FALL 2022
Fridays at 11:00 am, Rosenstiel School Auditorium / Virtual Auditorium
Aug 26: NO SEMINAR (ATM and OCE Faculty Meetings)
Sep 02: NO SEMINAR
Sep 09: NO SEMINAR
Sep 16: NO SEMINAR
Sep 23: NO SEMINAR
Sep 30: NO SEMINAR
Oct 07: STUDENT SEMINARS
Victoria Schoenwald (ATM)
Understanding Sea Level Rise Acceleration Along the East Coast of North America
The East Coast of North America has experienced rates of sea level rise (SLR) five times larger than the global average. This steep increase in SLR contributed to a higher frequency of coastal flooding events along the southeastern seaboard and the worst nuisance flooding event in Miami, FL during the last 20 years. Using tide gauge data from several stations, empirical mode decomposition (EMD) was used to understand sea level variability along the East Coast of the U.S., and its connectivity to atmospheric and oceanic circulation and thermosteric effects. This is a unique approach in identifying the "in phase" sea level variability and how it relates to the atmosphere and the ocean on varying timescales. The EMD modes were also used to understand the "out of phase" components of sea level variability such as the "hot spot" of SLR between Cape Hatteras, NC and Key West, FL where sea levels increased at rates of 25.5 mm/year compared to a global average of 4.5 mm/year. Similar techniques were then applied to climate model simulations using sea surface height at coastal locations as proxies for the tide gauge data. The EMD approach was applied at both ocean eddy parameterized and ocean eddy resolving scales. The goal was to determine if the natural variability in the models have similar characteristics to the observational estimates. And, to assess whether the modes associated with the trend in observations have appropriate analogues to the model simulations. By comparing pre-industrial simulations with historical simulations, we will be assessing whether a changing climate affects the natural variability.
Karen Papazian (ATM)
An Idealized Study of Cold Air Outbreaks
Cold air outbreaks (CAOs) have large societal and environmental impacts, such as agricultural losses, infrastructure damage, changes in atmospheric circulation, etc. As the Earth experiences a climate crisis, the focus on CAOs has been diminishing, but extreme CAOs continue to occur which society is gravely underprepared for. Many previous studies allude to a decrease in frequency and duration of CAOs as the climate system warms, but there are regional differences suggesting that there is dynamical significance. This study looks to explore how CAOs evolve in a changing climate through idealized modeling. Observational data from the NCEP-NCAR Reanalysis project is used in addition to model simulations from the Community Atmospheric Model - version 5 (CAM5), of the NCAR Community Earth System Model v1 (CESM1). In the model simulations, this study prescribes sea surface temperatures to a fully ocean-covered planet, an aquaplanet simulation. These aquaplanet simulations with differing pole to equator surface temperature gradients allow for a better understanding of the complex atmospheric dynamics of CAOs and were chosen to resemble various future climate change scenarios. By expanding the tropical region (TR) into the middle latitude region, there is an understanding that CAOs are still present on a synoptic scale. Additionally, as the pole to equator surface temperature gradient is eliminated, echoing an extreme climate change scenario where the poles warm more than the TR, CAOs continue to be present. With these differing aquaplanet simulations this study aims to further understand the physical mechanisms of CAOs, and the role dynamics plays in their future.
Oct 14: NO SEMINAR (Fall Recess and Rosenstiel School Faculty Meeting)
Oct 21: NO SEMINAR
Oct 28: NO SEMINAR
Nov 04: Dr. Nathan Laxague
Center for Ocean Engineering, University of New Hampshire, Durham
Surface Wave Mediation of Air-Sea Momentum Flux
Zoom Recording Available at COMPASS ON DEMAND
The interactions between the earth's atmosphere and ocean is a central driver of our daily weather patterns and large-scale climatic development. The phenomena which result from these interactions – from surface waves to extreme storms – are among the greatest constraints of human use of the ocean. In order to understand (and then model) these interactions, it is essential to parse the individual physical processes which mediate the exchange of momentum, heat, and mass across the air-sea interface. In this seminar, I will lay out some of the background theory for understanding wave-mediated air-sea momentum flux, describe techniques for characterizing ocean surface waves, and present some results from relevant field observations.
Nov 11: NO SEMINAR
Nov 18: STUDENT SEMINARS
Junfei Xia (MPO)
Machine Learning Applications in Ocean Datasets
With the explosive growth of available data and computing resources, recent advances in machine learning and artificial neural networks have yielded transformative results across diverse scientific disciplines. Machine learning has recently been widely applied in ocean science, such as in drifter observations and Synthetic Aperture Radar datasets. In this study, we focus on supervised learning. We perform error quantification on drifter datasets regression and use deep learning algorithms to classify SAR images. Drifter observations provide high-resolution surface velocity data, with which researchers often use different methods to project the Eulerian velocity fields. Gaussian Process Regression (GPR) is a machine learning method based on Gaussian probability distribution used in projection. The advantage of GPR is that it provides interpolation error estimates and benefits from smaller and larger separations of neighboring particles. There is no need to use low-pass or high-pass filters, which reduces human bias. We compare the GPR-predicted and model-provided velocity fields to quantify the error and reliability. SAR (Sentinel-1 Wave mode) provides data in 20 km × 20 km vignettes at 5 m × 5 m spatial resolution, which helps submesoscale studies. According to previous research, Xception deep learning architecture has the best performance in geophysical phenomena datasets. We will train a classification model based on Xception to identify oceanic fronts and determine when and where the highest frequency of oceanic fronts occurs.
Peisen Tan (OCE)
On the Modulation of Short Wind Waves by Long Waves Under Extreme Wind Conditions
The modulation of short wind waves by longer waves has been a popular topic of recent studies, since the ocean surface is usually a mixture of swell and wind waves and details of the effect of longer waves on shorter waves remain elusive. In particular, there has been a lack of systematic studies on the dependence of high wind forcing on long wave properties. To fill this gap, we carried out research at the SUrge-STructure-Atmosphere INteraction (SUSTAIN) facility of the University of Miami. Two experiments with initial paddle waves (monochromatic waves) with amplitudes of 2.5 cm and 5.0 cm and a wind-only experiment for comparison were conducted. We combined these background conditions with a wide range of wind forcing from a mild sea breeze to a Category 4 hurricane equivalent intensity. We found three regimes of wave development with increasing wind forcing: 1. The "warm-up" stage (U10 < 10 m/s, sea breeze): Capillary gravity waves begin to form uniformly on the paddle waves' surface. At this phase, no strong modulation from paddle waves is observed. 2. The "development" stage (10 m/s < U10 < 37 m/s, rough sea to tropical storm): Strong suppression of wind waves by paddle waves is observed. At this phase, monochromatic paddle waves begin to grow while the shorter wind waves remain suppressed. 3. The "saturation" stage (U10 > 37 m/s, hurricane): Intense breaking of paddle waves takes place, and wind wave energy begins to grow. We also found that with increasing fetch and the paddle wave amplitude, the suppression of the wind waves increases as well. Through this study, we will deepen the understanding of long wave - short wave - wind interactions and provide guidance for future field studies and the development of next generation wave models.
Nov 25: NO SEMINAR (Thanksgiving Recess)
Dec 02: STUDENT SEMINARS
Sisam Shrestha (ATM)
Weakening of Large-Scale Circulations in CMIP6 for Different SST Warming Patterns
The weakening of large-scale tropical circulations with warming has been explained through various dynamic and thermodynamic constraints. It has also been verified in observational datasets and Global Coupled Models (GCMs) for the twentieth century. However, studies for the recent decades show conflicting results regarding the change in the zonal and the meridional components of the circulation, namely the Walker Circulation and the Hadley Circulation. The growing conflict between the observed and model-projected changes in the Walker Circulation raise questions regarding the response of the circulation to warming and the ability of GCMs to simulate the correct response to external forcing. This discrepancy is closely tied to differences in the zonal gradient of SST change over the tropical Pacific. In our work, we quantify changes in large-scale atmospheric circulation using three different indices: changes in convective mass flux, the upward component of the mid-tropospheric vertical velocity and the spatial variance of vertical velocity. Our analysis shows a weakening of both the global and tropical-mean circulations. Similar weakening of large-scale circulations occurs in CMIP6 experiments using both observed or model projected SST changes. The similarity of the weakening of the mean circulation stands in sharp contrast with the spatial pattern of local circulation trends, which differ markedly between these two sets of simulations. Observational estimates also support a weakening of the mean circulation that is broadly consistent with the model projections.
Ivenis Pita (MPO)
An ARGO and XBT Observing System for the Atlantic Meridional Overturning Circulation
and Meridional Heat Transport (AXMOC) at 22.5°S
Changes in the Atlantic Meridional Overturning Circulation (AMOC) affect climate and weather patterns, regional sea levels, and ecosystems. However, despite its importance, direct observations of the AMOC are still limited spatially and temporally, particularly in the South Atlantic. Currently, only two observational arrays monitor the AMOC in the South Atlantic, TRACOS at 11.0°S and SAMOC at 34.5°S. The main goal of this study is to implement a cost-effective trans-basin section to estimate AMOC at 22.5°S, and to examine the seasonal and interannual variability of the AMOC at 22.5°S. For this, a monthly AMOC transport at 22.5°S from 2007 to 2020 is estimated from XBT transect and ARGO profile data. T and S profiles are averaged to the reference section at 22.5°S considering different temporal (30 to 180 days) and spatial (0.25 to 5 degrees) ranges. The mapped data was optimized using altimetry data as reference, and was used for the AMOC and MHT geostrophic calculation. ERA5 wind data was used to obtain the Ekman component. The high resolution XBT data available at the western boundary is vital for capturing the highly variable Brazil Current (–3.78±1.69 Sv), and our section shows a significant improvement if compared to gridded ARGO climatology (–1.64±0.51 Sv). The mean, interannual, and seasonal values of AMOC and MHT are compared with previous studies. In summary, these results highlight the importance of high resolution in situ based AMOC estimations, validate our AMOC / MHT time series, and contribute to a better understanding of AMOC and MHT in the South Atlantic.
Quinton Lawton (ATM)
Tropical Cyclogenesis and Kelvin Waves: Connections to Convection and Moisture
In recent years, research has illuminated a distinct relationship between convectively-coupled Kelvin waves (CCKWs), African easterly waves (AEWs), and tropical cyclone (TC) formation. This has been hypothesized to be the result of CCKW-related modifications to environmental factors and convection. However, little has been done to connect these environmental changes to the moisture- and convection- related processes involved in TC genesis. Here we leverage a 39-year database of AEWs and ERA5 reanalysis data to study the influence of CCKWs on TC genesis processes. Stand-alone results for TC genesis show it to be a multi-day process characterized by an increase in column- specific humidity ("preconditioning") and convective aggregation in the two days prior to TC genesis. Analysis of CCKW-relative composites illustrate a similar increase in convection and specific humidity as CCKWs pass AEWs, suggesting a possible pathway for CCKWs to influence TC genesis. We then compare composites of CCKW passages that result in TC genesis versus those that do not. The primary discriminator between these two outcomes appears to be convective coverage and diabatic heating at the time of CCKW passage. Ultimately, this analysis provides circumstantial evidence that CCKW-related modifications to convection and humidity could play an indirect role in preconditioning the AEW and a direct role in strengthening radiative-convective feedbacks – both of which could help explain why CCKWs promote TC genesis.
