His argument goes like this:
- "What we've seen several times is out here near Guam, of all places, we see a large microwave spiral shaped pulse that comes from the north and then extends to the south over Guam."
- "Following that, we see rotation begin to develop and form into these cyclones."
- "And, it's happened time after time now; it's happened like four times. And I've documented each one on my website."
There is no pulse
He fails to demonstrate that there is such a phenomenon as a "microwave pulse". (He attempts to at 3:33 and onward.) His example does not demonstrate a large microwave spiral shaped pulse. He doesn't demonstrate that anything originates from the north or Guam. What he claims is an example of this pulse (from 0:50 of this video):
http://tropic.ssec.wisc.edu/real-time/mimic-tpw/wpac/images/mosaic20131022T000000.gif
is only an expected artifact from the composition process:
As stated above, this technique is highly sensitive to biases between the different satellite instruments' calibration and/or retrievals of TPW. Even slight differences between retrievals can create two easily discernable artifacts in the animation: 1) The appearance of data swath edges throughout the image domain, and 2) The apparent "pulsing" of various high-TPW areas, especially in areas of precipitation. (From http://tropic.ssec.wisc.edu/real-time/mimic-tpw/prodDesc/)
These artifacts are frequent and easy to find. For example, here's one sequence filled with such artifacts:

And another more extreme example:
http://tropic.ssec.wisc.edu/real-time/mimic-tpw/natl/images/mosaic20130426T000000.gif
Post hoc ergo propter hoc
He then commits the post hoc ergo propter hoc fallacy. Even if, a "microwave pulse" happened before the formation of these storms, that does not suggest causation. In fact, he gets the causation backwards. The data provider says these artifacts will be found "especially in areas of precipitation".
Confirmation bias
Notwithstanding any of the above, his analysis also suffers from either confirmation bias (if not deliberate) or reporting bias (if deliberate). A proper analysis would include how many times a "microwave pulse" occurred and was not followed by a cyclone and how many times cyclones formed without such a "microwave pulse" occurring.