PNNL

Abstract

The spatial scales of land-atmosphere interactions over heterogeneous and realistic soil moisture patterns are not well quantified, although such knowledge is critical to improve our understanding of the processes involved with land-atmosphere interactions and their representation in large-scale models. We use high-resolution numerical experiments to quantify the aforementioned spatial scales for a selected day during the Holistic Interactions of Shallow Clouds, Aerosols, and Land-Ecosystems (HI-SCALE) campaign in Oklahoma. To focus on the local influence from soil moisture heterogeneity, a semi-idealized approach is taken by removing background winds and surface topography while retaining the real-world data for other boundary conditions. Simulations are then compared that apply uniform or variable soil moisture. The observation-based soil moisture pattern, with most of its variance in the mesoscale (> ~10 km), is found to enhance the variability of the surface sensible heat flux in the mesoscale and sub-mesoscale (~2–8 km). Atmospheric response to the mesoscale variation is hydrostatic without notable vertical motion. The sub-mesoscale variation includes stronger pressure gradient, buoyancy, convergence, and vertical winds occurring over dry soil patches and sharp sensible heat flux gradients. Characteristic scales of such “hotspots” are 2–3 km. While their areal coverage is small, hotspots lead to substantially larger convective cloud structure and domain-mean rainfall. Surface topography over this area does not significantly change the location nor the strength of the hotspots but blurs the associated spectral signal, making it difficult to pinpoint the resolutions required for models and observations to capture the variability due to hotspots.