Revisiting turbulent transport dissimilarities and coherent structure transitions: An observation study from subtropical paddy fields

• Turbulent transport dissimilarities are observed even on ideal underlying surfaces. • The dissimilarities are accompanied by the transitions of coherent structures. • Prevailing coherent structures and its role under each stability regime are depicted. Turbulent transport dissimilarities among momentum, heat, and passive scalars challenge the application of Monin-Obukhov similarity theory in land-atmosphere exchange parameterization, yet their intrinsic mechanisms remain insufficiently understood. To fill this gap, we revisited these dissimilarities and further investigated the role of coherent structures based on an observation over uniform paddy fields in subtropical regions. Compared with existing studies, the observation is characterized by weakened surface sensible heat forcing coupled with high ambient humidity. The results demonstrate that turbulent transport dissimilarities between different variables emerge even over the ideal underlying surface and under the specific configuration of surface thermal and dynamic forcings. The magnitude of dissimilarity/similarity varies with atmospheric stability, which is notably accompanied by changes in the type of coherent structures. Furthermore, the prevailing coherent structures under each stability regime (including thermal plumes, hairpin vortices, inverted ramps, and intermittent turbulence) and their effects were depicted from both macroscopic and microscopic perspectives. It is revealed that turbulent transport dissimilarities arise when specific coherent structures efficiently transport only certain variables (i.e., momentum, heat, or passive scalars) but not others. These findings not only support the generality of transport dissimilarities beyond complex underlying surfaces, but also reveal their intrinsic causes. The understanding is significant for improving the parameterization of land-atmosphere exchange processes involving various physical quantities in numerical models.