Deployment & Configuration
§1
Equipment was installed on 20 March while the subject was confirmed off-site at his place of employment. The property was fitted with a full sensor suite: thermal imaging across four rooms, standard and infrared video in continuous loop, ambient electromagnetic field monitors, barometric micro-sensors, and two broadband audio captures with a sensitivity floor of 12 Hz. Kitchen window received dedicated high-resolution coverage (two angles) given the reported focal behaviour. Total installation time was six hours and fourteen minutes. The client was briefed by telephone and confirmed no visible disruption to the property.
A monitoring post was established on the eastern side of the field in a patch of dense scrub approximately 340 metres from the property. Hard line was laid below ground to avoid RF exposure and to ensure signal integrity in the event of localised EM interference, which has been documented at comparable sites. The post was manned in rotating twelve-hour shifts (Osei/Chen nights, Villanueva/Chen days). Hargreaves on-site for nights two, five, and seven.
Observation Summary — Week 1
§2
Seven days of continuous monitoring produced no anomalous readings across any sensor modality. No electromagnetic deviation outside normal domestic variance. No thermal signatures unaccounted for by occupant presence. No infrasonic events. No barometric irregularities. Audio analysis identified no vocalisation, subvocalisation, or secondary acoustic presence during the subject's nocturnal episodes. By every available metric, the house is clean.
The subject's daytime behaviour was unremarkable in all respects. He left the property at consistent times, drove a predictable route to his office, ate lunch at or near his desk, and returned home by early evening. No deviations, no unaccounted stops, no secondary devices, no contact with unknown persons. There is no evidence of an affair. Pak's mandated alternative-explanation check can be closed out.
The nocturnal behaviour, however, was confirmed exactly as the client described. On every night of the observation period, the subject rose between 01:40 and 03:10, walked to the kitchen without turning on lights, and stood motionless at the east-facing window for a period ranging from nineteen to fifty-three minutes. He then returned to bed without apparent difficulty. On two occasions the client was observed to wake briefly; both times the client followed instructions and did not engage.
Directional Analysis
§3
During review of the compiled footage, Chen noted what appeared to be minor variation in the subject's gaze angle between nights. This observation was initially logged as incidental — the differences were small, well within what might be attributed to posture or fatigue. On the fourth review pass, Hargreaves requested frame-by-frame overlays.
The subject was not staring in the same direction each night. He was staring in a slightly different direction each night. The variation was precise, consistent, and — once the overlays were aligned — unmistakably deliberate. Or if not deliberate, then governed by something that does not care whether or not it appears deliberate.
Gaze-angle measurements were calculated to a precision of ±0.2° using the dual-camera triangulation from the kitchen installation. On each night of the observation period, the subject's orientation corresponded exactly to the azimuth of the following morning's sunrise, computed for the property's coordinates and corrected for terrain elevation. The correlation holds across all seven data points with no deviation outside the margin of error.
The subject is not staring east. He is staring at a point on the horizon that does not yet contain the sun.
| Night |
Date |
Gaze Az. (°) |
Solar Az. (°) |
Δ |
Duration |
| 1 |
21 Mar |
89.8 |
89.9 |
−0.1 |
19 min |
| 2 |
22 Mar |
89.5 |
89.6 |
−0.1 |
23 min |
| 3 |
23 Mar |
89.2 |
89.2 |
0.0 |
27 min |
| 4 |
24 Mar |
88.8 |
88.9 |
−0.1 |
31 min |
| 5 |
25 Mar |
88.5 |
88.5 |
0.0 |
35 min |
| 6 |
26 Mar |
88.1 |
88.2 |
−0.1 |
41 min |
| 7 |
27 Mar |
87.8 |
87.8 |
0.0 |
53 min |
Solar azimuth calculated for 51.████°N, 0.████°W at civil sunrise. Terrain correction +0.3° applied. Margin of error on gaze measurement: ±0.2°.
Duration Pattern & Extrapolation
§4
The observation period commenced on 21 March, which was the vernal equinox. This was coincidental — scheduling was driven by equipment availability and client coordination — but in retrospect may be significant. Beginning from that date, the duration of the subject's nightly episode has increased on each successive night, broadly in step with the lengthening of the day.
The rate of increase was uneven in the first three nights (19, 23, 27 minutes), but has shown increasing regularity from Night 4 onward. By Night 7 the increment was tracking the gain in daylight hours with notable precision. If this stabilisation continues, the pattern becomes fully predictable — and projectible.
Villanueva has modelled the duration curve against the photoperiod for the property's latitude through the end of June. If the current rate of stabilisation holds, episode duration will intersect total darkness-to-sunrise time — meaning the subject will begin staring at the moment the sun sets and continue until it rises — on or around 18 June. This is approximately three days before the summer solstice.
The model is preliminary and based on seven data points. It may not hold. But if it does, the subject's nocturnal behaviour is counting down to something, and that something falls on the longest days of the year. Monitoring will continue to confirm or disqualify.
Note — Hargreaves
Villanueva's model is conservative. It assumes linear stabilisation. If the curve is logarithmic — which frankly looks more likely to me from what I'm seeing in nights 5–7 — intersection may come several days earlier. Possibly on the solstice itself. I don't love that.
Supplementary Finding — Kellaway Field
§5
Pak flagged the Kellaway land in her intake notes and requested a cross-reference before the survey team was deployed. That cross-reference has now been completed.
The field between the Marchetti property and the Kellaway boundary contains the remains of a neolithic solar calendar. The site was partially excavated in 1971 by a team from ████████████ University and subsequently back-filled at the request of the landowner. Excavation records describe a stone alignment consistent with solstice and equinox observations, oriented along a primary axis running approximately ENE–WSW. Carbon dating placed the site's active use at roughly 3,200–2,800 BCE. The 1971 report classified it as a regional ritual site of moderate archaeological significance.
A supplementary paper published in 1974 and since withdrawn from the university's catalogue — Hargreaves located a physical copy through the Institution's archive — proposed that the site served as a locus of worship for a solar deity. The paper's author was unable to identify the deity by name. No inscriptions were recovered. No iconography survived. The name, if it was ever recorded, has been lost. The author noted this as unusual for a site of this scale and speculated — with appropriate academic hedging — that the absence might be intentional. That the name may have been removed rather than merely forgotten.
Note — Hargreaves
I want to be careful about what we're implying here. We have a man who stares at where the sun will rise, in a house that overlooks the remains of a place where people once worshipped something solar, whose name no longer exists. That is what we have. I am not yet prepared to say what it means.
What I will say is that the 1974 paper was withdrawn — not retracted, not corrected,
withdrawn, with no public explanation — and the author died eight months later in circumstances the Institution's file describes as "unresolved." I mention this for completeness and not because I think it is necessarily connected.
Recommend we extend monitoring through at least mid-June. Recommend we do
not inform the client about the field. Recommend Pak schedule a second interview to ask, very carefully, whether the subject has at any point mentioned the sun.
Operational Timeline
§6
Key Dates
14 Mar
Client intake (Pak). Case accepted.
17 Mar
Kellaway land cross-reference completed. Hargreaves assigned as lead.
20 Mar
Sensor installation and OP-1 established.
21 Mar
Observation begins. Vernal equinox.
24 Mar
Chen flags directional variation in gaze angle.
26 Mar
Sunrise-azimuth correlation confirmed.
28 Mar
Observation week concludes. Duration pattern noted.
29 Mar
Villanueva models duration extrapolation. 18 June intersection projected.
31 Mar
This report filed. Monitoring continues.
~18 Jun
Projected convergence date (±3 days). Pattern still manifesting.
Current Status
§7
Case remains active. Monitoring continues under existing configuration. No intervention is recommended at this time. The pattern is still manifesting and has not yet been observed to deviate, stall, or reverse. We do not know what happens when it completes.
Mundane-cause review is closed. Whatever this is, it is not mundane.