We discovered Swope Supernova Survey 2017a ( SSS17a ) in the LIGO/Virgo Collaboration ( LVC ) localization volume of GW170817 , the first detected binary neutron star ( BNS ) merger , only 10.9 hours after the trigger .
No object was present at the location of SSS17a only a few days earlier , providing a qualitative spatial and temporal association with GW170817 .
Here we quantify this association , finding that SSS17a is almost certainly the counterpart of GW170817 , with the chance of a coincidence being \leq 9 \times 10 ^ { -6 } ( 90 % confidence ) .
We arrive at this conclusion by comparing the optical properties of SSS17a to other known astrophysical transients , finding that SSS17a fades and cools faster than any other observed transient .
For instance , SSS17a fades > 5 mag in g within 7 days of our first data point while all other known transients of similar luminosity fade by < 1 mag during the same time period .
Its spectra are also unique , being mostly featureless , even as it cools .
The rarity of “ SSS17a-like ” transients combined with the relatively small LVC localization volume and recent non-detection imply the extremely unlikely chance coincidence .
We find that the volumetric rate of SSS17a-like transients is \leq 1.6 \times 10 ^ { 4 } Gpc ^ { -3 } year ^ { -1 } and the Milky Way rate is \leq 0.19 per century .
A transient survey designed to discover similar events should be high cadence and observe in red filters .
The LVC will likely detect substantially more BNS mergers than current optical surveys will independently discover SSS17a-like transients , however a 1-day cadence survey with LSST could discover an order of magnitude more events .