You're probably right that bird droppings could be a problem though, I would think one of the things you would want to avoid was nitrates.
I guess the easiest thing for you is to make sure the water that goes down there is already well filtered.

Since you're doing it by hand anyway you could add a set of valves to backflush valve A. Take water from your supply, for instance, and use a set of valves to isolate and inject water between valve A and the storage tank, then open A and B. You're reversing the flow to carry out the debris. You would have to have additional valves to do this when it was actually raining though, and I'm not sure how easy it is for you to add valves to isolate the different parts of your system. You might be able to do it with minimal digging. This is basically what you do to backflush an irrigation pump that gets water from a pond or something where it can pick up crud, or it's one way anyway.

The drawback of the downspout first flush systems is the difficulty in diverting enough water to do the job if you have a sizeable roof. You're lucky to divert 10 gallons or so if you just use one 4" diverter pipe, assuming a one story roof. I've heard you should try to divert 1 mm of rainfall, so I should be diverting 30 gallons or so from my 1200 sq ft. roof. It's better than nothing, though. I think that's right, it's been a while since I figured it out. I considered adding a small storage tank in parallel to the pipe, that could work if it is done right. I guess you could have 4 diverters, seems silly though.

I'm not a expert or anything but the fine dust ending up in the bottom of your tank is probably just tree pollen and crap like that, not horrible. Every tank has biosludge in the bottom eventually and in fact I think it helps keep the water clean. The trick is not to disturb it.

They say crabapple produces a lot of pollen. I agree with you, though.

Oh do they sell seeds as well? I know he has seedlings from "improved" trees, but I didn't find seed for sale. I believe I e-mailed him but didn't get an answer, busy time of year. I sort of wanted to try some specific varieties too, some old cultivars are said to produce large nuts but not fill well in the North, or not produce heavily.

Anyone interested in this (and perhaps there are none here on permies...), I have located a source at the University of Nebraska, they have an amazing collection of hickory varieties. 30 cents a seed, not awful considering the seeds are inspected and stratified.

I'm looking for Shagbark seed from trees like Porter, Grainger, Raudabaugh, Wilmouth. Also Shellbarks like Kreider, Keystone, or the hybrid Mitch Russell. There are a few places where scion wood is available for these guys but I'm having trouble sourcing seed.

Yes, it should work with little hazelnuts. The Davebilt it adjustable by removing or adding washers. It's nice if the nuts you are using are all similar in size.

You can eat it, but it's insipid and somewhat unpleasant unless it's perfectly ripe.

I know some people use it for walnuts, but it wouldn't be my choice. It works for hazelnuts and chestnuts, it won't work for black walnut or hickory.

I think you can trap and remove a raccoon or two that is a particular nuisance successfully. I had a raccoon kill 150 trees in one night because he had learned how to removed guards. I trapped two raccoons and released them in a conservation area about 10 miles away and have not had any further problems. My neighbor was out to shoot them for killing his ducks, so I might have done them a favor. Perhaps other raccoons have taken their place, but none were smart enough to pose a threat to my trees. There are ethical entanglements, perhaps, but in this case I represent the plants and my fence wasn't keeping them out.

Serviceberry will be mostly true to seed, especially if it is not near other Amelanchier. (And as long as it is not one of the named varieties that happens to be a hybrid).

There's more than one nursery owner on this forum that sells thousands of grafted Mulberries every year. If it has a name, it is grafted. It is impractical to clone Mulberry from cuttings, the success rate is generally very low. As I pointed out, Mulberry is almost always dioecious and the males do not produce fruit, so grafting is the only way to ensure the sex of the plant. Also, a grafted tree can fruit in a couple of years, whereas a seedling might take eight years or more to fruit. If you're wondering why someone would even buy a Mulberry, you might want to try comparing an improved cultivar to one of your 'weed' Mulberries.
However, you guys sure sound like you know what you're talking about so maybe I am just dead wrong.

I assume they are grafted? If they are "broken", do you mean broken at the graft? If the shoots are below the graft, they are not guaranteed to produce fruit. Most Mulberry is dioecious and a seedling rootstock has a 50% chance of being male. No fruit, just pollen.