1/f noise in electronics can be caused by defects in semiconductor fabrication processes (though actually first discovered in vacuum tubes), but 1/f noise has been demonstrated in many other places as well, including biological systems, and it seems to be a very common phenomenon. If you have IEEE access I would suggest reading the M.S. Keshner paper "1/f Noise." If not, the wikipedia article "Pink Noise" gives a decent overview.
Also remember that noise spectral density graphs are normalized to a 1Hz bandwidth, and there's a lot less Hz near dc, so unless you can synchronously demodulate or lock in, a dc measurement is likely to be less noisy (more sensitive) than an ac measurement.
