A researcher's desk in late afternoon — an external hard drive connected to a laptop, printed data sheets beside it, a tea glass, low evening light from a window. The atmosphere is something found in old data that was always there but not yet connected. No text, no signs, no writing visible anywhere. Photorealistic, cinematic, low afternoon light, documentary photography style, shallow depth of field, muted cool palette.

Thursday. June 11. Barometer: 1014 hPa. Stable.

I opened my 1998 data this morning.

July 1998

My measurement records from 1998 are on a hard drive that has been running continuously since 2004. I found the July entries.

On Tuesday, July 14, 1998, at 14:37, I recorded a deviation of −0.284 Hz. This is well outside my established range of −0.188 to −0.211 Hz. I had noted it at the time: “anomalous; possible equipment interference; not reproduced following week.” I did not investigate further. I had no context for it.

Three weeks after that Tuesday, there was a cascade failure at three substations in northern Kazakhstan. Duration: approximately three hours. Official cause: equipment failure during peak summer demand. MES — at that time two years old and holding its first Kazakhstan grid monitoring contract — provided post-incident analysis services. The monitoring contract was renewed six months later at improved terms.

The date of the cascade: July 28, 1998. I did not notice this connection for twenty-eight years. I had no reason to look.

March 2016

I searched for a second data point. March 2016: Tuesday, March 8, a deviation of −0.263 Hz. I had noted: “outlier; possible seasonal calibration deviation.”

Eight days later, Moldova experienced a grid cascade. Duration: approximately four hours. MES held a grid monitoring contract in Moldova from 2011. The contract was in active renegotiation at the time of the cascade. It was renewed six months later.

What I Have Not Said

Both events appear in my dataset as unexplained footnotes I never followed up. Both occurred shortly before a cascade failure. Both cascades happened in countries with active MES monitoring contracts. Both led to contract renewals on improved terms.

I am a physicist. I know that two events are not a pattern. A pattern requires three.

I called Ruslan at 17:31. The call lasted 22 minutes. He said:

“Anatoli. You need to decide what you are going to do with this. Not because you are in danger. Because Viktor was not, and it did not help him.”

Current status:

  • July 1998 outlier: −0.284 Hz; 14:37; Kazakhstan cascade 14 days later; MES contract renewed
  • March 2016 outlier: −0.263 Hz; Moldova cascade 8 days later; MES contract renewed
  • Two events confirmed in dataset; third not yet found
  • Ruslan: “Viktor was not, and it did not help him.”
  • Paper: day 59 in review; status unchanged
  • Barometer: 1014 hPa (stable)
  • Session 48: Tuesday, June 16; 14:37
  • Emotional state: two events are not a pattern

Previous post: Morev Energy Systems