README file for git repository bridge.collapses

Contents

1. Raw data from public and private sources related to collapsed bridges in the US
   • The NYSDOT database of "failed" (collapsed) bridges
   • The USDOT 2017 gis version of the US National Bridge Inventory
   • USGS station and streamflow data of various kinds
   • Related data on dams, hurricanes, etc.
2. Processed data from above in the form of .RData
3. R scripts compatible through R 3.2.1 and mostly compatible through 3.5.2 for
   • Loading and processing the raw data
   • Creating input and reading output from other software (e.g., software for flood frequency analysis)
   • Performing statistical and other analyses in R
   • Plotting results
4. R markdown notebooks for running performing a series of analyses

A related repository is available at https://code.vt.edu/mflint/Flint.et.al.2016.Historical.Hydraulic.Bridge.Collapses

Naming convention:

Derived (analysis) variable names are determined in the order followed, linked by underscores ("_")

• Variable type:
  • Q: flow or discharge, cfs
  • T: return period, years
  • DATE: YYY-MM-DD
  • BOOL: TRUE/FALSE or factor
  • COMMENT: string
  • COUNT: count, integer
• Event category:
  • FAIL
  • FAILPM2: fail +/- 2 days
  • MAX: max recorded
  • MAXPREFAIL: max recorded up to and including failure date in overlapping existence of bridge and gauge
  • MAXFAILYEAR: maximum in year of failure
  • FAIL_05: 5% confidence interval for failure
  • FAIL_95: 95% confidence interval for failure
• Data source:
  • D: daily mean
  • I: instantaneous
  • P: peak, usually instantaneous
  • IP: instantaneous and/or peak
• Analysis type:
  • HECD: HEC-SSP Bulletin 17B analysis using daily mean data
  • HECP: HEC-SSP Bulletin 17B analysis using annual peaks data
  • PARTDUR: partial duration analysis (uses daily mean data)
• Data source:
  • USGS

Most other variable names are taken directly from Fail, NBI, or USGS databases

Column descriptions: see tab-delimited "df.Fail.NBI.Gage.Colnames.txt"