You ask what I would expect on the lake bottom if diatoms are blooming and settling. Frankly I don't know what to expect because in the 10 or so papers I have read on this topic so far (including Guoqiang et al. 2005 - Lake Sihailongwan which I read yesterday), do not demonstrate this critical relationship. They just assume it. To tell the truth, what I would expect in a closed off lake such as Suigetsu would be exactly what we seem to have in the top layers - a mess. Kitagawa says it's flocculated and Kato can't see any distinct layers. I get the idea from you that what you think happens is that this soft, no-detectable-layer mess on the lake bottom gradually gets sorted into distinct layers over several years, gradually gets the water squished out and becomes compacted into these nice, neat half-mm layers. Now that's a fine story, but I don't see nature working that way. What I see from actual experiments is fine laminae forming from moving water, not still water. And I don't see this 1 lamina = 1 year thing either. I see a group of multiple laminae being deposited by episodic events.
What episodic events? Well ... Why not earthquakes, tsunamis, typhoons, etc? You pooh pooh these saying that I've got water running uphill and such, but you yourself admit the 1662 layer. Did water run uphill then? How about those 46 turbidites? Did those form from water running uphill? You see, the truth is, our points of disagreement are not that significant. They are only a matter of degree.
My key point of disagreement with you is this: I think fine laminae are formed episodically, possibly during the same events which formed each turbidite. Because historical records only record major events in the past 1000 years or so and do not record many smaller events which could have placed these turbidite/fine laminae assemblages, we cannot rely on laminae counting to give us an accurate chronology.
More to say on diatoms, drainage area and other issues later.