Neutrino decoherence from quantum gravitational stochastic perturbations

Thomas Stuttard and Mikkel Jensen Niels Bohr Institute, University of Copenhagen, DK-2100 Copenhagen, Denmark (Dated: July 2, 2020) Neutrinos undergoing stochastic perturbations as they propagate experience decoherence, damp- ing neutrino oscillations over distance. Such perturbations may result from fluctuations in space-time itself if is a quantum force, including interactions between neutrinos and virtual black holes. In this work we model the influence of heuristic neutrino-virtual interaction scenarios on neutrino propagation and evaluate the resulting signals in astrophysical and atmospheric neutrinos. We demonstrate how these effects can be represented in the framework of open quantum systems, allowing experimental constraints on such systems to be connected to quantum gravitational effects. Finally, we consider the energy-dependence of such Planck scale at energies observed in cur- rent neutrino experiments, and show that sensitivity to Planck scale physics well below the ‘natural’ expectation is achievable in certain scenarios.

I. INTRODUCTION The goal of this work is to investigate the character- istics of neutrino decoherence and other phenomena re- The mixing between neutrino mass and flavor eigen- sulting from the influence of on neutrino states produces the phenomena of neutrino oscillations, propagation, focusing on the promising case of neutrino where a neutrino produced as one flavor may be detected interactions with virtual black holes produced by space- some time later as another, and is well established exper- time fluctuations. To do so, we inject heuristic inter- imentally [1–3]. This is a quantum superposition effect action scenarios into a software implementation of neu- that is maintained over macroscopic distances due to the trino propagation to determine the resulting impact on feeble interactions between neutrinos and matter, allow- neutrino flavor transitions. ing the neutrino to propagate largely in isolation from We then demonstrate how the derived phenomena can its environment. Neutrino oscillations are thus generally be represented in the framework of open quantum sys- considered to be coherent, with the wavefunctions of two tems, which is commonly employed in experimental neu- neutrinos of identical energy travelling along an identical trino decoherence searches [11–27]. This framework is path evolving identically. very general, making constraints on the parameters of If however there is weak (and as yet undetected) cou- the open quantum system difficult to physically interpret. pling between neutrinos and the environment in which This work therefore allows neutrino decoherence experi- they propagate, the neutrinos may experience stochastic mental constraints to be directly interpreted in terms of perturbations to their wavefunctions as they travel, de- the underlying quantum gravitational phenomena con- grading or even completely destroying the coherence over sidered here. We also consider the energy dependence large distances1. By contrast, the known modifications of the physics tested, demonstrating that current experi- of neutrino oscillation probability due to the influence of ments are sensitive to Planck scale physics well below the matter2 such as the Mikheyev-Smirnov-Wolfenstein ef- ‘natural’ expectation in some scenarios, and compute the fect [5, 6] and parametric resonances [7, 8] are the net expected signal resulting from these effects in both astro- result of the influence of many matter particles on a physical and atmospheric neutrinos. propagating neutrino, producing consistent effects for all traversing neutrinos and thus preserving coherence. A stochastic environment is a frequent prediction of II. DECOHERENCE FROM STOCHASTIC quantum gravity models, with the postulated fluctuat- PERTURBATIONS arXiv:2007.00068v1 [hep-ph] 30 Jun 2020 ing nature of space-time at Planck scales (often referred to as space-time foam or ) perturbing the Neutrinos propagate as mass states, and the evolution propagating neutrino [9, 10]. Searches for neutrino de- of a relativistic neutrino mass state can be represented coherence thus potentially afford us a rare window on as a plane wave: Planck scale physics usually considered beyond the reach of current experiments. ( ) m2L |ν (L)i = exp −i j |ν (0)i , (1) j 2E j

1 This is a distinct phenomenon from wave packet decoherence [4] where |ν i is the neutrino mass state j (j = 1, 2, 3 in produced via the separation of the neutrino mass states over long j distances due to their differing masses. the 3v paradigm) of mass mj, with E being the neutrino 2 Non-Standard Interactions (NSI) also typically refers to coher- energy and L the distance travelled. ent effects on neutrino propagation resulting from interactions Neutrino mass states can be propagated according to between neutrinos and conventional matter via new forces. Equation (1), with the oscillation probability after a 2

Parameter Value )

l 1

# states 2 a

e unperturbed r

m1 0.1 eV ( perturbed √

r 0 unperturbed

m2 2m1 o

v perturbed a l

E 1 GeV f

Initial flavor να | 1 )

l 1

TABLE I. Parameters used for the propagating 2ν system. a

e 1 unperturbed r

The mass states are labelled 0, 1 and the flavor states α, β. ( 1 perturbed

0 unperturbed The parameter values are chosen to produce clear demon- s 2 s

a 2 perturbed

strations of the behaviour, rather than to represent realistic m neutrino parameters. | 1

1.0

) P( ) unperturbed Y given distance being determined by rotating the cur- P( ) perturbed 0.5 P( ) unperturbed

rent state to the neutrino flavor basis, as defined by the X

( P( ) perturbed Pontecorvo-Maki-Nakagawa-Sakata (PMNS) mixing ma- P trix [28, 29], and projecting onto the desired final flavor 0.0 state according to: 0 100 200 Propagation distance [km]

2 P (να → νβ) = | hνβ(L)|να(0)i | , (2) FIG. 1. Impact of a perturbation to the phase of the prop- agating neutrino states, where the neutrino is initially in a where α, β represent flavor indices (e, µ, τ in the 3v pure να state. The neutrino mass state phase is perturbed at paradigm). a randomized distance, in this example at L ∼ 10 km. The Decoherence can result from stochastic perturbations parameters defined in Table I are used, and the mixing an- to the mass states as they propagate, for example from gle, θ = 45°. Only the real components of the flavor/mass perturbations to the phase of one or more of the neutrino states are shown for clarity. For comparison, the dotted lines mass states. Such a phase perturbation can in included indicate the state evolution in the case of no perturbation. in Equation (1) as an additional term, δφj(L):

( 2 !) mj L 1.0 |ν (L)i = exp −i + δφ (L) |ν (0)i . (3) unperturbed j 2E j j 0.8 perturbed

) perturbed, average A example of the impact of such a phase perturbation 0.6 on the propagating neutrino states is shown in Figure 1, 3 where the perturbation to each mass state is injected ( 0.4 at a random distance, with the perturbation strength P 0.2 randomly sampled from the interval [0, 2π]. Following a perturbation, the mass states continue to evolve as be- 0.0 fore (with the same frequency and amplitude), but with 0 200 400 600 800 1000 1200 a shifted phase. When many neutrinos are considered, Propagation distance [km] the probability of each neutrino having undergone a per- turbation increases with distance, and the population be- FIG. 2. Oscillation (survival) probability for many perturbed comes increasingly out of phase. This results in damp- neutrinos, for the same system shown in Figure 1. Each light ing of the average oscillation probability with increasing red line shows the probability for a single neutrino undergo- distance, as shown in Figure 2, ultimately resulting in a ing stochastic phase perturbation, whilst the red dashed line total loss of coherence at large distances. We will show in shows the average oscillation probability of the whole popu- this work that this intuitive picture of perturbed phases, lation. The blue line shows the oscillation probability in the as well as a range of other types of stochastic pertur- absence of perturbations. The point in space at which the bations, are completely captured by the open quantum perturbation occurs is randomly chosen according to a mean system formalism of decoherence. free path of 250 km. Note that even after full decoherence, flavor transi- tions are still taking place in Figure 2 (in this case

P (να → να) ∼ 0.5 due to the maximal value of θ used), but the probability ceases to vary with time/distance. 3 Note that the phase perturbations must differ for each mass state, We will later show how the fully decohered behaviour of as neutrino oscillations are invariant to a global phase shift. the system varies for different forms of perturbation. 3

A. Neutrino perturbations from quantum gravity non-forward scattering [41]. Detector resolution also in- troduces a form of decoherence into measurements. The If gravity is a quantum force subject to the uncer- characteristics of any detected neutrino decoherence ef- tainty principle, it is hypothesised that the space-time fects must therefore be studied carefully to try to sepa- itself fluctuates at the Planck scale [9, 10, 30], often re- rate different scenarios, where strong energy-dependence ferred to as space-time foam [31, 32]. Such fluctuations in in particular may prove a useful handle for separating space-time curvature imply fluctuations in the travel dis- quantum gravity effects. tance/time between two points (e.g. the space-time met- ric), a phenomenon known as lightcone fluctuations [33– B. Modelling ν-VBH interactions 35]. In such a scenario, one might expect fluctuations in the time taken for neutrinos to propagate from a source to a detector, thus varying the neutrino mass state wave- We now evaluate the influence of ν-VBH interactions function at the point of detection. How strongly a parti- on neutrino propagation and oscillations. Given the ab- cle is influenced by Planck scale fluctuations would likely sence of an accepted model of quantum gravity, we test depend on the particle’s energy relative to the Planck a series of heuristic scenarios designed to capture the po- mass, e.g. how clearly it would ‘see’ features at this scale. tential microphysics of these interactions. Four potential At the extreme, fluctuations in the space-time foam of cases for the nature of the interaction/perturbation are sufficient magnitude could collapse to form black holes tested: of scale, which would almost immedi- Mass state selected: The interaction selects a single ately evaporate (at Planck time scales). These virtual neutrino mass state. The state is selected democratically, black holes (VBH) are analogous to the virtual - i.e. with equal probability for any state. pairs that form the phenomenon of vacuum po- Flavor state selected: The interaction selects a single larization in (QED). Neutri- definite flavor state, selected democratically as described nos encountering these black holes may experience loss for the mass state case. number is potentially of quantum information or other strong perturbations. violated in the interaction. As one example, a neutrino might be absorbed by the black hole, with the black hole subsequently evaporat- Large phase perturbation: The neutrino experiences ing/decaying to produce new particles altogether, con- large (but otherwise unspecified) perturbations to its serving only energy, charge and angular momentum (as mass state phases, which are essentially randomized. per the no hair theorem [36]) but not baryon or lepton Neutrino loss: The neutrino is lost in the interaction number. Such processes have been proposed as a source and not observed. This could result from the neutrino of decay, where the constituent of the oth- being swallowed by the black hole and either lost or re- erwise stable proton are absorbed by the VBH and re- emitted via as another (non-detected) emitted as other particles [37, 38]. Heuristically, the neu- particle type due to the lack of global symmetry conser- trino may be viewed as being stochastically absorbed and vation. An alternative picture would be that the outgo- (possibly) re-emitted by these VBH encounters during ing neutrino is simply re-emitted in a different direction propagation, with this stochasticity potentially resulting and thus not observed (particularly for a distant source). in decoherence. This is the only non-unitary case tested (where infor- Unlike lightcone fluctuations where significant effects mation is lost to the environment), and is phenomeno- would be expected to accumulate over very long propaga- logically similar to neutrino decay scenarios [42, 43] (al- tion distances [32, 39], ν-VBH interactions could produce though likely with differing energy-dependence). significant effects over more modest distances provided To determine the influence of these interactions on neu- they occur with sufficient frequency due to the poten- trino flavor transitions, we propagate neutrinos as de- tially strong perturbation experienced by the neutrino scribed in Section II and inject the interactions described during even a single VBH encounter. We focus on this above at randomised distances according to an interac- case in this work. tion mean free path. This mean free path is the lone free More mundane sources of decoherence in neutrino os- parameter of the system, and is the product of the VBH cillation measurements have also previously been identi- number density along the neutrino travel path and the fied that must not be confused with the effects of quan- interaction cross section. For each interaction scenario tum gravity or other new physics. For example, the we propagate many individual neutrinos and compute spatial extent of the neutrino source or other cases of the average behaviour of the neutrino ensemble. The re- variations in the source-detector distance in a neutrino sulting neutrino survival probabilities versus distance are experiment can produce decoherence effects, and occurs shown in Figure 3. As in Section II, a 2 flavor system is for instance due to variations in the height of cosmic ray shown with toy model parameters chosen for clarity. In air showers producing atmospheric neutrinos. Addition- particular, a non-maximal mixing angle θ is used. A three ally, conventional neutrino-matter effects feature some flavor system with realistic parameters is shown later in degree of decoherence, for example from sub-structure Section III A. in the Earth’s internal density distribution [40] or the In all scenarios, we observe the damping of the aver- 4

Unperturbed Averaged oscillations 1:1 flavors turbed final states are independent of the initial states 1.0 and thus the perturbations totally eliminate coherence for that neutrino. The damping rate is thus purely de-

) fined by the interaction mean free path, which controls the fraction of the neutrino ensemble that have experi- 0.5 enced one (or more) interactions after a given distance.

( -VBH: Mass state selected

P Individual All cases show clear deviation from standard (unper- Average turbed) oscillations, and can therefore be searched for 0.0 experimentally. 1.0 We now discuss the individual scenarios in more detail. For the ‘neutrino loss’ case, the neutrinos follow standard ) oscillation behaviour until they undergo an interaction, 0.5 at which point the neutrino is lost and the transition probability (to any flavor/state) immediately drops to 0.

( -VBH: Flavor state selected

P Individual The long distance behaviour of both individual neutri- Average nos and the ensemble is thus P (νX → νX ) = 0, where 0.0 νX represents any neutrino flavor. We note that although 1.0 this case shares the phenomenological characteristics of neutrino decoherence (and we will later see it can be ex- ) pressed in the same mathematical framework), it is not 0.5 strictly a form of decoherence as neutrinos are removed from the system, rather than losing coherence with the

( -VBH: Phase perturbation P Individual population. Average In the ‘phase perturbation’ scenario, once the neutrino 0.0 undergoes an interaction it becomes out of phase with 1.0 the neutrino population, but continues to oscillate. At large distance, eventually all neutrinos have experienced ) an interaction and coherence is totally lost in the en- 0.5 semble, resulting in an averaging of the oscillation be- -VBH: Neutrino lost haviour of the system. The long distance flavor transition ( Individual P Average probability is thus the averaged oscillation probability, P 2 2 P (να → νβ) = |Uαj| |Uβj| (where U is the PMNS 0.0 j 0 200 400 600 800 1000 1200 mixing matrix), and is mixing angle dependent. Propagation distance [km] In both the ‘mass state selection’ and ‘flavor state se- lection’ scenarios, the result of an interaction is that the FIG. 3. Neutrino flavor transition probability resulting from system is forced to align with a particular state vector, various ν-VBH interaction scenarios. Both individual neutri- in the mass or flavor basis respectively. This can be seen nos (translucent coloured lines) and the average behaviour of in the upper two panels of Figure 3, where the individ- the ensemble (opaque dashed coloured lines) are shown (note ual neutrinos separate into two distinct populations cor- that only the ensemble behaviour is observable). Note that responding to the two mass/flavor states in the system in some regions of the plots, many overlaid translucent lines (this would be three populations in a three neutrino sys- result in solid coloured lines. The neutrinos are initially in a tem). Neutrinos may switch between populations as they pure να state. A two flavor system is shown using the toy pa- continue to propagate and potentially encounter further rameters in Table I, with a non-maximal mixing angle θ = 20°. VBHs. The long distance behaviour of the ensemble in The interaction mean free path is 250 km. both these cases is equal numbers of neutrinos in each population, resulting in an average transition probability 1 of P (νX → νX ) = N (where N is the number of neu- age oscillation probability of the neutrino ensemble that trino states considered, in this case N = 2), independent is characteristic of neutrino decoherence, ultimately re- of the neutrino mixing angle(s). The behaviour of the en- sulting in a distance-independent flavor transition prob- semble is identical regardless of whether a mass or flavor ability at large distances. The main difference between state is selected in the interaction, and thus these cases the cases is the behavior at large distances, e.g. when the are indistinguishable through a neutrino oscillation mea- neutrino population has fully lost coherence or when all surement. For the flavor state case, individual neutrinos neutrinos have been lost (these differences are discussed continue to oscillate following the interaction (although in more detail below). These differences could poten- out of phase with each other since the interaction takes tially be used to discriminate between the scenarios in place at an random phase in the wavefunction evolution). the event of an experimental signal being observed. For the mass state case however, the selection of a sin- The rate of damping is identical in all cases, as the per- gle mass state destroys the superposition effect caused 5 by the co-evolution of multiple mass states that causes tional coherent matter effects appear in the Hamiltonian. the time-dependent oscillatory characteristics of the fla- The neutrino flavor transition probability can be deter- vor transitions. Flavor transitions are still possible for mined by: individual neutrinos though, e.g. P (νX → νX ) 6= 0, 1, due to the mixing of mass and flavor states, but in a time-independent manner. P (να → νβ) = Tr[ρα(t)ρβ(0)]. (6) An interesting observation from this study is that the ‘phase perturbation’ case can appear similar or even iden- The form of D[ρ] is dependent on the underlying physics producing the decoherence effect. A generalised tical to the other cases in certain να → νβ channels for particular mixing angles. For example, in a two state form of D[ρ] is [14, 45, 46]: system with maximal mixing (θ = 45°), the ‘phase per- turbation’ scenario produces identical ensemble damping N 2−1 effects to the ‘mass/flavor state selection’ scenarios (with 1 X  † †  D[ρ] = − [Vk, ρVk ] + ([Vkρ, Vk ] , (7) large distance behaviour of P (να → νβ) = 0.5). This is a 2 k good approximation of e.g. high-energy atmospheric neu- trino oscillations, limiting the distinguishability of these where N is the dimensionality of the SU(N) Hilbert scenarios in such cases. space defining the system (SU(3) for a system with 3 Now we have demonstrated the resulting signal for four neutrino flavors) and Vk are N × N complex matrices. ν-VBH interaction scenarios by injecting perturbations The general D[ρ] form shown in Equation (7) in prin- into a software model of neutrino propagation, we will ciple allows model-independent decoherence searches to now look to represent this physics in the open quantum be performed, but in practise contains far too many free system formalism often used to represent neutrino deco- parameters to be realistically testable. Studies have con- herence. sidered only a small number of non-zero parameters, ei- ther selected for simplicity or to target some particular physics case. Here, we seek to reproduce the effects of III. DECOHERENCE IN OPEN QUANTUM the ν-VBH interaction scenarios investigated in Section SYSTEMS II B using this open quantum system formalism. It is common to expand the D[ρ] operator in terms A neutrino coupled to its environment can be treated of the basis vectors, bµ of the SU(N) space defining the using an open quantum system formalism. Although the system [12, 13, 45, 46]: neutrino may be produced in a known state, the ran- dom nature of the perturbations discussed in this work µ mean that the observer becomes increasingly ignorant of D[ρ] = cµb , (8) the neutrino’s state as it propagates. The state can then only be expressed as an ensemble of possible states, each where cµ ≡ (D[ρ])µ, e.g. the µ’th coefficient of the D[ρ] with an associated probability, known as a mixed quan- expansion. The Einstein summation convention is used tum state. In the language of open quantum systems, here. decoherence is thus the transition from an initial pure For a 3 neutrino system, bµ are given by the SU(3) gen- quantum state to a mixed quantum state. erators, the Gell-Mann matrices, and the identity matrix: Mixed (and pure) quantum states can be mathemat- ically expressed using the density matrix formalism,       where the density matrix, ρ, for a system of j states of 1 0 0 0 1 0 0 −i 0 probability p is given by: b0 = 0 1 0 , b1 = 1 0 0 , b2 = i 0 0 , j 0 0 1 0 0 0 0 0 0       X 1 0 0 0 0 1 0 0 −i ρ = pj |ψji hψj| . (4) b3 = 0 −1 0 , b4 = 0 0 0 , b5 = 0 0 0  , j 0 0 0 1 0 0 i 0 0       The density matrix for a pure quantum state is thus 0 0 0 0 0 0 1 0 0 b = 0 0 1 , b = 0 0 −i , b = √1 0 1 0 . ρ = |ψi hψ|. Density matrices are suitable for describing 6   7   8 3   both the state of a single neutrino and an ensemble. 0 1 0 0 i 0 0 0 −2 The time evolution of an open quantum system ex- periencing decoherence is given by the Lindblad master To define the free parameters, we can express the de- equation [44]: coherence operator as:

ν µ ρ˙ = −i[H, ρ] − D[ρ], (5) D[ρ] = (Dµν ρ )b , (9) where H is the Hamiltonian of the system and D[ρ] is where ρν are the coefficients of the system’s density ma- µ an operator defining decoherence in the system. Conven- trix expanded in the SU(N) basis (e.g. ρ = ρµb ), and 6

2 2 Dµν are the elements of a (N × N ) matrix whose ele- All cases ultimately depend on a single free parameter, ν ments are the free parameters of the system (Dµν ρ = cµ the ν-VBH interaction mean free path, and thus we also as defined in Equation (8)). For a 3 neutrino system, D seek a single free parameter in the open quantum system is defined as4: description for each case. The following three D matrices reproduce the ν-VBH interaction cases in this work:   Γ0 β01 β02 β03 β04 β05 β06 β07 β08   β01 Γ1 β12 β13 β14 β15 β16 β17 β18 0 0 0 0 0 0 0 0 0   β02 β12 Γ2 β23 β24 β25 β26 β27 β28 0 Γ 0 0 0 0 0 0 0     β03 β13 β23 Γ3 β34 β35 β36 β37 β38 0 0 Γ 0 0 0 0 0 0     D = β04 β14 β24 β34 Γ4 β45 β46 β47 β48 , 0 0 0 Γ 0 0 0 0 0 β β β β β Γ β β β     05 15 25 35 45 5 56 57 58 Dstate selected = 0 0 0 0 Γ 0 0 0 0 , (11) β β β β β β Γ β β  0 0 0 0 0 Γ 0 0 0  06 16 26 36 46 56 6 67 68   β β β β β β β Γ β  0 0 0 0 0 0 Γ 0 0 07 17 27 37 47 57 67 7 78   β08 β18 β28 β38 β48 β58 β68 β78 Γ8 0 0 0 0 0 0 0 Γ 0 (10) 0 0 0 0 0 0 0 0 Γ where the diagonal parameters are indicated by Γµ and the off-diagonal elements by βµν (all are real scalars). Although there are a large number of free parameters 0 0 0 0 0 0 0 0 0 in D, fairly general conditions such as probability and 0 Γ 0 0 0 0 0 0 0   energy conservation can be imposed to reduce this ma- 0 0 Γ 0 0 0 0 0 0   trix [14, 45, 46]. For example, elements in the 0’th row 0 0 0 0 0 0 0 0 0   and column of D (those corresponding to the identity Dphase perturbation = 0 0 0 0 Γ 0 0 0 0 , (12) 0 0 0 0 0 Γ 0 0 0 matrix) must be zero for a unitary system where no prob-   ability is lost from the neutrino to the environment [12], 0 0 0 0 0 0 Γ 0 0   and thus are often omitted. Ultimately, the parameter 0 0 0 0 0 0 0 Γ 0 values are chosen to represent the particular physics case 0 0 0 0 0 0 0 0 0 of interest, or in some works a minimal set of non-zero pa- rameters is (often somewhat arbitrarily) chosen to allow the formalism to be tested against experimental data. Γ 0 0 0 0 0 0 0 0 0 Γ 0 0 0 0 0 0 0   0 0 Γ 0 0 0 0 0 0   A. Representing ν-VBH interactions in the open 0 0 0 Γ 0 0 0 0 0 quantum system formalism   Dneutrino loss = 0 0 0 0 Γ 0 0 0 0 , (13) 0 0 0 0 0 Γ 0 0 0   0 0 0 0 0 0 Γ 0 0 Now that we have a formalism for characterizing the   influence of the environment on neutrino propagation 0 0 0 0 0 0 0 Γ 0 within the context of an open quantum system, we seek 0 0 0 0 0 0 0 0 Γ to represent the ν-VBH interaction scenarios (specifically the average behaviour of the ensemble) examined in Sec- where in all cases there is a single non-zero free param- tion II in this framework by choosing appropriate forms eter, Γ, which has units of the inverse of distance, or for D. equivalently energy. All ν-VBH interaction scenarios tested in this work To understand these D matrices, it is useful to consider produce exponential damping behavior of the form e−αL, the resulting form of D[ρ]. Ultimately D[ρ] is a N × 5 where α represents a damping constant. Inspection of N matrix , e.g. 3 × 3 for a three neutrino system. It Equation (5) therefore implies D[ρ] terms of the general can be shown from Equation (9) that Γ3,8 determine the form αρ. The damping constants will be specified in the diagonal elements of D[ρ], whilst Γ1,2,4,5,6,7 determine the D matrix. off-diagonal elements. Where the scenarios differ is the large distance flavor In the mass basis, standard neutrino oscillations transition probability they tend to after full decoherence are driven by a diagonal H (resulting from non-zero or neutrino loss. The ‘mass state selected’ and ‘flavor mass splittings). These oscillatory terms appear as state selected’ cases produce identical results for the en- off-diagonal elements in the standard evolution term semble, and thus can be represented by a single D matrix. i[H, ρ] in Equation (5), and thus oscillations cause time- dependence in the off-diagonal elements of ρ. These

4 Care must be taken when comparing D between different studies, as the elements depend on the choice (and order) of basis vectors 5 This is can be seen in Equation (5), where it is evident that D[ρ] in which they are defined. has the same dimensions as H and ρ. 7 off-diagonal ρ elements are damped to zero by non-zero spond to equal populations of the mixed states. This Γ1,2,4,5,6,7, damping the oscillations but preserving the scenario is thus also sensitive to interactions selecting any diagonal ρ elements that yield the PMNS matrix de- new neutrino basis states (unrelated to the weak nuclear pendence of the large distance behaviour observed in force) resulting from new physics. the ‘phase perturbation’ scenario. Non-zero Γ3,8 instead It is useful to note that the D[ρ] operator resulting produce damping in the diagonal (non-oscillatory) ρ el- from Equation (12) is: ements, which tend to the value 1/N. In combination with the damped off-diagonal elements resulting from  0 Γρ Γρ  non-zero Γ1,2,4,5,6,7, this produces the 1/N large distance 10 20 behaviour observed for the ‘state selection‘ cases6. Fi- D[ρ] = Γρ01 0 Γρ21 , (14) Γρ Γρ 0 nally, the addition of non-zero Γ0 causes the diagonal ρ 02 12 elements to damp to 0 (instead of 1/N). In this case, all ρ elements tend to 0, resulting in the non-unitary ‘neutrino which is a common form that has been explored in the lit- loss’ scenario. erature [11, 15, 16], and so these limits can be interpreted in terms of the ν-VBH ‘phase perturbation’ interactions More generally, we note that the ‘state selection’ case considered here. More generally, the mapping of a diago- will represent state selection in any basis, as for unitary nal D matrix to D[ρ] when expressed as an N ×N matrix mixing an equal population of mass states must corre- is given by:

  Ω0 + Ω3 + Ω8 Γ1 Re{ρ01} − iΓ2 Im{ρ10} Γ4 Re{ρ02} − iΓ5 Im{ρ20}     D[ρ] =   , (15) Γ1 Re{ρ01} + iΓ2 Im{ρ10} Ω0 − Ω3 + Ω8 Γ6 Re{ρ12} − iΓ7 Im{ρ21}     Γ4 Re{ρ02} + iΓ5 Im{ρ20} Γ6 Re{ρ12} + iΓ7 Im{ρ21} Ω0 − 2Ω8

where the Ωµ terms are given by: Since the ν-VBH interaction cases considered in this work produce a total loss of coherence after a single in- teraction (e.g. the final state is independent of the initial Γ 0 state), Lcoh is equal to the interaction mean free path, Ω0 = (ρ00 + ρ11 + ρ22) , 3 and experimental constraints on Γ (and thus Lcoh) can Γ3 therefore be directly interpreted as constraints on the Ω = (ρ − ρ ) , (16) 3 2 00 11 mean free path of ν-VBH interactions. Γ Ω = 8 (ρ + ρ − 2ρ ) . 8 6 00 11 22

This mapping7 is useful for comparing forms of D[ρ] expressed with and without the SU(N) expansion de- scribed in Section III. To verify the D matrices in Equations (11) to (13) and The D matrices given by Equations (11) to (13) pro- −ΓL also the assertion that Lcoh can be interpreted as the ν- duce damping terms of the form e . To attribute phys- VBH interaction mean free path, in Figure 4 we show the ical meaning to the value of Γ, we define the coherence oscillation probabilities computed using both the open length, Lcoh, of the ensemble as the distance at which −1 quantum system formalism and by injecting perturba- damping terms have reached e , which implies: tions into our neutrino propagation model (as described in Section II B). A 3 neutrino system with realistic os- 1 cillation parameters is shown, with the injected coher- L = . (17) ence length shown being of relevance to quantum gravity coh Γ searches with atmospheric neutrinos. We observe per- fect agreement between the two approaches in all cases, and conclude that the open quantum system models pre- sented in this section do indeed correctly represent the ν- 6 The damping of non-oscillatory elements of ρ is sometimes re- ferred to as neutrino relaxation in the literature [24, 27]. VBH interaction scenarios investigated, and can be used 7 The specific case of Equation (14) results when Γ1,2,4,5,6,7 = Γ to experimentally search for quantum gravity. The open and Γ0,3,8 = 0, in addition to the properties Re{ρij } = Re{ρji} quantum system model is implemented in the nuSQuIDS and Im{ρij } = − Im{ρji}. software package [47, 48], and is solved numerically. 8

As an aside, an interesting observation is that the case No perturbation Standard oscillation (Lindblad) Phase perturbed (perturbation) Phase perturbed (Lindblad) of n = −1 coupled with the ‘neutrino loss’ D matrix State selected (perturbation) State selected (Lindblad) shown in in Equation (13) produces a signal that is phe- Neutrino loss (perturbation) Neutrino loss (Lindblad) Averaged oscillations Equally populated flavors nomenologically identical to neutrino decay with invisi- 1.0 ble decay products (where the energy-dependence results )

e from time dilation). Noting that Γ has units of energy, Equation (18) can 0.5 be re-written to express the Γ parameters with respect

( to an arbitrary energy scale, Λ: P 0.0 1.0 En

) Γ(E) = ζ , (19) Λn−1 0.5 where ζ is a dimensionless constant, and is a free pa- (

P rameter characterising the strength of the decoherence 0.0 effects. When considering decoherence from quantum grav- 1.0

) ity, the energy scale of interest is the Planck mass, 19 Λ ∼ MPlanck ' 1.2 × 10 GeV, and thus Γ can be ex- 0.5 pressed relative to the Planck scale as [19]: ( P 0.0 n 0 2 4 6 8 10 12 14 E Γ(E) = ζPlanck n−1 . (20) L/L MPlanck

Using Equation (17), Equation (20) can also be ex- FIG. 4. Oscillation probability resulting from ν-VBH inter- pressed as an energy-dependent coherence length relative actions, computed both by perturbing propagating neutrinos to the Planck length, L : and using the Lindblad open quantum system formalism. A Planck 3 neutrino system is shown in vacuum with the oscillation pa- rameters in Table II. L⊕ is the diameter of Earth. Both the  n LPlanck MPlanck ν-VBH interaction mean free path and 1/Γ are set to 3L⊕. Lcoh(E) = . (21) ζPlanck E

IV. ENERGY-DEPENDENCE OF Equation (21) yields physical insight into this energy- DECOHERENCE FROM PLANCK SCALE dependence parameterisation. From it, we see that a neu- PHYSICS trino with E = MPlanck would have a coherence length of −1 −1 ζPlanck Planck lengths, regardless of n. ζPlanck can thus be interpreted as the neutrino coherence length at the The general open quantum system approach out- n lined in Section III does not implicitly consider the en- Planck scale, whilst the (MPlanck/E) term encodes the ergy dependence of the physics producing the decoher- suppression of the decoherence effects at neutrino ener- ence effects, i.e. the ν-VBH interactions in this work. gies below from the Planck scale. In general, theories of This can be introduced however by defining the energy- quantum gravity predict significant effects at the Planck dependence of the free parameters in D. As previ- scale with large suppression at lower energy scales, which ously stated, there is currently no generally accepted the- can be represented using Equations (20) and (21) when ory of quantum gravity, and so we instead take a phe- n > 0. As such only positive n are considered for the nomenological approach and introduce a general form remainder of this section. for the energy-dependence of the Γ parameter control- A ‘natural’ Planck scale theory is expected to have ling the decoherence effects. A common approach in ζPlanck ∼ O(1) [19]. Figure 5 shows the coherence length the literature has been to assume a power-law energy- as a function of neutrino energy predicted by Equa- dependence [14, 22, 24, 27]: tion (21) under this assumption of naturalness for a range of n. For all n tested, coherence length decreases with increasing neutrino energy as the suppression of Planck  E n scale effects at low energies diminishes, and ultimately all Γ(E) = Γ(E0) , (18) E0 cases converge at the Planck scale, where the coherence length ∼ LPlanck. Lower n produces smaller coherence where E0 is a reference energy pivot and n is the power- lengths (e.g. stronger decoherence effects) at any given law index. Studies often test multiple cases for n, rather sub-Planck energy, as lower n represents weaker suppres- than assuming a specific model. sion. 9

n = 1 Earth diameter ural’ case of ζPlanck ∼ 1 implies Lcoh ∼ LPlanck for a n = 2 Earth-Sun distance neutrino with Planck scale energies, which according to n = 3 Milky Way diameter n = 4 Observable Universe the conclusions derived in Section III A implies a ν-VBH 1040 interaction occurs, on average, every Planck length trav- Planck = 1 1030 elled by a Planck scale neutrino. If ν-VBH interactions are less frequent than this, it would imply ζPlanck < 1 20 10 and thus weaker signals at the energies probed by neu- 10 ] 10 trino experiments, potentially evading detection thus far. m

[ Ultimately, ζPlanck is a free parameter that must be 100 h

o measured or constrained using experimental data. Ex- c 10 L 10 perimental constraints on Γ(E0) from analyses using M 10 20 Planck the energy-dependence parameterisation given by Equa- tion (18) can be converted to ζPlanck as follows: 10 30 LPlanck 10 40 100 106 1012 1018 1024 1030 n−1 MPlanck E [eV] ζPlanck = Γ(E0) n . (22) E0

8 −27 FIG. 5. Neutrino coherence length versus neutrino energy For example, the limit of Γ(E0) < 0.9 × 10 GeV resulting from a ‘natural’ (ζPlanck = 1) Planck scale source (n = 2) derived using data from the SuperKamiokande −8 of decoherence, for a range of possible n cases. Reference experiment [14] corresponds to ζPlanck < 1.1 × 10 . distance scales are shown for comparison.

V. DECOHERENCE IN ASTROPHYSICAL AND A number of reference distances are shown for compar- ATMOSPHERIC NEUTRINOS ison to the predicted natural coherence lengths in Fig- ure 5. We see that the weakest suppression, i.e. n = 1, Now that we have mathematical definitions for neu- case predicts microscopic coherence lengths for all ener- trino decoherence and other effects resulting from ν-VBH gies probed by neutrino experiments, and is thus strongly interactions, including their energy-dependence, we can excluded at this natural scale by the non-detection of evaluate the resulting potential signals in neutrino de- neutrino decoherence in any experiment to date. Note tectors. Given that decoherence effects accumulate over that exclusion at the natural scale does not exclude the distance (until coherence is fully lost) and that Planck model outright, but constrains it to ζPlanck < 1 (by many scale physics is expected to be suppressed at energies be- orders of magnitude in this case). At the other extreme, low the Planck scale, decoherence effects from quantum if n = 4 then neutrinos of up to ∼TeV energies have nat- gravity are expected to manifest most strongly in neutri- ural coherence lengths larger than the observable Uni- nos with high energies and long propagation baselines. verse, making decoherence effects essentially unobserv- The diffuse extragalactic high-energy neutrino flux dis- able. However, significant decoherence would occur for covered by the IceCube neutrino observatory [49] initially the high-energy extragalactic neutrino flux observed by seems an ideal hunting ground for such physics. Neu- neutrino telescopes such as IceCube [49] that extends into trinos of up to ∼PeV energies have been observed, and the PeV and even EeV range (detectable by radio neu- evidence has been found of neutrinos travelling ∼Gpc trino detectors and cosmic ray air shower detectors in the distances [53, 54]. The very fact that neutrinos from case of Earth skimming neutrinos). such distances have been observed at all significantly The case of n = 2 is of particular interest as this constrains the ‘neutrino loss’ scenario considered in this energy-dependence has been predicted by work probing work, but quantitative statements are however difficult quantum decoherence effects in string theory models, in- without a detailed knowledge of the nature and distri- cluding for particles encountering black holes in four di- bution of sources, not to mention the neutrino flux they mensions [50] and D-brane foam backgrounds [51, 52]. produce. We see from Figure 5 that a natural n = 2 Planck scale However, there is another fundamental limitation in theory predicts a neutrino coherence length smaller than observing ν-VBH interactions from the diffuse astro- the Earth’s diameter for E &10 MeV, and ∼1 km at physicical neutrino flux. Due to the large and unknown 1 GeV. Such strong decoherence effects have not been travel distances, as well as broad energy distributions and observed by long baseline accelerator and atmospheric finite detector resolution, the neutrinos are observed at neutrino experiments, constraining any such theory well Earth in an oscillation-averaged state [55]. This however below the natural scale (ζPlanck  1). We can comment on what a notional ‘natural’ theory really represents in the case of the ν-VBH interactions considered in this work. From Equation (21), the ‘nat- 8 Note that this result considers only 2 neutrino flavors. 10

and have a compact neutrino emission region. However, (1:0:0) Standard oscillations the identification of such sources at high energies, either (1:2:0) -VBH: Phase perturbation galactic or extragalactic, is still in its infancy, with only a (0:1:0) -VBH: State selection single source being identified thus far with O(10) neutri- 0 nos associated with it [53]. Additionally, neutrinos identi- 100 fied as originating from distant point-like sources are typ-

20 ically νµ events producing long muon tracks in detectors 80 in charged-current interactions, as only these have suffi- cient direction resolution. Neutrino source identification ] 40 % f is thus not sensitive at present to other neutrino flavours, [ r 60 a n c unless observed via time coincidence with a short time- o t i i t o scale flaring event. Coupled with the large uncertain- c 60 n a 40 [ r % ties in modelling the source neutrino flux (both steady- f ] state and in transient flaring emission periods) and finite 80 detector resolution, it is not yet possible to do robust 20 quantitative searches for new physics based on the flavor

100 composition of neutrino point source emission. Upper 0 bounds on the mean free path of the ‘neutrino loss’ sce- 0 20 40 60 80 100 nario however can be derived from the fact that neutrinos e fraction [%] are observed at all from a source of known distance [56]. The Sun or galactic supernovae both represent other FIG. 6. Astrophysical neutrino flavor triangle showing the astrophysical neutrino source candidates, but have neu- ratios of each neutrino flavor expected at Earth from the dif- trino emission of O(MeV), limiting sensitivity to Planck fuse astrophysical neutrino flux, for a range of possible initial scale effects. Solar atmospheric neutrinos [57–61] are source flux flavor ratios. Three different possible source fla- higher energy, but again their study is in its infancy. vor ratios are shown: (νe : νµ : ντ ) = (1 : 0 : 0) (neutron decay), (1 : 2 : 0) (pion decay), and (0 : 1 : 0) (muon-damped Closer to home, atmospheric neutrinos do offer a com- pion decay). Both the standard oscillation expectation and pelling source for neutrino decoherence studies. They of- ν-VBH decoherence cases are shown. Oscillation parameters fer a copious and relatively well understood flux of neu- from Table II are used. trinos reaching TeV energies, travel distances of up to ∼12700 km (the Earth’s diameter, L⊕) - more than an order of magnitude greater than any existing accelerator is also precisely the long distance result of the ‘phase per- experiment - and high statistics samples can be collected turbation’ ν-VBH interactions described in this work. A by large underground detectors. The range of baselines fully decohered diffuse astrophysical neutrino flux is thus and energies detected also allows the damping or energy- indistinguishable from the no-decoherence expectation. dependence of any detected signal to be explored. Whilst This is shown in Figure 6, which shows the expected ter- the distances involved are unlikely enough to produce de- restrial neutrino flavor ratio (presented as a flavor trian- tectable effects from e.g. lightcone fluctuations due to the gle) for a number of different source flux cases. fluctuating space-time metric, if some fraction of the neu- Figure 6 also demonstrates a major challenge in mea- trinos detected encounter VBHs as they cross the Earth suring the ‘mass/flavor state selection’ ν-VBH interac- a measurable signal could be produced. tions described in this work with diffuse astrophysical Figure 7 shows the ζPlanck value that produces deco- neutrinos. The long range behaviour in this case is herence with an Earth diameter scale neutrino coherence equally populated neutrino flavors, which produces a length, for a range of neutrino energies relevant for atmo- spheric neutrino experiments up to ∼100 TeV where the 1:1:1 flavor ratio at the Earth (assuming democratic fla- 9 vor selection and full loss of coherence) regardless of ini- astrophysical flux starts to dominate . We see that at- tial flux. This is almost identical to the standard oscil- mospheric neutrinos are sensitive to Planck scale physics lation expectation for a pion decay source, and thus this many orders of magnitude below the natural expectation case is also indistinguishable from the standard expecta- for the n = 1, 2 energy-dependence cases, and have lim- tion with our present level of uncertainty as to the mech- ited sensitivity to n = 3 too at the highest energies. anisms producing the astrophysical neutrino flux. More The atmospheric neutrino oscillation probability in the differentiable signals could exist for coherence lengths ap- presence of ‘state selection’ ν-VBH interactions is shown proximately commensurate with the neutrino propaga- tion distance, where coherence would not be completely lost at the Earth, or for undemocratic flavor scenarios. 9 We note that inverting the self-veto techniques [62, 63] used to re- Neutrinos from identified astrophysical objects (‘point move atmospheric neutrinos from astrophysical neutrino searches sources’) could in principal offer sensitive searches for in neutrino telescopes could provide an enriched high-energy at- neutrino decoherence, provided that they have a well un- mospheric neutrino sample above 100 TeV for decoherence stud- derstood initial neutrino flux, a well measured distance, ies. 11

n = 1 n = 2 n = 3 n = 4 1040 Lcoh = Earth diameter 1030 1020

k 10 c 10 n a l Natural ( Planck = 1) P 100 10 10 10 20 10 30 10 1 101 103 105 E [GeV]

FIG. 7. ζPlanck producing Earth diameter scale coherence lengths, shown for neutrinos energies relevant for atmospheric neutrinos and for a range of possible n cases. in Figure 8 for the well-motivated n = 2 case. The dom- inant νµ disappearance channel is shown, as a function of the neutrino energy and (cosine of the) zenith angle; a proxy for atmospheric neutrino baseline. The oscilla- tion parameters used throughout this section are given in Table II. Conventional matter effects are included as a potential in the Hamiltonian, with the matter density defined according to the Preliminary Reference Earth Model (PREM) [64]. The neutrino evolution is solved in the mass basis (using nuSQuIDS), and thus is not sub- FIG. 8. (Upper) Atmospheric νµ survival probability in ject to issues introduced by approximations employed in the presence of ‘flavor randomizing’ ν-VBH interactions with other works (see [18] for details). Lcoh(1 TeV) = L⊕ and n = 2. Oscillation parameters The n > 0 cases motivated by Planck scale physics from Table II are used. (Lower) The difference in oscilla- produce the strongest signals at higher energies, as can tion probability w.r.t. the standard (no decoherence case), be seen for the n = 2 cases in Figure 8. In this example, ∆P = Pdecoherence − Pstandard. ζPlanck is chosen such that a 1 TeV neutrino has a coher- ence length of one Earth diameter, producing a strong Parameter Value signal in a neutrino telescope such as IceCube and no sig- 2 −5 2 ∆m21 7.39 × 10 eV nificant signal in lower energy atmospheric or long base- 2 −3 2 ∆m31 2.528 × 10 eV line accelerator experiments. Full decoherence—in this θ 33.82 1 12 ° case equally populated flavors, i.e. P (νµ → νe,µ,τ ) = 3 — θ13 8.60° results at high energies, even for down-going neutrinos θ23 48.6° (cos(θzenith) = 1) travelling only O(10 km). This would δCP 221° be an extremely strong signal clearly visible in the Ice- Cube detector, and thus sensitivity to even significantly TABLE II. Neutrino oscillation parameters used for evaluat- weaker effects could be achieved. Note that at the highest ing atmospheric neutrino oscillations, taken from NuFit 4.1 energies and longest distances, Earth absorption becomes global fit results (normal mass ordering, SuperKamiokande significant, reducing the sensitivity to decoherence in pre- data included) [65]. cisely the region at which the signal would be strongest. For comparison, the ‘phase perturbation’ and ‘neutrino loss’ ν-VBH interaction cases for atmospheric neutrinos different apparent strengths. In particular the ‘phase per- are shown in Figure 9. As expected, similar trends are turbation’ and ‘state selection‘ cases are similar due to seen to the ‘mass/flavor state selection’ case but with dif- maximal mixing in the atmospheric sector. fering P (νµ → νµ) once full decoherence is reached, as Varying n changes the energy at which the decoher- was seen in Figure 4 in 1-dimension. Distinguishing be- ence effects become significant. For comparison, in Fig- tween the different ν-VBH cases may be difficult, as they ure 10 we also show the decoherence signal resulting from produce similar signatures to each other, although with energy-independent decoherence. Here the effects present 12

FIG. 9. Atmospheric νµ survival probability in the presence FIG. 10. Plot analogous to Figure 8 but with n = 0 (energy- of ‘phase perturbation’ (upper) and ‘neutrino loss’ (lower) ν- independent decoherence) and Lcoh(∀E) = L⊕. VBH interactions with Lcoh(1 TeV) = L⊕ and n = 2 (same parameters as for Figure 8). differentiable modifications to neutrino oscillation prob- ability. most strongly at long baselines but across all energies, re- We have shown how the ν-VBH interaction scenar- sulting in weaker oscillations in the standard oscillation ios considered can be represented in the formalism of region but introducing flavor transitions at higher ener- open quantum systems, allowing experimental decoher- gies where there are none in the standard picture. This ence constraints to be directly interpreted in terms of case would likely not be identified by long baseline accel- the mean free path of ν-VBH interactions. In turn such erator experiments due to their smaller baselines, with results can constrain the VBH number density and the ν- the corresponding oscillation probabilities for the NOνA VBH interaction cross section. The heuristic interaction experiment [66] shown in Figure 11 deviating little from cases considered are relatively simple, and can potentially the ‘no decoherence’ expectation. represent other new physics affects producing perturba- tions to propagating neutrino phases and other proper- ties. Tantalisingly, we see that depending on the energy- VI. SUMMARY AND CONCLUSIONS dependence of the suppression of Planck scale physics at the energies our experiments operate at, sensitivity to In this work we have seen that perturbations to propa- Planck scale theories well below the ‘natural’ scale can gating neutrino states can cause neutrino decoherence be achieved with current experiments, motivating further and other modifications to neutrino flavor transitions, experimental searches. with stochasticity of the perturbation being a key in- Finally we have shown that high-energy diffuse astro- gredient for decoherence. Quantum gravity is postulated physical neutrinos are (counter-intuitively) of somewhat to provide a fluctuating environment that could induce limited use in the search for ν-VBH interactions due to such effects in neutrino propagation, and we have evalu- degeneracies with the standard flux expectation at the ated the impact of a range of heuristic ν-VBH interaction Earth, but that atmospheric neutrinos do however still scenarios, showing varied and potentially detectable and offer power to probe decoherence signals many orders of 13

magnitude below the natural Planck scale for cases moti- 1.0 Standard oscillations NO A vated by string theory. With a new generation of atmo- 0.8 -VBH: Phase perturbation spheric neutrino experiments currently under construc- -VBH: Mass/flavor selected tion [67–69], the prospects for Planck scale experimental ) -VBH: loss 0.6 physics are looking bright.

( 0.4 P

0.2 ACKNOWLEDGEMENTS 0.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 E [GeV] The authors thank Jason Koskinen, Markus Ahlers, Car- los Arg¨uellesand Ben Jones for feedback on the paper

FIG. 11. νµ survival probability versus neutrino energy at the draft, and these same people plus Subir Sarkar, Jo˜ao NOνA experiment, shown for various ν-VBH interactions de- Coelho, Mauricio Bustamante, Shashank Shalgar, Mo- coherence cases with Lcoh = L⊕ and n = 0. Parameters from hamed Rameez, Peter Denton and Pilar Coloma for valu- Table II are used, and the experiment’s baseline is 810 km. able conversations. This work is supported by VILLUM Matter effects are not included. FONDEN (project no. 13161).

[1] Y. Fukuda et al. (Super-Kamiokande), Evidence for os- Phys. Rev. D 99, 075022 (2019). cillation of atmospheric neutrinos, Phys. Rev. Lett. 81, [14] E. Lisi, A. Marrone, and D. Montanino, Probing possible 1562 (1998), arXiv:hep-ex/9807003. decoherence effects in atmospheric neutrino oscillations, [2] Q. R. Ahmad et al. (SNO), Measurement of the charged Phys. Rev. Lett. 85, 1166 (2000). current interactions produced by b-8 solar neutrinos at [15] Y. Farzan, T. Schwetz, and A. Y. Smirnov, Reconciling the sudbury neutrino observatory, Phys. Rev. Lett. 87, results of LSND, MiniBooNE and other experiments with 071301 (2001), nucl-ex/0106015. soft decoherence, JHEP 07, 067, arXiv:0805.2098 [hep- [3] Q. R. Ahmad et al. (SNO), Direct evidence for neutrino ph]. flavor transformation from neutral-current interactions in [16] P. Coloma, J. Lopez-Pavon, I. Martinez-Soler, and the sudbury neutrino observatory, Phys. Rev. Lett. 89, H. Nunokawa, Decoherence in Neutrino Propagation 011301 (2002), nucl-ex/0204008. Through Matter, and Bounds from IceCube/DeepCore, [4] S. Nussinov, Solar neutrinos and neutrino mixing, Eur. Phys. J. C78, 614 (2018), arXiv:1803.04438 [hep- Physics Letters B 63, 201 (1976). ph]. [5] L. Wolfenstein, Neutrino Oscillations in Matter, Phys. [17] J. A. Carpio, E. Massoni, and A. M. Gago, Testing quan- Rev. D17, 2369 (1978), [,294(1977)]. tum decoherence at DUNE, Phys. Rev. D100, 015035 [6] S. P. Mikheev and A. Yu. Smirnov, Resonant amplifica- (2019), arXiv:1811.07923 [hep-ph]. tion of neutrino oscillations in matter and solar neutrino [18] J. A. Carpio, E. Massoni, and A. M. Gago, Revisiting spectroscopy, Nuovo Cim. C9, 17 (1986). quantum decoherence for neutrino oscillations in matter [7] M. Chizhov, M. Maris, and S. T. Petcov, On the oscil- with constant density, Phys. Rev. D97, 115017 (2018), lation length resonance in the transitions of solar and arXiv:1711.03680 [hep-ph]. atmospheric neutrinos crossing the earth core, (1998), [19] L. A. Anchordoqui, H. Goldberg, M. C. Gonzalez-Garcia, arXiv:hep-ph/9810501 [hep-ph]. F. Halzen, D. Hooper, S. Sarkar, and T. J. Weiler, Prob- [8] E. K. Akhmedov, Parametric resonance of neutrino os- ing Planck scale physics with IceCube, Phys. Rev. D72, cillations and passage of solar and atmospheric neutri- 065019 (2005), arXiv:hep-ph/0506168 [hep-ph]. nos through the earth, Nucl. Phys. B538, 25 (1999), [20] G. Balieiro Gomes, M. M. Guzzo, P. C. de Holanda, and arXiv:hep-ph/9805272 [hep-ph]. R. L. N. Oliveira, Parameter limits for neutrino oscil- [9] S. W. Hawking, The unpredictability of quantum grav- lation with decoherence in kamland, Phys. Rev. D 95, ity, Communications in Mathematical Physics 87, 395 113005 (2017). (1982). [21] B. J. P. Jones, Dynamical pion collapse and the coherence [10] J. A. Wheeler, Geons, Phys. Rev. 97, 511 (1955). of conventional neutrino beams, Phys. Rev. D 91, 053002 [11] J. a. A. B. Coelho and W. A. Mann, Decoherence, matter (2015). effect, and neutrino hierarchy signature in long baseline [22] G. L. Fogli, E. Lisi, A. Marrone, D. Montanino, and experiments, Phys. Rev. D 96, 093009 (2017). A. Palazzo, Probing nonstandard decoherence effects [12] L. Buoninfante, A. Capolupo, S. M. Giampaolo, and with solar and kamland neutrinos, Phys. Rev. D 76, G. Lambiase, Revealing neutrino nature and CPT viola- 033006 (2007). tion with decoherence effects, (2020), arXiv:2001.07580 [23] J. a. A. B. Coelho, W. A. Mann, and S. S. Bashar, Non- [hep-ph]. maximal θ23 mixing at nova from neutrino decoherence, [13] J. C. Carrasco, F. N. D´ıaz, and A. M. Gago, Probing Phys. Rev. Lett. 118, 221801 (2017). cpt breaking induced by quantum decoherence at dune, 14

[24] M. M. Guzzo, P. C. de Holanda, and R. L. Oliveira, 2000, 032 (2000). Quantum dissipation in a neutrino system propagat- [46] A. M. Gago, E. M. Santos, W. J. C. Teves, and R. Z. ing in vacuum and in matter, Nuclear Physics B 908, Funchal, A study on quantum decoherence phenomena 408 (2016), neutrino Oscillations: Celebrating the No- with three generations of neutrinos (2002), arXiv:hep- bel Prize in Physics 2015. ph/0208166 [hep-ph]. [25] D. Morgan, E. Winstanley, J. Brunner, and L. F. Thomp- [47] C. A. Arg¨uelles Delgado, J. Salvad´o, and C. N. son, Probing quantum decoherence in atmospheric neu- Weaver, A Simple Quantum Integro-Differential Solver trino oscillations with a neutrino telescope, Astropart. (SQuIDS), Comput. Phys. Commun. 196, 569 (2015), Phys. 25, 311 (2006), arXiv:astro-ph/0412618 [astro-ph]. arXiv:1412.3832 [hep-ph]. [26] R. Abbasi et al. (IceCube), Determination of the At- [48] C. A. Arg¨uellesDelgado, J. Salvad´o,and C. N. Weaver, mospheric Neutrino Flux and Searches for New Physics nuSQuIDS, https://github.com/arguelles/nuSQuIDS. with AMANDA-II, Phys. Rev. D79, 102005 (2009), [49] M. G. Aartsen et al. (IceCube), Evidence for High- arXiv:0902.0675 [astro-ph.HE]. Energy Extraterrestrial Neutrinos at the IceCube Detec- [27] A. Gomes, R. Gomes, and O. Peres, Quantum decoher- tor, Science 342, 1242856 (2013), arXiv:1311.5238 [astro- ence and relaxation in neutrinos using long-baseline data, ph.HE]. (2020), arXiv:2001.09250 [hep-ph]. [50] J. R. Ellis, N. E. Mavromatos, D. V. Nanopoulos, [28] B. Pontecorvo, Inverse beta processes and nonconserva- and E. Winstanley, Quantum decoherence in a four- tion of lepton charge, Sov. Phys. JETP 7, 172 (1958). dimensional black hole background, Mod. Phys. Lett. [29] Z. Maki, M. Nakagawa, and S. Sakata, Remarks on the A12, 243 (1997), arXiv:gr-qc/9602011 [gr-qc]. unified model of elementary particles, Prog. Theor. Phys. [51] J. R. Ellis, N. E. Mavromatos, and D. V. Nanopoulos, 28, 870 (1962). Quantum decoherence in a D foam background, Mod. [30] J. Ellis, N. Mavromatos, and D. Nanopoulos, String the- Phys. Lett. A12, 1759 (1997), arXiv:hep-th/9704169 ory modifies quantum mechanics, Physics Letters B 293, [hep-th]. 37 (1992). [52] F. Benatti and R. Floreanini, Non-standard neutral kaon [31] C. Misner, C. John Archibald Wheeler, U. Misner, dynamics from infinite statistics, Annals of Physics 273, K. Thorne, J. Wheeler, W. Freeman, and Company, 58 (1999). Gravitation, Gravitation No. pt. 3 (W. H. Freeman, [53] M. G. Aartsen et al. (IceCube), Neutrino emission from 1973). the direction of the TXS 0506+056 prior to [32] Y. J. Ng, Holographic quantum foam (2010), the IceCube-170922A alert, Science 361, 147 (2018), arXiv:1001.0411 [gr-qc]. arXiv:1807.08794 [astro-ph.HE]. [33] W. Pauli, Helv. Phys. Acta. Suppl. 4, No. 69. (1956). [54] M. G. Aartsen et al. (Liverpool Telescope, MAGIC, [34] L. H. Ford, Gravitons and light cone fluctuations, Phys- H.E.S.S., AGILE, Kiso, VLA/17B-403, INTEGRAL, ical Review D 51, 16921700 (1995). Kapteyn, Subaru, HAWC, Fermi-LAT, ASAS-SN, VER- [35] H. Yu, N. F. Svaiter, and L. H. Ford, Quantum light-cone ITAS, Kanata, IceCube, Swift NuSTAR), Multimessen- fluctuations in compactified , Physical Review ger observations of a flaring blazar coincident with high- D 80, 10.1103/physrevd.80.124019 (2009). energy neutrino IceCube-170922A, Science 361, eaat1378 [36] J. D. Barrow, Cosmic no-hair theorems and inflation, (2018), arXiv:1807.08816 [astro-ph.HE]. Physics Letters B 187, 12 (1987). [55] M. Ahlers, M. Bustamante, and S. Mu, Unitarity bounds [37] F. C. Adams, G. L. Kane, M. Mbonye, and M. J. Perry, of astrophysical neutrinos, Phys. Rev. D 98, 123023 , black holes, and large extra dimensions, In- (2018). ternational Journal of Modern Physics A 16, 2399 (2001). [56] K. J. Kelly and P. A. Machado, Multimessenger astron- [38] S. Alsaleh, A. Al-Modlej, and A. F. Ali, Virtual black omy and new neutrino physics, Journal of Cosmology and holes from the generalized and pro- Astroparticle Physics 2018 (10), 048. ton decay, EPL (Europhysics Letters) 118, 50008 (2017). [57] D. Seckel, T. Stanev, and T. K. Gaisser, Signatures of [39] E. S. Perlman, S. A. Rappaport, W. A. Christiansen, cosmic ray interactions on the solar surface, in Contri- Y. J. Ng, J. DeVore, and D. Pooley, New constraints butions to the 21st international cosmic ray conference, on quantum gravity from x-ray and gamma-ray observa- Adelaide, Australia, jan 6-19, 1990 (1990) pp. 463–466. tions, The Astrophysical Journal 805, 10 (2015). [58] C. Arg¨uelles,G. de Wasseige, A. Fedynitch, and B. Jones, [40] B. Jacobsson, T. Ohlsson, H. Snellman, and W. Win- Solar atmospheric neutrinos and the sensitivity floor for ter, Effects of random matter density fluctuations on the solar dark matter annihilation searches, Journal of Cos- neutrino oscillation transition probabilities in the earth, mology and Astroparticle Physics 2017 (07), 024. Physics Letters B 532, 259 (2002). [59] K. C. Y. Ng, J. F. Beacom, A. H. G. Peter, and C. Rott, [41] J. F. Nieves and S. Sahu, Neutrino decoherence in an elec- Solar atmospheric neutrinos: A new neutrino floor for tron and nucleon background (2020), arXiv:2002.08315. dark matter searches, Phys. Rev. D 96, 103006 (2017). [42] M. Lindner, T. Ohlsson, and W. Winter, A combined [60] J. Edsjo, J. Elevant, R. Enberg, and C. Niblaeus, Neu- treatment of neutrino decay and neutrino oscillations, trinos from cosmic ray interactions in the sun, Journal of Nuclear Physics B 607, 326 (2001). Cosmology and Astroparticle Physics 2017 (06), 033. [43] J. M. Berryman, A. [de Gouvˆea], D. Hern`andez,and [61] A. Gutlein et al., Solar and Atmospheric Neutrinos: Lim- R. L. Oliveira, Non-unitary neutrino propagation from itations for Direct Dark Matter Searches, Proceedings, neutrino decay, Physics Letters B 742, 74 (2015). 24th International Conference on Neutrino physics and [44] G. Lindblad, On the generators of quantum dynamical astrophysics (Neutrino 2010): Athens, Greece, June 14- semigroups, Comm. Math. Phys. 48, 119 (1976). 19, 2010, Nucl. Phys. Proc. Supl. 229-232, 536 (2012), [45] F. Benatti and R. Floreanini, Open system approach arXiv:1009.3815 [hep-ph]. to neutrino oscillations, Journal of High Energy Physics 15

[62] T. K. Gaisser, K. Jero, A. Karle, and J. van Santen, Gen- sions in the determination of θ23, δCP , and the mass or- eralized self-veto probability for atmospheric neutrinos, dering, Journal of High Energy Physics 2019, 106 (2019). Phys. Rev. D 90, 023009 (2014). [66] D. S. Ayres et al. (NOvA), The NOvA Technical Design [63] C. A. Arg¨uelles, S. Palomares-Ruiz, A. Schneider, Report 10.2172/935497 (2007). L. Wille, and T. Yuan, Unified atmospheric neutrino [67] A. Ishihara (IceCube), The IceCube Upgrade – De- passing fractions for large-scale neutrino telescopes, Jour- sign and Science Goals (2019) arXiv:1908.09441 [astro- nal of Cosmology and Astroparticle Physics 2018 (07), ph.HE]. 047. [68] S. Adrian-Martinez et al. (KM3Net), Letter of in- [64] A. M. Dziewonski and D. L. Anderson, Preliminary refer- tent for KM3NeT 2.0, J. Phys. G43, 084001 (2016), ence earth model, Phys. Earth Planet. Interiors 25, 297 arXiv:1601.07459 [astro-ph.IM]. (1981). [69] K. Abe et al., Letter of Intent: The Hyper-Kamiokande [65] I. Esteban, M. C. Gonzalez-Garcia, A. Hernandez- Experiment — Detector Design and Physics Potential —, Cabezudo, M. Maltoni, and T. Schwetz, Global analysis (2011), arXiv:1109.3262 [hep-ex]. of three-flavour neutrino oscillations: synergies and ten-