Various embodiments herein relate to systems and methods for detecting seismic events. Systems may include inertial sensors distributed on or in communication with a network of optically switchable windows in the building. In some systems, inertial sensors are located within a window controller, within an insulated glass unit, or in some way rigidly attached to the structure of a building. Logic is described for leveraging sensed inertial data to predicted a seismic event and/or evaluate the structural health of the building. In some cases, logic may be used to issue an alert to building occupants about impending shear waves that will arrive at the building's location. In some cases, a window network may respond to a detected seismic event by, e.g., changing the optical state of windows and/or providing occupants with evacuation instructions.

Apparatus and methods are described, including identifying an arrival of a first arriving S-wave emitted from a seismic source at an array (120) of sensors (129, 140) in real-time, by continuously analyzing waveforms received by the sensors (120, 140), and continuously monitoring back-azimuth and slowness data within the detected waveforms. Arrival of a first arriving P-wave emitted from the seismic source at the array (120) of sensors (129, 140) is identified, based upon the back-azimuth and slowness data. Slowness and back azimuth of the first arriving P-wave are determined, by analyzing a waveform of the P-wave, and based upon the determined slowness of the first arriving P-wave, the arrival of the first arriving S-wave at the array (120) of sensors (129, 140) is identified. Other applications are also described.

A retrofit valve shutoff device comprises a coupling key for coupling with an actuator of a shutoff valve, an accelerometer for making acceleration measurements in three directions, a motor, and a processing unit. The processing unit determines the arrival of seismic P-waves when the ratio of vibrations' power in the vertical direction with respect to a sum of the vibrations' power in the three directions exceeds a first threshold. The processing unit then determines the arrival of seismic S-waves when the sum of the vibrations' power in the three directions exceeds a second threshold. The processing unit then determines the arrival of seismic surface waves when the sum of the vibrations' power in the three directions exceeds a third threshold. The processing unit then sends a signal to the motor to rotate the coupling key and the actuator of the shutoff valve to close the shutoff valve.

A system for collecting and managing seismic data via an external communications network comprises one or more seismic stations, each including a seismic measurement apparatus producing seismic signals, a station processor converting the signals to seismic data, a station memory securely storing the seismic data on site and a station communication interface transmitting the seismic data onto an external network. The system further comprises one or more data servers, each including a server computing device, a server communication interface receiving the seismic data from the seismic stations and a server memory storing the received seismic data. The data server can determine if the received seismic data satisfies predetermined conditions for certification and/or triggering a payout in accordance with a contract, and can thereafter transmit the appropriate data signals to another location on the external communications network.

A method can include, responsive to receipt of input characterizing a geologic environment, utilizing a trained machine learning model to identify a number of geologic environments that include corresponding data stored in at least one database; analyzing one or more of the number of geologic environments; and outputting a result based at least in part on the analyzing.

Methods, systems, and devices for operating emergency prevention sensor systems are described. Devices can include a plurality of components including a sensor component, a processor, and memory. In an example, a method can include receiving at a processor signaling from a plurality of environmental sensing devices, each having at least one biodegradable component, in an area of concern, wherein the area of concern corresponds to a particular set of coordinates in a database, determining environmental characteristics of an emergency associated with the area of concern based, at least in part, on the signaling, and determining a preventive action based on the determined characteristics. In another example, a number of components of the sensing devices are biodegradable.

A seismic observation device includes an input unit receiving input of time-series data of measurement values of a vibration, a processing target determination unit determining a time period of the time-series data that is a processing target, and a type determination unit acquiring a likelihood of classifying a cause of the vibration indicated in the time-series data in the time period into each of types of cause.

An earthquake event classification method using an attention-based neural network includes: preprocessing input earthquake data by centering; extracting a feature map by nonlinearly converting the preprocessed earthquake data through a plurality of convolution layers having three or more layers; measuring importance of a learned feature of the nonlinear-converted earthquake data based on an attention technique in which interdependence of channels of the feature map is modeled; correcting a feature value of the measured importance value through element-wise multiply with the learned feature map; performing down-sampling through max-pooling based on the feature value; and classifying an earthquake event by regularizing the down-sampled feature value. Accordingly, main core features inherent in many/complex data are extracted through attention-based deep learning to overcome the limitations of the existing micro earthquake detection technology, thereby enabling earthquake detection even in low SNR environments.

A method for automatically identifying an active source azimuth of a planet seismometer, comprising: intercepting a three-channel original time sequence in a duration before and after an active source first arrival signal of each support leg received by a planet seismometer in turn; converting the three-channel original time sequence to a horizontal plane based on a pitch angle and a roll angle of the planet seismometer after being deployed to a surface of a planet; converting a time sequence of the horizontal plane to RTZ coordinates and calculating the maximum amplitudes of components of a vibration signal; constructing a target function based on the maximum amplitudes of components of a vibration signal, and scanning an azimuth of the planet seismometer at preset angle intervals, wherein when the target function reaches the minimum, the corresponding azimuth of the planet seismometer is the optimal estimation.

A method and system for multi-trigger parametric data management and associated transactions is provided. The parametric triggers can include measured wind speed occurring at predetermined geographic locations, calculated wind speed at a predetermined geographic locations, reported storm track and wind speed/category at a predetermined distance from a reference location and/or measured tide levels occurring at predetermined geographic locations. The multiple triggers are evaluated by the disclosed process to evaluate whether indicated threshold levels are exceeded such that conditions for full or fractional payout are achieved