What would cause a low volume 120 gph water pump to surge full-low volume in a regular 5 second cycle as it recirculates water in a huge volume filled container with a very restricted neck opening?

To help answer my own question and give you more information:

• The "huge volume, filled container" is a 2,500 gallon outdoor fishpond filled with very clean (very little particulate matter, probably no/very little free-floating solids like string algae, leaves, etc.) water and capped with about 15+ inches of ice. The ice is frozen to the pond liner walls, making it, effectively, a "container."
• The "very restricted neck opening" is an over-winter aeration device (an Allied Precision Pond Breather, see http://www.bestnest.com/bestnest/RTProduct.asp?SKU=API-8PB ) that provides a 2 inch diameter, 24" long column that isice-free/liquid-water connecting the deeper liquid pond water (temp 33.5F, 1C) with the outer atmosphere. Since this is the only opening with the atmosphere, the device has a clever, heated 'overflow' hole to allow for the expansion of below-ice water when the temperature decreases and more in-pond ice is formed; this extra volume of water-turned-to-ice is forced to the surface through the column where it overflows as water and soon freezes on top of the other ice on the pond. (If you followed this last part you're probably qualified to conjecture about the problem. ;-)
• The interesting thing is when a second Pond Breather is operating in this system - identical except the pump is only 40 gph, usually off, only turned on to thaw out its water column and collect water samples for water quality testing weekly) - the surging effect STOPS and the 120 gph pumps 120 gph continuously without a flutter. Also, the 40 gph pump has NEVER done this surging thing, even when it was on 24/7 and the only Pond Breather. (I had to drill a hole through the icecap to install the 120 gph unit mid-winter.)
• I believe that the answer is not simply a restricted pump intake, though this is possible and the most likely explanation without considering all they dynamics of this system. Based on the very regular 5-second period of this full/low pump surge I think it has something to do with the dynamics of the whole system...
• My best explanation so far: I was thinking a small vacuum may have been being created with the 120 gph pump in an otherwise 'almost-sealed container', and since fluids do not respond well to vacuums this negative pressure difference effectively restricted the water available to this impeller-driven pump. As the negative pressure is relieved (through the 2" column or small leaks around the edge of the ice) it allows the full amount of water to be available to the impeller - for a few seconds - and then slows it down again as the pressure is reduced. The 40 gph model may be low-volume enough that this significant a vacuum never builds up, and so it is not an issue, but with all the volume of water trying to pass up and down the column of the 120 gph model, passive-gravity-down can't keep up - the active-pump-up volume exceeds gravity-down - and a small vacuum is formed. As this vacuum slows down the pump, the water in the column can return to equilibrium up-and-down, the vacuum is broken and the pump can return to full blast again - only to be reduced a few seconds later when this happens again. By adding the lower volume 40 gph unit - which doubles the water column handling capacity - there is always enough gravity-down (return) to keep the vacuum from forming and this pressure issue never becomes significant, allowing the 120 gph to operate at 120 gph. What do you think of this explanation?
• What would your explanation be for the mechanism of this issue? Thanks!

First answer by Dragon9. Last edit by Dragon9. Contributor trust: 5 [recommend contributor]. Question popularity: 2 [recommend question]