The introduction of manmade structures in the etymology of harbor

In a Stackexchange post, a commenter offers a distinction between ports and harbors:

Or to put it another way, "harbor" is a description of the natural geography, while "port" is something made by people.

Emphasis mine

I can wholeheartedly affirm that a port is manmade in any nautical sense of its definition, but I have a question about the claim that harbor is an intrinsically natural phenomenon.

The current usage of *harbor* seems to favor manmade structures without foreclosing the possibility of natural harbors:

A place on the coast where vessels may find shelter, especially one protected from rough water by piers, jetties, and other artificial structures:

OED (American English)Emphasis Mine

Various popular posts make the distinction between natural harbors and manmade harbors. So there is no doubt the current usage accepts both.

This question of manmade structures in harbor is not simplified by the apparent broadening of port to converge as a synonym of harbor:

NOUN

1 A town or city with a harbor where ships load or unload, especially one where customs officers are stationed.

1.1 A harbor:

OED (American English)

Examining the etymology of harbor with my limited tools doesn't create a definitive notion of manmade structures, but seems to imply them strongly in the types of manmade shelters described by OE herebeorg:

"lodging for ships," early 12c.,

probably from Old English herebeorg "lodgings, quarters,"

from here "army, host" (see harry) + beorg "refuge, shelter" (related to beorgan "save, preserve;" see bury);

perhaps modeled on Old Norse herbergi "room, lodgings, quarters."

Sense shifted in Middle English to "refuge, lodgings," then to "place of shelter for ships."

In an unrelated Stackexchange post, reference is made to more etymological data, which seems to confirm my theory: manmade structures were the English implication well before the word received a nautical application, but I have not been able to confirm the source material of that claim either.

Accepting the current dual usage as the status quo, what linguistic evidence might support the claim that harbor is rooted in natural phenomena, and if there is evidence for that, what linguistic event or process introduced the current emphasis on manmade structures.

Let me emphasize again, this is not a question about acceptable current usage!

Solution 1:

A geographer would would describe the character of the boundary between land and sea in terms of capes, headlands, bays and inlets. A harbour suggests manmade adjustments to the terrain to facilitate docking. A port implies not only the presence of a harbour but also commercial activity relating to the harbour.