Looks like NOAA may be using automated conversion software to create ENC charts from satellite images. It works great most of the time, but sometimes things don’t go as planned. In this chart (US5NY43) Mid Hudson Suspension Bridge seems to be charted based on a shade of the bridge rather than bridge itself, moving it somewhat North. Stated accuracy of 1:40000 scale chart is about 140′ so this one is probably just about within the limit.
Good navigators always remember that chart is not a substitute for a proper lookout.
P.S. A bit more on this – the following is my guess, correct me if I am wrong.
So, NOAA may be trying to do the right thing. Note that the satellite image is taken from position slightly due North of the bridge, rather than vertically. From that point of view bridge as tall as this one (134 ft) would be visually shifted South relative to anything at “sea”/water level. On the other hand, if the sun was at high noon and directly overhead, shade from the bridge would be precisely under it at exactly the sea level, and would provide a perfect outline of the bridge on a chart. I think this method works “most of the time”, in particular for images taken during summer months in Southerly latitudes.
Here we have a relatively high Northern latitude, and photo likely not taken at noon, so the sun is low in the southern sky. Because of that, the shade is located North of the bridge. The “real” bridge position should probably be charted in the middle between the shade and the bridge on the satellite photo.