What the station is doing against what was expected: the forecast curve as it was issued, last night at the same hour, the nights on record, the mast against the station. It says what is happening and leaves the why to the person reading it. Every reading of the last 7 days is on the Station's Readings page; the forecast itself is on Today. The record of every graded day is further down this page.
Today reading 16 min ago
At 12:15 PM the station reads 62.8°. The hour-by-hour forecast, as published at 7:05 AM, had 57.5° for this hour: the station is 5.3° over it, and that gap has opened 4.6° since 9:15 AM.
The published high for today is 63.4°; the NWS says 62°, WxSim 64.9°, the four-model average 61.8°.
The main model has 63.6°, its physics alone 64.6°.
The 51 runs put 8 in 10 between 62 and 65°.
So far today the station has reached 61.7°; on how the station is running, the Today page now expects 65.3°.
Baselines: the hour-by-hour forecast as published when the window began (the Hour by hour page's curve, which starts from the station's reading), or until the Tower has kept one, the curve frozen for the Today page's own grading; last night from the same readings file; the main model's run restarted each hour from the latest reading (Physics page).
The levels
The air 115 inches up the mast reads the same as the station's: no inversion across the mast.
The grass is 8.2° warmer than the station's air. It sits 24.5° above the dew point, so dew is not forming.
The soil is 0.2° warmer at 8 inches than at 2, so heat is flowing up toward the surface.
Over the last week the top foot of soil has been losing heat, 1.2° a day.
From the ground sensor's latest reading and its record of nights (Ground page).
What happened last 12 hours
- The grass reached 32° at 1:20 AM.
- The dew point rose 2.0° between 5:00 AM and 5:30 AM: moister air arrived.
- The air 115 inches up crossed above the station's at 5:30 AM: the inversion began.
- The air reached saturation (97% humidity) at 5:30 AM.
- Cooling stalled by 5:50 AM: the hour before fell 1.8°, the last hour 0.0°. Cloud, moist air or mixing from above.
- The air 115 inches up fell below the station's at 7:25 AM: the inversion broke.
Rules over the 5-minute readings: the first wind after half an hour of calm at night, a warm-up or a dew point rise of 2° or 1.5° within 30 minutes, a fall of 5° within 30 minutes away from sunset, a pressure jump of 0.7 mb in 10 minutes, cooling that stalls after a falling hour, the mast crossing the station, saturation, the grass at 32°, rain, gusts of 15 mph.
The rest of the day
Sunset is at 5:29 PM.
Tonight's published low is 40.8°; the NWS says 41°.
The last verdict
Tue Nov 10 night's low was 34.2°. WxSim 36.0° (+1.8); Ours 36.1° (+1.9); Four-model average 38.4° (+4.2); NWS 39.0° (+4.8). Closest: WxSim.
The most recent completed period in the record below, each source against the station's actual.
Background
All 30 parts of the Tower report normal.
The ground sensor sent 286 of 288 readings in the last 24 hours; battery 4.01 V, radio -69 dBm.
The Studio's latest research report, relayed through the Tower Wed 10:28 AM: The night model with a proper soil column under the grass fits the 474 training nights 1 percent worse than the current night physics (and 5 percent worse than its best rival, 1G). (all of them: the Engine page).
The record so far
For the forecasts made the day before, over the last 28 days:
Daytime highs
Nighttime lows
"Off by" means how many degrees the forecast high or low was from what the station measured, averaged over the days graded. Smaller is better. Every forecast is graded on what it said at the same time ahead, so changing its mind later never improves its score. 28 days is not many yet, and these numbers will move as more days are added. The year-long test further down this page, on about 359 days, is the steadier guide.
Every range, every forecast
How many degrees each forecast was off, on average. The same day or night is graded three ways: on the forecast that was out just before it began (for the high, before 8 AM that morning; for the low, before 8 PM that evening), on the forecast made the day before, and on the one made two days before. The number in parentheses is how many days or nights each figure covers.
Daytime highs
| Forecast | Made just before the day or night began | Made the day before | Made two days before |
|---|---|---|---|
| Our forecast (the statistical forecast the site shows) | 1.7° (27 days) | 1.7° (26 days) | 2.1° (25 days) |
| The National Weather Service | 2.2° (27 days) | 2.3° (26 days) | 2.4° (25 days) |
| WxSim | 3.5° (26 days) | 2.6° (25 days) | 2.6° (24 days) |
| The average of the four weather models | 2.5° (27 days) | 2.8° (26 days) | 3.2° (25 days) |
| Our physics forecast, corrected (the candidate, on trial, L-038) | 1.4° (20 days) | 1.8° (19 days) | not made |
Nighttime lows
| Forecast | Made just before the day or night began | Made the day before | Made two days before |
|---|---|---|---|
| Our forecast (the statistical forecast the site shows) | 1.2° (28 nights) | 1.3° (27 nights) | 2.2° (26 nights) |
| The National Weather Service | 2.6° (28 nights) | 2.5° (27 nights) | 2.8° (26 nights) |
| WxSim | 2.9° (26 nights) | 3.0° (25 nights) | 2.8° (24 nights) |
| The average of the four weather models | 1.9° (28 nights) | 2.1° (27 nights) | 3.1° (26 nights) |
| Our physics forecast, corrected (the candidate, on trial, L-038) | 1.3° (20 nights) | 1.3° (19 nights) | not made |
The closest forecast in each column is in bold. Since November 1, 2026 a second complete forecast built on our physics, corrected with the station's own record, is made at the same times as ours and graded the same way; if it wins it becomes the forecast (the plan and what counts as winning: L-038). It forecasts today and tomorrow only, so it has no "two days before" forecast. Which model is which, with today's numbers from every one of them side by side: The models. How the physics works: Physics.
Recent days
| Day or night | The station measured | Our forecast | The National Weather Service | WxSim | The average of the four models |
|---|---|---|---|---|---|
| Tuesday, November 10, the low that night | 34.2° | 36.1°+1.9 | 39.0°+4.8 | 36.0°+1.8 | 38.4°+4.2 |
| Tuesday, November 10, the high | 66.0° | 66.4°+0.4 | 64.0°−2.0 | 70.7°+4.7 | 63.4°−2.6 |
| Monday, November 9, the low that night | 40.5° | 42.4°+1.9 | 43.0°+2.5 | 37.0°−3.5 | 43.3°+2.8 |
| Monday, November 9, the high | 61.2° | 61.7°+0.5 | 59.0°−2.2 | 66.4°+5.2 | 59.1°−2.1 |
| Sunday, November 8, the low that night | 36.9° | 36.2°−0.7 | 35.0°−1.9 | 37.5°+0.6 | 37.0°+0.1 |
| Sunday, November 8, the high | 59.7° | 59.1°−0.6 | 57.0°−2.7 | 61.7°+2.0 | 56.7°−3.0 |
Each forecast as it stood just before the day or night began. The number after each forecast shows how far off it was: +2.2 means 2.2 degrees too warm, and −2.3 means 2.3 degrees too cold. The closest forecast on each row is in bold.
The year-long test
Before going live, our forecast was tested on a full year of days it had never seen (October 2025 to October 2026), against the average of the four weather models on the same days.
| Days ahead | Highs: our forecast | Highs: the models | Highs: our forecast within 2° | Lows: our forecast | Lows: the models | Lows: our forecast within 2° |
|---|---|---|---|---|---|---|
| 1 | 1.6° | 2.0° | 68% of days | 1.7° | 1.9° | 69% of days |
| 2 | 2.0° | 2.5° | 61% of days | 2.0° | 2.3° | 62% of days |
| 3 | 2.4° | 2.8° | 53% of days | 2.3° | 2.6° | 53% of days |
| 4 | 2.8° | 3.2° | 49% of days | 2.5° | 2.8° | 51% of days |
| 5 | 3.2° | 3.6° | 46% of days | 3.1° | 3.4° | 44% of days |
| 6 | 4.0° | 4.3° | 37% of days | 3.8° | 4.0° | 35% of days |
| 7 | 4.8° | 4.9° | 29% of days | 4.7° | 4.9° | 27% of days |
"Our forecast" and "the models" are how many degrees each was off, on average. "Within 2°" is the share of days our forecast was within 2 degrees of what the station measured. "Days ahead" counts the day itself as 1, so 1 is the forecast made that morning or evening, and 2 is the one made the day before. Our forecast was closer than the models' average at every range.
Rain chances
| Forecast | Daytime, made the day before | Overnight, made the day before |
|---|---|---|
| Our forecast | 0.089 (26 days) | 0.096 (27 nights) |
| The National Weather Service | 0.090 (26 days) | 0.122 (27 nights) |
A score for rain chances: 0 would be perfect (saying 100% when it rained and 0% when it did not), and lower is better. With few rainy days so far, these mostly show dry days called dry.
More ways to look at it
- Day by day: one row per day, every source beside the station's high and low, with a goal to count the hits.
- How it went: the last completed period replayed hour by hour against the forecast as issued.
- Against the station, hour by hour: every weather model and ours drawn against the station's readings, for the last two days and today.
- Misses: the biggest misses, and what was going on.
- The forecast adjusted during the day: how much the same-day adjustments help. They are graded separately and never count on this page.
- Our physics model: how it works, and its other tests.