Skip to content

GFDL Past Events & Seminars - 2020

Click a row to display that event's description/abstract.

Date Speaker Affiliation Title of Presentation
Jan. 8Lunchtime Seminar Series - Johannes QuaasUniversit├Ąt Leipzig, Institute for Meteorology, Leipzig University, GermanyProgress in quantifying the effective radiative forcing due to aerosol-cloud interactions
The effective radiative forcing due to aerosol-cloud interaction, ERFaci, is composed of the radiative forcing due to aerosol-cloud interactions, RFaci (Twomey effect) that is the immediate response of cloud albedo to an increase in droplet number concentration, Nd. Previous satellite-based quantifications of this effect were hampered by deficiencies in the retrieval of aerosol and also Nd. The talk will firstly discuss progress in this regard, which leads to a stronger estimated RFaci than previous satellite-based approaches. The other component of ERFaci is in the cloud adjustments. These can be split into adjustments of cloud fraction, f, and liquid water path, L. In terms of the latter, statistical relationships between L and Nd show on average negative adjustments of L (a positive forcing component). In turn, the analysis of ship-, volcano- and industry tracks leads to an estimated small overall effect on L; these results are trustworthy since a cause-effect relation is assured. In terms of the f adjustment, the current results point to an increase in cloud fraction at larger Nd. It is unclear which processes lead to this result. The talk will also briefly discuss how cloud-resolving simulations may help to better understand the remaining uncertainties. In the last part, a brief discussion will be presented on initial steps towards an estimate of the response of cirrus to anthropogenic aerosols.
Jan. 9Formal Seminar - Bob KoppRutgers UniversityLinking climate science, economics, and Big Data to estimate climate change impacts and endogenous adaptation
Understanding the likely global economic impacts of climate change is of tremendous practical value to both policymakers and researchers. Yet the economics literature has struggled both to provide empirically founded estimates of the economic damages from climate change and to provide quantitative insight into what climate change will mean at the local level for diverse populations. The Climate Impact Lab (a collaboration among Rutgers, UC-Berkeley, the University of Chicago, and the Rhodium Group) is advancing a method based on combining: (1) probabilistic simple climate model projections of the global mean response to forcing, downscaled and pattern-scaled based on CMIP-class models to translate global mean to local responses, and (2) empirical econometric estimates of the historical response of human systems to weather variability, derived from massive, standardized data sets and incorporating cross-sectional variability to estimate the benefits and costs of climate adaptation. This talk will focus on the example of temperature-related mortality and associated adaptation using sub-national data from 40 countries. Our results demonstrate that the temperature-related mortality impacts fall disproportionately on low-income populations, with high-income counties projected in the median to experience a decline in mortality through 2100, even under RCP 8.5, although the economic benefits of this decline are outweighed by the costs of adaptation. Even moderate emissions reductions result in substantial benefits, with median projected global mortality risk in RCP 4.5 (SSP 3 Socioeconomics) about 85% lower than that under RCP 8.5. Contact: robert.kopp@rutgers.edu
Jan. 15Lunchtime Seminar Series - Yongfei ZhangAssimilation of sea ice observations in MOM6/SIS2 and prospects for improved summer Arctic sea ice predictions
The seasonal prediction of Arctic sea ice, especially in the summertime, is vital to human activities and environment protections. The lack of constraint on sea ice initial conditions is one of the major hurdles for predicting summer Arctic sea ice several months ahead of time. This study exploits data assimilation (DA) to generate a better sea ice reanalysis and study the potential benefits of a more accurate initial condition. The GFDL Sea-Ice Simulator version 2 (SIS2) is coupled with the GFDL Modular Ocean Model version 6 (MOM6) and forced by a single atmosphere from the JRA-55 reanalysis. We link SIS2 and the data assimilation research testbed (DART) to conduct DA experiments. The sea ice concentration (SIC) observations from NSIDC are assimilated every 5 days from 1982 to 2017 through the Ensemble adjustment Kalman filter (EAKF). Before applying DA, we restore the sea surface temperature (SST) to the daily Optimum Interpolation Sea Surface Temperature (OISST), which improves our model background of SIC and also ameliorates an over-shooting problem arisen from SIC DA. We test the influences of different localization cutoffs, observation errors, and DA frequencies on the results. Our best DA experiment increases the September pan-Arctic sea ice extent (SIE) correlation and better reproduces the decreasing trend of pan-Arctic September SIE. Performances of SIC DA at regional scales are also discussed in our study. At the end of the talk, we show that the improved initial conditions of SIC and SIE have prospects for advancing short-lead time predictions of the summer Arctic sea ice.