My reading of the highlights from the initial FAA
Report on Flt BA038 (below) is that:
a. There were center-tank water warnings on the previous two flights - as well as on the accident flight ex Beijing.
b. These were not unusual events and usually gave rise to a routine "sumping" to clear any collected water within range of the low-point drains.
c. Water warnings usually indicated (by design) a significant amount of water in the tanks.
d. Because the nature of the warnings were usually that they would appear and then self-clear, the 777's system had gained a reputation of intermittency and fallibility. It was generally believed to be flaky, unreliable and therefore no great cause for concern - beyond generating a requirement for a sumping post-flight.
However what may have been the case was that water in the fuel was circulating (courtesy of pumping and refueling) alternately into and then away from areas covered by the water warning sensors. Additionally (and frequently) the water would also (either) become dissolved in the fuel or freeze in non-designed water-trap areas and no longer be detectable as water. This would explain the (perhaps undeserved) reputation of the water detecting system as being unreliable and inconsistent. The attitude adopted by engineers would have been to respond to reported/recorded "water in tank" warnings by simply sumping (draining) from tank low-points at the next turnaround servicing. Self-evidently this would be totally ineffective if done soon after a long-haul high-level leg (water still frozen) OR soon after refueling (water redistributed by the refueling operation to areas well away from tank low-point sumping drain-valves - as well as being to some extent re-dissolved hygroscopically in the swirling warmer fresh fuel. It should be noted that water-detection IS traditionally done soon after refueling - and that may not be any longer such a good idea. Because the Beijing overnight was in sub-zero conditions, sumping would've been quite unproductive, even though fuel would have deceptively emerged quite water-free from the drain-points. At Heathrow, after the accident, the center-tank was ruptured (see FAA report below) and all the residual fuel was lost - including, it's presumed, any/all water evidence. Without clear data, the UKAAIB's bottom line would necessarily have to be quite conjectural. Boeing wouldn't abide by (and would vigorously contest) any proposal to redesign tanks, pumps, piping, filters and water-traps. The bottom line is most likely to be a greater constraint upon TAT's at which the 777 can operate. Longer legs in low outside temps will be flown at lower, less economic, altitudes.
The highest hazard area during a "latent" (dormant) water-carrying flight would be precisely as in the BA038 scenario, i.e. the center-tank fuel level is low and the water melting during descent and the detached lumps of ice moving towards engine-feed pump inlet ports courtesy of pump-suction, higher engine fuel demands (spool up during aircraft attitude changes - both on descent and on finals when configuring with gear and flap).
One further thought. How will the formula for fuel absorption of water change once tanks are nitrogen inerted? Will it be a catalyst effect (i.e. dissolved water being more prone to precipitate out and collect and freeze at low points) - or will there be a positive effect? Would it be possible to capture the (presently wasted) oxygen by-product of the NEA process and utilize it for fuel-warming? My guess is that the result of BA038's accident will be a token SOP change, a limitation on cruising at low ambient temperatures and an overall inconclusive and unsatisfying bottom line.