Complex Numbers

Coasting Along

I'm not entirely sure why I saved this particular Live Science article, and it's only been like three weeks.

    What's the 'coastline paradox'?  
Coastlines are challenging to measure, but why?

Yeah... it's not really a paradox.

Flanked with fjords and inlets, Alaska is the state with the most coastline in the United States.

Easy to accomplish when you're a giant peninsula with many craggy islands offshore.

But what is the length of its oceanic coast?

It depends on whom you ask. According to the Congressional Research Service, the number is 6,640 miles (10,690 kilometers). But if you consult the National Oceanic and Atmospheric Administration (NOAA), the coastal edges of the state total 33,904 miles (54,563 km).

Yep, that's a big difference, all right. Not to brag, but I knew the answer. Still, I want to say that, obviously, the former number is incorrect, because Congress always lies.

The coastline paradox occurs because coasts are not straight lines, and this makes them difficult, or impossible, to measure definitively. From an aircraft, you can see that the coast has many features, including bays, inlets, rocks and islands. And the closer you look, the more nooks and crannies you'll find.

Oh, now I remember why I saved it. It's related to fractals like the Julia set or Mandelbrot set, which involve complex numbers, and, well... you know.

As a result, the length of a coastline depends on the size of the ruler you use.

This isn't just a coastline issue. Lots of survey boundaries follow the thread of a river (or, in the case of VA/MD, the low-tide shoreline of the Potomac), which has similar characteristics.

But if you used a smaller ruler, you'll capture more complexity, resulting in a longer measurement. Hence, a paradox.

Okay, I suppose, for some definitions of paradox. Regardless, that's what it's called, so I'll run with it.

According to work published in 1961, English mathematician Lewis Fry Richardson noted how different countries had different lengths for the same shared border because of differences in scales of measurement. In 1967, mathematician Benoit Mandelbrot expanded on Richardson's work, writing a classic Science paper on the length of Britain's coastline. This later led him to discover and conceptualize the shape of fractals, a curve that has increased complexity the more you zoom in.

This is also related to why no one can agree whether China or the US is the third-largest country by area: it depends how you measure some of the boundaries, including the coasts. Also, vertical differences get thrown in; this is the same fractal problem, only in two dimensions (surface), not one (boundary line).

The concept of "dimensions" also gets modified when you're dealing with fractals; you can get fractional dimensions. Which, it should come as no surprise, gave fractals their name.

I object, however, to the idea of "increased complexity the more you zoom in." I'd argue that you get the same complexity, just at different scales. I guess the sentence can be read like zooming in reveals greater complexity.

The article also features some nice Mandelbrot set zoom animations, which I always find fascinating.

This can hold true for coastlines. You could technically measure a coastline down to the grain of sand or atomic level or smaller, which would mean a coastline's length could be close to infinity, Sammler said.

Another nitpick: no such thing as "close to infinity." In the real world, as opposed to a purely mathematical construct, there's a minimum length (it's very, very small). That minimum length implies an upper bound to how long a fractal boundary can be. It can be very, very long... but that's still not infinity.

Coastlines are also shifting entities. Tides, coastal erosion and sea level rise all contribute to the fluctuating state of coastlines. So maps from the 1900s, or even satellite imagery from a few years ago, may not resemble what coastlines really are today.

And if you want to get really technical, it changes from moment to moment, as portions are eroded and others built up. Not to mention general changes in sea level. As I've noted before, you can't step into the same river once.

So how much coastline does Alaska, the United States, or our entire planet, have? We may never know the accurate number. It's a paradox, and like many things in nature, escapes our ability to define it.

Which shouldn't mean we throw up our hands and give up. I like to think of it as a metaphor for life itself: always approaching an answer, never quite getting there. But learning more and more along the way.