PI: Gavin Schmidt


Allegra LeGrande, Anastasia Romanou, Gary Russell, Timothy Hall, Craig Rye, Michael J. Puma, Marco Tedesco, Radley Horton, Jacqueline Austermann, Gregory Garner, Robert Kopp, Maxwell Kelley, Vivien Gornitz.


Existing projections of sea level change are an amalgam of multiple sources of information which have not always been developed coherently. For instance, steric or dynamic effects come from GCMs that have not used the freshwater additions from ice sheets/shelves, estimates of future groundwater removals or reservoir construction have been made independently of local climate changes that might affect the need for irrigation, local storm surge calculations have not used regional variations in sea level change or storm climate. While allowance for these interactions can be made in the final syntheses (such as FACTS), a more coherent modeling system based on a full Earth System Model will allow for a reduction in uncertainty. Similarly, modeled estimates of risk require coherence in regional sea level projections, changes in storm climate and vertical land motions.

Expected Significance

The credibility of any projection system is heavily based on its coherence, inclusion of known factors and performance in hindcasts. We propose to demonstrate with historical simulations of GISS ModelE that we can improve on hindcasts of regional sea level rise, and by developing and integrating projections of all relevant factors relevant for coastal risk assessments, we will attempt to constrain scenarios of future change. The resulting projections of sea level change will directly support the NASA SLCT’s sea level change projections and various assessment frameworks like the IPCC and the NCA.


  1. Extend coupled GISS ModelE historical simulations and projections from CMIP6 to include groundwater removal, mountain glacier loss, reservoir construction, ice sheet and ice shelf mass loss and their impact on circulation and steric change.
  2. Develop and use interactive ice sheet, ice shelf and sub-ice shelf cavities components within the GISS ModelE system.
  3. Use high resolution storm surge modeling with inputs from GCMs, including extratropical storm climate and Tropical Cyclone statistical emulators plus information on vertical land movements and GRD effects, to improve estimates of risk.
  4. Embed new connections into FACTS

Deliverables (e.g. Datasets, Tools)


  • New historical simulations with additional forcings for CMIP6 for evaluation against observed changes (1850 onwards, with focus on 1990-2020)
  • Constrained projections based on SSPs with uncertainties, using externally generated projections for freshwater additions.
  • Fully interactive projections using dynamic ice sheet models and other interactive components - New storm surge projections for select locations

Open-source code

  • GISS ModelE Earth System Model
  • TC emulator

Datasets Used

  • GRACE-derived mountain glacier loss (ie. Wouters et al, 2019)
  • GRACE/IceSat/IMBIE estimates of historical ice sheet and ice shelf mass loss
  • Artificial water impoundment (Hawley et al, 2020)
  • Irrigation data (Cook et al, 2014)
  • ISMIP6 freshwater addition projections
  • CMIP6 estimates of steric and dynamic effects
  • VLM (via NSLCT)