May 11, 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.

Delivery events

Plan writes

No-change acknowledgements

Pending

Event	Delivered	Type	Status	Note
Ad-hoc planning cycle via MCP plan_run(mode=smoke)	03:41	MANUAL	delivery_failed	Delivery recorded without a public plan write.
Ad-hoc planning cycle via MCP plan_run(mode=smoke)	03:44	MANUAL	acked	Acknowledged: Validation mode acknowledge-only smoke: operator request explicitly instructed no set_plan or set_tunable; closing manual planning SLA without behavior change. Resolved 03:45 MDT.
Ad-hoc planning cycle via MCP plan_run(mode=smoke)	03:57	MANUAL	acked	Acknowledged: Validation mode acknowledge-only smoke requested; no plan or tunable changes made. Resolved 03:58 MDT.
Operator recovery request: write a real full 72-hour plan now to restore the active future setpoint horizon after the planning gateway audit repair. Use set_plan, not acknowledge_trigger. Include every tactical Tier 1 param in every transition, with 5-8 waypoints spanning dawn, morning ramp, solar max, peak stress, decline, evening, and overnight. Before planning, evaluate the previous completed plan if enough outcome data exists, call knowledge_search over lesson,plan,site_doc,playbook,observation using today's forecast and stress headline, and use retrieved historical plans, observations, site docs, and lessons as evidence. Close the audit loop with the trigger_id and planner_instance from this prompt.	04:46	MANUAL	plan_written	Wrote public plan iris-20260511-0448. Resolved 04:49 MDT.
validation ack-only: semantic retrieval end-to-end check after MCP OpenAI key repair. Call knowledge_search with source_types lesson,plan,site_doc,playbook,observation for hot dry high solar misting outcomes, call lessons_search for the same condition, then acknowledge_trigger only if both return non-error results. Do not call set_plan or set_tunable during this validation.	04:52	MANUAL	acked	Acknowledged: Validation ack-only semantic retrieval check completed: lessons_search returned non-error results, and knowledge_search with source_types lesson,plan,site_doc,playbook,observation returned non-error results on retry after an initial transient MCP session termination. No set_plan or set_tunable called per operator request. Resolved 04:54 MDT.
Morning planning cycle	05:49	SUNRISE	plan_written	Wrote public plan iris-20260511-0556. Resolved 05:51 MDT.
Solar peak planning checkpoint	12:57	SOLAR_MAX	plan_written	Wrote non-archive plan iris-oneshot-20260511-1258. Resolved 12:58 MDT.
Peak Stress	14:57	TRANSITION	acked	Acknowledged: Peak-stress checkpoint read: current greenhouse is hot/dry (86.6F, VPD 2.21 kPa) but tracking the severe hot-dry plan posture. Controller is already in VENTILATE with both fans, vent, and fog on; active setpoints are already near the aggressive dry-day floor (mister_engage_kpa 1.05, gap 15s, fog_escalation_kpa 0.15, bias_cool -2, dwell gate enabled). Dew-point margin is safe at 21.3F, so no condensation constraint... Resolved 14:57 MDT.
Forecast deviation	17:18	FORECAST_DEVIATION	plan_written	Wrote non-archive plan iris-oneshot-20260511-1719. Resolved 17:19 MDT.
Decline	19:06	TRANSITION	plan_written	Wrote non-archive plan iris-oneshot-20260511-1907. Resolved 19:07 MDT.
Evening planning cycle	20:05	SUNSET	plan_written	Wrote public plan iris-20260511-2005. Resolved 20:07 MDT.

🌅 Morning Cycle (4:49 AM) — `iris-20260511-0448`

Status

validated

Outcome score

6/10

Changed parameters

none recorded

Reflection

Validating previous cycle: iris-20260510-2058

Previous hypothesis: Validation smoke plan: compact 6-hour overnight posture from ~21:00-03:00 MDT. Current indoor conditions are cool and humid enough that the safe objective is to avoid unnecessary mist/fog, preserve dew-point margin, and reduce heater-to-vent oscillation before the later Monday dry ramp. Result: Backlog evaluation at MIDNIGHT 2026-05-20. Plan was a short recovery/coverage window (05-11 04:49-05:51 per context). The current prompt does not surface a dedicated plan-window scorecard; no specific hypothesis can be validated beyond restoring coverage. Score is neutral-positive for coverage repair, with no durable climate lesson extracted. Score: 6/10

Hypothesis

Testing: Recovery plan after planning gateway audit repair: restore full active future horizon and test whether aggressive fog/mist with dwell gate on can reduce severe dry-day VPD-high without causing the previous evening VPD-low carryover. Expected outcome: Plan horizon restored with 8 complete Tier-1 waypoints. For Monday severe dry window, target VPD-high stress under 3h, dew-point risk 0h, and avoid water-budget bottleneck; accept some heat stress as physics-limited above 85F.

Conditions

Outdoor peak

88.0°F

RH minimum

5.0%

Solar peak

901.0 W/m²

Cloud average

18.0%

Monday starts cool and humid indoors, then becomes severe hot-dry with RH 5-14% and solar around 900 W/m2; Tuesday is moderate dry; Wednesday repeats severe heat/VPD before evening clouds.

Expected stress windows

vpd_highhigh · 2026-05-11T09:00:00-06:00 to 2026-05-11T19:30:00-06:00

early ramp from engage 1.30 to 1.05, 15-20s gaps, fog_escalation_kpa 0.15, short sealed windows, south/west weighting

heathigh · 2026-05-11T12:00:00-06:00 to 2026-05-11T18:30:00-06:00

bias_cool -1 to -2, d_cool_stage_2 3, do not extend sealed mist beyond 240s because THERMAL_RELIEF must flush heat

vpd_lowmedium · 2026-05-11T20:30:00-06:00 to 2026-05-12T07:30:00-06:00

evening unwind to engage 1.45, all 2.10, gap 35, fog escalation high while fog is time-window blocked

vpd_highhigh · 2026-05-13T09:00:00-06:00 to 2026-05-13T18:00:00-06:00

repeat severe-day posture for 90F/RH 12-21% with aggressive fog and short thermal relief cycles

Parameter rationale

mister_engage_kpa1.7 → 1.05; Mon 11:00-18:00 RH 5-14%, VPD 2.7-4.25 kPa, solar 554-901 W/m2

start misting before the steep dry ramp and hold VPD-high stress under 3h despite physics-limited heat

mister_pulse_gap_s60.0 → 15.0; dry <20% RH validated lesson recommends 15-25s gaps

increase evaporative recovery during peak without relying on longer sealed windows

fog_escalation_kpa0.8 → 0.15; post-PR-A fog can assist inside VENTILATE once VPD exceeds band plus escalation; forecast outdoor VPD exceeds 3 kPa for six hours

bring fog assist online during VENTILATE and reduce stubborn VPD-high versus Sunday

mist_max_closed_vent_s240.0 → 240.0; lesson says not to extend sealed max when indoor heat risk is high; solar gain outpaces mist cooling

avoid trapping heat while allowing repeated humidity recovery

sw_dwell_gate_enabled1.0 → 1.0; recent equipment shows high transition counts and PR #35 exempts THERMAL_RELIEF from dwell gate

reduce non-safety whipsaw without blocking heat flushes

mister_vpd_weight1.5 → 2.8; south/west dry first under solar and south misters are highest effectiveness

spend water on most productive mist zones while center crop visual stress is monitored separately

bias_cool3.0 → -2.0; peak outdoor 85-88F today and 90F Wednesday, physics-limited cooling

prefer earlier ventilation/fans during peak heat and accept unavoidable heat stress rather than sealing too long

Setpoints

Monday May 11

Primary crop-band changes:

Time	Values	Note
05:00	hyst 0.5	Dawn guarded posture: indoor VPD is low-normal and dew margi
08:30	hyst 0.4	Morning ramp: forecast VPD jumps from 0.6 to 1.9 by 10:00, s
11:00	hyst 0.35	Solar-max onset: outdoor RH falls below 15% and VPD exceeds
14:00	hyst 0.3	Peak stress: 85-88F, RH 5-8%, VPD near 4 kPa. Aggressive VPD
18:30	hyst 0.4	Decline: solar falls but outside air remains extremely dry;
21:30	hyst 0.45	Evening/overnight: fog is blocked and the prior plan missed

Tactical tunable changes:

Time	Values	Note
05:00	d_cool 3; engage 1.7; all 2.35; pulse 45; gap 60; wt 1.5	Dawn guarded posture: indoor VPD is low-normal and dew margi
08:30	d_cool 3; engage 1.3; all 1.85; pulse 60; gap 25; wt 2.2	Morning ramp: forecast VPD jumps from 0.6 to 1.9 by 10:00, s
11:00	d_cool 3; engage 1.1; all 1.65; pulse 60; gap 20; wt 2.8	Solar-max onset: outdoor RH falls below 15% and VPD exceeds
14:00	d_cool 3; engage 1.05; all 1.55; pulse 60; gap 15; wt 3	Peak stress: 85-88F, RH 5-8%, VPD near 4 kPa. Aggressive VPD
18:30	d_cool 3; engage 1.35; all 2; pulse 45; gap 30; wt 2.2	Decline: solar falls but outside air remains extremely dry;
21:30	d_cool 3; engage 1.45; all 2.1; pulse 45; gap 35; wt 1.8	Evening/overnight: fog is blocked and the prior plan missed

Tuesday May 12

Primary crop-band changes:

Time	Values	Note
08:00	hyst 0.4	Tuesday moderate dry day: RH 19-31% and solar near 884 W/m2

Tactical tunable changes:

Time	Values	Note
08:00	d_cool 3; engage 1.3; all 1.9; pulse 60; gap 25; wt 2.3	Tuesday moderate dry day: RH 19-31% and solar near 884 W/m2

Wednesday May 13

Primary crop-band changes:

Time	Values	Note
09:00	hyst 0.35	Wednesday severe ramp: forecast reaches 90F with VPD above 4

Tactical tunable changes:

Time	Values	Note
09:00	d_cool 3; engage 1.15; all 1.65; pulse 60; gap 20; wt 2.8	Wednesday severe ramp: forecast reaches 90F with VPD above 4

Changed secondary parameters:

Time	Parameter	Change
05:00	bias_cool	initial 3
05:00	bias_heat	initial 1
05:00	d_heat_stage_2	initial 5
05:00	dwell_gate_ms	initial 300000
05:00	enthalpy_close	initial 1
05:00	enthalpy_open	initial -2
05:00	fog_escalation_kpa	initial 0.8
05:00	heat_hysteresis	initial 1.2
05:00	min_fog_off_s	initial 90
05:00	min_fog_on_s	initial 45
05:00	min_heat_off_s	initial 180
05:00	min_heat_on_s	initial 120
05:00	min_vent_off_s	initial 60
05:00	min_vent_on_s	initial 60
05:00	mist_backoff_s	initial 1200
05:00	mist_max_closed_vent_s	initial 240
05:00	mist_thermal_relief_s	initial 90
05:00	mist_vent_close_lead_s	initial 15
05:00	mist_vent_reopen_delay_s	initial 45
05:00	mister_all_delay_s	initial 360
05:00	mister_engage_delay_s	initial 90
05:00	mister_water_budget_gal	initial 500
05:00	outdoor_staleness_max_s	initial 600
05:00	summer_vent_min_runtime_s	initial 180
05:00	sw_dwell_gate_enabled	initial 1
05:00	sw_fog_closes_vent	initial 1
05:00	sw_fsm_controller_enabled	initial 1
05:00	sw_mister_closes_vent	initial 1
05:00	sw_summer_vent_enabled	initial 1
05:00	temp_hysteresis	initial 1.8
05:00	vent_prefer_dp_delta_f	initial 5
05:00	vent_prefer_temp_delta_f	initial 5
05:00	vpd_watch_dwell_s	initial 60
08:30	bias_cool	3 → 0
08:30	bias_heat	1 → 0
08:30	fog_escalation_kpa	0.8 → 0.3
08:30	heat_hysteresis	1.2 → 1
08:30	min_fog_off_s	90 → 60
08:30	min_fog_on_s	45 → 60
08:30	mist_backoff_s	1200 → 900
08:30	mist_vent_reopen_delay_s	45 → 30
08:30	mister_all_delay_s	360 → 240
08:30	mister_engage_delay_s	90 → 60
08:30	mister_water_budget_gal	500 → 600
08:30	temp_hysteresis	1.8 → 1.5
08:30	vpd_watch_dwell_s	60 → 45
11:00	bias_cool	0 → -1
11:00	fog_escalation_kpa	0.3 → 0.15
11:00	min_fog_off_s	60 → 45
11:00	mist_backoff_s	900 → 600
11:00	mister_all_delay_s	240 → 180
11:00	mister_engage_delay_s	60 → 45
11:00	temp_hysteresis	1.5 → 1.2
11:00	vpd_watch_dwell_s	45 → 30
14:00	bias_cool	-1 → -2
14:00	mist_vent_reopen_delay_s	30 → 20
14:00	mister_all_delay_s	180 → 120
14:00	mister_engage_delay_s	45 → 30
14:00	temp_hysteresis	1.2 → 1
18:30	bias_cool	-2 → 0
18:30	fog_escalation_kpa	0.15 → 0.5
18:30	min_fog_off_s	45 → 90
18:30	min_fog_on_s	60 → 45
18:30	mist_backoff_s	600 → 900
18:30	mist_max_closed_vent_s	240 → 180
18:30	mist_vent_reopen_delay_s	20 → 30
18:30	mister_all_delay_s	120 → 240
18:30	mister_engage_delay_s	30 → 60
18:30	temp_hysteresis	1 → 1.5
18:30	vpd_watch_dwell_s	30 → 45
21:30	bias_cool	0 → 2
21:30	bias_heat	0 → 0.5
21:30	fog_escalation_kpa	0.5 → 0.8
21:30	heat_hysteresis	1 → 1.2
21:30	min_fog_off_s	90 → 120
21:30	mist_backoff_s	900 → 1200
21:30	mist_vent_reopen_delay_s	30 → 45
21:30	mister_all_delay_s	240 → 360
21:30	mister_engage_delay_s	60 → 90
21:30	mister_water_budget_gal	600 → 500
21:30	temp_hysteresis	1.5 → 1.8
21:30	vpd_watch_dwell_s	45 → 60
08:00	bias_cool	2 → 0
08:00	bias_heat	0.5 → 0
08:00	fog_escalation_kpa	0.8 → 0.3
08:00	heat_hysteresis	1.2 → 1
08:00	min_fog_off_s	120 → 60
08:00	min_fog_on_s	45 → 60
08:00	mist_backoff_s	1200 → 900
08:00	mist_max_closed_vent_s	180 → 240
08:00	mist_vent_reopen_delay_s	45 → 30
08:00	mister_all_delay_s	360 → 240
08:00	mister_engage_delay_s	90 → 60
08:00	mister_water_budget_gal	500 → 550
08:00	temp_hysteresis	1.8 → 1.5
08:00	vpd_watch_dwell_s	60 → 45
09:00	bias_cool	0 → -1.5
09:00	fog_escalation_kpa	0.3 → 0.2
09:00	mist_backoff_s	900 → 600
09:00	mist_vent_reopen_delay_s	30 → 25
09:00	mister_all_delay_s	240 → 180
09:00	mister_engage_delay_s	60 → 45
09:00	mister_water_budget_gal	550 → 600
09:00	temp_hysteresis	1.5 → 1.2
09:00	vpd_watch_dwell_s	45 → 30

🌅 Morning Cycle (5:51 AM) — `iris-20260511-0556`

Status

validated

Outcome score

3/10

Changed parameters

none recorded

Reflection

Validating previous cycle: iris-20260510-2058

Result: Partial SUNSET evaluation using 2026-05-11 scorecard and plan-window scorecard before midnight. The plan expected severe hot-dry VPD/heat stress and tried aggressive mist/fog with capped 180-240s sealed windows plus early evening unwind. Actual daily performance so far: planner_score 33.9, both-axis compliance 22.2%, temp compliance 48.6%, VPD compliance 29.9%. Stress was dominated by VPD-high 9.48h and heat 9.18h, with VPD-low 4.58h and cold 1.13h. Plan-window scorecard for iris-20260511-0556 shows compliance 18.9%, temp 41.4%, VPD 25.5%, heat 4.64h, cold 0.57h, VPD-high 4.79h, VPD-low 2.32h, cost USD 1.46. Dew-point safety held: dp_margin_min 5.7F and dp_risk 0h. The capped sealed-window / dew-point-safety part worked, but VPD-high and heat control did not meet the expected effect; the day was physics-limited and aggressive mist/fog still could not maintain band compliance. Score: 3/10

New finding: On severe clear hot-dry days with outdoor RH near 10%, solar near or above 900 W/m², and peak outdoor temperatures around 88-91F, capped sealed windows and aggressive fog/mist can preserve dew-point safety but may still leave both heat and VPD-high stress near 9h; evening plans should prioritize recovery and avoid carrying peak aggressiveness overnight rather than trying to erase structural daytime stress. → Added to Lessons Learned

Hypothesis

Testing: Test a sharper time-gated dry-day posture: maximum fog/mist only after live/forecast VPD ramp, capped sealed windows for heat safety, and an earlier evening unwind to reduce VPD-low carryover. Expected outcome: For 2026-05-11, hold VPD-high stress under 3h despite RH 5-14%, keep dew-point risk near 0h with min dp margin above 5F, keep water budget from bottlenecking, and reduce overnight VPD-low below yesterday’s 13.23h while accepting some physics-limited heat stress.

Conditions

Outdoor peak

88.0°F

RH minimum

5.0%

Solar peak

901.0 W/m²

Cloud average

6.0%

Clear severe dry Monday: outdoor RH falls to 5-14% from 11:00-19:00, VPD peaks 4.25 kPa, solar peaks near 901 W/m2; current indoor dawn is cool/humid with VPD 0.51 and safe 7.2F dew-point margin.

Expected stress windows

vpd_lowmedium · 2026-05-11T05:55:00-06:00 to 2026-05-11T08:30:00-06:00

guard dawn with high thresholds and 55s gap until live VPD rises

vpd_highhigh · 2026-05-11T09:30:00-06:00 to 2026-05-11T19:30:00-06:00

ramp engage 1.35 to 1.05, 15-20s gaps, fog_escalation 0.15-0.20, dwell gate on, south/west weighting

heathigh · 2026-05-11T12:00:00-06:00 to 2026-05-11T18:30:00-06:00

bias_cool -1 to -2, d_cool_stage_2 3, mist_max_closed_vent_s capped 240 then 180 so THERMAL_RELIEF can flush heat

vpd_lowmedium · 2026-05-11T18:30:00-06:00 to 2026-05-12T07:30:00-06:00

earlier unwind to engage 1.55 then 1.75, gaps 45-60s, fog escalation 0.6-0.9, shorter sealed window

vpd_highhigh · 2026-05-13T09:00:00-06:00 to 2026-05-13T18:00:00-06:00

repeat severe-day but slightly moderated posture for forecast 90F / 12-21% RH with afternoon cloud risk

Parameter rationale

mister_engage_kpa1.7 → 1.05; Mon 11:00-18:00 RH 5-14%, VPD 2.72-4.25 kPa, solar 554-901 W/m2

limit VPD-high stress under 3h while avoiding pre-misting at humid dawn

mister_pulse_gap_s60.0 → 15.0; dry <20% RH and solar >800 W/m2 at peak; validated dry-day lesson recommends 15-25s gaps

speed evaporative recovery during the confirmed dry window

fog_escalation_kpa0.8 → 0.15; post-PR-A fog assists VENTILATE once VPD exceeds band plus escalation during 11:00-17:00 fog window

use fog for stubborn VPD during ventilation rather than extending sealed heat traps

mist_max_closed_vent_s240.0 → 180.0; retrieval and high-confidence lessons warn SEALED_MIST at >85F traps heat; solar gain outpaces mist cooling

keep dew-point risk near 0h and limit heat trapping while allowing repeated short humidity recovery

mist_backoff_s1200.0 → 600.0; short sealed windows and severe outdoor VPD require repeated attempts, but not immediate reseal loops

allow more dry-window recovery cycles without extending any single sealed event

vpd_hysteresis0.5 → 0.35; VPD ramp is steep after 09:00 and dry-day posture needs tighter response

reduce missed VPD-high excursions during peak while restoring 0.5 overnight

bias_cool3.0 → -2.0; forecast high 88F with peak solar 901 W/m2 and physics-limited cooling

prefer earlier ventilation/fans during peak heat; accept some structural heat stress rather than sealing too long

mister_vpd_weight1.5 → 3.0; south/west dry first under clear solar; south misters are highest effectiveness

spend water on most productive dry zones and protect orchids/cannas without center-zone waste

Setpoints

Monday May 11

Primary crop-band changes:

Time	Values	Note
05:55	hyst 0.5	Guarded dawn: live indoor VPD is still low-normal (0.51) wit
08:30	hyst 0.4	Morning ramp: forecast VPD climbs from 0.60 to 1.91 by 10:00
10:30	hyst 0.35	Dry window confirmed by forecast: RH falls below 20%, solar
13:00	hyst 0.35	Peak stress: RH 8-9%, VPD 3.45-3.94, solar near 900 W/m2. Us
16:30	hyst 0.4	Late peak and fog-window exit: outdoor air remains extremely
18:30	hyst 0.45	Early evening unwind: previous failures were VPD-low carryov
21:30	hyst 0.5	Night posture: suppress mist/fog carryover; retain bias_cool

Tactical tunable changes:

Time	Values	Note
05:55	d_cool 3; engage 1.65; all 2.3; pulse 45; gap 55; wt 1.5	Guarded dawn: live indoor VPD is still low-normal (0.51) wit
08:30	d_cool 3; engage 1.35; all 1.9; pulse 55; gap 30; wt 2.3	Morning ramp: forecast VPD climbs from 0.60 to 1.91 by 10:00
10:30	d_cool 3; engage 1.15; all 1.7; pulse 60; gap 20; wt 2.8	Dry window confirmed by forecast: RH falls below 20%, solar
13:00	d_cool 3; engage 1.05; all 1.6; pulse 60; gap 15; wt 3	Peak stress: RH 8-9%, VPD 3.45-3.94, solar near 900 W/m2. Us
16:30	d_cool 3; engage 1.2; all 1.8; pulse 55; gap 25; wt 2.5	Late peak and fog-window exit: outdoor air remains extremely
18:30	d_cool 3; engage 1.55; all 2.1; pulse 45; gap 45; wt 1.8	Early evening unwind: previous failures were VPD-low carryov
21:30	d_cool 3; engage 1.75; all 2.35; pulse 45; gap 60; wt 1.5	Night posture: suppress mist/fog carryover; retain bias_cool

Tuesday May 12

Primary crop-band changes:

Time	Values	Note
08:00	hyst 0.4	Tuesday moderate dry day: keep support, but avoid Monday-lev

Tactical tunable changes:

Time	Values	Note
08:00	d_cool 3; engage 1.4; all 2; pulse 55; gap 30; wt 2	Tuesday moderate dry day: keep support, but avoid Monday-lev

Wednesday May 13

Primary crop-band changes:

Time	Values	Note
09:00	hyst 0.35	Wednesday repeats severe heat/VPD but with cloud/precip unce

Tactical tunable changes:

Time	Values	Note
09:00	d_cool 3; engage 1.1; all 1.7; pulse 60; gap 20; wt 2.8	Wednesday repeats severe heat/VPD but with cloud/precip unce

Changed secondary parameters:

Time	Parameter	Change
05:55	bias_cool	initial 3
05:55	bias_heat	initial 1
05:55	d_heat_stage_2	initial 5
05:55	dwell_gate_ms	initial 300000
05:55	enthalpy_close	initial 1
05:55	enthalpy_open	initial -2
05:55	fog_escalation_kpa	initial 0.8
05:55	heat_hysteresis	initial 1.2
05:55	min_fog_off_s	initial 90
05:55	min_fog_on_s	initial 45
05:55	min_heat_off_s	initial 180
05:55	min_heat_on_s	initial 120
05:55	min_vent_off_s	initial 60
05:55	min_vent_on_s	initial 60
05:55	mist_backoff_s	initial 1200
05:55	mist_max_closed_vent_s	initial 180
05:55	mist_thermal_relief_s	initial 90
05:55	mister_all_delay_s	initial 360
05:55	mister_engage_delay_s	initial 90
05:55	mister_water_budget_gal	initial 500
05:55	outdoor_staleness_max_s	initial 600
05:55	sw_dwell_gate_enabled	initial 1
05:55	sw_fog_closes_vent	initial 1
05:55	sw_fsm_controller_enabled	initial 1
05:55	sw_mister_closes_vent	initial 1
05:55	sw_summer_vent_enabled	initial 1
05:55	temp_hysteresis	initial 1.8
05:55	vent_prefer_dp_delta_f	initial 5
05:55	vent_prefer_temp_delta_f	initial 5
05:55	vpd_watch_dwell_s	initial 60
08:30	bias_cool	3 → 0
08:30	bias_heat	1 → 0
08:30	fog_escalation_kpa	0.8 → 0.35
08:30	mist_backoff_s	1200 → 900
08:30	mist_max_closed_vent_s	180 → 240
08:30	mister_all_delay_s	360 → 300
08:30	mister_engage_delay_s	90 → 60
08:30	temp_hysteresis	1.8 → 1.5
08:30	vpd_watch_dwell_s	60 → 45
10:30	bias_cool	0 → -1
10:30	fog_escalation_kpa	0.35 → 0.2
10:30	min_fog_off_s	90 → 60
10:30	min_fog_on_s	45 → 60
10:30	mist_backoff_s	900 → 600
10:30	mister_all_delay_s	300 → 240
10:30	mister_engage_delay_s	60 → 45
10:30	temp_hysteresis	1.5 → 1.3
13:00	bias_cool	-1 → -2
13:00	fog_escalation_kpa	0.2 → 0.15
13:00	mister_all_delay_s	240 → 180
13:00	temp_hysteresis	1.3 → 1.2
13:00	vpd_watch_dwell_s	45 → 30
16:30	bias_cool	-2 → -1
16:30	fog_escalation_kpa	0.15 → 0.25
16:30	min_fog_off_s	60 → 90
16:30	min_fog_on_s	60 → 45
16:30	mist_backoff_s	600 → 900
16:30	mist_max_closed_vent_s	240 → 180
16:30	mister_all_delay_s	180 → 300
16:30	mister_engage_delay_s	45 → 60
16:30	temp_hysteresis	1.2 → 1.5
16:30	vpd_watch_dwell_s	30 → 45
18:30	bias_cool	-1 → 0
18:30	fog_escalation_kpa	0.25 → 0.6
18:30	min_fog_off_s	90 → 120
18:30	mist_backoff_s	900 → 1200
18:30	mister_all_delay_s	300 → 360
18:30	mister_engage_delay_s	60 → 90
18:30	temp_hysteresis	1.5 → 1.8
18:30	vpd_watch_dwell_s	45 → 60
21:30	bias_cool	0 → 2
21:30	bias_heat	0 → 0.5
21:30	fog_escalation_kpa	0.6 → 0.9
21:30	mister_all_delay_s	360 → 420
08:00	bias_cool	2 → 0
08:00	bias_heat	0.5 → 0
08:00	fog_escalation_kpa	0.9 → 0.4
08:00	min_fog_off_s	120 → 90
08:00	mist_backoff_s	1200 → 900
08:00	mist_max_closed_vent_s	180 → 240
08:00	mister_all_delay_s	420 → 360
08:00	mister_engage_delay_s	90 → 75
08:00	temp_hysteresis	1.8 → 1.6
09:00	bias_cool	0 → -1.5
09:00	fog_escalation_kpa	0.4 → 0.2
09:00	min_fog_off_s	90 → 60
09:00	min_fog_on_s	45 → 60
09:00	mist_backoff_s	900 → 600
09:00	mister_all_delay_s	360 → 240
09:00	mister_engage_delay_s	75 → 45
09:00	temp_hysteresis	1.6 → 1.3
09:00	vpd_watch_dwell_s	60 → 45

🌆 Evening Cycle (8:07 PM) — `iris-20260511-2005`

Status

validated

Outcome score

9/10

Changed parameters

none recorded

Reflection

Validating previous cycle: iris-20260511-0556

Previous hypothesis: SUNRISE plan for Monday severe hot-dry ramp. Yesterday’s plan scored 49.6 daily with 39.9% both-axis compliance; VPD was the bottleneck (44.3% vs 65.5% temp), and VPD-low dominated (13.23h) after successful VPD-high suppression. plan_evaluate wrote outcome_score 5 vs anchor_score 3, no deviation warning. Lessons 88 and 105 were validated: aggressive mist/fog is justified only for today’s true extreme window, and the evening unwind must start before humidity rebounds. Retrieval found prior extreme-dry plans with engage 1.3, 15s gaps, low fog escalation, but also repeated VPD-low carryover; current high-confidence hot-day lessons warn not to extend sealed time under >85F heat, so this plan keeps short 180-240s sealed windows and uses post-PR-A vent+fog rather than 900s heat traps. Result: SUNSET recovery plan performed very well through the overnight-to-sunrise window. Window scorecard for iris-20260511-2005: planner_score 93.9, both-axis compliance 93.7%, temp compliance 98.5%, VPD compliance 94.6%. Stress stayed low: heat 0.01h, cold 0.02h, VPD-high 0.03h, VPD-low 0.09h by window summary; current partial-day scorecard shows total stress 0.38h with VPD-low 0.22h and dew-point safety preserved (dp_margin_min 6.6F, dp_risk 0h). The hypothesis was correct: moderate overnight recovery avoided peak-day carryover, fog remained suppressed outside its useful window, bias_cool +3.5/bias_heat +0.5 prevented meaningful cold/vent chatter, and the Tuesday handoff starts from a stable cool/humid baseline. Cost remained low relative to 7-day average: partial today cost USD 0.76 vs 7d avg USD 5.47, with water 139 gal vs 217 gal average and gas use sharply below average. Score: 9/10

New finding: Recovery-first sunset posture after a severe hot-dry failure can reset the greenhouse safely: use moderate evening mist thresholds, high overnight fog escalation, short sealed windows, and anti-chatter bias_cool +3 to +3.5; expect high overnight compliance and zero dew-point risk when dawn dp_margin remains above 6F. → Added to Lessons Learned

Hypothesis

Testing: Test a recovery-first sunset posture after a failed severe hot-dry day: moderate evening VPD recovery, no overnight peak aggressiveness, and a built-in Tuesday dry ramp plus evening unwind. Expected outcome: Overnight dp_risk_hours remains 0, additional VPD-low stays under 1h, cold_stress from heater/vent chatter stays under 1h, and Tuesday moderate dry-window VPD-high stays under 3h if the SUNRISE cycle is missed.

Conditions

Outdoor peak

79.0°F

RH minimum

11.0%

Solar peak

882.0 W/m²

Cloud average

23.0%

Post-sunset outdoor air remains very dry tonight, VPD about 2.2 falling to 1.1 by dawn; indoor dew-point margin is safe at 13.4F now. Tuesday is moderate dry, 79F and 18-23% RH with high solar, not a repeat of Monday’s 88F/near-10% extreme.

Expected stress windows

vpd_highmedium · 2026-05-11T20:05:00-06:00 to 2026-05-12T03:30:00-06:00

moderate mist thresholds engage 1.55-1.60 with 45-50s gaps; fog escalation high because fog is time-window blocked; sealed window capped at 180s

coldlow · 2026-05-12T04:30:00-06:00 to 2026-05-12T07:30:00-06:00

bias_cool +3.5 and bias_heat +0.5 only near predawn to avoid heater-to-vent oscillation without chasing unnecessary warmth

vpd_highmedium · 2026-05-12T09:30:00-06:00 to 2026-05-12T19:30:00-06:00

moderated dry-day posture: engage 1.35, gap 25s, fog escalation 0.40, short sealed windows, then evening unwind

Parameter rationale

mister_engage_kpa1.75 → 1.55; current indoor VPD is 0.91 and outdoor VPD stays 1.9-2.2 kPa through midnight

recover remaining VPD-high stress without using Monday's peak 1.05 kPa setting overnight

mister_pulse_gap_s60.0 → 45.0; dry evening air but no solar and safe 13.4F dew-point margin

permit moderate VPD recovery while keeping dp_risk_hours near 0

fog_escalation_kpa0.15 → 0.9; fog is firmware-blocked overnight; high escalation prevents accidental early fog before morning dry ramp is confirmed

avoid overnight over-humidification and VPD-low carryover

bias_cool-2.0 → 3.5; overnight cooling from 70F to 56F with possible pre-dawn heating; retrieved sunset plans favor +3 to +4 against heater/vent chatter

keep cold_stress from vent oscillation under 1h overnight

mist_max_closed_vent_s240.0 → 180.0; lessons 95 and 89 warn not to extend sealed heat traps after hot-dry stress; dew safety is the overnight priority

keep dew-point risk at 0h while allowing repeated short humidity recovery

mister_engage_kpa1.55 → 1.35; Tuesday 10:00-18:00 forecast VPD 1.58-2.80 kPa, RH 18-31%, solar 584-882 W/m2

hold Tuesday VPD-high stress under 3h without the stronger Monday extreme posture

Setpoints

Monday May 11

Primary crop-band changes:

Time	Values	Note
20:10	hyst 0.45	evening_settle: recover residual dry stress moderately while
23:30	hyst 0.45	midnight_posture: keep mist conservative as humidity retenti

Tactical tunable changes:

Time	Values	Note
20:10	d_cool 3; engage 1.55; all 2.25; pulse 45; gap 45; wt 1.8	evening_settle: recover residual dry stress moderately while
23:30	d_cool 3; engage 1.6; all 2.3; pulse 45; gap 50; wt 1.6	midnight_posture: keep mist conservative as humidity retenti

Tuesday May 12

Primary crop-band changes:

Time	Values	Note
05:30	hyst 0.45	pre_dawn: suppress unnecessary mist, add small heat support,
09:30	hyst 0.4	tuesday_dry_ramp_fallback: moderate dry-day VPD support if s
19:30	hyst 0.45	tuesday_evening_unwind_fallback: remove dry-day aggressivene

Tactical tunable changes:

Time	Values	Note
05:30	d_cool 3; engage 1.7; all 2.35; pulse 45; gap 60; wt 1.5	pre_dawn: suppress unnecessary mist, add small heat support,
09:30	d_cool 3; engage 1.35; all 1.95; pulse 60; gap 25; wt 2.4	tuesday_dry_ramp_fallback: moderate dry-day VPD support if s
19:30	d_cool 3; engage 1.75; all 2.35; pulse 45; gap 60; wt 1.5	tuesday_evening_unwind_fallback: remove dry-day aggressivene

Changed secondary parameters:

Time	Parameter	Change
20:10	bias_cool	initial 3
20:10	bias_heat	initial 0
20:10	d_heat_stage_2	initial 5
20:10	dwell_gate_ms	initial 300000
20:10	enthalpy_close	initial 1
20:10	enthalpy_open	initial -2
20:10	fog_escalation_kpa	initial 0.9
20:10	heat_hysteresis	initial 1.2
20:10	min_fog_off_s	initial 120
20:10	min_fog_on_s	initial 45
20:10	min_heat_off_s	initial 180
20:10	min_heat_on_s	initial 120
20:10	min_vent_off_s	initial 60
20:10	min_vent_on_s	initial 60
20:10	mist_backoff_s	initial 900
20:10	mist_max_closed_vent_s	initial 180
20:10	mist_thermal_relief_s	initial 90
20:10	mister_all_delay_s	initial 360
20:10	mister_engage_delay_s	initial 90
20:10	mister_water_budget_gal	initial 500
20:10	outdoor_staleness_max_s	initial 600
20:10	sw_dwell_gate_enabled	initial 1
20:10	sw_fog_closes_vent	initial 1
20:10	sw_fsm_controller_enabled	initial 1
20:10	sw_mister_closes_vent	initial 1
20:10	sw_summer_vent_enabled	initial 1
20:10	temp_hysteresis	initial 1.8
20:10	vent_prefer_dp_delta_f	initial 5
20:10	vent_prefer_temp_delta_f	initial 5
20:10	vpd_watch_dwell_s	initial 60
23:30	bias_cool	3 → 3.5
05:30	bias_heat	0 → 0.5
05:30	mist_backoff_s	900 → 1200
09:30	bias_cool	3.5 → 0
09:30	bias_heat	0.5 → 0
09:30	fog_escalation_kpa	0.9 → 0.4
09:30	min_fog_off_s	120 → 90
09:30	mist_backoff_s	1200 → 600
09:30	mist_max_closed_vent_s	180 → 240
09:30	mister_all_delay_s	360 → 300
09:30	mister_engage_delay_s	90 → 60
09:30	temp_hysteresis	1.8 → 1.5
19:30	bias_cool	0 → 2.5
19:30	fog_escalation_kpa	0.4 → 0.9
19:30	min_fog_off_s	90 → 120
19:30	mist_backoff_s	600 → 1200
19:30	mist_max_closed_vent_s	240 → 180
19:30	mister_all_delay_s	300 → 360
19:30	mister_engage_delay_s	60 → 90
19:30	temp_hysteresis	1.5 → 1.8

End-of-Day Summary

Climate

Temperature

61.8–89.0°F; avg 71.2°F

VPD

0.35–2.35 kPa; avg 0.99 kPa

Relative humidity

38.3–81.9%

Stress Hours

Heat stress (>85°F): 10.9h
VPD stress (>2.0 kPa): 11.4h
Cold stress (<55°F): 1.1h

Economics

Electric

USD 2.22

Gas

USD 1.39

Water

USD 1.240

Total

USD 4.85

Equipment Runtimes

Equipment	Runtime	Reading
Fan 1	593 min	Primary exhaust runtime.
Fan 2	618 min	Secondary exhaust runtime.
Vent	787 min	Intake vent runtime.
Fog	253 min	Fogger runtime.
Heat 1 electric	476 min	Electric heater runtime.
Heat 2 gas	134 min	Gas heater runtime.
Grow lights	0 min	Supplemental lighting runtime.
Mister south	0.38h	South mister runtime.
Mister west	0.30h	West mister runtime.
Mister center	2.78h	Center mister runtime.

Water

Total: 257 gal
Mister: 136 gal

Crop Health (Gemini Vision)

Crop	Zone	Health	Observations	Note
Canna Lilies	south	78%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.
Vanda Orchids	center	63%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.
lettuce	east	78%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.
pepper	east	73%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.
strawberry	east	73%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.

Vision observation notes

Canna LiliesGemini Vision notes

Canna lilies on the floor appear generally healthy, though visibility is somewhat limited in the dark.

Vanda OrchidsGemini Vision notes

[browning roots, dry appearance] Hanging roots appear somewhat dry and brown, possibly indicating insufficient humidity or watering frequency.

lettuceGemini Vision notes

Seedlings appear healthy and are establishing well.

pepperGemini Vision notes

Seedlings appear healthy and green.

strawberryGemini Vision notes

Seedlings appear healthy and green.

Hourly Pattern

Hour	Temperature	VPD	RH
12:00	76.9°F	1.18 kPa	63.0%
13:00	78.7°F	1.41 kPa	58.2%
14:00	79.7°F	1.43 kPa	59.3%
15:00	79.5°F	1.45 kPa	58.4%
16:00	76.4°F	1.23 kPa	60.4%
17:00	74.2°F	1.18 kPa	59.5%
18:00	71.8°F	1.09 kPa	59.0%
19:00	68.0°F	1.12 kPa	52.2%
20:00	65.1°F	0.80 kPa	62.8%
21:00	64.8°F	0.45 kPa	78.4%
22:00	63.6°F	0.39 kPa	80.8%
23:00	63.5°F	0.53 kPa	74.0%
00:00	63.9°F	0.67 kPa	67.0%
01:00	64.0°F	0.71 kPa	65.1%
02:00	63.8°F	0.58 kPa	71.3%
03:00	63.8°F	0.44 kPa	78.4%
04:00	63.3°F	0.41 kPa	79.4%
05:00	64.2°F	0.44 kPa	78.6%
06:00	65.3°F	0.59 kPa	72.7%
07:00	68.4°F	0.70 kPa	70.7%
08:00	71.2°F	0.90 kPa	65.7%
09:00	74.0°F	1.02 kPa	64.6%
10:00	76.6°F	1.26 kPa	60.0%
11:00	76.4°F	1.16 kPa	62.7%

7-Day Stress Context

Date	Heat	VPD high	Cold
2026-05-05	0.0h	2.8h	7.7h
2026-05-06	1.0h	3.6h	7.1h
2026-05-07	4.9h	5.9h	1.8h
2026-05-08	6.5h	7.1h	1.0h
2026-05-09	4.4h	5.6h	1.0h
2026-05-10	7.3h	8.3h	0.9h
2026-05-11	10.9h	11.5h	1.1h

Lab

May 11, 2026

Planner Execution Ledger

🌅 Morning Cycle (4:49 AM) — iris-20260511-0448

Reflection

Hypothesis

Setpoints

Monday May 11

Tuesday May 12

Wednesday May 13

🌅 Morning Cycle (5:51 AM) — iris-20260511-0556

Reflection

Hypothesis

Setpoints

Monday May 11

Tuesday May 12

Wednesday May 13

🌆 Evening Cycle (8:07 PM) — iris-20260511-2005

Reflection

Hypothesis

Setpoints

Monday May 11

Tuesday May 12

End-of-Day Summary

Climate

Stress Hours

Economics

Equipment Runtimes

Water

Crop Health (Gemini Vision)

Hourly Pattern

7-Day Stress Context

🌅 Morning Cycle (4:49 AM) — `iris-20260511-0448`

🌅 Morning Cycle (5:51 AM) — `iris-20260511-0556`

🌆 Evening Cycle (8:07 PM) — `iris-20260511-2005`