- With one notable exception, the presentations were very careful to distinguish model output form measured data, and nearly every observation had error bands.
- Computer models are used extensively to test hypotheses for reasonableness. Many models were discussed that failed to honor the data and these models were shown to be dead ends that did not help with assessing when and where Oil & Gas operations were impacting seismicity.
- When the scientists did not know something, they seemed very comfortable with "my research has not examined that" or "the data does not support a conclusion on that topic".
- The exception was the one guy who categorically declared that frac'ing was causing a massive increase in the rate and severity of earthquakes in Oklahoma (several others had strongly indicated that the data did not support a conclusion that frac'ing or production was related to seismic activity, but there was a correlation between produced water injection and seismicity). When he was cornered by experts, he did admit that out of 50,000 frac jobs in Oklahoma during his study he could only find 2 frac jobs that had a temporal and geographic coincidence with earthquakes, and both of those were measured on quite distant stations with low resolution and no ability to correlate with depth. It was clear to all in the room that his work was of a completely different quality and honesty than the rest of the workshop and it was further clear that he had carefully spun the data for a political agenda. The sense of the room was that his data does in fact demonstrate a significant increase in seismic events in recent years. Further, that no conclusion was supported in data.
David Simpson, PE
Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat