Advanced searches left 3/3

Hazard - Astrophysics Data System

Summarized by Plex Scholar
Last Updated: 05 August 2022

* If you want to update the article please login/register

On the Considerations of Using Near Real Time Data for Space Weather Hazard Forecasting

In recent years, numerical models have been developed that can forecast volatile intervals, but they almost exclusively use post-processed "science" solar wind data from upstream of the Earth. In general, the data available in NRT matches well with post-processed results, but there are three main areas of concern: increasing short-term variability in the NRT results, occasional anomalous values, and frequent data gaps. Some space weather models are able to account for these shortcomings if they are also present in the model's results, while others will need additional inspections to be carried out in order to produce high quality forecasts. While the DSCOVR NRT records are generally more consistent, they have been available for a small fraction of a solar cycle, and therefore DSCOVR has only experienced a narrow range of solar wind conditions, causing us to be more detailed.

Source link: https://ui.adsabs.harvard.edu/abs/2022SpWea..2003098S/abstract


Generation of seismic hazard maps for Assam region and incorporation of the site effects

Probabilistic seismic hazard evaluation, which includes a site-specific amplification analysis, is needed to determine soil-specific spectra of soil samples in order to help determine the soils' behavioural characteristics under earthquake excitation. From Gutenberg-Richter's new homogenized and de-clustered earthquake database from 1735 to 2021, the earthquake recurrence parameters of each zone are determined. 0. 24 and 0. 37 g, 0. 42 and 0. 91 g, respectively, with a 0. 34 and 0. 90 g, 0. 3 and 0. 1 g, 0. 42 and 0. 91 g, respectively. The peak ground acceleration value obtained at the rock outcrop of Assam's 10-, 2, and 0. 5 g, 0. 3 and 0. 38 g, 0. 42 and 0. 91 g, 0. 39 and 0. 91 g, respectively. For various earthquake return periods of 475,2475, and 9975 years, a site-specific response spectra at bedrock level for major cities of Assam state has been developed.

Source link: https://ui.adsabs.harvard.edu/abs/2022AcGeo.tmp..170B/abstract


Landslide's analysis and Hazard mapping based on ANALYTIC HIERARCHY PROCESS (AHP) using GIS, in Lawas, Sabah-Sarawak

This study was carried out to create a landslides hazard map in a Malaysian district called Lawas, using the AHP techniques. As determined from the spatial data analysis, weightage values for the conditioning factors were assigned considering the influence of each factor. Overall, the report revealed that all factors collected were highly relevant to landslides in the study area, and that the AHP technique was also discovered to be effective.

Source link: https://ui.adsabs.harvard.edu/abs/2022E&ES.1064a2031I/abstract


Learning to Assess Danger from Movies for Cooperative Escape Planning in Hazardous Environments

To solve the first challenge, we suggest we use the massive amount of visual content available in the form of movies and TV shows to create a database that can depict dangerous environments encountered in the real world. The images are annotated with high-end danger ratings for realistic disaster photos, as well as corresponding keywords that summarize the scene's text. We recommend a multi-modal danger estimation pipeline for collaborative human-robot escape scenarios in response to the second challenge. Our Bayesian framework improves risk estimation by utilizing information from robot's camera sensor and human voice inputs. In addition, we updated the estimation module with a risk-aware planner that helps in locating safer routes out of the dangerous environment. We demonstrate the benefits of our multi-modal perception framework, as well as increased success rates in a collaborative human-robot mission through extensive experiments.

Source link: https://ui.adsabs.harvard.edu/abs/2022arXiv220713791S/abstract


A Hazard Analysis Framework for Code Synthesis Large Language Models

We use a hazard analysis framework developed at OpenAI to identify risks or safety risks that deployment of Codex models such as Codex might pose politically, socioeconomically, politically, and economically. The evaluation is based on a novel evaluation framework that evaluates the ability of advanced code generation techniques against the breadth and expressivity of specification prompts, as well as their ability to recognize and execute them according to human capability.

Source link: https://ui.adsabs.harvard.edu/abs/2022arXiv220714157K/abstract


Urban Particulate Matter Hazard Mapping and Monitoring Site Selection in Nablus, Palestine

In the State of Palestine, only four air pollution experiments have been carried out, and all revealed an increase in particulate matter concentrations above WHO recommendations. To obtain an air pollution hazard map for Nablus, Palestine, a GIS-based weighted overlay summation procedure for the potential sources of air pollution was used. The findings of the hazard map indicated that 82% of Nablus is vulnerable to a high and medium risk of PM pollution. Sensors' reports indicated a good match between the hazard intensity and PM concentrations.

Source link: https://ui.adsabs.harvard.edu/abs/2022Atmos..13.1134S/abstract


Procedurally generated simulated datasets for aerial explosive hazard detection

In addition, the best-performing ML models are black versus glass boxes. Herein, we use modern video game engine technology to help recognize and help create better ML solutions by tackling the real world annotated data bottleneck problem. In particular, we address the Unreal Engine's procedural environment and data processing process for explosive hazard detection. In addition, we describe a process for generating reports at various levels of visual abstraction to prepare ML algorithms, encourage better functionality, and examine ML model generalizability.

Source link: https://ui.adsabs.harvard.edu/abs/2022SPIE12116E..11K/abstract


Research and application of accident hazard prediction based on matrix decomposition

Based on cloud services, the example of a safety product hazard investigation is investigated in this paper. The matrix decomposition technique is used in second to determine the company's hazard degree with the company's hazard analysis. It is shown that the matrix decomposition strategy can be used in hazard prediction in the industry to help avoid hidden risks by comparing the forecast results with the real hidden data.

Source link: https://ui.adsabs.harvard.edu/abs/2022SPIE12256E..31W/abstract


Infrared backscatter imaging spectroscopy for standoff detection of hazardous materials

We discuss the latest advancements in a cart-based backscatter imaging spectroscopy that can detect and analyze trace amounts of hazardous materials at proximal stand-off distances. To illuminate samples contaminated with analyte, a four-chip quantum cascade laser system quickly scans through the mid- to long-wave infrared wavelength range. Each frame taken with the MCT camera corresponding to the laser's release at the time of capture is assigned a wavelength to it. Every pixel in the photo is pixelized in this process. This process produces a hyperspectral image cube with spectral reflectance data for every pixel in the image. Any pixel of the 128x128 camera array's single detection event can be completed in less than a second, and every pixel of the 128x128 camera array produces an individual spectrum. The IBIS cart-based measurements were performed using a high-resolution FTIR to verify the highly sensitive and chemically specific nature of the target analytes.

Source link: https://ui.adsabs.harvard.edu/abs/2022SPIE12116E..0QF/abstract

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions