Out of curiosity this morning, and to prepare for a SF National Meeting presentation, I did a quick sanity check of the data trends in the Hazmat headlines I've been collecting since 2010. The numbers are generally consistent with those I reported in 2011 and 2012, which is the last time I carefully cleaned up data inconsistencies before generating the numbers. In case other people are interested in the overall numbers since 2010, they are listed below. The "Total included of reported" indicates the percentage of incidents which had expected data in the right locations; the early days of the system had more inconsistency than today's semi-automated system.

One trend that I notice is that there are a lot more "follow up" stories included in the press, particularly those related to the UCLA fire, the West Virginia leak and the Wes, Texas ammonium nitrate explosion. Meth lab reports seem to have decreased somewhat.

Time period: 4.5 years; 1611 days; 230 weeks; 

Number of hazmat events included:
7780 total events (5/day); 938 lab events (4/week)

Geographic distribution: 74% reported in US

Sector:
industrial 29%
public 22%
transportation 13%
laboratory 10%
education 4%
Total included of reported 78%

Type distribution:
release 32%
explosion 15%
fire 15%
follow-up 8%
discovery 7%
Total included of reported 77%

Lab event type:
release 40%
fire 17%
explosion 15%
follow-up 13%
discovery 8%
Total included of reported: 93%

Event Extent:
response 47%
injury 24%
environmental 6%
death 6%
follow-up 3%
Total included of reported events: 86%

Chemicals involved are generally unclear: 

15% unknown; 

3% each of the usuals (chlorine, ammonia, petroleum, meth labs, etc.)

The rest less than 1%

Let me know if you have any questions about this.

- Ralph

Ralph Stuart

secretary**At_Symbol_Here**dchas.org

Secretary

Division of Chemical Health and Safety

American Chemical Society