Shadow-Stick Method
Two shadow marks separated by time give you a real east-west line — one mark doesn't.
A straight stick roughly 30cm long, flat open ground, and direct sun.
The sun's shadow moves west to east through the day, so two shadow-tip marks give you an east-west line.
Plant the stick, mark the shadow tip, wait 15–20 minutes, mark the new tip; the line between marks runs roughly east-west, first mark west.
A perpendicular line through the midpoint of your two marks points roughly north-south — check it against a known landmark if you have one.
Using a single shadow mark. One mark only gives a direction from the stick, not a reliable line.
Night Sky — Locating True North by Stars (Northern Hemisphere)
Polaris barely moves while the sky rotates around it.
The Big Dipper (Plough) or Cassiopeia, both usable to locate Polaris, the North Star.
Polaris sits almost directly above the North Pole and stays nearly fixed while other stars rotate around it.
Trace a line through the two outer stars of the Big Dipper's bowl, extend it about five times their separation to reach Polaris; the ground point beneath it is true north.
Watch Polaris over ten to fifteen minutes — unlike other stars, it should barely appear to move.
Relying on this method south of the equator, where Polaris isn't visible — a different star method is required there.
Reading Landscape and Vegetation Cues (Supporting, Not Primary)
These cues sanity-check a real method — they don't replace one.
Prevailing wind-shaped tree growth, snow melting faster on one slope aspect, general terrain orientation if already mapped.
These are secondary, regional-specific cues that support a primary method — they are not reliable enough to use alone.
Use these only to sanity-check a shadow-stick or star reading, not as a standalone method.
Compare against your shadow-stick or star result; large disagreement means re-run the primary method rather than trusting the vegetation cue.
Trusting "moss grows on the north side of trees." Moss grows wherever it's damp and shaded, which is often not north-facing at all — a persistent myth, not a reliable method.
Daytime: use the shadow-stick method. Nighttime with a clear sky: use star location. Use vegetation and terrain cues only to sanity-check, never as your primary method.
Treating any single natural sign — moss, one shadow mark, one leaning tree — as sufficient on its own. Cross-check with a second method whenever the terrain allows.
Two data points make a line. One data point is a guess.