pdf reprint - Berkeley Seismological Laboratory

of 21

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
21 pages
0 downs
PUBLICATIONS Geochemistry, Geophysics, Geosystems RESEARCH ARTICLE 10.1002/2014GC005355 Key Points: Arc front migration occurs globally in continental and some…
PUBLICATIONS Geochemistry, Geophysics, Geosystems RESEARCH ARTICLE 10.1002/2014GC005355 Key Points: Arc front migration occurs globally in continental and some oceanic settings Crustal thickening causes arc front migration and truncates mantle melting Tectonic extension may balance crustal thickening for stationary arc fronts Correspondence to: L. Karlstrom, a class= __cf_email__ href= /cdn-cgi/l/email-protection data-cfemail= 117d7478777a20516265707f777e63753f747564 [email protected] /a script data-cfhash='f9e31' type= text/javascript /* ![CDATA[ */!function(t,e,r,n,c,a,p){try{t=document.currentScript||function(){for(t=document.getElementsByTagName('script'),e=t.length;e--;)if(t[e].getAttribute('data-cfhash'))return t[e]}();if(t&&(c=t.previousSibling)){p=t.parentNode;if(a=c.getAttribute('data-cfemail')){for(e='',r='0x'+a.substr(0,2)|0,n=2;a.length-n;n+=2)e+='%'+('0'+('0x'+a.substr(n,2)^r).toString(16)).slice(-2);p.replaceChild(document.createTextNode(decodeURIComponent(e)),c)}p.removeChild(t)}}catch(u){}}()/* ]] */ /script Citation: Karlstrom, L., C.-T. A. Lee, and M. Manga (2014), The role of magmatically driven lithospheric thickening on arc front migration, Geochem. Geophys. Geosyst., 15, doi:10.1002/2014GC005355. Received 24 MAR 2014 Accepted 30 MAY 2014 Accepted article online 5 JUN 2014 The role of magmatically driven lithospheric thickening on arc front migration L. Karlstrom1, C.-T. A. Lee2, and M. Manga3 1 Department of Geophysics, Stanford University, Stanford, California, USA, 2Department of Earth Science, MS-126 Rice University, Houston, Texas, USA, 3Department of Earth and Planetary Science, University of California at Berkeley, Berkeley, California, USA Abstract Volcanic activity at convergent plate margins is localized along lineaments of active volcanoes that focus rising magma generated within the mantle below. In many arcs worldwide, particularly continental arcs, the volcanic front migrates away from the interface of subduction (the trench) over millions of years, reflecting coevolving surface forcing, tectonics, crustal magma transport, and mantle flow. Here we show that extraction of melt from arc mantle and subsequent magmatic thickening of overlying crust and lithosphere can drive volcanic front migration. These processes are consistent with geochemical trends, such as increasing La/Yb, which show that increasing depths of differentiation correlate with arc front migration in continental arcs. Such thickening truncates the underlying mantle flow field, squeezing hot mantle wedge and the melting focus away from the trench while progressively decreasing the volume of melt generated. However, if magmatic thickening is balanced by tectonic extension in the upper plate, a steady crustal thickness is achieved that results in a more stationary arc front with long-lived mantle melting. This appears to be the case for some island arcs. Thus, in combination with tectonic modulation of crustal thickness, magmatic thickening provides a self consistent model for volcanic arc front migration and the composition of arc magmas. 1. Introduction One of the most distinctive geographic features on Earth is the series of long arcuate chains of volcanoes on the upper plate of subduction zones, where cold and hydrothermally altered oceanic lithosphere descends into the Earth’s deep interior. Arc volcanism forms one of the primary connections between longterm climate, landscape, and mantle dynamics. For example, convergent margin igneous activity is one of the main drivers of crustal differentiation and the formation of continents. Arc volcanoes create mountain ranges high and long enough to influence large-scale atmospheric circulation. They are also a significant source of volatiles, such as H2O, CO2, and SO2, to the atmosphere and hydrosphere. Surface topography, composition of erupted magmas, and the volume of volatiles released all depend on the nature of magma generation in the mantle and how these magmas interact with the upper plate. Active volcanism in subduction zones is spatially focused into a narrow, 10–30 km wide, lineament called the arc front, which varies in distance from several tens to several hundreds of kilometers from the trench. The distance of the arc front from the trench must be a manifestation of the thermal state of the mantle wedge or subducting slab and is thus of particular interest. One view is that arc magmatism is driven by hydrous flux melting of the mantle wedge, the fluids being derived from dehydration of the subducting slab as it undergoes prograde metamorphism over a narrow temperature interval [Grove et al., 2009]. If arc magmatism is limited by dehydration reactions in the slab, the position of the arc front may depend primarily on the diffusive time scale for slab heating. This time scale sets the temperature structure in the overlying mantle wedge and determines whether rising melts freeze or ascend into the crust. Another view is that arc magmas originate from the hot nose of the mantle wedge, melting via decompression that is independent of dehydration reactions in the slab [England and Katz, 2010]. Both scenarios predict some dependency of volcano location on subduction parameters slab velocity V and dip angle d that control flow in the wedge, and indeed, a number of studies have shown that the positions of modern arc fronts correlate negatively with slab dip or some product of slab dip and plate velocity [Syracuse and Abers, 2006; Grove et al., 2009; England and Katz, 2010]. KARLSTROM ET AL. C 2014. American Geophysical Union. All Rights Reserved. V 1 Geochemistry, Geophysics, Geosystems a. 10.1002/2014GC005355 Relative distance to trench (km) 150 87 Sr/ 86 Sr b. 100 50 0 0.71 0.708 0.706 0.704 c. 80 Sierra Nevada Peninsular Ranges La/Yb 60 Andes 0-25 Ma Andes 25-69 Ma 40 Andes 69-120 Ma Lesser Antilles Marianas 20 0 0 20 40 60 80 100 120 Time (Ma) Figure 1. (a) Crystallization ages of volcanic and plutonic rocks and relative distances to the trench for three continental (Andes, Sierra Nevada, Penninsular Ranges) and two oceanic arcs (Lesser Antilles, Marianas). Andean volcanic data have been divided into three episodes and detrended to focus on the cycles of volcanic migration. Corresponding geochemical indices through time for continental arcs covary with spatial migration: (b) Initial 87Sr/86Sr isotopic ratio and (c) ratio of trace elements La/Yb. Independent of the mechanism of melting, a longstanding extension of this correlation between subduction parameters and arc front location has been that changes in slab dip with time will move the front position relative to the trench [Dickinson and Snyder, 1978]. For example, the eastward migration of arc magmatism in the Sierra Nevada Batholith, California during the Late Cretaceous is widely attributed to progressive shallowing of the angle of eastward subduction of the Farallon oceanic plate beneath North America [Coney and Reynolds, 1977; Dickinson and Snyder, 1978; Lipman, 1992; Humphreys et al., 2003]. Similar arguments have been used to explain the migration of arc fronts in Tibet [Chung et al., 2005], Southeast China [Li and Li, 2007], and the Andes [Haschke et al., 2002]. It has also been argued that mechanical erosion by the downgoing plate may drive the migration of arc fronts [Scholl and von Huene, 2007]. Any successful model of subduction zones, however, must satisfy some key observations related to arc front migration. Some arcs migrate, some do not, and in those that do, migration is not always continuous (Figure 1a). Continental arc volcanism generally migrates away from the trench [Dickinson and Snyder, 1978], sometimes in cycles of spatial advance and retreat of volcanic activity with intervening temporal gaps in magmatism [Haschke et al., 2002]. Some oceanic arc fronts remain stationary relative to the trench or migrate without temporal gaps in eruptive output [Stern et al., 2003]. These differences appear to correspond to variations in the overall tectonic state of the overriding plate: oceanic arcs (e.g., Mariana, Tonga) are often strongly extensional, to the point of back-arc basin seafloor spreading, while some continental arcs (e.g., Andes) evolve in the presence of tectonic shortening and subduction erosion of the accretionary wedge [Uyeda and Kanamori, 1979; von Huene and Scholl, 1991]. Spatial migration of arcs also involves changes in the nature of magma transport, differentiation, and interaction with the upper plate as evidenced by evolving geochemistry as the arc front migrates. For example, the isotopic ratio 87Sr/86Sr (Figure 1b) and bulk silica content increase as continental arcs migrate away from the trench, suggesting longer magma transport times and crustal storage (we note that it is the combination of these factors that imply increased transport times rather than magma interacting with older crust). Increases in trace element ratios such as La/Yb (Figure 1c), which are sensitive to the pressure-temperature KARLSTROM ET AL. C 2014. American Geophysical Union. All Rights Reserved. V 2 Geochemistry, Geophysics, Geosystems 10.1002/2014GC005355 conditions for garnet stability, suggest thickening of crust [Haschke et al., 2002; Lee et al., 2007; DeCelles et al., 2009]. Furthermore, in the Sierra Nevada, California, where shallowing of the slab is widely accepted to have driven migration of the arc front, xenolith data show that the Sierran arc root extended to depths of at least 90 km, approaching or even exceeding the average depth to the slab beneath modern arcs [Ducea and Saleeby, 1998; Saleeby, 2003]. This thick Sierran arc root apparently developed during the peak of arc magmatism, due to a combination of magmatic thickening and lithospheric shortening [Barth et al., 2012; Chin et al., 2012]. Thermobarometric studies indicate that this thickening root impinged directly against a normally dipping slab [Chin et al., 2012]. None of the above observations require that arc front migration is caused by changing dip of the downgoing slab. A number of mechanisms have been proposed for the transient flattening of slabs: subduction of oceanic plateaux [Saleeby, 2003], overthrusting or suction from deep continental roots [van Hunen et al., 2002], and evolving rheology of upper or downgoing plate [Billen and Hirth, 2007]. However, these mechanisms do not naturally explain the ubiquity and variability of arc front migrations or the unsteady magmatic output, and do not naturally explain the consistency between migration and geochemical data. Here we present a new model for arc front migration. Rather than relying on time-varying dip angle of the downgoing slab, we hypothesize that arc front migration occurs by the thickening of overlying crust and lithosphere due to melt extraction from the mantle wedge. To evaluate this hypothesis, we first present a new compilation of arc front migration geochemical data that suggest thickening of crust during arc front migration. We then develop a mathematical model of arc front migration that, by including the possibility of tectonic thickening/thinning and erosive thinning of the overlying plate, can explain the presence of long-lived and stationary arc fronts in tectonic settings where large magnitude extension occurs. We focus on the geodynamic consequences of a few interacting subduction zone components, primarily the effect of thickening or thinning the overlying plate on evolving kinematic confinement of the mantle wedge flow field. This end-member approach is similar in spirit to that which underlies the longstanding model of arc front migration as due to slab dip changes, providing a quantitative template with which to evaluate observations. In particular, this model predicts a progressively decreasing melt supply into the crust and eventual shutoff of magmatism as thickening truncates the mantle melt column, consistent with geochemical evidence for increasing magma-crust interaction over the timespan of arc front migration and eventual cessation of surface volcanism in some settings. 2. Evidence for Arc Front Migration We compile data from five arc segments in a variety of tectonic settings, the Sierra Nevada, Peninsular Ranges, Andean, Lesser Antilles, and Izu-Bonin-Mariana arcs. These examples demonstrate both the prevalence of arc front migration and the strong variability in migration that reflects regional history. Data for the Peninsular Ranges Batholith in Southern California were obtained from Kistler et al. [2003] and Lee et al. [2007]. The major and trace element data are based on samples averaged over a 10 by 10 foot outcrop, minimizing sampling bias (details described in Lee et al. [2007]). Ages are U-Pb zircon ages or Rb-Sr whole-rock isochrons. Relative distance to the trench was determined by projecting sample locations to a transect perpendicular to the current trace of the arc (the Western edge of the Coast Ranges and to the South coastal California, assuming this region represents the now extended accretionary prism). The orientation of this transect is N 51.3 E. Data for the Sierra Nevada Batholith, California were obtained from the Navdat database based on compilations updated on 2 January 2013 (www.navdat.org). Only data for the contiguous Sierra Nevada batholith were used. Data for displaced Sierran blocks, such as the Salinia terrane, were excluded. There may be some sampling bias in this compilation because of heterogeneous sampling of the batholith in the compiled data set. Relative distance to the trench was determined by projecting sample locations to a transect perpendicular to the current trace of the arc structures and accretionary structures in the Coast Range in California. Orientation of this transect is N 48.3 E. In both the Sierra Nevada and Peninsular Ranges Batholiths, the arc-trench distance appears to increase between 120 and 80 Ma, after which there is a sudden cessation in magmatism. Because these two arc KARLSTROM ET AL. C 2014. American Geophysical Union. All Rights Reserved. V 3 Geochemistry, Geophysics, Geosystems 10.1002/2014GC005355 fronts were accompanied by emplacement of a large accretionary complex, now preserved in the California Coast Ranges and southern California ‘‘borderlands,’’ this apparent increase in arc-trench distance is not an artifact of eroding or shortening of the leading edge of the upper plate. Data for the Andes were taken from Haschke et al. [2002] for Chile between 21 S and 26 S. Unlike the Sierra Nevada and Peninsular Ranges Batholiths, the evolution of the Andean arc between 0 and 200 Ma preserves four 20–50 Myr long episodes of arc migration and compositional evolution of the arc magmas. Each of these short-term episodes appears to be characterized by gradual migration of the arc away from the trench, culminating in a magmatic gap, which in turn is followed by return of the magmatic arc toward the trench. In each episode, 87Sr/86Sr and La/Yb ratios increase, only to return to baseline values with the renewal of each cycle. Over the 200 Myr history of the Andean arc, there appears to be a drift in the absolute location of the arc relative to the current trench [Haschke et al., 2002]. However, no long-term drift in the geochemistry is seen. Superimposed on a 200 Myr systematic eastward drift of the arc [Kay and Mpodozis, 2001; Haschke et al., 2002] are the shorter-term episodes of arc migration we consider to be indicative of crustal thickening. The long-term drift in absolute arc front location is likely due to subduction erosion or tectonic shortening of the western edge of the South American plate [von Huene and Scholl, 1991], making it difficult to know exactly the location of the arc relative to the trench in the past. We have thus normalized the beginning of each arc migration segment to a common distance specific to the given segment so that what is reported is the arc-trench distance relative to the initial arc-trench distance for a given segment of arc migration. For the Andes, we have assumed that the orientation of the current trench relative to the past arcs has not changed. In this setting, unsteady convergence, episodes of shortening that resulted in nonmagmatic crustal thickening [Pardo-Casas and Molnar, 1987; Haschke et al., 2002], and subduction erosion [von Huene and Scholl, 1991] complicate the volcanic history. Despite these processes, observed trends in spatial migration and geochemistry are quite similar to the Sierran arc. Migration data for the Lesser Antilles were compiled from the geologic map of Macdonald et al. [2000] and ages from Bouysse et al. [1990]. We have assumed that the trench location has remained constant, and study only the last 40 Ma, after the Aves ridge to the west stopped being active [Neill et al., 2011], and after Eocene extension in the Grenada basin ended [Speed and Walker, 1991; Aitken et al., 2011; Manga et al., 2012]. Data for the Izu-Bonin-Mariana come from Stern et al. [2003] and Stern et al. [2012], who argue that the arc front has maintained a nearly constant distance from the trench since the onset of magmatism after subduction initiation at 52 Ma [Reagan et al., 2013]. Tectonic erosion, back-arc extension, and rotation of the trench have occurred since the inception of subduction and some migration of the volcanic front away from the trench may have occurred. However, the presence of a large 200 km wide fore arc since the Eocene [Stern et al., 2012], and seafloor spreading reconstructions [Faccenna et al., 2009] indicate small overall motion. Figure 2 summarizes our data compilations, an estimate for the range of probable arc front migration for the five arc segments plotted in Figure 1. Maximum and minimum migration distances and times were chosen by constructing two metrics for picking the arc front in space and time. First, we develop a point density metric for volcanism (Figure 3), which locates the highest density of activity from the trench. We then compare this location to the single closest approach of volcanism to the trench to define the error bars in Figure 2. Our compilation is not exhaustive, and there are a number of other locations not included in this study where arc front migration away from the trench has been observed (e.g., the Cascades [du Bray and John, 2011], Japan [Kimura et al., 2005], Southeast China [Li and Li, 2007]), as well as those in which little migration appears to have occurred (e.g., the Aleutian arc [Jicha et al., 2006]), and those in uncommon tectonic environments that record more complicated migrations (e.g., trenchward migration in Nicaragua [Plank et al., 2002]). We however focus here on five data sets for which migration and geochemical data are most readily available, in relatively well-constrained tectonic environments. For these arc examples, Figures 1a and 2 show spatial migration of the volcanic front away from the trench is a prominent feature of the volcanic history (except for the stationary Marianas). Although with considerable scatter, geochemical indices 87Sr/86Sr and La/Yb covary with arc position in continental arcs (Figures 1b and 1c). This scatter likely reflects the nature of available data (the Peninsular KARLSTROM ET AL. C 2014. American Geophysical Union. All Rights Reserved. V 4 Geochemistry, Geophysics, Geosystems Ranges Batholith alone represents unbiased grid sampling) as well as our incomplete knowledge of the spatial overprinting of intrusions in time. This covariance of geochemistry with arc front position is also reflected in major elements such as bulk silica contents (a representative example from the Andes is presented later). Maximum migration distance (km) 160 140 Sierra Nevada 120 Peninsular Ranges 100 Andes 69-120 Ma 80 Andes 0-25 Ma Andes 25-69 Ma 60 Lesser Antilles 40 3. Implications for Crustal Thickening 20 Marianas (no migration since initiation at ~52 Ma) 0 15 10.1002/2014GC005355 20 25 30 35 40 45 The chemical composition of arc lavas varies systemati
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks