The help file clearly states that, for complex z = (x,y),
|z| = sqr(x) + sqr(y).
It does not say that, for float x, |x| = sqr(x), but it should.
Seems self-explanatory to me. 
Ultra Fractal author
Hence why it is the square modulus...
A more formal way of describing |z| is as trivial in trigonometry: using the Pythagorean theorem, sqrt(x^2 + y^2), we get the distance, modulus or magnitude of a complex number. The argument, or phase, theta, of a complex number, is given by atan2(z), or atan(y/x). And the magnitude is given as sqrt(cos(theta)^2 + sin(theta)^2), and z = rexp(itheta), where r is the magnitude of z, and i = sqrt(-1). It also equals to r(cos(theta) + isin(theta)), with Euler's identities, yielding the polar coordinates.
