1. Reference Earth model: heat-producing elements & geoneutrino flux *Yu Huang, Roberta Rudnick, and Bill McDonough Geology, U Maryland *Slava Chubakov, Fabio Mantovani Physics, U Ferrara, Italy

2. Geoneutrino Flux on Earth Surface Earth structure (  and L) and chemical composition (a) Activity and number of produced geoneutrinos Volume of source unit Survival probability function Distance between source unit and detector

3. Constructing a 3-D reference model Earth assigning chemical and physical states to Earth voxels 1 o x 1 o x “z” tile mapping of elements Huang et al (2013) G-cubed arXiv:1301.0365 arXiv:1301.0365

4. Global Earth Reference Model -7 layers for the top 200 km -Integrate 3 global models for the crust -New crust model with uncertainties Huang et al (2013) G-cubed

5. Global Earth Reference Model -7 layers for the top 200 km -Integrate 3 global models for the crust -New crust model with uncertainties Huang et al (2013) G-cubed

6. Defining Crustal Thickness: 3 models  Refraction and Reflection seismic waves: CRUST 2.0 1  Surface seismic waves: CUB 2.0 2  Gravitational potential field and gradiometer: GEMMA 3 1 Bassin et al., 2000. 2 Shapiro and Ritzwoller, 2002. 3 Negretti et al., 2012

7. Defining Uncertainties Crustal Thickness  ~10% uncertainty in continents  Larger uncertainty in oceans and continental margins  First uncertainty estimate of global crustal thickness!

8. Seismic Velocities of Deep Crustal Rocks Linear relationship (Vp vs. SiO 2 ) Two components mixing in MC and LC: felsic and mafic SiO 2 (wt. %) Middle CC

9. Seismic Velocities of Deep Crustal Rocks Two components mixing in MC and LC: felsic and mafic Distinguishable by Vp (1-sigma) Close to linear relationship (Vp vs. SiO 2 ) SiO 2 (wt. %) Amphibolite Granulite

10. Non-Gaussian distributions Uranium Abundance Composition of Mafic & Felsic Components

12. How to Track Uncertainty? Monte Carlo simulation : highly desired for the propagation of asymmetric uncertainties Requirement : the PDFs of all inputs are known Generate random samples for inputs, including correlation Calculate output variables Statistical analysis

13. Uranium Abundance in Middle Continental Crust layer Average middle Cont. Crust U abundance is  g/g Rudnick and Gao (2003) 1.3  g/g U MCC (  g/g)

14. Uranium Abundance in Lower Continental Crust layer Average lower Cont. Crust U abundance is  g/g U LCC (  g/g)

15. Upper crust Middle crust Lower crust Upper mantle ~35km Geological model – Continental Crust Moho Surfaces of each layer is defined by geophysical data (i.e., gravity and seismic) crust – mantle boundary U (ppm) 2.7 1.0 0.2 0.01

16. Predicted Global geoneutrino flux based on our new Reference Model Huang et al (2013) G-cubed arXiv:1301.0365arXiv:1301.0365 10.1002/ggge.2012910.1002/ggge.20129 Geoneutrino detection rate --TNU: Terrestrial Neutrino Unit -- 1 TNU = one geoneutrino event per 10 32 free protons per year

17. Near Field: six closest 2° × 2° crustal voxels Far Field = bulk crust – near field crust

18. What’s hidden in the mantle? Seismically slow “red” regions in the deep mantle Can we image it with geonus? Ritsema et al (Science, 1999)

19. Testing Earth Models Mantle geoneutrino flux ( 238 U & 232 Th) Šrámek et al (2013) EPSL 10.1016/j.epsl.2012.11.001; arXiv:1207.085310.1016/j.epsl.2012.11.001arXiv:1207.0853

20. Predicted geoneutrino flux Mantle flux at the Earth’s surface Šrámek et al (2013) EPSL 10.1016/j.epsl.2012.11.001; arXiv:1207.085310.1016/j.epsl.2012.11.001arXiv:1207.0853 Yu Huang et al (2013) G-cubed arXiv:1301.0365 arXiv:1301.0365 10.1002/ggge.20129 Total flux at surface dominated by Continental crust dominated by deep mantle structures

21. Geology input Physics input

22. Geology input Physics input

23. Physics input Geology input

24. SNO+ Hanohano Physics input Geology input