Caption: "A plot of the cumulative shift in the periastron-to-periastron orbital period in seconds for the binary pulsar PSR B1913+16 as the system loses energy (specifically gravitational potential energy) by gravitational radiation (AKA gravitational waves). Red points are experimental data, and the blue line is the shift predicted by general relativity. Error bars are shown, but are generally too small to be seen as distinct for most points." (Slightly edited.)

Features:

1. A binary pulsar is a pulsar (highly magnetized rotating neutron star or, very rarely, a white dwarf: Wikipedia: Pulsar: Milestones) in a binary system.

   The binary companion can be a star or a compact star (AKA compact remnant) (i.e., a white dwarf, neutron star, or black hole). There is at least one known case where both members of the binary system are pulsars: double pulsar PSR J0737-3039.

   The binary companion of binary pulsar PSR B1913+16 is a neutron star that is NOT a pulsar (relative to Earth: Wikipedia: Neutron star).

2. Because pulsars emit radio pulses with intrinsically very precise regularity, variations in pulse rate due to the doppler effect allow the indirect detection of the binary companion and the determination of the orbital period of the binary system.

   The orbital period can, in fact, be determined to very high accuracy/precision.

   This in turn allows the change in orbital period to be determined to accuracy/precision.

   There should be orbital decay (which includes the decrease in orbital period) due to the loss of specific orbital energy due to gravitational waves which carry away energy. Gravitational waves (and, thus orbital decay for an ideal gravitationally bound 2-body system are predicted by general relativity, but NOT by Newtonian physics.

3. The orbital decay of binary pulsar PSR B1913+16 (the first binary pulsar known) has been followed nearly since discovery in 1974.

   The plot shows the cumulative periastron-to-periastron orbital period shift since circa the start of 1975.

   The cumulative shift is the amount of time between when a periastron occurs and when it should occur if there were NO orbital decay.

   The observed cumulative shift is agrees to high accuracy/precision with the prediction of general relativity as the plot illustrates.

   This strong indirect evidence for gravitational waves. It is NOT direct evidence for gravitational waves since they themselves are NOT detected. The first direct detection of gravitational waves was GW150914 in 2015.

4. The orbital decay (i.e., orbital inspiral) of binary pulsar PSR B1913+16 will end in a neutron star merger in, it is estimated, ∼ 300 megayears (see Wikipedia: Binary pulsar PSR B1913+16: Star system).

   Since the total binary system mass is ∼ 2.8 M_☉ (see Wikipedia: Binary pulsar PSR B1913+16: Characteristics), the merged binary system will almost certainly become black hole.

   Recall if neutron stars grow over Tolman-Oppenheimer-Volkoff mass of ∼ 2.2 M_☉ (which is the best estimate of this uncertain value circa 2017), then nothing we know of can prevent collapse to a black hole.

   One wonders who will notice the neutron star merger of binary pulsar PSR B1913+16.