High energy photons from blazars can initiate electromagnetic pair cascades interacting with the extragalactic photon background. The charged component of such cascades is deflected and delayed by extragalactic magnetic fields (EGMF), reducing thereby the observed point-like flux and leading potentially to multi degree images in the GeV energy range. We calculate the fluence of 1ES 0229+200 as seen by Fermi-LAT for different EGMF profiles using a Monte Carlo simulation for the cascade development. The non-observation of 1ES 0229+200 by Fermi-LAT suggests that the EGMF fills at least 60% of space with fields stronger than O(10 −16 − 10 −15 ) G for life times of TeV activity of O(10 2 − 10 4 ) yr. Thus the (non-) observation of GeV extensions around TeV blazars probes the EGMF in voids and puts strong constraints on the origin of EGMFs: Either EGMFs were generated in a space filling manner (e.g. primordially) or EGMFs produced locally (e.g. by galaxies) have to be efficiently transported to fill a significant volume fraction, as e.g. by galactic outflows.

In the dense-neutrino region at 50 -400 km above the neutrino sphere in a supernova, neutrino-neutrino interactions cause large flavor transformations. We study when the multiangle nature of the neutrino trajectories leads to flavor decoherence between different angular modes. We consider a two-flavor mixing scenario between e and another flavor x and assume the usual hierarchy F e > F e > F x F x for the number fluxes. We define F e ÿ F e =F e ÿ F x as a measure for the deleptonization flux which is the one crucial parameter. The transition between the quasi-single-angle behavior and multiangle decoherence is abrupt as a function of . For typical choices of other parameters, multiangle decoherence is suppressed for * 0:3, but a much smaller asymmetry suffices if the neutrino mass hierarchy is normal and the mixing angle small. The critical depends logarithmically on the neutrino luminosity. In a realistic supernova scenario, the deleptonization flux is probably enough to suppress multiangle decoherence.

A few seconds after bounce in a core-collapse supernova, the shock wave passes the density region corresponding to resonant neutrino oscillations with the "atmospheric" neutrino mass difference. The transient violation of the adiabaticity condition manifests itself in an observable modulation of the neutrino signal from a future galactic supernova. In addition to the shock wave propagation effects that were previously studied, a reverse shock forms when the supersonically expanding neutrinodriven wind collides with the slower earlier supernova ejecta. This implies that for some period the neutrinos pass two subsequent density discontinuities, giving rise to a "double dip" feature in the average neutrino energy as a function of time. We study this effect both analytically and numerically and find that it allows one to trace the positions of the forward and reverse shocks. We show that the energy dependent neutrino conversion probabilities allow one to detect oscillations even if the energy spectra of different neutrino flavors are the same as long as the fluxes differ. These features are observable in theν e signal for an inverted and in the ν e signal for a normal neutrino mass hierarchy, provided the 13-mixing angle is "large" (sin 2 ϑ 13 ≫ 10 −5 ).

We have reconsidered the atmospheric neutrino anomaly in light of the 1289-day data from Super-Kamiokande contained events and from Super-Kamiokande and MACRO up-going muons.We have reanalysed the proposed solution to the atmospheric neutrino anomaly in terms of nonstandard neutrino-matter interactions (NSI) as well as the standard ν µ → ν τ oscillations (OSC).Our statistical analysis shows that a pure NSI mechanism is now ruled out at 99%, while the standard ν µ → ν τ OSC mechanism provides a quite remarkably good description of the anomaly. We therefore study an extended mechanism of neutrino propagation which combines both oscillation and non-standard neutrino-matter interactions, in order to derive limits on flavour-changing (FC) and non-universal (NU) neutrino interactions. We obtain that the off-diagonal flavour-changing neutrino parameter ε and the diagonal non-universality neutrino parameter ε ′ are confined to −0.05 < ε < 0.04 and |ε ′ | < 0.17 at 99% CL. These limits are model independent and they are obtained from pure neutrino-physics processes. The stability of the neutrino oscillation solution to the atmospheric neutrino anomaly against the presence of non-standard neutrino interactions establishes the robustness of the near-maximal atmospheric mixing and massive-neutrino hypothesis. The best agreement with the data is obtained for ∆m 2 = 2.4 × 10 −3 eV 2 , sin 2 (2θ) = 0.99, ε = −9.1×10 −3 and ε ′ = −1.9×10 −3 , although the χ 2 function is quite flat in the ε and ε ′ directions for ε, ε ′ → 0. A revised analysis which takes into account the new 1489-day Super-Kamiokande and final MACRO data is presented in the appendix; the determination of ∆m 2 and θ is essentially unaffected by the inclusion of the new data, while the bounds on ε and ε ′ are strongly improved to −0.03 ≤ ε ≤ 0.02 and |ε ′ | ≤ 0.05 at 99.73% CL.3 The NSI may, however, be rather small [34].

One of the robust features found in simulations of core-collapse supernovae (SNe) is the prompt neutronization burst, i.e. the first ∼ 25 milliseconds after bounce when the SN emits with very high luminosity mainly νe neutrinos. We examine the dependence of this burst on variations in the input of current SN models and find that recent improvements of the electron capture rates as well as uncertainties in the nuclear equation of state or a variation of the progenitor mass have only little effect on the signature of the neutronization peak in a megaton water Cherenkov detector for different neutrino mixing schemes. We show that exploiting the time-structure of the neutronization peak allows one to identify the case of a normal mass hierarchy and large 13-mixing angle ϑ13, where the peak is absent. The robustness of the predicted total event number in the neutronization burst makes a measurement of the distance to the SN feasible with a precision of about 5%, even in the likely case that the SN is optically obscured.