Seminars

Kathy Pegion, George Mason U. (AOES), and Noushin Behboudi
The COLA Datasets Catalog: An Open Source Data Catalog that Supports Internal Use and FAIR Data Standards 
Wed, 1:30-2:30pm

We address several issues for data management in AOES/COLA:   
•    The need to manage large shared climate datasets across many users without a single coordinator.
•    The need to manage remote datasets with adaptability to include datasets in the cloud.   
•    The need to meet FAIR data standards for research projects for large datasets.

To address these issues, we are developing an open source data cataloging system and producing a catalog that documents the metadata for our large internal data collection.  The system is in Github and the catalog is rendered as a website.  Internal users can easily determine what datasets are available, how to access them and where they were obtained from.  Datasets can be added or updated by submitting issues and pull requests to Github, thus providing tracking of changes to the catalog. 

The system is intended to be adaptable to any collection of datasets.  For example, a small research group can generate a catalog to keep track of datasets being used by the group.  A catalog of the datasets used for a journal article can be submitted as supplementary materials with the article to meet data access requirements. The catalog code is available in Github and version 1 is planned for release in early 2021.  
 

Kathy Benison, West Virginia University

Evaporites on Mars

Thu, 4:30-5:45pm

Join Zoom Meeting:
https://gmu.zoom.us/j/94556653985

Meeting ID: 945 5665 3985

Image

Chia-Ying Lee, Columbia U.

Subseasonal Prediction of Tropical Cyclone Activity

Wed, 1:30-2:30

I will present results from our NOAA S2S project that examines the probabilistic predictions of  tropical cyclone (TC) activity -- genesis, occurrence, intensity (accumulated cyclone energy, ACE), and landfall-- at lead times of week 1- 4, in the Subseasonal to Seasonal (S2S) dataset. I will show that most S2S models evaluated have skill in predicting TC genesis and occurrence 4 weeks in advance. However, only the ECMWF model has skill in predicting the anomaly of TC activity from the seasonal climatology beyond week 1 in most global TC basins. Errors in genesis prediction largely limit models’ skill in predicting TC occurrence and ACE; S2S models are more skillful in predicting TC occurrence during favorable Madden–Julian oscillation phases. Calibrating forecasts through a linear regression method can better improve prediction skills when compared to mean-bias removal techniques. Lastly, I will briefly discuss our recent work on examining the monthly TC track forecasts from ECMWF at North Atlantic. Specifically, I will show how well the ECMWF model captures the observed track patterns through cluster analysis, and how tracks at each cluster are modulated by climate modes, such as ENSO and MJO.

Barry Klinger (coordinator), George Mason U. (AOES)

Current Climate Conversation

Fri, 10-11am

Quasi-monthly AOES discussion of notable recent climate and weather events. Come with an interesting news item. Come to add your two cents. Or just come to listen.

Image

Michael Scheuerer, NOAA Physical Sciences Laboratory

Using artificial neural networks for generating probabilistic subseasonal precipitation forecasts over California
Wed, 1:30-2:30pm

Ensemble weather predictions from global forecast systems require statistical postprocessing in order to remove systematic errors and to obtain reliable probabilistic forecasts. In this seminar we present two new postprocessing approaches based on artificial neural networks. The first approach uses rather basic neural network to link ensemble precipitation forecasts from a numerical weather prediction system to observed precipitation amounts. The second approach uses ensemble forecasts of large scale predictors like geopotential height at 500 hPa and total column water, and uses a convolutional neural network to link the associated weather patterns to observed precipitation amounts. Both methods are demonstrated with week-2, week-3, and week-4 forecasts of precipitation accumulations over California.

 Mark Puckett, Univ. of Southern Miss.

Gulf Coast quantitative biostratigraphy

Thu, 4:30-5:45pm

Image

Peitao Peng, Wanqiu Wang, Yun Fan, NOAA Climate Prediction Center

Significant improvement of dynamical ENSO forecasts using an artificial neural network
Wed, 1:30-2:30pm

A common deficiency of operational dynamical El Niño-Southern Oscillation (ENSO) forecasts in the past decade is the recurrent ENSO false alarms issued in late boreal spring. Ultimate solutions to this problem rely on model improvements, which may take years or even longer. Another approach is correcting forecast errors through post-processing of the existing output from dynamical models. This study explores the value of an artificial neural network (ANN) in post-processing to the enhancement of dynamical model performance in ENSO forecasts. By applying the ANN to the North American Multi-model Ensemble (NMME) hindcast and forecast data, it is found that the false alarms can be effectively silenced, and the overall forecast skill of all leads can be significantly improved as well. Further analyses show that the information contributing to the ANN correction is primarily from the model forecasts in the Northwestern Tropical Pacific region, where forecasted sea surface temperatures (SSTs) are well correlated with the observed Niño 3.4 SST index. This finding suggests that correctly representing the relationship between SSTs in the Northwestern Tropical Pacific and that in the tropical central-eastern Pacific in dynamical models is critical to an improved dynamical ENSO forecast.

Soumaya Belmecheri, National Science Foundation

Trees, fire and climate change

Thu, 4:30-5:45pm

Joke Lübbecke, Helmholtz Centre for Ocean Research, Kiel
Does the Tropical Atlantic Show Signs of a Weakening AMOC?
Wed, 1:30-2:30pm

Alberto Malinverno, LDEO

Magnetic anomalies and the GPTS

Thu, 4:30-5:45pm

Barry Klinger (coordinator), George Mason U. (AOES)

Current Climate Conversation

Fri, 10-11

Image

Quasi-monthly AOES discussion of notable recent climate and weather events. Come with an interesting news item. Come to add your two cents. Or just come to listen.

Francina Dominguez, U Illinois,
with Jorge Eiras-Barca and Zhao Yang
“Beating” of the Amazonian Hydroclimate and Possible Future Changes
Wed., 1:30-2:30

Between 20 and 40% of the precipitation that falls over the Amazon basin comes from evapotransportation within the forest itself.  Water vapor tracers embedded within an atmospheric model allow us to “tag” and track the moisture originating in the Amazonian basin.  This analysis led us to the discovery of a characteristic “beating” of the atmospheric moisture of Amazonian origin.  While the timescale of total water vapor is dominated by the annual cycle, that of Amazonian origin ebbs and flows with a very important diurnal cycle instead.  The characteristic timescales of evapotranspiration, precipitation and moisture transport by winds drive the variability.  I also discuss using tracers to analyze the possible impacts of future Amazonian deforestation on future hydroclimate patterns.  During the dry season, deforestation leads to decreased evapootranspiration, higher temperature and no changes in precipitation.  During the wet season however, an increase in moisture export reduces precipitation in the Eastern Amazon. Contrary to expectations, the tracers show no evidence that precipitation decreases are due to recycling or changes in stability.

Robin Trayler, University of California-Merced

Bayesian geochronology

Thu, 4:30-5:45pm

Derrick Lampkin, U Maryland

Title TBA

Wed 1:30-2:30

J. Brendan Murphy, St. Xavier University

Global tectonics: Pannotia and Pangea

Thu, 4:30-5:45pm

AOES PhD Students

Wed, 1:30-2:30

Finley Hay-Chapman

David Benson

Paul Buchmann

Bronwen Konecky, Washington University

Tropical paleoclimate

Thu, 4:30-5:45pm

Kai Huang, George Mason University (AOES)

Process-Oriented Diagnosis of Four Types of MJO

Wed., 1:30-2:30

We present a process-oriented diagnosis of four types of Madden-Julian Oscillation (MJO) with different propagating features, including standing, jumping, fast and slow events. The moist static energy budget, cloud-radiation feedback, and boundary layer moisture convergence in the subseasonal time scale are diagnosed. The pre-moistening process and high-level radiation heating are found essential to MJO propagation. Also, the moisture tendency indicates a slowly westward-propagating wave structure over the central-east Pacific during the standing and jumping MJO life spans, mainly related to the zonal moisture advection. Therefore, MJO propagation characteristics are also likely affected by interaction between other convectively coupled equatorial waves.

Clark Wilson, University of Texas at Austin

GRACE satellite geodesy

Thu, 4:30-5:45pm

AOES PhD Students

Wed, 1:30-2:30
Scott Knapp
Mary Korendyke
Nicholas Lydeen

Image

Liz Chamberlain, Vanderbilt University

Climate-delta-tectonic interactions

Thu, 4:30-5:45pm

Barry Klinger (coordinator), George Mason U. (AOES)

Current Climate Conversation

Fri, 10-11

Image

Quasi-monthly AOES discussion of notable recent climate and weather events. Come with an interesting news item. Come to add your two cents. Or just come to listen.

Abdullah al-Fahad, Geoge Mason University (AOES)

The influence of tropical diabatic heating and midlatitude static stability on subtropical anticyclones

Wed, 1:30

Arghya Goswami, Northwestern Missouri State Univ.

Tectonic evolution of India

Thu, 4:30-5:45pm

Rodrigo Bombardi, Texas A & M U

Monsoon prediction: sub-seasonal to seasonal forecasts and climate change projections

Wed, 1:30-2:30

Over half of the world's population lives in monsoonal regions. Monsoonal regions experience very pronounced rainy seasons for which the timing varies significantly from area to area. Due to the importance of monsoons to the water-energy-food nexus, skillful predictions would have a major impact on society. State-of-the-art global coupled climate models’ skillful predictability of summer monsoon rainfall is largely confined to equatorial regions. I will present work to investigate monsoon predictability and model forecasts skill on a range of timescales using a combination of numerical simulations and statistical methods. We evaluate the predictability of the subseasonal timing of monsoons, the predictability of summer monsoon precipitation, and climate change projections (CMIP6) for future monsoon precipitation.

Jerry Burgess, Johns Hopkins University

Geo-ecosystem structure and dynamics

Thu, 4:30-5:45pm

Barry Klinger (coordinator), George Mason U. (AOES)

Current Climate Conversation

Fri, 10-11

Image

Quasi-monthly AOES discussion of notable recent climate and weather events. Come with an interesting news item. Come to add your two cents. Or just come to listen.

The Sensitivity of the Tropical Atmosphere to Changes in Mixed-Layer Depth and Ocean Heat Flux Convergence Over the Indian Ocean

Wed., 1pm, via Webex [contact bklinger@gmu.edu for link)

The Madden-Julian Oscillation (MJO) is a tropical phenomenon that influences high-impact events such as hurricane activity and Asian monsoons on intraseasonal (2-8 week) time scales.  However, the essential dynamical mechanisms responsible for the MJO are highly debated. Specifically, the role of the ocean on the MJO is unclear. Ocean dynamics has a negligible impact on atmospheric intraseasonal variability. Therefore, to gain insight into MJO-ocean interactions, we conduct climate model experiments where the ocean is replaced by a motionless slab whose thickness, called the mixed-layer-depth (MLD), varies in space but not in time. This slab-model configuration allows the atmosphere and ocean to communicate through surface fluxes of heat and moisture while suppressing interactive ocean currents. Consistent with previous studies, the MJO is nearly indistinguishable between the fully-coupled and slab-models. However, drastic changes in the MLD and ocean heat convergence over the Indian Ocean had no discernible impact on MJO propagation, predictability, or variability. This suggests that atmosphere-ocean coupling may not be critical to the MJO over the Indian Ocean.  A discriminant analysis technique used to optimize the differences between experiments reveals that differences caused by changing Indian Ocean MLD were restricted mostly to surface fluxes and could be explained by simple physics. Despite precautions to preserve the climate, the control slab has a warmer climate than the fully coupled model.  Generally,  differences between fully coupled and slab models were consistent with changes associated with global warming, particularly the ``wet-gets-wetter" mechanism in a warmer climate.