Earth models and primordial heat... and geonu emission from deep mantle piles

Earth models and primordial heat... and geonu emission from deep mantle piles Ondřej Šrámek University of Maryland Collaborators: Bill McDonough & Vedran Lekić (Univ. Maryland) Edwin Kite (Caltech) Steve Dye (Hawaii Pacific Univ.) Shijie Zhong (Univ. Colorado at Boulder) presented at Neutrino Geoscience 13, Takayama (Japan), 21 23 March 13

Present-day Earth Present-day dynamics Quantify major energy terms Earth s bulk composition Spatial distribution of radioactivity Geoneutrino constraints?????? U Th K

Solar System formation time Present-day Earth Building blocks Solar nebula Planetary formation Early differentiation processes First few 1 My Core formation Magma ocean Ongoing dynamic evolution Chemical fractionation vs. homogenization Crust formation Subduction Mantle, core dynamics

Planetary formation Solar nebula composition Big Bang (H, He, Li) + stellar (higher Z) nucleosynthesis Compositional gradients, heterogeneity in the cooling solar nebula? Condensation temperature, volatility U & Th refractory, K moderately volatile

Planetary formation Last stage of planetary formation. Outcome of 4 simulations planet position, radius, composition in terms of original heliocentric distance of material. Chambers, EPSL 11 color = original radial location of material Some memory of compositional gradient in the nebula

Planetary formation Solar nebula composition Big Bang (H, He, Li) + stellar (higher Z) nucleosynthesis Compositional gradients, heterogeneity in the cooling solar nebula? Condensation temperature, volatility U & Th refractory, K moderately volatile Dust & gas to grains to planetesimals Planetesimals to planetary embryos ~1 My Last stage ~1 My Giant impacts Planet radial migration

The Grand Tack Scenario Walsh et al. 11 Inward-then-outward migration of Jupiter, Saturn Jupiter formed at ~3.5 AU Spiraled inward to ~1.5 AU (presentday Mars orbit) due to gas drag Then migrated outward to current location at 5.2 AU Saturn experienced a similar sequence Can explain the small size of Mars Consistent with the existence of asteroidal belt May have rid the inner Solar System of volatiles

Planetary formation Solar nebula composition Big Bang (H, He, Li) + stellar (higher Z) nucleosynthesis Compositional gradients, heterogeneity in the cooling solar nebula? Condensation temperature, volatility U & Th refractory, K moderately volatile Dust & gas to grains to planetesimals Planetesimals to planetary embryos ~1 My Last stage ~1 My Giant impacts Planet radial migration Hot beginning Gravitational energy released upon accretion, differentiation Extensive melting, magma ocean(s) Incompatibility, solid vs. melt partitioning U, Th, K incompatible, enter the melt

Gravitational energy of accretion forming a homogeneous body E g =4 Z R g(r) (r)r 3 dr dispersed in space planet Eacr = GM 2 /R Earth: ΔTacr = GM 2 /R / (MCP) ~ 5 1 4 K of differentiation forming the core ΔTdiff ~ K undifferentiated differentiated

Loosing primordial heat Mantle cooling rate petrological constraint based on inference of melting temperature of primitive mantle melts Abbott et al. JGR 1994 inferred cooling rate of 5 ± 25 K/Gy translates into 8 ± 4 TW mantle heat output due to cooling

Giant impacts, Moon formation, magma ocean From Hf W chronology (Kleine et al. GCA 9): Core formation at ~ 1 My after beginning of Solar System Moon formation Deep magma ocean (metal silicate equilibration at >4 km depth; Wood et al. Nature 6) Siderophile elements go in the core Core formation major chemical differentiation event Lithophile elements go in the silicate phase: negligible in U, Th, K in the core Makes sense to talk about Silicate Earth

U Th K Composition of Silicate Earth (BSE) Geochemical estimate Cosmochemical estimate Geodynamical estimate

Composition of CI chondrites matches solar photospheric abundances in refractory lithophile, siderophile, and volatile elements. Most primitive, highest volatile abundance of chondrites Best representative of primordial nebular material Wood Phys.Today 11

Geochemical BSE estimate(s) Ratios of Refractory Lithophile Element abundances constrained by CI chondrites Absolute RLE abundances and non-refractory element abundances inferred from available mantle & crustal rock samples Uses petrological models on peridotite xenoliths, massif peridotites, primitive melts McDonough & Sun (1995)... ± 4 ppb U Allègre (1995)... 21 ppb U Hart & Zindler (1986)....8 ppb U Palme & O Neill (3)... 22 ± 3 ppb U Results in ~ TW radiogenic power in BSE

Cosmochemical BSE estimate Isotopic similarity between Earth rocks and enstatite chondrides Similarity in oxidation state Build the Earth from E-chondrite material Fri AM: Claude Jaupart Javoy et al. EPSL 1... 12 ppb U Gives ~11 TW radiogenic power in BSE Also collisional erosion model (O Neill & Palme 8) Earth formed with geochemical abundances (22 ppm U in BSE) Differentiated early crust (enriched in U, Th, K) This crust was lost during a giant impact collision 1 ppm U in BSE

Geodynamical BSE estimate Based on energetics of mantle convection Parameterized convection models: heat loss = radiogenic heating + secular cooling Classical scaling between Qs (Nusselt number) and vigor of convection (Rayleigh number) Q s (t) =H rad (t) C Nu / Ra(T ) 1/3 dt (t) dt Need a large proportion of radiogenic heating to account for mantle heat flow, otherwise thermal catastrophe in the Archean Requires mantle Urey.6 (geochemical =.3, cosmochemical =.1) Therefore needs higher abundance of U, Th, K Radiogenic heating TW in BSE Thu PM: John Crowley

U Th K Composition of Silicate Earth (BSE) Geochemical estimate Ratios of RLE abundances constrained by C1 chondrites Absolute abundances inferred from Earth rock samples McDonough & Sun (1995), Allègre (1995), Hart & Zindler (1986), Palme & O Neill (3), Arevalo et al. (9) Cosmochemical estimate Isotopic similarity between Earth rocks and E-chondrides Build the Earth from E-chondrite material Javoy et al. (1) also collisional erosion models (O Neill & Palme 8) TW radiogenic power BSE Mantle ±4 11±2 12±4 3±2 Geodynamical estimate Based on a classical parameterized convection model Requires a high mantle Urey ratio, i.e., high U, Th, K 33±3 25±3 CRUST2. thickness BSE = Mantle + Crust Oceanic:.22 ±.3 TW Continental: 7.8 ±.9 TW Tomorrow: New crustal model by Yu Huang et al. CC = 6.8 (+1.4/-1.1) TW?

How is radioactivity distributed in the mantle?? Educated hypothesis Geoneutrino flux prediction Testing with geoneutrino measurement

Mantle geoneutrino emission Uniform mantle Mantle geoneutrino U+Th signal prediction (r) = n Z hp i 4 A(r ) (r ) r r 2 dr Flux Φ at position r from a given radionuclide distributed with abundance A in volume Ω 25 "UNIF" mantle TNU 15 1 5 Cosmochemical Geochemical Geodynamical

Shallow mantle composition Mid-Oceanic Ridge Basalt composition source rock abundances in shallow mantle? MORB-source mantle abundance estimates Workman & Hart (5) 2.8 ±.4 Salters & Stracke (4) Arevalo & McDonough (1) 4.1 ± 1.2 7.5 ± 1.5 TW if occupying entire mantle Bulk mantle (=BSE Crust) Cosmochemical Geochemical Geodynamical 3.3 ± 2. 12 ± 4 25 ± 3 TW Require enriched material in the mantle

Mantle geoneutrino emission Spherically symmetric U & Th distribution in the mantle Uniform mantle Mantle geoneutrino U+Th signal prediction Enriched layer (1% of mantle) TNU 25 15 "UNIF" mantle "EL" with A&McD DM "EL" with S&S DM "EL" with W&H DM 1 5 N/A Cosmochemical Geochemical Geodynamical

Mantle geoneutrino emission Spherically symmetrical U & Th distribution in the mantle Uniform mantle Mantle geoneutrino U+Th signal prediction Enriched layer (1% of mantle) TNU 25 15 1 "UNIF" mantle "EL" with A&McD DM "EL" with S&S DM "EL" with W&H DM Measurements Combined analysis of detection at KamLAND and Borexino (Bellini et al. arxiv:13.2571) 5 N/A Cosmochemical Geochemical Geodynamical + new KamLAND data??

Seismic tomography image of present-day mantle Seismic shear wave speed anomaly Tomographic model SRTS (Ritsema et al.) Two large scale seismic speed anomalies below Africa and below central Pacific Anti-correlation of shear and sound wavespeeds + sharp velocity gradients suggest a compositional component piles or LLSVPs or superplumes Candidate for an distinct chemical reservoir Bull et al. EPSL 9 Sat AM: Ed Garnero

Origin of deep mantle reservoir Continuously recycled subducted slabs buried at core mantle boundary? Next talk: Dapeng Zhao Remnant of dense basal magma ocean at the bottom of the mantle? Boyet & Carlson EPSL 6 Early differentiation Fri AM: of primordial Sujoy Mukhopadhyay mantle? Sat AM: Matt Jackson Labrosse et al. EPSL 6

Thermochemical piles in deep mantle? LLSVPs Assume these piles represent an enriched reservoir. δlnvs isocontours shape Mantle geoneutrino U+Th signal prediction Pyhasalmi TNU 1.5 Gran Sasso Baksan Kamioka Homestake Sudbury 1. Hawaii 9.5 9. 8.5 8. Can we detect such variation in mantle geonu flux? O.Š., McDonough, Kite, Lekić, Dye, Zhong, EPSL 13

35 35 35 25 Crust + Mantle geoneutrino emission 15 Crustgeoneutrino + mantle signal Crust+mantle Crust+mantle geoneutrino signal Geochemical BSE To constrain mantle U,DMwe A&McD Mantle / Total Th, need to measure in the ocean. % TNU TNU 55 5 55 45 5 4 45 7 6 35 4 35 25 5 Site #1 Mantle Mantle / Total Mantle / Total contribution to A&McD DM total signal 8 25 6 5 Site #1 4 Site #2 Site #2 Site #1 Longitude = 161 W geodynamical Site #1 Longitude =locations: 161 Mantle, W Crust Continental not more Mantle, geodynamical Crust A&McD DM S&S DMDM than ~25% A&McD W&H DM S&S DM from mantle W&H Site #2 DM of geonu signal coming 25 25 O.Š., McDonough, Kite, Lekić, Dye, Zhong, EPSL 13 TNU 7 6 5 4 Site #1 Longitude = 161 W Mantle, geodynamical Crust 8 7 A&McD DM S&S DM W&H DM Site #2 Mantle, geochemical 15 1 5 Mantle, cosmochemical 9 6 Mantle geoneutrino signal thermochemical piles Pyhasalmi TNU 1.5 Gran Sasso Homestake Baksan Kamioka Hawaii Sudbury Latitude in degrees 1. 9.5 9. 8.5 Site #2 Prediction along 161 W meridian 35 % Site #1 Site #2 15 % 4 Hanohano (proposed) 25 15 Geochemical BSE A&McD DM BSE Geochemical 8 8. 6 9

5 45 4 35 TNU 25 Crust+mantle geoneutrino signal 55 5 15 45 4 Geochemical BSE Mantle contribution signal A&McD DM Mantle / Total to total 35 % 25 8 7 15 6 Geochemical BSE A&McD DM Mantle / Total Site #1 5 % 4 8 7 6 Site #2 central value 24 +/ 1σ 16 A&McD DM central value S&S DM +/ 1σ W&H DM A&McD DM S&S DM 16 12 W&H DM 12 8 not resolved 8 4 4 Prediction along 161 W meridian Site #2 25 35 Site #1 Longitude = 161 W Mantle, geodynamical Crust Mantle, geochemical 15 25 A&McD DM S&S DM W&H DM Site #2 5 TNU 1 Mantle, cosmochemical Mantle, geochemical 15 9 1 6 6 9 Latitude in degrees 5 9 Mantle, cosmochemical 6 6 Latitude in degrees O.Š., McDonough, Kite, Lekić, Dye, Zhong, EPSL 13 9 r e ffe di c en r s ile p le le t b n va ma Geodynamical p ol s ee Reel d es il ep m ntl a O M resolved! p m Geodynamical TO ee d l Geochemicalode m O M TO 4 8 12 detection uncertainty 16 24 28 Predicted event(crustal, rate at site reactor, #1 in TNU other bg) Cosmochemical 32 4 8 12 32 1 5 1 2 5 1 2 Cosmo chemical.5 1.5 16 24 28 Predicted event rate at site #1 in TNU 1 kiloton detector (Hanohano) uniform mantle uniform mantle Site #2 A&McD DM S&S DM W&H DM 32 32 DetectorDetector size in 1 size free in 1 protons free protons 4 e l ab v ol ffe di Geochemical Cosmochemical Site #1 Site #1 Longitude = 161 W Mantle, geodynamical Crust s Re e c en od 5 35 Sp Sp he he ric ric al al ly ly sy sy m m m m et et ric ric m m an an tle tle 55 not resolvable not resolvable Proposed two-site measurement in the Pacific TNU Predicted Predicted event rate event at site rate#2atinsite TNU #2 in TNU Crust+mantle geoneutrino signal 1 kiloton detector (Hanohano) 1 kiloton detector Geochemical Geodynamical 4 8 12 16 24detector 28 1 kiloton Cosmo Mean event rate in TNU chemical Geochemical Geodynamical live time in months 4 8 1 12 1 116 12 24 13 28 32 1432

Summary Process of planetary formation Hot beginning of the Earth Early differentiation events core formation early fractionation of silicate Earth? Three classes of BSE compositional models [ Enstatite chondrite model inconsistent with geonu measurements (1σ)? ] Ocean site measurement needed to significantly constrain mantle radioactivity. Multiple-site to resolve lateral variations.