Methodology

What this site is: a neutral comparison of viewing locations for the 2 August 2027 total solar eclipse. What it is not: a tour seller. Nothing here is ranked by commission; clearly-labelled affiliate links may appear next to genuinely relevant needs (stays, certified glasses) and change no recommendation.

The full technical detail — how durations are computed from NASA's Besselian elements, the validation table against independent sources, and the effort-score formula — lives on /methodology. The short version of the sources:

Sources

• Eclipse geometry and durations: computed from NASA GSFC's published Besselian elements (greatest duration 6m23.2s for 2027; 2m18.2s for 2026) and validated against timeanddate.com and published path summaries — agreement within seconds; two corrections (Cádiz longer than the old quote; Aden/Berbera partial-only) documented on /methodology. Path width 258 km.
• August climatology: Jay Anderson's satellite-derived statistics at eclipsophile.com — the reference eclipse meteorology source (Cádiz 88% sun-visible; strait sites ≈74%; Melilla 61% of max sunshine; ≈0–5% cloud Benghazi→Saudi; Luxor 0.7%; Siwa/Bahariya flagged best-on-path).
• Safety: risk labels summarise the stance of major government travel advisories; check your own government's current advisory — they change.

The honesty rules

• No invented figures. Computed values are labelled computed with their method; quoted values carry their source; anything else is a dash. A wrong duration sends someone to the wrong country.
• Sun altitude and per-location contact times are now published — computed from NASA's Besselian elements and validated against independent tables (method & validation), never estimated.
• High-risk ≠ hidden. Libya, Yemen and Somalia stay on the list because the shadow crosses them — with plain statements that they are not ordinary travel options, and no planning numbers that would imply otherwise.

Corrections: hello@eclipsewhere.com.