Large Blackouts in North America: Historical Trends and Policy Implications

Large Blackouts in North America: Historical Trends and Policy Implications.

This is a page of supplementary information related to the following publication:

P. Hines, J. Apt, and S. Talukdar. Large Blackouts in North America: Historical Trends and Policy Implications. Energy Policy, v. 37, pp. 5249-5259, 2009.
Pre-print version.

Raw data

Well, enough time has gone by that I have decided to release my raw, unprocessed data for this work. If you end up using these data for research please cite the paper above. Let me know if you find additional errata, and I will try to update my data.

Here is the original csv file that contains all of the raw data after coding and a variety of judgement calls. Zeros or -1 are just as likely to mean missing data as they are to mean that actually zero customers were affected. Durations are generally mean restoration times, however there is a ton of uncertainty in the duration data.
Here is a revised version of the csv file. I don't recall the difference between the two versions at the moment, but the most recent results use this file.
Here is the matlab script that I used to read and pre-process the data. It's not elegant, but it works.

Errata

Since the publication of this paper one error and one potentially confusing item have come to my attention.

First the error. Figure 7 shows the (lack of) correlation between blackout sizes and and blackout durations. The label on the horizontal axis is "Blackout size in y2k MW." It should have been "Blackout size in y2k customers." Clearly looking at the numbers MW is not the correct messure.

Now the potentially confusing item. Table 1 shows the 15 largest blackouts, but does not include the great Toronto ice storm of 1998. Obviously this item is among the largest blackouts during this time period. However, the original NERC DAWG data for 1998 that I used for this paper did not include size information for this particular event. When the NERC data did not show size information I did not include the event, for consistency sake. Clearly this was a judgement that could have gone the other way, but c'est la vie. In fact all of the information for 1998 is goofy for some reason (a fact that is noted in the paper).

References to this work in media and government

JR Minkel. The 2003 Northeast Blackout--Five Years Later. Scientific American. Aug. 13, 2008. [local pdf]

Stan Mark Kaplan, Electric Power Transmission: Background and Policy Issues, Congressional Research Service, April 14, 2009.