I wonder if this would work on the open sea, in choppy water. It seems that the interference would mess up the specific pattern needed to pull objects inward. If it is possible to make this work in non-laboratory environments, i definitely can think of uses for the tractor beam.
It works by constructive interference, which means it works on water tables in physics labs. If you read the article you discover that they're basically leveraging lobing: Those maxima are still pushing out. They're pointing the minima at objects using complex math. Which means you can get sort of a sense of its applicability (as it stands, anyway) by modeling it as a concert. Walk around an empty room with two speakers and you hear the maxima and minima - we call it "comb filtering". but the louder the crowd is, the less combing you have because the environment gets chaotic. So - if your wave generator is such that its orders and orders of magnitude higher than the wave strength from chop, it'll work. But if you point a maxima at the thing you're trying to suck in, it'll push it away vigorously. And also kill all the sea life for miles. Pumping this much energy into the water is generally pretty rough on anything in it.