August 14, 2026

Generated lab notebook from daily_summary, plan_journal, plan_delivery_log, and setpoint audit data. It is intentionally chronological and may include in-progress cycles before validation.

Planner Execution Ledger

Planner checkpoints can acknowledge that the active plan is still suitable without writing a new public plan ID. Those no-change decisions are part of the audit trail.

🌅 Morning Cycle (12:19 AM) — iris-20260814-0016

Reflection

Validating previous cycle: iris-20260813-0611

Previous hypothesis: SUNRISE plan for Thursday 2026-08-13. Wednesday closed at score 58.8 versus the 69.2 seven-day average. Legacy temperature and VPD compliance were tied at 49.9%; heat stress slightly dominated at 10.00 h, followed by 9.35 h VPD-high, with 0 cold, 0.31 h VPD-low, minimum dew margin 8.3F, and 0 dew-risk hours. Reliable electric/gas/total-cost fields are unavailable. Water was 168 gal, including 43 gal misting, below the 276 gal seven-day average, so resource use was not the constraint. I evaluated iris-20260812-0610 at 5/10, exactly matching deterministic anchor 5. The formal evaluator reported zero guardrail events, while the assembled broader transition audit reports many holds/misses; this scope discrepancy is treated as safe dew outcome but uncertain materialization, not silently folded into the anchor. At 06:11 MDT the house is safe and inside its dispatcher-owned corridor at 72… Result: Plan governed approximately 00:19-06:18 MDT, so this grade covers the completed humid pre-dawn portion only; its later hot/dry goals were unobserved because the plan was replaced at sunrise. Governed-window score/compliance was 62.7, with legacy temperature compliance 57.0% and VPD compliance 47.6%. Stress was 1.69 h heat, 1.50 h VPD-low, 0.46 h VPD-high, and 0 h cold. The predicted high VPD-low window did occur, but suppression/stable dehumidification only partly contained it: dew safety held for the completed day at 6.7F minimum margin and 0 dew-risk hours, while VPD remained the weaker axis and accumulated 1.50 h low-side stress. The no-cold outcome and safe dew margin support the conservative overnight posture. The planned 08:30 hot/dry handoff, fog escalation, and daytime water-budget effects were outside the governed window and are unproven, not failed. Reliable electric/gas/total cost and mist-water fields are unavailable, so resource effects cannot be graded. Score: 6/10

Hypothesis

Testing: Test a two-stage handoff: stable high-dew-floor humid-night suppression until live sunrise conditions, then forecast-discounted but fully band-coupled thermal/moisture authority before the high-solar ramp. Expected outcome: From 00:15-08:30 MDT, keep dew-risk hours at 0, minimum dew margin at or above 7F, cold stress at 0, and VPD-low stress below 0.5 h while avoiding unnecessary wet runtime. By 09:45-11:00, provide live-edge-triggered band-relative assist and thermal lead before the forecast hot/dry ramp; do not pre-wet an in-corridor house. The 14:30 fallback should taper only if storm humidity arrives, with the sticky VPD-high guardrail retaining moisture authority if measured recovery has not occurred.

Conditions

Very humid pre-dawn, rapid hot/dry solar ramp from 09:00, then a possible convective temperature and humidity reversal. Forecast VPD is historically high-biased, while yesterday’s solar forecast undercalled the live peak by more than 3x.

Expected stress windows

Parameter rationale

Setpoints

Friday August 14

Primary crop-band changes:

Tactical tunable changes:

Changed secondary parameters:

🌅 Morning Cycle (6:18 AM) — iris-20260814-0613

Reflection

Validating previous cycle: iris-20260813-2001

Previous hypothesis: SUNSET plan for Thursday 2026-08-13. Today is 57.1 versus a 70.4 seven-day average. Graded attributable fields are unavailable; legacy temperature compliance is the narrower axis at 50.9% versus 54.6% VPD. Heat dominated at 7.73 h, followed by 5.81 h VPD-high, 0.80 h VPD-low and 0 cold. Dew safety held at 7.4F minimum margin and 0 risk hours. Reliable utility and cost fields are null; runtime was cooling/fog/vent-heavy, while context water was about 140 gal versus 274 gal over seven days. I evaluated iris-20260813-0017 at 4/10 versus deterministic anchor 2. I credited the directionally correct humid recovery and dew/cold safety; the anchor more appropriately penalized 272 guardrail events, 267 held, and 57.1 window compliance. This plan responds by starting with guardrail-compatible moisture because live VPD is 1.10 kPa above the 0.95 kPa served edge and alert 10302 is active… Result: Replacement-boundary evaluation through about 20:00 MDT, using the plan-window scorecard rather than attributing the later night to this plan. Window score/compliance was 60.8; legacy VPD compliance was the weaker axis at 44.1% versus 54.3% temperature compliance. Stress was 4.46 h heat, 4.34 h VPD-low, 0.93 h VPD-high, and 0 h cold. The predicted dawn VPD-low, daytime heat, and evening low-side risks materialized; the predicted severe VPD-high window did not, and the goal of keeping VPD-high below 4 h succeeded. Forecast temperature peak was close (89.0F predicted versus 87.7F actual), but solar was materially undercalled (767 versus 1233 W/m2) and the RH/VPD dryness forecast was substantially overcalled (27% forecast minimum versus roughly 60% indoor proxy), so live-edge rather than forecast-triggered wetting was correct. The plan preserved dew safety (5.8F minimum margin, 0 risk hours) and avoided cold stress, but VPD-low persisted and corridor performance stayed poor. Daily water was 86 gal with 18 gal misting versus 237 gal seven-day average, so the 275 gal water budget was not binding; reliable gas/electric/total-cost fields are unavailable. The requested fog and shorter pulse cadence cannot be credited as causal successes because the broader transition audit reports held/missed materialization, while the formal window evaluator scope may differ. Overall: safe dew/cold outcome and limited high-side VPD, but weak low-side VPD recovery, heat exposure, and dirty materialization justify 5/10. Score: 5/10

Hypothesis

Testing: Hold full live-edge-triggered thermal and wet-assist authority through the 14:30 convective uncertainty window, then taper only after measured recovery; test whether this avoids yesterday’s solar-underforecast failure without pre-wetting the VPD-low dawn. Expected outcome: Keep dew-risk hours at 0 and minimum dew margin at or above 5F. Clear the active VPD-low drift after sunrise without adding wet runtime, then keep VPD-high stress below 4 h and heat stress below 6 h through the forecast 89F/767 W/m2 ramp. Hold misting water below 150 gal unless persistent, safe high-edge error makes the 275 gal cap necessary. Avoid a repeat moisture-gap alert and avoid tapering peak authority solely on the 15:00 cloud forecast.

Conditions

Humid VPD-low dawn, rapid clear hot-dry ramp, then an uncertain convective cloud/humidity reversal. Forecast VPD is positively biased and yesterday’s solar peak exceeded forecast by more than 3x.

Expected stress windows

Parameter rationale

Setpoints

Friday August 14

Primary crop-band changes:

Tactical tunable changes:

Changed secondary parameters:

🌆 Evening Cycle (8:03 PM) — iris-20260814-2000

Reflection

Validating previous cycle: iris-20260814-0613

Previous hypothesis: SUNRISE plan for Friday 2026-08-14. Thursday closed at planner score 57.2 versus a 70.4 seven-day average. Graded attributable fields and reliable electric/gas/total-cost fields remain unavailable. Legacy temperature compliance was the narrower axis at 49.6% versus 53.3% VPD compliance. Heat dominated at 9.95 h, followed by 7.23 h VPD-high, 1.36 h VPD-low and 0 cold. Water was 140 gal, including 28 gal misting, versus a 274 gal seven-day average, so water use was down sharply and was not the constraint; a gas-versus-electric cost diagnosis cannot be made from null utility fields. previous_plan_validation: iris-20260813-2001 scored 6/10 versus deterministic anchor 3. I credited the successful evening VPD-high recovery, 7.4F minimum dew margin, 0 dew-risk hours, 0 cold stress, and only 0.06 h VPD-high in its short window… Result: Elapsed replacement-boundary grade for the completed 2026-08-14 local-day portion only, approximately 20:03-00:00 MDT; the 04:00, 07:30, and 10:30 Saturday waypoints and the predicted daytime VPD-high/heat windows were unobserved and receive no credit or penalty. Friday closed at planner score/compliance 64.1 versus a 68.5 seven-day average, with legacy VPD compliance 45.9% weaker than temperature compliance 55.5%. Daily stress was heat 8.60 h, VPD-low 8.24 h, VPD-high 1.86 h, cold 0; water was 86 gal including 18 gal misting, below the 237 gal seven-day average, and reliable cost fields were unavailable. Within the sunset window, the predicted high-risk humid/VPD-low regime clearly materialized: hourly VPD declined from about 0.61 to 0.44 kPa, alert 10309 remained active, and the midnight live state was 68.7F, 0.44 kPa VPD, 82% RH, with a 5.8F dew margin. The plan met the zero-fog objective and all climate/wet equipment was off at midnight; dew-risk hours remained 0 and cold stress remained 0. It narrowly missed the requested >=6F minimum dew margin, and the sustained low-side drift indicates the below-3h VPD-low objective was not demonstrated and likely missed across most of the roughly four-hour governed window. Rationale grading: higher mister engagement, 60 s pulse gap, and 0.50 fog escalation suppressed heavy wetting/fog but did not center VPD; the requested 19:00 wet cutoff and 11-12F direct-wet dew guard did not materialize (effective latest hour 21 and dew floor 8), while the 120 gal budget was nonbinding. The transition audit reports 34 matched, 15 already-at-value, and 7 missed writes, including the direct-wet guard changes. The assembled per-plan window scorecard reports all-zero compliance/stress despite fresh hourly telemetry and the active low-side alert, so it is treated as a platform rollup gap rather than a successful zero-stress outcome. Overall: correct regime diagnosis and safe dew/cold/fog outcome, but inadequate VPD-low recovery and dirty materialization; later dry-ramp handoff remains unproven. Score: 4/10

Hypothesis

Testing: Use zero overnight wet duty plus a 12-13F requested dew floor and stable dehumidification posture through the 95-96% RH forecast, then reopen only live-edge-triggered assist at 10:30; test whether this limits VPD-low without creating a delayed VPD-high gap. Expected outcome: Keep dew-risk hours at 0, keep minimum dew margin at or above 6F, hold overnight VPD-low stress below 3 hours, avoid all overnight fog/mist duty, and provide band-coupled moisture/cooling authority by 10:30 MDT if Saturday’s live dry ramp materializes.

Conditions

Very humid 61-66F overnight with 89-96% RH and raw VPD 0.07-0.24 kPa, followed by a variable-cloud late-morning jump toward 81-83F and raw VPD above 2 kPa. VPD forecast is positively biased, so live crop-band edges govern the morning handoff.

Expected stress windows

Parameter rationale

Setpoints

Friday August 14

Primary crop-band changes:

Tactical tunable changes:

Saturday August 15

Primary crop-band changes:

Tactical tunable changes:

Changed secondary parameters:

End-of-Day Summary

Climate

Interior Light Evidence

• Crop DLI: unavailable
• Reason: source_reading_missing
• Provenance: legacy_invalid_exterior_proxy_plus_fixture_estimate
• Validity: dli-validity-v1; 2024-01-01T00:00:00+00:00 → open
• Qualified-light minutes and photoperiod remain independent of unavailable DLI.

Stress Hours

• Heat stress (>85°F): 4.0h
• VPD stress (>2.0 kPa): 4.6h
• Cold stress (<55°F): 10.8h

Economics

Equipment Runtimes

Water

• Total: 86 gal
• Mister: 18 gal

Crop Health (Gemini Vision)

Hourly Pattern

7-Day Stress Context