A complete analysis of Corey Kell's "Operation Resolute Time" (2020) and "Operation North American Sun" (2021–23) solar elevation measurements — compared against standard globe-model predictions.
Corey Kell claims that on the globe model, the sun's elevation cannot exceed 45° when measured 3 hours (one 45° sector) from solar noon. He calls this "the limit" and argues that any measurement above 45° at the 3-hour mark proves the globe model fails.
This claim is based on a fundamental misunderstanding of solar geometry. The standard formula for solar elevation is:
sin(altitude) = sin(lat) × sin(dec) + cos(lat) × cos(dec) × cos(hour_angle)
Where hour_angle at 3 hours from noon = 45°. Latitude and solar declination determine the elevation — and it can easily exceed 45° at mid-latitudes during summer. Below, we compute the predicted globe-model elevation for every one of Kell's recorded measurements and compare.
| Date | Time | E/W | Solar Decl. | Hour Angle | Globe Pred. | Kell Meas. | Δ (Diff) | >45° Limit? |
|---|
After ORT 2020 in Kabul, Kell conducted "Operation North American Sun" (ONAS 21-22-23) in Fayetteville, North Carolina — at almost the same latitude (~35°N) but on the opposite side of the globe. He claims this second dataset confirms the globe model "fails" his 45° limit test with a 56% failure rate.
His most revealing dataset is from 19 June 2023 — hourly readings from 07:00 through 13:12, just days before the summer solstice. This gives us a complete solar elevation curve to compare against the globe model prediction.
An earlier version of this analysis used Kell's self-reported "SSP" (Solar South Pass / solar noon) times to compute hour angles. We took those values at face value without independently verifying them. @helpingflatearthers spotted that our globe predictions were off for specific data points, which led us to cross-check against the NOAA Solar Calculator, PySolar, and SunCalc. It turned out Kell's SSP times were systematically early — by 5 to 48 minutes depending on the date — inflating hour angles and making globe predictions appear lower than they should be.
The page now computes true solar noon from first principles using the Jean Meeus algorithm (the same method used by NOAA). This corrected the ONAS RMSE from ~4.3° down to ~0.9°, with afternoon readings now matching within fractions of a degree. Credit to @helpingflatearthers for catching the discrepancy. Lesson learned: always validate your source data independently, even when it comes from the person you're debunking.
Kabul (34.56°N, 69.31°E, 1817m elev.) and Fayetteville (34.97°N, 79.01°W, 56m elev.) are at nearly identical latitudes but separated by ~148° of longitude and 1,760m of elevation difference. On the globe model, they should produce very similar solar elevations for the same dates/times — and they do. Kell's own data from both sites matches globe predictions to within a few degrees.
On a flat earth with a local, nearby sun, two locations at the same "latitude" but on opposite sides of the disc would see dramatically different sun angles depending on their distance from the subsolar point. The consistency of his readings across both sites is itself evidence for the globe.
| Date | Time | E/W | Solar Decl. | Hour Angle | Globe Pred. | Kell Meas. | Δ (Diff) | >45° Limit? |
|---|
This section contains the original pages from Corey Kell's book alongside transcriptions of key content, including his theoretical framework, methodology, original data tables, and conclusions. Click any image to enlarge.
The excerpts reproduced below are included for purposes of criticism, comment, scholarship, and research, as expressly permitted under the fair use doctrine codified at 17 U.S.C. § 107. Our use is non-commercial and transformative: we reproduce only the specific pages necessary to verify quoted claims, data tables, and methodology, and we accompany each excerpt with substantive commentary, independent calculation, and critical analysis. We reproduce no more of the work than is reasonably necessary to support that analysis, and our use has no adverse effect on the potential market for the original work — readers interested in the complete argument are encouraged to obtain the book directly from the author.
Copyright in the underlying text and figures remains with Corey Kell. No claim of ownership is made or implied. If you are the rights holder and have concerns about any specific excerpt, please open an issue on GitHub and we will respond promptly.
Kell divides the 360° solar day into eight 45-degree sectors, each representing 3 hours. He argues both the globe and flat earth models can be tested by measuring the sun's vertical angle (elevation) at the 3-hour marks from solar noon.
Kell's central claim is that the sun is always at 90 degrees to the observer in both models, and therefore the maximum measurable sun height at 3 hours from noon (one 45° sector) is 45°. His math: 45 + 45 = 90, therefore 45° is "the limit."
This reasoning treats the hour angle as if it directly determines the elevation — which would only be true if the sun were on the celestial equator AND the observer were at the equator AND the surface were a flat plane. In reality, solar elevation depends on three variables together: latitude, declination, and hour angle. At 34–35°N during summer (declination ~+23°), the sun's elevation at a 45° hour angle routinely exceeds 50°.
Period: 7 February – 8 August 2020
Location: Kabul, Afghanistan — N34°33'22.88", E69°18'34.88"
Elevation: 5,963 feet above sea level
GPS: 42S WD 29946 23898
This was described as a non-military operation conducted in a war zone. Testing ended on 7 August 2020 partly due to events within Afghanistan during the on-going war.
The observer station (OS) used a leveled tripod platform base set at 0.5m height, with a digital protractor and radial arm positioned plumb vertically over the sun survey point (SSP). The sun survey point was established along the line of magnetic using a compass and MARX device at solar noon.
| ORT Totals | Tests | Failed | Failure % |
|---|---|---|---|
| Globe Model | 87 | 50 | 58% |
Kell defines "failure" as any measurement exceeding his 45° limit. As shown in our analysis, the globe model predicts these values — they are not failures but expected results.
Red-circled values in Kell's original tables indicate readings that exceed his "globe model limit." N/R = No Reading.
| Date | E/W Time | Easterly (°) | Az (mils) | Westerly (°) | Az (mils) | SSP |
|---|---|---|---|---|---|---|
| 7 Feb 20 | W-15:37 | N/R | N/R | 21.2 | N/R | 12:37 |
| 9 Feb 20 | E-09:37 | 29.1 | N/R | N/R | N/R | 12:37 |
| 10 Feb 20 | E-09:37 | 29.1 | N/R | N/R | N/R | 12:37 |
| 21 Feb 20 | E-09:38 | 33.0 | N/R | N/R | N/R | 12:37 |
| 20 Mar 20 | E-09:20 | 40.0 | 2340 | N/R | N/R | 12:37 |
| 28 Mar 20 | E-09:02/09:30 | 38.7 / 44 | 2340 | N/R | N/R | 12:02 |
| 2 Apr 20 | E-09:02/09:32 | 38.7 / 45.6 | 2280 | N/R | N/R | 12:02 |
| 4 Apr 20 | E-09:02/09:32 | 40.5 / 45.5 | 2280 | N/R | N/R | 12:02 |
| 8 Apr 20 | E-09:02/09:32 | 42.4 / 47.1 | N/R | N/R | N/R | 12:02 |
| 10 Apr 20 | W-15:02 | N/R | N/R | 40.5 | N/R | 12:02 |
| 11 Apr 20 | E-09:02/09:32 | 42.6 / 47.9 | 2000 | N/R | N/R | 12:02 |
| Date | Azimuth (MILS/Deg Mag) | Vertical Angle | Surface Dist (SD)+3E | Over Limit (VA+SD-90) | Remarks |
|---|---|---|---|---|---|
| 19 Jun 20 | 1560/87 | 51.3 | 50 | +11.3 | 101.3 |
| 21 Jun 20 | 1580/88 | 51.3 | 48 | +9.3 | 99.3 |
| 30 Jun 20 | 1500/84 | 50.3 | 59 | +19.3 | 109.3 |
| 2 Jul 20 | 1450/81 | 50.1 | 66 | +26.1 | 116.1 |
| 4 Jul 20 | 1580/88 | 50.1 | 55 | +15.1 | 105.1 |
| 14 Jul 20 | 1600/89 | 49.4 | 54 | +13.4 | 103.4 |
Period: July 2021 – June 2023
Location: Fayetteville, North Carolina, United States
Coordinates: N34°58'11", W79°0'29"
Elevation: 184 feet above sea level
GPS: 17S PU 81842 71503
| ONAS Totals | Attempts/Timing | Completed Tests | Globe Failures | Globe Fail % | Flat Earth Failures | FE Fail % |
|---|---|---|---|---|---|---|
| Totals | 58 | 46 | 26 | 56% | 0 | 0% |
Kell's most detailed single-day dataset — hourly readings from 07:00 through 13:12 (solar noon):
| Time | Easterly Reading (°) | Azimuth (mils) | Note |
|---|---|---|---|
| 07:00 | 10.7 | 1360 | Hourly check |
| 08:00 | 22.5 | 1550 | Hourly check |
| 09:00 | 34.6 | 1700 | Hourly check |
| 10:00 | 45.7 | 1850 | Hourly check |
| 10:12 | 48.4 | 1920 | SSP 3-hour mark |
| 11:00 | 57.8 | 2100 | Hourly check |
| 12:12 | 73.5 | 2420 | Hourly check |
| 13:12 | 79.8 | 3360 | Solar noon |
This hourly curve traces out exactly the solar elevation arc the globe model predicts — a smooth climb from ~11° near sunrise to ~80° at solar noon. This is the standard behavior of a sun 93 million miles away being observed from a rotating globe.
Response: This treats hour angle as directly determining elevation. In reality: sin(alt) = sin(lat)·sin(dec) + cos(lat)·cos(dec)·cos(HA). At 34.5°N with declination +23.4° and HA=45°, the predicted elevation is ~49.4° — well above 45°.
Response: The globe model predicts the majority of his summer measurements should exceed 45°. His "failure rate" is the globe model's success rate — his data matches predictions.
Response: The apparent speed difference is an artifact of using magnetic azimuths without proper declination correction, combined with the observer's position relative to the subsolar point and atmospheric refraction effects. The sun's apparent motion is constant at 15°/hour.
Response: The flat earth model has 0% failure rate in his test because his "limit" is defined in a way that only tests the globe model. He never applies equivalent geometric constraints to the flat earth model. On a flat earth with a sun at ~3,000 miles altitude, simple trigonometry would predict very different elevation angles than what he measures — angles his own data contradicts.
Response: When your data perfectly matches the predictions of the model you're trying to disprove, invoking simulation theory is an admission that the globe model works.
| Kabul (ORT 2020) | Fayetteville (ONAS 21-23) | |
|---|---|---|
| Latitude | 34.556°N | 34.970°N |
| Longitude | 69.310°E | 79.008°W |
| Elevation | 1,817m (5,963 ft) | 56m (184 ft) |
| Separation | ~148° longitude, 1,761m elevation difference | |
| Summer peak elevation | ~51.3° (21 Jun) | ~79.8° at noon, ~48-51° at 3hr |
| Globe prediction match | Yes — within ~2-3° | Yes — within ~2-3° |
Both sites at ~35°N produce nearly identical solar elevation patterns, exactly as the globe model predicts. This cross-continental consistency is a powerful confirmation — and something a flat earth model with a local sun cannot explain.