When Stars Attack! Recent Near-Earth Supernovae Revealed by 60Fe

Thank You Brian Fields Organizers! Astronomy & Physics, U Illinois

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Nearby Supernova Collaborators Themis Athanassiadou

Scott Johnson

Kathrin Hochmuth

John Ellis CERN

Brian Fry

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Recent Near-Earth Supernovae Revealed by 60Fe

★ Supernovae are Radioactivity Factories if near: a unique laboratory…and a unique threat ★ The Smoking Gun supernova radioactivities on Earth ★ Geological Signatures sea sediments as telescopes

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Supernovae are Radioactivity Factories

Ø medium-lived radioactivities: 60Fe, Core-Collapse 60Fe: Theoretical Yields Tur+ 2010; Limongi & Chieffi 2006 26Al, 53Mn, 41Ca, 97Tc(?), 146Sm(?)

Ø 60Fe: made by neutron captures

“weak s-process” Fe 60 56Fe(n,g)57Fe(n,g)58Fe(n,g)59Fe(n,g)60Fe

large theoretical uncertainties in yield ejected sensitive to stellar evolution, nuke rates

accuracy ~order of magnitude SN mass

Ø r-process? 182Hf, 244Pu

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 The Smoking Gun: Supernova Debris on the Earth Ellis, BDF, & Schramm 1996 Explosion launched at ~few% c Slows as plows thru interstellar matter

SOHO Earth “shielded” by solar wind

If blast close enough: ✓ overwhelms solar wind ✓ SN material dumped on Earth ✓ Accumulates in natural “archives” sea sediments, ice cores

Q: How would we know? Need observable SN “fingerprint” Nuclear Signature X Stable nuclides: don’t know came from SN ü Live radioactive isotopes: none left on Earth If found, must come from SN! also Korschinek+ 96 Chandra

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 The Fury of Aerial Bombardment: Supernova Blast Passage--Global View BDF, Athanassiadou, Johnson 2008

Dense Shell ~10 to 100 kyr transit time

10 pc

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Supernova at 20 pc: Zoom to Solar System Impact

Sun & Solar Wind

1 AU

Incoming blast

BDF, Athanassiadou, & Johnson 2008 Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 60Fe Time Profile: Sedov SN

Fry, BDF, Ellis 2015 Fe abundance Fe

60 0.2 Myr

time before present [Myr]

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Debris Delivery via Dust Athanassiadou & BDF 11; Benitez+ 02; Fry, BDF, & Ellis 2016

What if d SN > 10 pc r shock > 1AU ? ‣ gas-phase SN debris excluded from Earth But SN radioisotopes all are refractory elements dust grains

SN1987A: ‣ ~100% (!) of Fe in dust after 20 years

SN dust reaches Earth even if gas does not ‣ dust decouples from gas at shocks SN1987A dust: Matsuura+ 2011 ‣ radioisotope delivery efficiency set by dust survival fraction

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Geological Signatures

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Deep Ocean Crust

Knie et al. (1999) –ferromanganese (FeMn) crust –Pacific Ocean –growth: ~ 1 mm/Myr

AMS live 60Fe, !

Expect: one radioactive layer

1999: 60Fe in multiple layers!? ‣detectable signal exists ‣but not time-resolved Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 60Fe Confirmation Knie et al (2004)

Advances New crust from new site ✓Better geometry (planar) ✓better time resolution ✓10Be radioactive timescale Woo hoo!

Isolated Signal t = 2.8 0.4 Myr Background: 60Ni A Landmark Result± Fe abundance Fe

★ Isolated pulse identified 60 ★ Epoch quantified ★ Consistent with original crust time before present [Myr] Note fantastic AMS

sensitivity!Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Whodunit? The Astrophysical Journal,798:1(17pp),2015??? Fry, BDF, & Ellis 2015 Fry, Fields, & Ellis Turn the problem around: [pc]

“radioactivity distance” from 60Fe yield

What makes 60Fe? • core-collapse supernovae • Type Ia supernovae • AGB stars • kilonovae

SN distance: Figure 2. Evidence from Knie et al.d(SN) (2004)andFitoussietal.(∼ 20 − 1002008 pc)foran anomalous peak in the 60Fe isotope fraction 2.2 Myr ago, compared with simulations of a possible signal from a SN explosion.∼ We plot the results using “Radioactivity Distance” to Supernova “Radioactivity ECSN yields; other progenitorsEncouraging: yield similar results. ★astronomical distances not built in! ★ 60 the signal arrival. In addition,d( Fe the) valued( forSN the 880-kyrEarth) time dSN(3 Myr) resolution was less than the 440-kyr sample,⇡ as expected! due⇡ to nontrivial consistency! Mass of Progenitor the additional stable Fe in the wider sample. Figure 3. Estimated distances for possible progenitors, for U 0.5. SNe 60 Fe Using the decay-corrected Knie et al. (2004)fluenceof Fe candidates are circles and SAGB candidates are squares. The solid= error bars 60 (Section 5.1), and Fe yields from various source candidatesBrian Fields | Cosmicrepresent Rays uncertainty @ APC | Dec in the9, 2016 fluence measurement (Knie et al. 2004). The dashed (Section 2), we have solved Equation (1)forthedistancetothe error bars represent additional uncertainty in 60Fe yields due to nuclear reaction rates in SNe (Tur et al. 2010)andadelayedsuper-windphaseinSAGBs(Doherty source. Distances and other parameters for some of the possible et al. 2013). Of particular note are the TNSN/Type Ia SN and the KN/NS–NS sources appear in Table 3 and Figure 3.Weseethat,forsources merger models, which are too close to have produced the detected 60Fe signal. at distances 100 pc that are typical of our subsequent estimated distances, the∼ en route time and the signal width are (Myr), so it is possible that the signal could be time-resolvedO in future measurements, and thus it is of interest to model the signal 6.3. Kilonovae shape. Our calculations give a possible KN distance of 5pc.Of the little that is known observationally or even theoretically∼ 6.1. Core-Collapse and Electron-Capture Supernovae about KNe, we are unaware of any estimates of their ionizing radiation output. In addition, the strength and shape of the shock Figure 3 shows the calculated distances for our examined from ejected material is highly dependent on the orientation of CCSNe and ECSN; they range from 60–130 pc. All CCSNe the merger. Thus, we are unable to estimate the corresponding from our set lie outside of the kill∼ distance and within the kill distance either by direct exposure or descreening. The ejecta fadeaway distance for both their average fluence values and from KNe are certainly energetic (explosive velocities 0.3 c, errors. Similarly, the ECSN lies outside the kill distance and Goriely et al. 2011), and one might imagine decompressing∼ within the fadeaway distance (the ECSN kill and fadeaway neutron star matter initially emitting in the UV or at shorter distances are shorter due to its lower explosive energy). The wavelengths. However, the observed radiation for the KN ECSN upper error is outside the fadeaway distance, but because candidate associated with GRB 130603B is very red at times SN dust can still travel great distances after decoupling, this 8hr(Bergeretal.2013). Moreover, while the KN shock is not an absolute limitation. Based on these distances, either ! and radiation is expected to be much more isotropic than the aCCSNoranECSNcouldhaveproducedthemeasured60Fe GRB, more study of the geometry of the resulting blast is signal. needed to determine a definitive kill distance like that used for TNSN. Consequently, a biohazard argument cannot rule out 6.2. Thermonuclear Supernovae aKNexplosionasthesourceofthe60Fe anomaly. 60 However, a much better discriminator for a KN source TNSN produce so little Fe that it would require a TNSN to 244 60 244 have been at a distance of 0.6pcinordertoproducethesignal would be the Pu/ Fe ratio. The single Pu atom detected measured by Knie et al.∼ (2004). This is an implausibly short by Wallner et al. (2000, 2004)yieldsasurfacefluenceof 3 104 atoms cm2 for the period 1–14 Myr ago. Looking at distance and, any uncertainty in the fluence measurement would × not change this determination. At that range, the TNSN would the yields from Goriely et al. (2011)again,wecaninferthe yield for A 244 should be at least on the order of the have killed nearly all life on Earth, so we can exclude a TNSN = 244 60 10 60 yield for A 60 (i.e., ( Pu/ Fe)KN " 1). Based on this as the source of the Fe signal (in this case, the descreening = 60 kill distance for a TNSN is 10 pc and the ionizing radiation assumption and the surface fluence for Fe during the signal kill distance from 1048 erg of∼ γ -rays is 20 pc, Smith et al. ∼ 7 2004). Adopting the largest yield (Mej,60Fe 10− M )from 10 ∼ ⊙ More likely, A 244 yields are 10–100 times larger than A 60 yields Seitenzahl et al. (2013)extendsthedistanceto 6pc,whichis given the A 240= yields and the fact that the fission recycling= sources are still inside the kill radius and does not change this∼ conclusion. centered around∼ A 280–290 region, Goriely et al. (2011). ≃ 11 2016

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Wallner+ 2016 Nature ̣confirmation of 60Fe crust signal at ~3 Myr ̣sedimentary time profile: ~1 Myr width?! ̣indication of second 60Fe pulse ~8 Myr

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 ionization medium. Individualionization60Fe medium. ions Individual can thus60Fe ionsbe can thus identified be identified by bytheir energy their de- energystatistics, de- and our 60Fe/Festatistics, results are displayed and in our Fig. 2.60 OwingFe/Fe to results are displayed in Fig. 2. Owing to position (Bragg curve), whichposition is described (Bragg curve), which in more is described detail in more detail in inSupportingSupporting Information Information. the availability. of severalthe near-surface availability (0 Ma to 1 Ma) of and several very deep near-surface (0 Ma to 1 Ma) and very deep 60 60 54 9+ 54 9+ For measurements of theFor atom measurements ratio of the atomFe/Fe, ratio Fe/Fe, the the stable stable beam beam of Fe is of (7 MaFe to 8is Ma) samples(7 in Ma core 848, to the 8 presence Ma) of samples a distinct 60Fe in core 848, the presence of a distinct 60Fe tuned into a Faraday cup intuned front into a Faraday of the cup in GAMS,front of the GAMS, where where the the typical typical electrical electricalsignal could be clearsignally identified could (Fig. 2A). The be data clear are com-ly identified (Fig. 2A). The data are com- current is 30 nA to 150 nAcurrent (sample-dependent). is 30 nA to 150 nA (sample-dependent). Then, Then, the tandem tandem terminal terminal 606010+ 10+ plemented by the observationplemented of a similar by signal the in core observation 851 (Fig. of a similar signal in core 851 (Fig. voltage and the injector magnetvoltage and the are injector switched magnet are switched to allow to allow Fe Feto pass, andto pass, and 60 60 60 60 2B), which is characterized2B), by a which∼1.5 times is lower characterizedFe/Fe ratio. The by a ∼1.5 times lower Fe/Fe ratio. The the Faraday cup is retracted.the Faraday The cup isFe retracted. ions The areFe ions then are then individually individually identified identified60 60 6060 onset of the Fe signalonset occurs at of2.7 the± 0.1 MaFe and is signalcentered at occurs at 2.7 ± 0.1 Ma and is centered at and counted in the ionizationand counted chamber, in the ionization while chamber, whilethe the NiNi background background count count ð Þ ð Þ rate is only about 10 Hz torate100 is only about Hz. 10 In Hz to this 100 Hz. manner, In this manner, a a highly highly selective selective dis- (2.2 ± 0.1) dis- Ma. The(2.2 signal termination± 0.1) is Ma. not as clear, The because signal it termination is not as clear, because it crimination of 60Fe against crimination60Ni and of 60Fe other against 60Ni background and other background sources sources is achieved, is achieved, as remains slightly as aboveremains the 1σ blank slightly level until around above 1.5 Ma, the 1σ blank level until around 1.5 Ma, demonstrated by the extremelydemonstrated low by the extremelyblank low levels blank levels obtained obtained with different with differentaccording to the dataaccording grouping used in Fig. to 2A the.Adetailedanalysis data grouping used in Fig. 2A.Adetailedanalysis blank materials, such as processingblank materials, such blanks as processing and blanks environmental and environmental samples. samples. The averaging The over both sedimentaveraging cores and several over data both groupings sediment yields cores and several data groupings yields concentration of 60Fe/Fe is calculatedconcentration of 60Fe/Fe from is calculated the from number the number of of 6060Fe eventsFe events counted a counted more conservative estimatea more for the conservative termination time of (1.7 estimate± 0.2) for the termination time of (1.7 ± 0.2) 54 9+ 54 9+ by the detector, the measurementby the detector, time, the measurement and time, the and average the average current current of Fe ofin Ma;Fe this resultsin in aMa; (1.0 ± 0.3) this Ma long results exposure of in the Eartha (1.0 to ± 0.3) Ma long exposure of the Earth to front of the GAMS. The transmissionfront of the GAMS.efficiency The transmission efficiency between between the the Faraday Faraday cup and the cup influx and of 60Fe. Thethe60Fe flux, influx concomitant of with60Fe. the data The of Fig.60 2, Fe flux, concomitant with the data of Fig. 2, the ionization chamber is canceled out by relating the measured concentra- the ionization chamber is canceled out by relating the measured concentra- isSupporting shown Information in Fig. S7 (see Supporting Information for its derivation). tion to the known concentration of a standard sample, which is measuredis shown in Fig. S7 (see for its derivation). tion to the known concentration of a standard sample, which is measured An overview of the collected AMS data split into three age re- periodically during a beamtime.periodically during For a this beamtime. work, For this work,the the standard standard sample sample PSI-12, An PSI-12, overview of the collected AMS data split into three age re- with a concentration 60Fe=Fewith= a concentration1.25 ±600.06Fe=Fe = 1.25× 10± 0.06−12× 10,wasused(see−12,wasused(seeSupportingSupportinggions (<1.8 Ma, 1.8 Magions to 2.6 Ma, (<>2.61.8 Ma) Ma, is shown 1.8 in Table Ma 1. An to 2.6 Ma, >2.6 Ma) is shown in Table 1. An ð Þð Þ 60 60 Information). The transmissionInformation efficiency). The transmission of efficiency the of entire the entire system (including (including ion estimate ion for the significanceestimate of the Fe for signal the in the peak significance region can of the Fe signal in the peak region can source yield, stripping yield,source ion yield, stripping optical yield, iontransmissions, optical transmissions, and and software software cuts) be obtained cuts) by applyingbe the obtained procedure suggested by by applying ref. 45 for the the procedure suggested by ref. 45 for the during 60Fe measurementsduring is in60Fe the measurements range is in (1 the–4) range× (110–4) ×−104.−4. superposition of two Poissonsuperposition processes (i.e., signal of and two background). Poisson processes (i.e., signal and background). As a control region (noAs expected a control signal), we selected region (i)thepro- (no expected signal), we selected (i)thepro- Results Results cessing blank and, forcessing comparison, ( blankii)the<1.8-Ma and, age interval for comparison,of (ii)the<1.8-Ma age interval of A total of 111 sedimentA total samples of 111 sediment (67 samples from (67 from core core 848 848 and 44 and from 44each from respective sedimenteach core. Threspectiveesecondchoiceisratherconser- sediment core. Thesecondchoiceisratherconser- core 851), each with a masscore 851), of each∼ with35 a mass g, of were∼ 35 g, were treated treated with with our CBD ourvative, CBD because this overestimatevative,sthebackgroundsignal.Inthecase because this overestimatesthebackgroundsignal.Inthecase protocol, yielding ∼ 5 mgprotocol, AMS yielding samples∼ 5 mg AMS samples consisting consisting of Feof2O Fe3. Each2O3of.i,theprocedureyieldsasignificance(inmultiplesof Each of i,theprocedureyieldsasignificance(inmultiplesofσ)of5.2and σ)of5.2and AMS sample was measuredAMS sample for was an measured average for an average of of 4 4 h until until the sample the sample3.9 for cores 848 and3.9 851, respectively. for cores In the case 848 of ii,thesevalues and 851, respectively. In the case of ii,thesevalues PHYSICS material was exhausted,material yielding was exhausted, one yielding60Fe one event,60Fe event, on on average, average, and a become and 7.2 a and 2.0 (afterbecome subtraction 7.2 of the and blank level). 2.0 The (after sig- subtractionPHYSICS of the blank level). The sig- total of 86 events integratedtotal of 86 over events integrated both over cores both cores (42 (42 in core core 848 and 848 47 nificance and47 for core 851nificance is low in the case forof ii,becauseonlylittledata core 851 is low in the case of ii,becauseonlylittledata in core 851). Details ofin core the 851).60 DetailsFe ofevent the 60Fe selectionevent selection analysis analysis can be were can collected be in thewere control region. collected in the control region. found in Supporting Informationfound in Supportingand InformationFigs.and S5Figs. S5andand S6S6. Thus,. Thus, several severalA useful measure for theA intensity useful of the measure total 60Fe exposure for is the intensity of the total 60Fe exposure is Radioactive Fossil Bacteria 60 60 AMS samples have beenAMS samples grouped have been together grouped together to to increase increase counting countinggiven by the terrestrialgiven (Φter) fluence by oftheFe. terrestrialΦter represents the (Φter) fluence of Fe. Φter represents the Ludwig, Bishop, et al 2016 time-integrated, decay-correctedtime-integrated, flux of 60Fe into a givendecay-corrected terres- flux of 60Fe into a given terrestrial reservoir overtrial the entire reservoir exposure time. To over sum up the the entire exposure time. To sum up the contributions of all sedimentcontributions layers in the signal of range, all the sediment fol- layers in the signal range, the fol- A 10 A 10 lowing integral is computedlowing to calculate integral the average is concentration computed to calculate the average concentration 60 60 60 ★ Core 848 848Fe/Fe Fe/Feof 60Fe/Fe in the signalof range:Fe/Fe in the signal range: Deep-ocean 8 8 Blank level level ) ) 1-σ u.l.σ blanku.l. level blank level t2 t2 -16 -16 −1 60 −1 60 sediments 6 6 = t2 − t1 Fe=Fe dt = [1]t2 − t1 Fe=Fe dt [1] C ð Þ Cð Þ C ð Þ Cð Þ Zt1 Zt1 4 4 60 60 ★Select small grains The average terrestrialTheFe fluence, averageΦter, is then terrestrial given by Fe fluence, Φter, is then given by Fe/Fe (10 Fe/Fe (10 60 2 60 2 Φter = ρrsed t2 − t1 YCBD NA=WFe, Φter = [2]ρrsed t2 − t1 YCBD NA=WFe, [2] of magnetite Fe3O4 C ð Þ 0 0 C ð Þ 0.0 0.5 1.0 1.50.0 0.5 2.0 1.0 1.5 2.5 2.0 2.5 3.0 3.0 7.0 7.5 7.0 8.0 7.5 8.0 ★ where ρ is the dry sedimentwhere density,ρ rissed is the the sedimentation dry sediment rate, density, rsed is the sedimentation rate, Fossilized remains Age (Ma)Age (Ma) YCBD is the efficiency-correctedYCBD is CBD the yield ofefficiency-corrected extracted iron per CBD yield of extracted iron per Fig. S2. Scanning electron microscopy images of magnetic extracts obtained from representative sediment samples of core 848 over the 2.41-Ma to 2.62-Ma unit mass of dry sediment,unitNA massis Avogadro of’s number, dry sediment, and WFe is NA is Avogadro’s number, and WFe is B 8 B 8 age interval. (A) Overview of the extract showing prominent features including (i) the copper sample-holding grid, (ii) large grains consisting mainly of CaCO3 the molecular= weight± of iron. Using t t = 1.0 ± 0.3 Ma as a of magnetotacticand SiO2,and(iii) diatoms. (B) Image of a titanomagnetite grain (dark gray octahedron in the center) of most likely lithogenic origin. (C) Image of an oc- the molecular weight of iron. Using t2 − t1 1.0 0.3 Ma as a 2 − 1 60 60 ð Þ ð Þ tahedral titanomagnetite (dark, in the center) with small-grained Fe-bearing minerals adhering on its surface (bright spots). All images have been obtained Core 851 851Fe/Fe Fe/Fe nominalð signalÞ ð durationÞ results in a terrestrial fluence into our sed- with a JSM5900LV SEM (Jeol). nominal signal duration results in a terrestrial fluence into our sed- Blank level 848 5 848 5

) 6 Blank level

) 6 iments of Φter = 4.7iments± 1.6 × 10 atoms of Φ perter square= centimeter4.7 ± 1.6 × 10 atoms per square centimeter bacteria 2 851 2 -16 2 851 2 ð Þ 5 -16 σ ð Þ 5 1-σ u.l.u.l. blank level blank level(at/cm )andΦ = (at/cm8.8 ± 2.9 × 10)andat/cm forΦ cores 848= and8.8 851, ± 2.9 × 10 at/cm for cores 848 and 851, ter ð Þ ter ð Þ 4 4 respectively. The slightlyrespectively. different values and The larger errors slightly com- different values and larger errors compared with Table 1 resultpared from an with averaging Table over different 1 exposureresult from an averaging over different exposure times t2 − t1 ,whereasthefluencesinTable1werecalculatedforatimes t2 − t1 ,whereasthefluencesinTable1werecalculatedfora Fe/Fe (10 2 Fe/Fe (10 2 ð Þ ð Þ

60 = 60 fixed t2 − t1 = 0.8 Ma.fixed For the followingt2 − t1 discussion,0.8 we use Ma. the For the following discussion, we use the error-weightedð Þ meanerror-weighted of the fluencesð determinedÞ in mean both sediment of the fluences determined in both sediment sed 5 2 sed 5 2 0 0 cores, which is Φter =cores,5.6 ± 1.8 × which10 at/cm (taking is Φ intoter account= 5.6 ± 1.8 × 10 at/cm (taking into account 1.50 1.75 2.00 2.251.50 1.75 2.50 2.00 2.25 2.75 2.50 2.75 3.00 3.00 3.25 3.25 3.50 3.75 3.50an additional 3.75 10% uncertaintyanð additional forÞ the signal duration). 10% uncertaintyð forÞ the signal duration). Age (Ma)Age (Ma) To compare this resultTo with the compare fluence obtained this by ref. result 11 for with the fluence obtained by ref. 11 for the Pacific Ocean FeMnthe crust, Pacific several correction Ocean factors FeMn must first crust, several correction factors must first 60 Fig. 2. The Fe/Fe atom ratioFig. 2. The determined60Fe/Fe atom ratio determined in all in sediment all sediment samples samples of cores be of applied. cores Since thosebe results applied. were published, Since the half-life those of 60Fe results were published, the half-life of 60Fe (A)848and(B)851.They error bars indicate 1 statistical uncertainties. The 10 10 Brian Fields | Cosmic Rays @ APC | Dec 9, 2016(A)848and(B)851.They error bars indicateσ 1σ statistical uncertainties. The has undergone a revisionhas (2, undergone 3); the half-life of Be, a revisionwhich was (2, 3); the half-life of Be, which was Fig. S3. (A–C) Transmission electron microscopy images of magnetic extracts obtainedx axis from ranges representative were sediment chosen samplesx accordingaxis of ranges core 848 were over chosen to the theaccording 2.41-Ma availability to to the availability of of sample sample material. material. 2.62-Ma age interval showing abundant magnetofossils. All images have been obtained with a JEM2011 (Jeol). used to convert the crustused growth to rate convert to geological time, the has crust also growth rate to geological time, has also The x error bars represent coreThe x error depth bars represent of core sample depth of sample material material used used for each for data each data 60 60 point. Each data point containspoint. Each datathree point contains to six three adjacent to six adjacent individual AMS sam- AMSbeen sam-redetermined (46,been 47). A redetermined correction for the different (46,Fe 47). A correction for the different Fe ples grouped together. Theples grouped blank together. level The blank and level its and 1 itsσ 1σ upperupper limit (u.l.) limit are (u.l.)standard are samples usedstandard (which became samples available due to used advanced (which became available due to advanced shown for a chemical processingshown for a chemical blank. processing Note blank. that Note thatA Ahashas a brokena broken time axis timecross-calibration axis measurements)cross-calibration must also be taken into measurements) account. must also be taken into account. Ludwig et al. www.pnas.org/cgi/content/short/1601040113 relative to B. relative to B. 5of9 The resulting, decay-corrected,The resulting, terrestrial fluence decay-corrected, derived from terrestrial fluence derived from

Ludwig et al. Ludwig et al. PNAS Early Edition | 3of6 PNAS Early Edition | 3of6 The Moon! Lunar Soil ★ consistency check for deep- ocean signal ★ but: nontrivial background: cosmic-ray activation of lunar regolith Fe abundance Fe

60 cosmic-ray production Alan Bean, Apollo 12 (1969)

Fimiani+ 2016 PRL ★ 60Fe excess in top layer of lunar drill core 53 ★ signal (surface density) radioactive Mn abundance consistent with deep ocean

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Whodunit? The Moon as a Telescope Fry, BDF, & Ellis (2016) ★ 60Fe dust grains nearly undeflected in Solar System ★ Earth: – stratosphere scrambles ★ Moon is airless: – encodes direction! – 60Fe pattern points to source!

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Cosmic-Ray Corroboration?

★excess of antiprotons & positrons Kachelriess, Neronov, & Semikoz 2015 – requires local & recent source – d~100 pc t~2-4 Myr

Fig. 2. Mass histograms of the observed iron and cobalt nuclei. (A) The mass histogram of iron 60 nuclei detected during the first 17 ★ Fe detectedyears in in cosmic orbit is plotted. rays Clear Binns+ 2016, Binns talk peaks are seen for masses 54, 55, 56, and 58 amu, with a shoulder at - requires localmass 57 amu. & Centeredrecent at 60 amu are 15 events that we identify as source the very rare radioactive 60Fe nuclei. There are 2.95 × 105 events in the 56Fe peak. From these data Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 we obtain an 60Fe/56Fe ratio of (4.6 ± 1.7) × 10−5 near Earth and (7.5 ± 2.9) × 10−5 at the acceleration source. (B) The mass histogram of cobalt isotopes from the same data set are plotted. Note that 59Co in panel B has roughly the same number of events as are in the 58Fe peak in panel A. To the right of 59Co on April 22, 2016 there is only a single event spaced two mass units to the right of 59Co, while there are 15 events in the location of 60Fe, which is two mass units from 58Fe. This is a strong argument that most of the 15 nuclei identified as 60Fe are really 60Fe, and not a tail of the 58Fe distribution.

http://science.sciencemag.org/ Downloaded from

First release: 21 April 2016 www.sciencemag.org (Page numbers not final at time of first release) 6

Aftermath: The Local Bubble ★ The Sun lives in region of hot, rarefied gas – The Local Bubble – hot cavity ~50 pc huge

★ Nearby SN needed – we live inside SN remains – bubble requires >> 1 SN in past 10 Myr Smith & Cox 01 – 60Fe event from nearest massive star cluster? Benitez et al 00 – Bubble wall as source of ~1 Myr 60 Fe pulse width? Breitschwerdt+ 2016, Lallement and Schulreich talks

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 A Near Miss? Thomas+ 2016, Knie+ 2004, BDF+ 2005 ...but barely: "near miss" ¿ TeV cosmic-ray boost: 20x muon irradiation ¿ cosmic-ray winter? ¿ bump in extinctions? Image: Mark Garlick www.markgarlick.com If true: implications for astrobiology tightens Galactic habitable zone

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Outlook Live 60Fe seen globally and on the Moon ★ signal in deep ocean crusts, nodules, sediments find ★ confirmed pulse ~2-3 Myr ago ★ evidence for pulse at ~8 Myr ★ evidence for lunar signal ★ Source of Local Bubble?

Birth of “Supernova Archaeology" Implications across disciplines: cosmic rays, nucleosynthesis, stellar evolution, bio evolution, astrobiology

Future Research ‣ Supernova(e) origin and direction ̣ lunar distribution Thank You! ̣ cosmic-ray anisotropies ̣ neutron star/pulsar correlation ‣ more, different samples: ✓ other isotopes ✓ other media (fossil bacteria) ✓ other sites: Moon! ‣ other epochs? Mass extinction correlations? ‣ stay tuned… BDF Euro sabbatical AY 2017-2018

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016 Cosmic WMD: Rates

★ How often? Depends on how far! Shklovskii 68 ★ Rate of Supernovae inside d: 3 V (< d) d – disk 1 Galactic(< supernovad) = rate today: SN = (0.3 Myr) (1) Vdisk,total R 100pc⇥

3 Vdisk(< d) 1 d (< d) = SN = (10 Myr) (1) Vdisk,total R 30pc⇥ – corrections: spiral arms, molecular clouds,

exponential disk... Talbot & Newman 77 – multiple events < 10 pc in the last 4.5 Gyr!

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016

1 Nachbarsternsupernovaexplosionsgefahr or Attack of the Death Star! Ill efects if a supernova too close possible source of mass extinction • Shklovskii; Russell & Tucker 71; Ruderman 74; Melott group

Ionizing radiation • initial gamma, X, UV rays destroy stratospheric ozone Ruderman 74; Ellis & Schramm 94 • solar UV kills bottom of food chain Crutzen & Bruhl 96; Gehrels etal 03; Melott & Thomas groups; Smith, Sclao, & Wheeler 04 • cosmic rays arrive with blast, double whammy • ionization damage, muon radiation ~8 pc Neutrinos distance: • neutrino-nucleon elastic scattering: safe “linear energy transfer” Minimum DNA damage Collar 96, but see Karam 02

Brian Fields | Cosmic Rays @ APC | Dec 9, 2016