MOND might be several phenomena including multipole gravitational potential flows near the speed of light
https://en.wikipedia.org/wiki/Modified_Newtonian_dynamics Their a0 is 1.2E-10 meter/second^2 which is the sensitivity needed to follow the sun moon tidal gravity signal at the earth’s surface. Potential changes at the sun and moon (and on earth) “diffuse” at the speed of light and change the local gradients. Also, electrons and protons from suns – a few charges easily mimic gravity. It is very simple, if you use the positions of the masses, calculate the potentials based on where the masses were when the gravitational potential changes were emitted, and calculate the resultant local forces – remembering to account for center of mass. It is the sun’s acceleration at the detector minus the sun’s effect on the center of mass, plus the moon’s acceleration on the detector minus the moon on the center of mass, plus rotational acceleration. Maybe that is possible to model for solar wind that far out from many bodies.
My current thinking is that the relevant “gravity” signals are multipole electromagnetic signals which carry energy and are only lightly interacting. Monopole and dipole terms are strong, then smaller multipole terms which greatly affect absorption. Like tuning, except the signal requires 3D orientation and timing to be absorbed.
The 1/r potentials are statistical from dynamic sources at time scales larger than the transit times for the size bodies involved – at the speed of light. Maybe I will find time. I am just writing it here to remind myself. There are many more local things happening than gravitational radiation effects on observed hydrogen concentrations and velocities in distant galaxies.
The neutrinos might have a mixture of massless and massive multipole particles with a size distribution from natural (and now fabricated) sources. Lots of massless ones which are essentially electromagnetic solitons and then larger ones which take longer to change and have internal structure and dynamics. Neutrinos mostly are not spherically symmetric but can have any nonlinear Schrodinger shapes and wave-functions. Resonant collisions and resonance gives them measurable mass, momentum, and energy transfer abilities.