Understanding Multiscale Tropical Cyclones-Environment Interactions: A Step Toward Bridging Knowledge Gaps Across the Weather-Climate Continuum

Abstract

Landfalling Tropical Cyclones (TCs) are extremely impactful weather phenomena that com- bine strong winds, extreme precipitation and extensive flooding, resulting in billion-dollar losses and more than 2500 deaths in the United States from 1963 to 2012. The interaction between TCs and the environment is a highly complex, multi-scale problem that directly affects the TC genesis, track, intensity, activity and impacts on a broad range of timescales. TC-environment interactions remain a grand challenge for weather forecasters and disaster managers alike, with widespread implications for the safety and economy of coastal communities. This dissertation aims to better understand TC-environment interactions across the weather-climate continuum – from hourly to interannual variability – bridging knowledge gaps across scales, with an emphasis on TC genesis and rainfall, to improve TC forecast lead time and societal preparedness.A key to TC genesis is the initial development of a warm core within an emergent cyclonic vortex, a process which occurs on small spatial scales and is often difficult to observe. Us- ing a combination of aircraft-based observations from the Convective Processes Experiment (CPEX) and CPEX-Aerosols and Wind (CPEX-AW) field campaigns and high-resolution numerical simulations, we examine how the dynamic and thermodynamic properties of the near-storm environment influenced the genesis of Tropical Storm (TS) Cindy (2017) and Kate (2021). The genesis of TS Cindy occurred within a shallow cyclonic circulation, embedded in a highly sheared environment due to the west-southwesterly flow associated with an upper-level trough. Within the disturbance, a lower-tropospheric warm and dry anomaly near the center of the cyclonic circulation can be observed both in the CPEX dropsondes and in the model simulation. Backward trajectory analysis shows that subsidence is primarily associated with a thermally indirect circulation along the western flank of the storm, where air parcels descend more than 1000 m while warming up by 9-12◦ C. The subsidence-induced virtual temperature perturbation accounts for 50 % of the sea-level pressure depression, thus playing a key role in the genesis of TS Cindy. During TC genesis, convective activity and the warm core formation are highly sensitive to the entrainment of dry and cold air from the environment. Leveraging unique CPEX-AW in-situ aircraft observations, in combination with a high-resolution numerical simulation, we identify two compound dry air intrusions during the genesis of TS Kate (2021) from a precursor African easterly wave. Passive atmospheric tracers reveal that low equivalent potential temperature air associated with the SAL and with the subtropical high is entrained within the AEW along two pathways: i) lateral entrainment following the wave-relative inflow, ii) downward into the boundary layer and subsequently upward within deep convection. A second compound intrusion is observed at the tropical depression stage, in a highly-sheared environment. Subsidence-induced dry air from the mid-latitude combines with the remnant SAL front in the northwest quadrant of the storm, resulting in strong radial ventilation of the developing inner warm core with low equivalent potential temperature air from the environment. As such, compound dry air intrusions can negatively impact TC genesis both by reducing moist convection within the parent AEW and by impeding the warm core formation. Synoptic-scale processes influence TC genesis within a background state that is largely controlled by lower-frequency variability in the coupled atmosphere-ocean system, from sub-seasonal to multi-decadal time scales. Understanding the large-scale environmental control of spatio-temporal TC variability beyond a few weeks thus remains a challenge with significant implications for societal impacts. This dissertation focuses on the association between the North Atlantic Oscillation (NAO) and year-to-year changes in TC activity and rainfall using observational data and reanalysis products. We leverage Poisson regression models to show that low-frequency NAO (LF-NAO) variability is associated with a distinct pattern of TC activity, which extends across the western North Atlantic, the Caribbean Basin, and the Gulf of Mexico. In these regions, the negative LF-NAO phase is typically characterized by 30-40 % more TCs than the positive one, in response to significant modulation of large-scale environment. The negative LF-NAO phase is associated with significantly higher Sea-Surface Temperature (SST) and weaker deep-layer wind shear across the tropical North Atlantic Ocean. The LF-NAO modulation of TC activity dictates basin-scale variations of TC-related rainfall impacts. By developing annual TC rainfall composites, we show that TC rainfall is strongly enhanced in the Caribbean and in the Gulf of Mexico during the negative LF-NAO phase. These multiscale interactions are further examined in TC rainfall properties at landfall across widely-used Quantitative Precipitation Estimates (QPEs) datasets, focusing on both extreme events and the climatology. We examine five datasets over an 18-year span (2002-2019). The products include three satellite-based products, CPC MORPHing technique (CMORPH), Integrated Multi-satellitE Retrievals for GPM (IMERG), Tropical Rainfall Measuring Mission - Multisatellite Precipitation Analysis (TRMM-TMPA), the ground-radar and rain-gauge- based NCEP Stage IV, and a state-of-the-art, high-resolution reanalysis (ERA5). Annual TC rainfall is typically highest along the coastal region from northeast Florida to North Carolina and in the New Orleans and Houston metropolitan areas. Along the East Coast, TC can contribute up to 20% of the warm-season rainfall and more than 40% of all daily and 6-hourly extreme rain events. Our analysis shows that the Stage IV detects far higher precipitation rates in landfalling TCs relative to IMERG, CMORPH, TRMM and ERA5. As a result, satellite- and reanalysis-based QPEs underestimate both the TC rainfall climatology and extreme events, particularly in the coastal region. This uncertainty is further reflected in the TC flooding potential measured by the Extreme Rain Multiplier (ERM) values, whose magnitudes are substantially underestimated and misplaced compared to the NCEP Stage IV.