Interpreting consequences of inadequate sampling of oceanic motions

Abstract

Abstract Inadequate sampling of oceanic motions, which commonly occurs for both oceanic measurements and simulations, can cause peculiar spectral features and potentially leads to misinterpretations. Here, we combine an extremely high‐frequency moored velocity record and a high‐resolution numerical simulation with the basic signal‐processing theory to quantitatively explore how varying sampling rates affect the ability to represent oceanic motions, especially internal gravity waves (IGWs). The moored measurements and simulations demonstrate that hourly sampling is sufficient to capture horizontal internal tidal velocities, but inadequate to faithfully characterize the vertical velocity for nearly all frequencies. The daily‐sampled model simulation shows a complicated frequency‐wavenumber spectral pattern of IGWs. Due to contrasting periodicities in time and space, temporal subsampling tends to retain the total variance of original IGWs, while spatial subsampling directly induces spectral energy loss. This study sheds light on data applications of the upcoming satellite altimetry missions.