mirror of
https://github.com/acamarata/pray-calc-ml.git
synced 2026-07-01 11:24:26 +00:00
ada08e7ec4
Add 6 new data collection pipelines and their processed outputs: Sources added: - TESS/Stars4All photometer network: 37 months (Jun 2017-Aug 2020), ~40k raw events from 100+ European stations via Zenodo archives - Globe at Night citizen science: 26k twilight observations (2006-2024), filtered from 308k total observations for solar depression 6-22 deg - GaN-MN continuous monitoring: 45 months (Jan 2022-Sep 2025), ~12.5k twilight events from 88 stations across 20+ countries - Galicia SQM network: 14 stations, 1-min resolution, 7.5k events - Madrid/Majadahonda SQM: multi-year continuous monitoring, 3.1k events - washetdonker.nl Netherlands: 7 stations, 3.3k morning events - Academic papers: Jordan (Abed 2015), Fayum Egypt, India photometer Pipeline changes: - ingest.py: add all new files to APPROVED_RAW_CSVS allowlist, fix filter to use allowlist instead of hardcoded exclusions - .gitignore: exclude bulk raw data directories (BSRN, TESS, GaN-MN, washetdonker, Globe at Night downloads) Final dataset: 56,668 Fajr + 34,763 Isha = 91,431 total records Previous: 5,871 Fajr + 46 Isha = 5,917 total records
62 lines
3.6 KiB
Markdown
62 lines
3.6 KiB
Markdown
# Determination of the True Dawn by Several Different Ways at Fayum in Egypt
|
|
|
|
**Authors:** M.G. Rashed, Yasser A. Abdel-Hadi, M.S. El-Nawawy, M.Y. Amin, U.A. Rahoma, M.A. Semeida, A. Abul-Wafa, G.A. Goma, M. Sedek, M.M. Hussien, A.A. Elminawy, M.A. Shahat, A.H. Ibrahim, A.R. Mouner, A.H. Hassan, and Hussein M. Farid
|
|
**Year:** 2022
|
|
**Journal:** International Journal of Mechanical Engineering and Technology (IJMET), 13(10): 8-24
|
|
**DOI:** https://doi.org/10.17605/OSF.IO/9K3MJ
|
|
**URL:** https://iaeme.com/Home/article_id/IJMET_13_10_002
|
|
**Sites studied:** Wadi al Hitan, Fayum, Egypt (29 17'N, 30 03'E, 50m elevation)
|
|
**Observation method:** SQM (LU-DL, Unihedron) in horizontal (H) and zenith (Z) directions, plus naked eye by groups of 9+ astronomers
|
|
**Date range:** December 9-11, 2018 (Group 1) and December 19, 2019 (Group 2)
|
|
**Records:** 4 nights of SQM data, 2 separate naked-eye group observations
|
|
**Mean angle (Fajr):** D0 = 14.0 to 14.8 depending on method
|
|
|
|
## Location Details
|
|
|
|
- Wadi al Hitan, Fayum desert, Egypt
|
|
- 29 17'N, 30 03'E, 50m elevation
|
|
- Desert area far from urban light pollution
|
|
- Chosen specifically for astronomical observation quality
|
|
|
|
## Summary
|
|
|
|
A multi-method study combining SQM photometry and naked-eye observations over 4 nights at Fayum desert. The paper applies four different criteria to determine true dawn:
|
|
|
|
1. Eye threshold method (0.83 mag from full night): D0 = 14.7 degrees
|
|
2. Naked eye group observation (9+ trained observers, tayakkun): D0 = 14.8 degrees
|
|
3. Kocifaj relation H = pi*Z: D0 = 14.0 degrees
|
|
4. Luminance threshold 0.015 cd/m2 (mesopic region): D0 = 13.75 degrees (clear), 10 degrees (cloudy)
|
|
|
|
Key SQM findings:
|
|
- Full night illuminance: 18.76 mag (east direction), 16.92 mag (zenith) for clear nights
|
|
- Full night illuminance: 22 mag for cloudy night
|
|
- The H/Z ratio equals pi at D0 = 14 degrees (Kocifaj criterion)
|
|
- Clear vs cloudy comparison shows clouds delay apparent dawn by ~4 degrees
|
|
|
|
Table 4 provides SQM readings at D0 = 6, 12, 14, 14.8, 18 degrees for Dec 19, 2019 in both east and zenith directions. This is valuable calibration data.
|
|
|
|
Clouds absorb 19% of the night sky energy during full night and 2.5% during twilight.
|
|
|
|
## Data Availability
|
|
|
|
**LIMITED PER-NIGHT DATA.** The paper presents:
|
|
- SQM curves as figures for 4 nights (Dec 9, 10, 11 2018 + Dec 19 2019)
|
|
- Table 4: SQM readings at specific D0 values for Dec 19, 2019
|
|
- Tables 2-3: Statistical values of twilight luminance energy per degree for Dec 10-11, 2018
|
|
- No per-minute time series data published
|
|
|
|
The SQM curves in the figures could be digitized to extract approximate readings at specific D0 values, but the paper does not provide raw time-stamped SQM data.
|
|
|
|
For ML training, the useful data points are:
|
|
- Dec 9, 2018: clear night, SQM threshold at D0 = 14.6 degrees
|
|
- Dec 10, 2018: clear night, mesopic threshold at D0 = 13.75 degrees
|
|
- Dec 11, 2018: cloudy night, threshold at D0 = 10 degrees (excluded from ML, clouds)
|
|
- Dec 19, 2019: clear night, D0 = 14.0 (H=piZ method), D0 = 14.8 (naked eye group)
|
|
|
|
## Comprehensive Literature Table
|
|
|
|
Table 1 (page 12) provides an excellent summary of all published twilight studies chronologically from 1909 to 2022, covering Egypt, Libya, Saudi Arabia, Yemen, Indonesia, and Malaysia. This is a useful cross-reference for the aggregate D0 database.
|
|
|
|
## For ML Training
|
|
|
|
Can extract 4 data points (one per clear night) with SQM-derived D0 values. The naked-eye group observation on Dec 19, 2019 gives D0 = 14.8 degrees. Only useful as aggregate/representative data since exact dawn times are not given (only D0 values derived from SQM threshold analysis).
|