Gravitational waves redshift in an expanding universe, so as they propagate out to lightlike infinity, their wavelengths stretch so much that there is no hope of detecting them.
In a flat universe, they go out to lightlike infinity without redshifting, but with the amplitude falling off 1/r rather than the 1/r^2 of light. Eventually they become too weak to be detected unless we can detect a single graviton of a particular frequency (which is going to be unimaginably difficult technologically).
> can GWs hit the edge of the universe?
This is an excellent question! Neither of the above types of universe has an edge; there is a lightlike infinity that electromagnetic and gravitational waves head towards. What's there? Who knows! For all practical purposes, these waves just depart from us and everything we can see, never to return.
We can have an edge to a universe that is contracting rather than expanding, or one that eventually contracts.
This is really hard to think about and I had to look around, and found http://jetp.ac.ru/cgi-bin/dn/e_042_06_0943.pdf [I sadly have had almost no previous exposure to the authors, and somewhat regret it; this paper is from the "Third Soviet Gravitational Conference" in 1972. Wow!]
It will take more time to digest it than is reasonable for a real response, but section 3 is relevant for a universe that is like ours in most ways except that it won't expand forever but rather will eventually collapse. However, essentially (and this matches my intuition), for a typical source the gravitational waves will radiate outwards from the source, slow down, and come back. Depending some choices of Friedmann parameters, they could even return as convergent waves focusing on whatever moved purely timelike from the source (e.g. the remnant of the merger) during the GWs lightlike journey; more likely they'd smear out in a larger region; and possibly they might not actually converge (or smear into the same general region) until very near the future singularity having undergone gravitational blueshifting. After converging they'd depart, and rapidly get lost in nonlinearities. This is not super-different from ripples on a pool that spread out, bounce back and lose structure.
This also raises the question of what significantly blueshifted gravitational waves might do. Drag matter a bit faster towards the future singularity is my guess. It also raises questions about constructive interference, but in that direction lies strong gravity, and there's not going to be an answer to those questions in the linearized gravity that we use to model gravitational waves.
I answered some of your other questions in other comments on this thread, I think. If not, ask again!
In a flat universe, they go out to lightlike infinity without redshifting, but with the amplitude falling off 1/r rather than the 1/r^2 of light. Eventually they become too weak to be detected unless we can detect a single graviton of a particular frequency (which is going to be unimaginably difficult technologically).
> can GWs hit the edge of the universe?
This is an excellent question! Neither of the above types of universe has an edge; there is a lightlike infinity that electromagnetic and gravitational waves head towards. What's there? Who knows! For all practical purposes, these waves just depart from us and everything we can see, never to return.
We can have an edge to a universe that is contracting rather than expanding, or one that eventually contracts.
This is really hard to think about and I had to look around, and found http://jetp.ac.ru/cgi-bin/dn/e_042_06_0943.pdf [I sadly have had almost no previous exposure to the authors, and somewhat regret it; this paper is from the "Third Soviet Gravitational Conference" in 1972. Wow!]
It will take more time to digest it than is reasonable for a real response, but section 3 is relevant for a universe that is like ours in most ways except that it won't expand forever but rather will eventually collapse. However, essentially (and this matches my intuition), for a typical source the gravitational waves will radiate outwards from the source, slow down, and come back. Depending some choices of Friedmann parameters, they could even return as convergent waves focusing on whatever moved purely timelike from the source (e.g. the remnant of the merger) during the GWs lightlike journey; more likely they'd smear out in a larger region; and possibly they might not actually converge (or smear into the same general region) until very near the future singularity having undergone gravitational blueshifting. After converging they'd depart, and rapidly get lost in nonlinearities. This is not super-different from ripples on a pool that spread out, bounce back and lose structure.
This also raises the question of what significantly blueshifted gravitational waves might do. Drag matter a bit faster towards the future singularity is my guess. It also raises questions about constructive interference, but in that direction lies strong gravity, and there's not going to be an answer to those questions in the linearized gravity that we use to model gravitational waves.
I answered some of your other questions in other comments on this thread, I think. If not, ask again!