I’ve written about High Bridge – the structure carrying the Croton Aqueduct across the Harlem River into Manhattan – before, but there’s always more to say. In this case, looking at the difference between an ordinary bridge and an aqueduct bridge, the difference between modern and ancient aqueducts, and the peculiarities of the Harlem River.
The Romans were experts at using big elevated structures consisting of a series of arches to carry aqueducts over low areas, and High Bridge looks a lot like those predecessors built up to 2000 years before it. (Note that a gravity feed aqueduct has to maintain a uniform slope if you want to avoid pressure increases and leaks at low points and pressure decreases and decreases in flow at high points. So the underground Croton Aqueduct, by maintaining its straight and almost-level line, comes up above ground at the Harlem River valley as the land drops down.)
The big difference between an elevated masonry aqueduct and an elevated masonry viaduct is loading. If High Bridge were carrying a road, there would be changes in the loading over time, and unbalanced load on the various arches as traffic bunched up and thinned out. If it were carrying a railroad, those changes and unbalanced loading would be significantly greater in magnitude. I’ve learned a lot about how masonry viaducts handle unbalanced load from Bill Harvey, and the short answer is that they don’t always work the way they were intended to. They do work – the UK and other European countries are crisis-crossed with old masonry viaducts carrying trains – but in complicated ways. High Bridge, with its inherently even and nearly unchanging load of water, is not subject to those problems. The main arches and piers have weathered, of course, but the recent restoration was focussed on the top pedestrian deck, because the basic structure has performed well.
In the 1900 picture above, the tall tower right in line with the end of the bridge was built some twenty years after the aqueduct as part of a system to increase pressure. The original design had the slope of the water uninterrupted from its source at the Croton River to the storage reservoir in central Manhattan. The new design added a reservoir at the the top of the hill next to the tower. Water was pumped up to the reservoir (the pumphouse is the building to the right of the bridge’s far end, with the red smokestack for its steam engine) for storage, and then pumped to a tank at the tower top before heading back to the aqueduct, with its pressure increased to the new head of the tower’s 200-foot height. This is a modern design, using an engine to make water run uphill so that it could gravity-feed downhill at high pressure.
Finally, note that there bridge piers basically go straight into the river. River piers of bridges usually have “starlings” to protect them: pointed upstream extensions to prevent flood water or high currents from damaging the piers. The Harlem River, like the East River, is not actually a river. It’s a tidal strait. It connects the East River at Hell Gate to the Hudson River at the northwestern tip of Manhattan, and the East River is connecting New York Bay with the Long Island Sound. As a result, the Harlem’s current changes direction with changes in the tides. More importantly, its current is relatively slow because of the bottlenecks at its ends (Hell Gate and Spuyten Duyvil), and it is not subject to riparian flooding. It can flood if there is ocean flooding in the East and Hudson Rivers – as happened during Hurricane Sandy – but this is a very rare occurrence and, again, this is constrained by the bottlenecks. In short, there’s no need to protect the piers against a form of flooding that doesn’t happen.