JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 91, NO. B6, PAGES 6263-6276, MAY 10, 1986 Cenozoic Thermal, Mechanical and Tectonic Evolution of the Rio Grande Rift PAUL MORGAN1 Departmentof Geosciences,Purdue University,West Lafayette, Indiana WILLIAM R. SEAGER Departmentof Earth Sciences,New Mexico State University,Las Cruces MATTHEW P. GOLOMBEK Jet PropulsionLaboratory, CaliforniaInstitute of Technology,Pasadena Careful documentationof the Cenozoicgeologic history of the Rio Grande rift in New Mexico reveals a complexsequence of events.At least two phasesof extensionhave been identified.An early phase of extensionbegan in the mid-Oligocene(about 30 Ma) and may have continuedto the early Miocene (about 18 Ma). This phaseof extensionwas characterizedby local high-strainextension events (locally, 50-100%,regionally, 30-50%), low-anglefaulting, and the developmentof broad, relativelyshallow basins, all indicatingan approximatelyNE-SW •-25ø extensiondirection, consistent with the regionalstress field at that time.Extension events were not synchronousduring early phase extension and were often temporally and spatiallyassociated with major magmatism.A late phaseof extensionoccurred primarily in the late Miocene(10-5 Ma) with minor extensioncontinuing to the present.It was characterizedby apparently synchronous,high-angle faulting givinglarge verticalstrains with relativelyminor lateral strain (5-20%) whichproduced the moderuRio Granderift morphology.Extension direction was approximatelyE-W, consistentwith the contemporaryregional stress field. Late phasegraben or half-grabenbasins cut and often obscureearly phasebroad basins.Early phase extensionalstyle and basin formation indicate a ductilelithosphere, and this extensionoccurred during the climax of Paleogenemagmatic activity in this zone.Late phaseextensional style indicates a more brittle lithosphere,and this extensionfollowed a middle Miocenelull in volcanism.Regional uplift of about1 km appearsto haveaccompanied late phase extension, andrelatively minor volcanism has continued to thepresent. We have estimated geotherms and calculated lithosphericstrength curves for the two phasesof extension,using geologic data to constrainearlier events andgeophysical data to constrainthe moderngeotherm and crustalstructure. A highgeotherm was deduced for early phaseextension, resulting in a shallowcrustal brittle-ductile transition and negligiblemantle strength.The lithospherecooled after early phaseextension, resulting in a deepercrustal. brittle-ductile transition,and, perhapsmore significantly,a considerablezone of mantle strengthimmediately beneath the Moho. Theseresults indicate that early phaseextensional style was controlled by a crustaldecollement nearthe brittle-ductiletransition, which was prevented during late phaseextension by significantstrength in the uppermostmantle. Late Cenozoicuplift of the rift zonecannot be explainedby crustalthinning duringextension and geothermevolution predicted from simplecooling. However, this uplift doesnot appearto be restrictedto the rift zone, and Plioceneto Recentvolcanism and heat flow data suggest that uplift may be causedby magmaticthickening of the crust,perhaps unrelated to rifting. The complex interrelationshipamong regional and local prerifting,synrifting, and postriftingevents in the Rio Grande rift suggeststhat rifting, at least in this region,should not be consideredin isolationof other geologic events. INTRODUCTION tectonicstyles. We havecarefully documented these events with relevant geological and geophysicaldata to attempt to Much attention has recently been focused upon the understandthe factorswhich controlledtectonic style during mechanismsand processesof rifting. In this studywe present each extensionalevent through an analysisof the thermal and a relativelydetailed analysis of two areasof New Mexico which mechanicalevolution of the lithospherein theseareas. As these exhibit complex but similar Cenozoichistories of extensional data are crucialto the thermomechanicalmodeling presented tectonism.The locationsof theseareas are shownin Figure in this contribution,we presentdetailed documentation of the 1. The first study area is the Basin and Range provinceand geologichistory of the southernRio Grande rift study area southernRio Grande rift in southernNew Mexico; the second below, followed by a more concisehistory of rift evolutionin studyarea is the centralRio Granderift in centraland northern New Mexico, the southernSan Luis basin,the Espafiolabasin, northernNew Mexico [see also Morgan and Golo•bek, !984]. It is beyondthe scopeof thispaper to presentsimilarly detailed and the Albuquerque-Belenbasin. At least two Cenozoic documentationfor the Cenozoicevolution of all,.areasof the extensionalevents have occurred in eacharea with contrasting Rio Granderift: however,the general similarities in evolutionary historythat we have deduced for ourtwo study are• suggest that our resultsmay be generallyapplicable to the Rio Grande •Alsoat Lunarand Planetary Institute, Houston, Texas. rift. • Our analysisof extensionin the Rio Granderift differsfrom Copyright1986 by the AmericanGeophysical Union. manyprevious rift modelingstudies in that we ignorethe causes Paper number 5B5473. of extension.Through this analysis,we show that, at least in 0148-0227/ 86/ 005B-5473505.00 our study areas,the extensionevents cannot be understoodin 6263 6264 MORGAN ET AL.: CENOZOIC EVOLUTION OF THE RIO GRANDE RIFT I • SouthernSan andesiticflows and lahars from widespreadcentral volcanoes. By 35 Ma, calc-alkalinerhyolite and rhyodacitewere erupted I I in hugevolumes, largely as ash flow tuff sheetsfrom cauldron complexes.Individual sheets spread across thousands of square I Alamo. LuisBasi_•n kilometersof southwestern New Mexico, suggestingthat neither activeextension nor compressionwas producingtopography jNEW MEXICO •..•.:.:.:.11 Espan01a I that might control the distributionof volcanicrocks [e.g., Chapin,1974; Chapin and Seager,1975; Seager et al., 1984). Belen Basin entoFe Elston [1984a] and Engebretsonet al. [1984] believemiddle Albuquerque- Tertiaryarc magmatismwas accompaniedby stronginterarc extension.On theother hand, Price and Henry [1984] and Henry . • I• rque , I and Price,[this issue] present data from westTexas that indicates ENE compressionfrom Laramidetime throughthe main pulse of middle Tertiary volcanism 39-32 Ma. Becauseof these •IZ.• •RIO•RANDE RIFT contradictions,it is unclear what kind or orientation of stress I :"'•• ' ' •ocorro field accompaniedmiddle Tertiary volcanismin southwestern New Mexico. What does seemclear is that by about 30 q- 2 _ ";•'-'-j-• SouthernRio Ma changingpatterns of magmatism,sedimentation, and deformationsignify establishment of regionalextension as a • '.'•.•:'( C:.• • ' •'Gra nde Riftstable, dominating stress field. Early PhaseExtension Magmatism. Several kinds of evidence document the existenceof an extensionalregime in southernNew Mexico in late Oligocenetime. One of theseis the strikingand locally thick sequence of "basaltic andesite" that overlies middle Oligoceneash flow tuffsor associatedrocks almost everywhere. Foults odive in ' .'• Ranging in age from about 29 to 18 Ma, these marie flows loreQu•ernery :••••idio locally comprisecentral shield volcanoes or are associatedwith Fig. 1. Index map for study arc•. The southernRio Grande •t VOLCANIC dct•lcd study area is outlined by the box. The •buqucrquc-Bclcn, TIME STYLE VOLCANIC PRODUCTS TECTONIC STYLE Espafiolaand southernSan Luis b•ins comprisethe ccntr• Rio Grande m.y.b.p. •t studyarea (b•c map from Seagerand Morgan [1979]). RiftIAikali-Oiivine BasaltCulmination of Rifting isolation of other geologicevents. We attempt to show that the interrelationshipbetween regional and local eventsmay be t RiftExtension: Rapid complex but that it is fundamental to the understandingof --10-- extensionaltectonics. The interrelationshipbetween local and Magmatic regionalevents may be the key to understandingsome of the Lull Slow Extension ? complexitiesand diversitiesof continentalrifting. --20-- CENOZOIC GEOLOGIC HISTORY OF SOUTHERN RIO GRANDE AlkaliRhyolite RIFT & Basaltic Andesite- Datii Prerift Tectonicsand Magmatism Backarc Extension: Backarc? Rapid During the last 65 m.y., southwesternNew Mexico has been - --- 29- 30 -- -- --30-- Mogollon•-- -- -- 29 -- - 30 ..... the site of almost constanttectonic and/or magmaticactivity Backarc? I Calc-A!kaline or Arc? I Basaltic Andesire (Figure 2). Most of southernNew Mexico was a cratonicback -- ----33 I arc region during the Laramide which failed under regional Calk-Alkaline No strong regional compressionby forming a seriesof northwesttrending, Wind Arc? stress (Arc?) River style basementblock uplifts and complementarybasins --40-- I J Rhyolite [Seaget, 1983; Seaget and Mack, 1986a]. Only in the I •Calk-Aikaline southwesterncorner of the state did a magmatic arc extend Andesire into the region,although isolated volcano-plutonic centers occur Laramide Compression: Foreland-style basement outsidethe arc at Tyrone,Pinos Altos, SantaRita, and Hillsboro. block uplifts and basins; All of the magmatismis Late Cretaceousto Paleocenein age. overthrusting (?) Middle Tertiary magmatism affected virtually all of --50-- southwesternNew Mexico. Although locally sheetsof rhyolitic Fig. 2. Cenozoicvolcano-tectonic history of southwesternNew Mexico ash flow tuff are prominent, earliesteruptions were generally basedon geologicdata. MORGANET AL.: CENOZOICEVOLUTION OF THE RIO GRANDERIFT 6265 the remains of cinder cone fields or diatremes. At least one