Connecting Risk and Resilience for a Power System Using the Portland Hills Fault Case Study

Round 1

Reviewer 1 Report

The paper is very pleasant to read, however if the authors just to took a picture of the current Portland General Electric network and calculate an economic scenario if an M6.8 earthquake were to occur tomorrow, the paper would have been quite interesting.

My point of view of seismologist is that the authors embarked on a nice exercise that might be an expensive one if Portland General Electric were to follow their approach. As they correctly point out, the chance of this earthquake occurrence is rather small. Unless I grossly misunderstood the approach, in no case they considered the expected temporal operative duration of their assets: I would have included a substitution analysis that would take into account the natural modernization rate of the network, eventually taking into account the extra costs of seismic improvements. I would have preferred to see the computation performed for the most probable event in the next two decades, since I strongly doubt that the equipment currently in operation will last more than this period of time.

Moreover (but I do not expect this from the authors since they are not expert seismologists) I would have considered the time dependent probability of an event occurrence, since real earthquakes are not poissonianly distributed, provided that enough data are available in the area.

In my opinion the minimum effort to undertake to publish this paper, is to discuss the above issues and explain why the authors prefer to stick to their approach (technically time independent and economically time dependent – twenty years). I would strongly suggest however to compare their results with at least the most probable event in the next twenty years, since the paper could greatly benefit from this and it could be a useful tool for Portland General Electric to set up an internal strategy with respect to seismic hazard: the final disclaim in the conclusion is in fact too strong since it completely empties the paper meaning.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

This paper analyzed the impact of an earthquake originating in the Portland Hills Fault on the surrounding power grid. A 100,000 round Monte Carlo simulation was used for analysis, with each round simulating how a single earthquake would affect each of the 71 substations in the surrounding area. The effect of the substation was determined by fragility curves and the resulting effect on the power network was determined by a load flow analysis. Based on the network outage/load loss, the economic impact of the earthquake was determined for each simulated round. By aggregating the results of the 100,000 Monte Carlo simulation runs, a distribution of load loss and consequence cost was generated. The authors then considered how an "upgraded" network (where substations were less likely to fail/better fragility curves) would perform. 

The paper is well motivated and written. The paper appears technically correct.

To consider:
1) A distribution of the "consequence cost"/economic impact is generated, however in the analysis only the mean value of 180 million is discussed. Since analysis is motivated by understanding high-impact low probability events, I think it also makes sense to also consider the "near-worst case" impact of the earthquake. That is, also consider/discuss the economic impact of the 99th percentile as well as the mean (for both upgraded and non-upgraded networks).

Typos/formatting:

1) title of paper, "using" --> "Using"

2) figure 3, font of axis does not match rest of paper

3) figure 5 seems a bit too large

4) page 10 line 246, consider using ":" for the enumeration

5) Consider combining figures 10 and 11 as a side by side figure since they are representing the same information

Author Response

Please see the attachment

Author Response File: Author Response.docx