IF (big if) the current is sinusoidal, and equal in all three phases, then the net current is ~1.7 (square root of three) times the current in one leg. In a DC circuit you would have, say, 5 amps in the positive lead, and 5 amps in the negative lead, and 5 amps flowing through the load...5 amps is the answer, not 10, even though there are two wires each carrying 5 amps. Similarly, in a three phase AC circuit, you can't just add up measurements from each wire...1.7 is the correct factor, not 3.
That is almost certainly the wrong answer for a wind turbine or other alternator directly charging batteries though, because measuring AC with accuracy is a fussy business, and there are many ways to get the wrong answer. Poorly done AC measurement is what is behind a number of "over unity" (perpetual motion) claims, in fact.
There are several different ways that meters can measure current (and voltage) with AC. "True RMS" is the only method that is valid for any waveform, but only higher priced meters will do this for voltage, and only the very best will do it for current (amps). If Tektronics, HP, or Fluke claim "true RMS" it is probably OK. Other instrument makers will vary on how "true" their "true RMS" measurements are, and there is probably a strong relationship between true and price.
All the other methods of metering AC assume that the waveform is sinusoidal and use a conversion factor to convert what is actually measured into an "RMS" value. If the waveform is not sinusoidal, that value will be wrong. How wrong depends on exactly how the meter makes the measurement, and how much the waveform departs from a sine wave. In practical terms, there is no way of knowing.
Even if your alternator puts out a faithful sine wave when unloaded, the current will still not be sinusoidal when you are charging batteries through a rectifier. There will be a large current pulse near the + and - peaks of the voltage waveform, and nothing for the rest of the cycle. Sort of like what a "modified sine" inverter produces, only more so.
Additionally, many "amp clamp" meters are only calibrated at 60 Hz, and can have significant errors at higher and lower frequencies...and those non-sinusoidal waveforms have quite a bit of harmonic content at 3,5, and 7 times the fundamental, so even if you knew a correction factor for frequency, you wouldn't know how to apply it for the various harmonics.
ETA: ampmeters on the AC lines are stil useful for relative measurements. If you add resistance to the line and observe higher current readings, then you are seeing a real improvement, and it will probably be fairly close to proportional to the increase in readings. You can also make readings on the DC side and come up with a calibration chart that will equate your AC and DC readings, but these will change a bit depending on your battery state of charge.
