OBSERVATIONS AND SIMULATIONS OF THE INTERACTIONS BETWEEN COASTAL BREEZES AND THE ATMOSPHERIC BOUNDARY LAYER IN THE COASTAL HOUSTON ENVIRONMENT

Abstract

The atmospheric boundary layer (ABL) in the coastal environment is constantly evolving due to gradients in temperature, moisture, and surface roughness. The ABL influences regional air quality and weather forecasts through low-level stability and thermodynamic fluxes. Even though the ABL is intrinsic to daily life, it is historically under-sampled and poorly simulated in numerical models, because it evolves on a high spatiotemporal scale. The sea breeze (SB) is a contributor to the rapid coastal ABL evolution and develops because of the land-sea thermal gradient and advects the marine airmass onshore during the day. As a result, the SB can contribute to convection initiation and alter air quality in coastal areas. The TRacking Aerosol Convection interactions ExpeRiment (TRACER) field campaign gathered observations in the coastal-urban environment of Houston, Texas. With the goal of understanding the convective cloud lifecycle, a dense network of ABL measurements was gathered to understand how coastal breezes influence cloud evolution. Ground-based remote sensors and small uncrewed aerial systems (UAS) were deployed throughout the campaign and gathered a high temporal and vertically resolved dataset throughout numerous SB events, some of which triggered convection initiation. Vertical profiles of temperature, humidity, and winds were collected by the OU CopterSonde UAS from June to September in the coastal region of Houston, Texas. These profiles offer 5 m vertical resolution, on average, every 30 min through diurnal transitions, SB events, and nearby deep convection. During the campaign, CopterSonde observations were gathered through 17 SB events, 6 of which led to convection initiation. The UAS data can resolve the thermodynamic evolution and interactions between the SB and the pre-existing convective boundary layer. Using the UAS observations, SB impacts and interactions with the ABL are investigated. The range of SB thermodynamic impacts is found to be broad and depends on the time of SB passage and influence from additional water bodies, like Galveston Bay. The SB is also found to convectively destabilize the ABL by advecting an airmass with enhanced equivalent potential temperature. The rate at which equivalent potential temperature increases in response to the SB provides insight into how conducive the environment is to convection initiation. The resolution of UAS observations also provides a unique opportunity to evaluate numerical weather prediction on temporal scales consistent with the PBL evolution during coastal breeze passages, and beyond the surface layer. A case study is evaluated using observations during a bay breeze to sea breeze transition to understand how model parameterizations influence the forecast. The Warn-on-Forecast System (WoFS) is a convection-allowing ensemble designed to predict high-impact weather by combining rapidly updating data assimilation cycles with varying PBL and radiation parameterizations. While the model is not geared toward PBL studies, it provides a benchmark to evaluate commonly used parameterizations for NWP to offer insight into best-performing combinations and potential improvements. The biases of state variables calculated from the UAS observations show distinct differences from biases calculated using surface meteorological station observations, suggesting that traditional ABL evaluation practices may underestimate errors in the mixed layer. Additionally, the SB is variably represented in terms of the depth, arrival time, and intensity with different parameterization configurations. Only a subset of members simulate the preceding Galveston Bay breeze, and none do so accurately. The bay breeze is very sensitive to initial conditions, especially for members with local mixing schemes. As a result, initial cloud cover induced by the radiation scheme affects the ability to simulate a bay breeze, as well as the SB onset time. The mixing scheme for ABL parameterization lends differences in SB depth, intensity, and evolution. This variability across simulated SBs motivated analyzing how these differences impact simulated convection. Two events are analyzed to understand how simulated convection initiation differs with the ABL parameterization. Forecast performance is evaluated using storm object identification and matching with gridded reflectivity observations. One case's performance is consistent across all members, but the other is greatly dependent on the ABL parameterization. Local mixing scheme members tend to have more moisture across the region and disperse rising motion ahead of the SB front, which leads to an overestimation of storm objects inland. Nonlocal mixing members have more isolated lifting and unstable air close to the SB front that causes an underestimation of convective storms. These differences are less impactful when the environment is very unstable upon initialization. Then, all EMs result in an overestimation of convective storm objects ahead of the SB. These results suggest that the simulation of the convective boundary layer is more critical to accurately depicting convection initiation than the sea breeze characteristics.