshifts estimated fromα850

1.4

with the spectroscopic observations of proposed optical counterparts of the submm sources.We?nd that the majority of the‘uncer-tain’spectroscopic identi?cations from Barger et al.(1999a)are likely to be incorrect.We conclude that the true counterparts lie at higher redshifts and are intrinsically very faint,I>～24,mak-ing the prospects for a complete optical spectroscopic survey of the submm population bleak.

ACKNOWLEDGEMENTS

We thank Amy Barger,Chris Carilli,Len Cowie,Glenn Mor-rison,Jason Stevens and Min Yun for useful conversations and help.

REFERENCES

Barger,A.J.,et al.,1998,Nature,394,248.

Barger,A.J.,Cowie,L.L.,Smail,I.,Ivison,R.J.,Blain,A.W.,Kneib,J.-P., 1999a,AJ,117,2656.

Barger,A.J.,Cowie,L.L.,Sanders,D.B.,1999b,ApJL,518,L5.

Blain,A.W.,Kneib,J.-P.,Ivison,R.J.,Smail,I.,1999a,ApJL,512,L87. Blain,A.W.,Smail,I.,Ivison,R.J.,Kneib,J.-P.,1999b,MNRAS,302,632. Blain,A.W.,M?ller,O.,Maller,A.H.,1999,MNRAS,303,423.

Blain,A.W.,1999,MNRAS,in press.(B99)

Carilli,C.L.,Yun,M.S.,1999,ApJL,513,L13.

Downes,D.,et al.,1999,preprint.

Eales,S.,et al.,1999,ApJ,515,518

Edge,A.C.,Ivison,R.J.,Smail,I.,Blain,A.W.,Kneib,J.-P.,1999,MNRAS, 306,599.

Fixsen D.J.,Dwek E.,Mather J.C.,Bennett C.L.,Shafer R.A.,1998,ApJ, 508,123.

Frayer,D.T.,Ivison,R.J.,Scoville,N.Z.,Yun,M.,Evans,A.S.,Smail,I., Blain,A.W.,Kneib,J.-P.,1998,ApJ,506,L7.

Frayer,D.T.,Ivison,R.J.,Scoville,N.Z.,Yun,M.,Evans,A.S.,Smail,I., Barger,A.J.,Blain,A.W.,Kneib,J.-P.,1999,ApJL,514,L13.Holland,W.S.,et al.,1999,MNRAS,303,659.

Hughes,D.,et al.,1998,Nature,394,241.

Ivison,R.J.,Smail,I.,Le Borgne,J.-F.,Blain,A.W.,Kneib,J.-P.,et al.,1998, MNRAS,298,583.

Ivison,R.J.,Smail,I.,Barger,A.J.,Kneib,J.-P.,Blain,A.W.,Owen,F.N., Kerr,T.H.,Cowie,L.L.,1999,MNRAS,submitted.

Ivison,R.J.,et al.,2000,in prep.

Lilly,S.J.,et al.,1999,ApJ,518,641.

Morrison,G.,et al.,1999,in prep.

Richards,E.A.,1999,ApJL,513,L9.

Sanders,D.B.,Mirabel,I.F.,1996,ARAA,34,749.

Smail,I.,Ivison,R.J.,Blain,A.W.,1997,ApJL,490,L5.

Smail,I.,Ivison,R.J.,Blain,A.W.,Kneib,J.-P.1998,ApJ,507,21L. Smail,I.,Ivison,R.J.,Kneib,J.-P.,Cowie,L.L.,Blain,A.W.,Barger,A.J., Owen,F.N.,Morrison,G.,1999,MNRAS,in press.

Soucail,G.,Kneib,J.-P.,B′z ecourt,J.,Metcalfe,L.,Altieri,B.,Le Borgne,J.-F., et al.,1999,A&A,343,L70.

Steidel,C.C.,Adelberger,K.L.,Giavalisco,M.,Dickinson,M.,Pettini,M., 1999,ApJ,519,1.

黄自艺术歌曲钢琴伴奏及艺术成就

【摘要】黄自先生是我国杰出的音乐家，他以艺术歌曲的创作最为代表。而黄自先生特别强调了钢琴伴奏对于艺术歌曲组成的重要性。本文是以黄自先生创作的具有爱国主义和人道主义的艺术歌曲《天伦歌》为研究对象，通过对作品分析，归纳钢琴伴奏的弹奏方法与特点，并总结黄自先生的艺术成就与贡献。 【关键词】艺术歌曲；和声；伴奏织体；弹奏技巧 一、黄自艺术歌曲《天伦歌》的分析 （一）《天伦歌》的人文及创作背景。黄自的艺术歌曲《天伦歌》是一首具有教育意义和人道主义精神的作品。同时，它也具有民族性的特点。这首作品是根据联华公司的影片《天伦》而创作的主题曲，也是我国近代音乐史上第一首为电影谱写的艺术歌曲。作品创作于我国政治动荡、经济不稳定的30年代，这个时期，这种文化思潮冲击着我国各个领域，连音乐艺术领域也未幸免――以《毛毛雨》为代表的黄色歌曲流传广泛，对人民大众，尤其是青少年的不良影响极其深刻，黄自为此担忧，创作了大量艺术修养和文化水平较高的艺术歌曲。《天伦歌》就是在这样的历史背景下创作的，作品以孤儿失去亲人的苦痛为起点，发展到人民的发愤图强，最后升华到博爱、奋起的民族志向，对青少年的爱国主义教育有着重要的影响。 （二）《天伦歌》曲式与和声。《天伦歌》是并列三部曲式，为a+b+c，最后扩充并达到全曲的高潮。作品中引子和coda所使用的音乐材料相同，前后呼应，合头合尾。这首艺术歌曲结构规整，乐句进行的较为清晰，所使用的节拍韵律符合歌词的特点，如三连音紧密连接，为突出歌词中号召的力量等。 和声上，充分体现了中西方作曲技法融合的创作特性。使用了很多七和弦。其中，一部分是西方的和声，一部分是将我国传统的五声调式中的五个音纵向的结合，构成五声性和弦。与前两首作品相比，《天伦歌》的民族性因素增强，这也与它本身的歌词内容和要弘扬的爱国主义精神相对应。 （三）《天伦歌》的伴奏织体分析。《天伦歌》的前奏使用了a段进唱的旋律发展而来的，具有五声调性特点，增添了民族性的色彩。在作品的第10小节转调入近关系调，调性的转换使歌曲增添抒情的情绪。这时的伴奏加强和弦力度，采用切分节奏，节拍重音突出，与a段形成强弱的明显对比，突出悲壮情绪。 c段的伴奏采用进行曲的风格，右手以和弦为主，表现铿锵有力的进行。右手为上行进行，把全曲推向最高潮。左手仍以柱式和弦为主，保持节奏稳定。在作品的扩展乐段，左手的节拍低音上行与右手的八度和弦与音程对应，推动音乐朝向宏伟、壮丽的方向进行。coda 处，与引子材料相同，首尾呼应。 二、《天伦歌》实践研究 《天伦歌》是具有很强民族性因素的作品。所谓民族性，体现在所使用的五声性和声、传统歌词韵律以及歌曲段落发展等方面上。 作品的整个发展过程可以用伤感――悲壮――兴奋――宏达四个过程来表述。在钢琴伴奏弹奏的时候，要以演唱者的歌唱状态为中心，选择合适的伴奏音量、音色和音质来配合，做到对演唱者的演唱同步，并起到连接、补充、修饰等辅助作用。 作品分为三段，即a+b+c+扩充段落。第一段以五声音阶的进行为主，表现儿童失去父母的悲伤和痛苦，前奏进入时要弹奏的使用稍凄楚的音色，左手低音重复进行，在弹奏完第一个低音后，要迅速的找到下一个跨音区的音符；右手弹奏的要有棱角，在前奏结束的时候第四小节的t方向的延音处，要给演唱者留有准备。演唱者进入后，左手整体的踏板使用的要连贯。随着作品发展，伴奏与旋律声部出现轮唱的形式，要弹奏的流动性强，稍突出一些。后以mf力度出现的具有转调性质的琶音奏法，要弹奏的如流水般连贯。在重复段落，即“小

我国艺术歌曲钢琴伴奏-精

我国艺术歌曲钢琴伴奏-精 2020-12-12 【关键字】传统、作风、整体、现代、快速、统一、发展、建立、了解、研究、特点、突出、关键、内涵、情绪、力量、地位、需要、氛围、重点、需求、特色、作用、结构、关系、增强、塑造、借鉴、把握、形成、丰富、满足、帮助、发挥、提高、内心 【摘要】艺术歌曲中，伴奏、旋律、诗歌三者是不可分割的重 要因素，它们三个共同构成一个统一体，伴奏声部与声乐演唱处于 同样的重要地位。形成了人声与器乐的巧妙的结合，即钢琴和歌唱 的二重奏。钢琴部分的音乐使歌曲紧密的联系起来，组成形象变化 丰富而且不中断的套曲，把音乐表达的淋漓尽致。 【关键词】艺术歌曲；钢琴伴奏；中国艺术歌曲 艺术歌曲中，钢琴伴奏不是简单、辅助的衬托，而是根据音乐 作品的内容为表现音乐形象的需要来进行创作的重要部分。准确了 解钢琴伴奏与艺术歌曲之间的关系,深层次地了解其钢琴伴奏的风 格特点,能帮助我们更为准确地把握钢琴伴奏在艺术歌曲中的作用 和地位，从而在演奏实践中为歌曲的演唱起到更好的烘托作用。 一、中国艺术歌曲与钢琴伴奏 “中西结合”是中国艺术歌曲中钢琴伴奏的主要特征之一，作 曲家们将西洋作曲技法同中国的传统文化相结合，从开始的借鉴古 典乐派和浪漫主义时期的创作风格，到尝试接近民族乐派及印象主 义乐派的风格，在融入中国风格的钢琴伴奏写作，都是对中国艺术 歌曲中钢琴写作技法的进一步尝试和提高。也为后来的艺术歌曲写 作提供了更多宝贵的经验，在长期发展中，我国艺术歌曲的钢琴伴 奏也逐渐呈现出多姿多彩的音乐风格和特色。中国艺术歌曲的钢琴

写作中，不可忽略的是钢琴伴奏织体的作用，因此作曲家们通常都以丰富的伴奏织体来烘托歌曲的意境，铺垫音乐背景，增强音乐感染力。和声织体，复调织体都在许多作品中使用，较为常见的是综合织体。这些不同的伴奏织体的歌曲，极大限度的发挥了钢琴的艺术表现力，起到了渲染歌曲氛围，揭示内心情感，塑造歌曲背景的重要作用。钢琴伴奏成为整体乐思不可缺少的部分。优秀的钢琴伴奏织体，对发掘歌曲内涵，表现音乐形象，构架诗词与音乐之间的桥梁等方面具有很大的意义。在不断发展和探索中，也将许多伴奏织体使用得非常娴熟精确。 二、青主艺术歌曲《我住长江头》中钢琴伴奏的特点 《我住长江头》原词模仿民歌风格，抒写一个女子怀念其爱人的深情。青主以清新悠远的音乐体现了原词的意境，而又别有寄寓。歌调悠长，但有别于民间的山歌小曲；句尾经常出现下行或向上的拖腔，听起来更接近于吟哦古诗的意味，却又比吟诗更具激情。钢琴伴奏以江水般流动的音型贯穿全曲，衬托着气息宽广的歌唱，象征着绵绵不断的情思。由于运用了自然调式的旋律与和声，显得自由舒畅，富于浪漫气息，并具有民族风味。最有新意的是，歌曲突破了“卜算子”词牌双调上、下两阕一般应取平行反复结构的惯例，而把下阕单独反复了三次，并且一次比一次激动，最后在全曲的高音区以ff结束。这样的处理突出了思念之情的真切和执著，并具有单纯的情歌所没有的昂奋力量。这是因为作者当年是大革命的参加者，正被反动派通缉，才不得不以破格的音乐处理，假借古代的