Abstract
We present the results of a search for nonlinear gravitational wave memory in
the NANOGrav 15-year data set. We find no significant evidence for memory
signals in the dataset, with a maximum Bayes factor of 3.1 in favor of a model
including memory. We therefore place upper limits on the strain of potential
gravitational wave memory events as a function of sky location and observing
epoch. We find upper limits that are not always more constraining than previous
NANOGrav results. We show that it is likely due to the increase in common red
noise between the 12.5-year and 15-year NANOGrav datasets.