### Abstract

Environmental risk may be defined as the probability of environmental or health loss. Often it is necessary to estimate probability distributions for loss magnitude, for planning and design. Data for such event size probability distributions for industrial gas explosions are compared here to various empirical relations and distributions. A form of the Pareto distribution is derived using information theory and Bayesian statistics, and compared to the data, as well. Where the probability of extremely large events is of interest, the Bayesian distribution provides the closest empirical fit.

Original language | English (US) |
---|---|

Title of host publication | National Conference on Environmental Engineering |

Publisher | Publ by ASCE |

Pages | 232-237 |

Number of pages | 6 |

State | Published - 1992 |

Event | 1992 National Conference on Environmental Engineering - Water Forum '92 - Baltimore, MD, USA Duration: Aug 2 1992 → Aug 6 1992 |

### Other

Other | 1992 National Conference on Environmental Engineering - Water Forum '92 |
---|---|

City | Baltimore, MD, USA |

Period | 8/2/92 → 8/6/92 |

### Fingerprint

### ASJC Scopus subject areas

- Environmental Engineering

### Cite this

*National Conference on Environmental Engineering*(pp. 232-237). Publ by ASCE.

**Event size probability distribution for risk analysis.** / Englehardt, James Douglas.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*National Conference on Environmental Engineering.*Publ by ASCE, pp. 232-237, 1992 National Conference on Environmental Engineering - Water Forum '92, Baltimore, MD, USA, 8/2/92.

}

TY - GEN

T1 - Event size probability distribution for risk analysis

AU - Englehardt, James Douglas

PY - 1992

Y1 - 1992

N2 - Environmental risk may be defined as the probability of environmental or health loss. Often it is necessary to estimate probability distributions for loss magnitude, for planning and design. Data for such event size probability distributions for industrial gas explosions are compared here to various empirical relations and distributions. A form of the Pareto distribution is derived using information theory and Bayesian statistics, and compared to the data, as well. Where the probability of extremely large events is of interest, the Bayesian distribution provides the closest empirical fit.

AB - Environmental risk may be defined as the probability of environmental or health loss. Often it is necessary to estimate probability distributions for loss magnitude, for planning and design. Data for such event size probability distributions for industrial gas explosions are compared here to various empirical relations and distributions. A form of the Pareto distribution is derived using information theory and Bayesian statistics, and compared to the data, as well. Where the probability of extremely large events is of interest, the Bayesian distribution provides the closest empirical fit.

UR - http://www.scopus.com/inward/record.url?scp=0026976742&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026976742&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0026976742

SP - 232

EP - 237

BT - National Conference on Environmental Engineering

PB - Publ by ASCE

ER -