Hearing the heat: Most of the excess heat that causes global warming is absorbed by the oceans. Quantifying that heat increase is challenging because it requires many different temperature measurements over both the vertical and horizontal extent of the oceans. Wu et al. report success in this effort through the use of a different method: They inferred temperature changes from sound waves generated by repeating earthquakes (see the Perspective by Wunsch). The travel time of these earthquakes from source to receiver reflects changes in the average water temperature that they encounter. This technique should substantially enhance our ability to monitor ocean warming. Science , this issue p. 1510 ; see also p. 1433

Strong waveform complexities, including multipathing of the S diffracted phase and rapid changes in differential ScS‐S times, are observed for multiple deep Fiji earthquakes recorded at the USArray. The complexities occur at the northeastern edge of the Pacific Large Low Shear Velocity Province (LLSVP), about 12 degrees southeast of present‐day Hawaiʻi. Waveform modeling of the multipathing provides good constraints on an ultra‐low velocity zone (ULVZ) with a width of 5 degree located near the inner edge of the LLSVP. Based on the mineralogical‐modeling of the ULVZ as a solid iron‐rich magnesiowüstite‐bearing assemblage with compatible morphology predicted from geodynamical simulations, a ULVZ model with a thickness of 30 km and a shear wave velocity reduction of 18% is preferred. The rapid change in differential ScS‐S travel time is best explained by having both the aforementioned ULVZ and an adjacent high velocity structure near the LLSVP. Furthermore, a low‐velocity plume‐like structure could potentially explain the observed S travel time delay independent of ScS. These seismic features are proposed to be a ULVZ driven toward the edge of the LLSVP while potentially pushed by a subducted slab. This configuration may trigger plume generation due to strong thermal instabilities and is in the same vicinity where mantle flow models place the present‐day Hawaiian plume source. Multiple ScS can potentially be used to verify vertical plume structure in tomographic models but the accuracy of upper mantle structure, which is a key reflection point, needs to be considered.

Internal structures of the Moon are key to understanding the origin and evolution of the Earth–Moon system and other planets. The Apollo Passive Seismic Experiment detected thousands of lunar seismic events and vastly improved our understanding of the Moon’s interior. However, some critical questions like the state and composition of the core remain unsolved largely due to the sparsity of the Apollo seismic stations and the strong scattering of seismic waves in the top layer of the Moon. In this study, we propose the concept of a fiber seismic network on the Moon and discuss its potential in overcoming the challenges in imaging deep Moon structures. As an emerging technique, distributed acoustic sensing (DAS) can provide a cost‐efficient solution for large‐aperture and dense seismic network deployment in harsh environments. We compute lunar synthetic seismograms and evaluate the performance of DAS arrays of different configurations in retrieving the hidden core reflected seismic phase ScS from the strong scattered waves. We find that, compared to a sparse conventional seismic network, a fiber seismic network using tens of kilometers of cable can dramatically increase the chance of observing clear ScS by array stacking. Our results indicate that DAS could provide new opportunities for the future lunar seismic surveys, but more efforts and further evaluations are required to develop a space‐proof DAS.