This paper describes the Polarization Spectroscopic Telescope Array (PolSTAR), a mission proposed to NASA's 2014 Small Explorer (SMEX) announcement of opportunity. PolSTAR measures the linear polarization of 3-50 keV (requirement; goal: 2.5-70 keV) X-rays probing the behavior of matter, radiation and the very fabric of spacetime under the extreme conditions close to the event horizons of black holes, as well as in and around magnetars and neutron stars. The PolSTAR design is based on the technology developed for the Nuclear Spectroscopic Telescope Array (NuSTAR) mission launched in June 2012. In particular, it uses the same X-ray optics, extendable telescope boom, optical bench, and CdZnTe detectors as NuSTAR. The mission has the sensitivity to measure ∼1% linear polarization fractions for X-ray sources with fluxes down to ∼5 mCrab. This paper describes the PolSTAR design as well as the science drivers and the potential science return.

The Guerrero seismic gap is presumed to be a major source of seismic and tsunami hazard along the Mexican subduction zone. Until recently, there were limited observations at the shallow portion of the plate interface offshore Guerrero, so we deployed instruments there to better characterize the extent of the seismogenic zone. Here we report the discovery of episodic shallow tremors and potential slow slip events in Guerrero offshore. Their distribution, together with that of repeating earthquakes, seismicity, residual gravity and bathymetry, suggest that a portion of the shallow plate interface in the gap undergoes stable slip. This mechanical condition may not only explain the long return period of large earthquakes inside the gap, but also reveals why the rupture from past M < 8 earthquakes on adjacent megathrust segments did not propagate into the gap to result in much larger events. However, dynamic rupture effects could drive one of these nearby earthquakes to break through the entire Guerrero seismic gap.

No abstract

Either the triggering of large earthquakes on a fault hosting aseismic slip or the triggering of slow slip events (SSE) by passing seismic waves involve seismological questions with important hazard implications. Just a few observations plausibly suggest that such interactions actually happen in nature. In this study we show that three recent devastating earthquakes in Mexico are likely related to SSEs, describing a cascade of events interacting with each other on a regional scale via quasi-static and/or dynamic perturbations across the states of Guerrero and Oaxaca. Such interaction seems to be conditioned by the transient memory of Earth materials subject to the “traumatic” stress produced by seismic waves of the great 2017 (Mw8.2) Tehuantepec earthquake, which strongly disturbed the SSE cycles over a 650 km long segment of the subduction plate interface. Our results imply that seismic hazard in large populated areas is a short-term evolving function of seismotectonic processes that are often observable.

A thorough study of the admittance of TiN/Ti/HfO 2 /W bipolar resistive memories [resistance random access memory (RRAM)] was carried out under different bias conditions and in a wide range of ac signal frequencies. We demonstrate that a continuum of intermediate states can be obtained by applying appropriate dc bias waveforms. Cumulative writing and erasing admittance cycles were performed by applying triangular voltage waveform of increasing amplitude. The influence of the initial conditions on the variation of the real (conductance) and imaginary (susceptance) components of the admittance is described. An accurate control of the memory state is achieved both in terms of the conductance and the susceptance by means of an adequate selection of the voltage values previously applied. A method to obtain three-dimensional voltage-conductance-susceptance state-plots is described in detail. Memory maps of admittance parameters as a function of the programming voltage are made by sensing the memory state at 0 V, without static power consumption. The multilevel nature of RRAM devices and their suitability for neuromorphic computation are demonstrated.