Content uploaded by Christopher P. Weaver
Author content
All content in this area was uploaded by Christopher P. Weaver on Sep 13, 2017
Content may be subject to copyright.
NOAA Technical Report NOS CO-OPS 083
noaa National Oceanic and Atmospheric Administration
U.S. DEPARTMENT OF COMMERCE
National Ocean Service
Center for Operational Oceanographic Products and Services
GLOBAL AND REGIONAL SEA
LEVEL RISE SCENARIOS FOR THE
UNITED STATES
Silver Spring, Maryland
January 2017
Photo: Ocean City, Maryland
Center for Operational Oceanographic Products and Services
National Ocean Service
National Oceanic and Atmospheric Administration
U.S. Department of Commerce
The National Ocean Service (NOS) Center for Operational Oceanographic Products and Services (CO-
OPS) provides the National infrastructure, science, and technical expertise to collect and distribute
observations and predictions of water levels and currents to ensure safe, efficient and environmentally
sound maritime commerce. The Center provides the set of water level and tidal current products required to
support NOS’ Strategic Plan mission requirements, and to assist in providing operational oceanographic
data/products required by NOAA’s other Strategic Plan themes. For example, CO-OPS provides data and
products required by the National Weather Service to meet its flood and tsunami warning responsibilities.
The Center manages the National Water Level Observation Network (NWLON), a national network of
Physical Oceanographic Real-Time Systems (PORTS®) in major U. S. harbors, and the National Current
Observation Program consisting of current surveys in near shore and coastal areas utilizing bottom mounted
platforms, subsurface buoys, horizontal sensors and quick response real time buoys. The Center: establishes
standards for the collection and processing of water level and current data; collects and documents user
requirements, which serve as the foundation for all resulting program activities; designs new and/or
improved oceanographic observing systems; designs software to improve CO-OPS’ data processing
capabilities; maintains and operates oceanographic observing systems; performs operational data
analysis/quality control; and produces/disseminates oceanographic products.
NOAA Technical Report NOS CO-OPS 083
U.S. DEPARTMENT OF COMMERCE
Penny Pritzker, Secretary
National Oceanic and Atmospheric Administration
Dr. Kathryn Sullivan, NOAA Administrator and Under Secretary of
Commerce for Oceans and Atmosphere
National Ocean Service
Dr. Russell Callender, Assistant Administrator
Center for Operational Oceanographic Products and Services
Richard Edwing, Director
Global and Regional Sea Level Rise Scenarios
for the United States
William V. Sweet
National Oceanic and Atmospheric Administration, Center for Operational Oceanographic
Products and Services, Silver Spring, MD, USA
Robert E. Kopp
Department of Earth & Planetary Sciences, Rutgers Energy Institute and Institute of Earth, Ocean
& Atmospheric Sciences, Rutgers University–New Brunswick, New Brunswick, NJ, USA
Christopher P. Weaver
U.S. Environmental Protection Agency, Office of Research and Development, Research Triangle
Park, NC, USA
Jayantha Obeysekera
South Florida Water Management District, West Palm Beach, FL
Radley M. Horton
Center for Climate Systems Research, Columbia University Earth Institute, New York, NY, USA
E. Robert Thieler
U.S. Geological Survey, Woods Hole, MA, USA
Chris Zervas
National Oceanic and Atmospheric Administration, Center for Operational Oceanographic
Products and Services, Silver Spring, MD, USA
January 2017
NOTICE
Mention of a commercial company or product does not constitute an
endorsement by NOAA. Use of information from this publication for
publicity or advertising purposes concerning proprietary products or
the tests of such products is not authorized.
iii
TABLE OF CONTENTS
LIST OF FIGURES ............................................................................................................................... IV
LIST OF TABLES ................................................................................................................................. V
EXECUTIVE SUMMARY ..................................................................................................................... VI
1.0 INTRODUCTION .....................................................................................................................1
2.0 SEA LEVEL RISE: HISTORIC INSIGHTS AND RECENT OBSERVATIONS ..........................................7
2.1 GLOBAL MEAN SEA LEVEL CHANGES ................................................................................................... 8
2.2 REGIONAL SEA LEVEL CHANGES .......................................................................................................... 9
2.3 RELATIVE SEA LEVELS ........................................................................................................................ 9
3.0 FUTURE SEA LEVELS: SCENARIOS AND PROBABILISTIC PROJECTIONS ..................................... 11
3.1 PROBABILISTIC GMSL RISE PROJECTIONS ........................................................................................... 12
3.2 UPPER AND LOWER GMSL RISE SCENARIO BOUNDS ............................................................................ 13
4.0 REGIONALIZATION OF THE GMSL RISE SCENARIOS ................................................................ 15
4.1 PROCESSES AFFECTING REGIONAL RSL CHANGE .................................................................................. 15
4.2 REGIONALIZATION METHOD ............................................................................................................ 15
5.0 RESULTS .............................................................................................................................. 21
5.1 GLOBAL MEAN SEA LEVEL RISE SCENARIOS ........................................................................................ 21
5.2 GMSL RISE RATES THIS CENTURY AND RISE BEYOND 2100 .................................................................. 22
5.3 REGIONAL CLIMATE-RELATED RSL CHANGES....................................................................................... 24
5.4 NONCLIMATIC BACKGROUND RSL AND GPS VLM TRENDS ................................................................... 27
5.5 SCENARIO PROJECTIONS OF RELATIVE SEA LEVEL (RSL) ........................................................................ 29
6.0 USAGE OF SCENARIOS WITHIN A RISK-BASED CONTEXT ........................................................ 33
6.1 GENERAL GUIDELINES FOR SCENARIO SELECTION ................................................................................. 33
6.2 SCENARIO PROJECTIONS OF RSL AND TIDAL FLOOD FREQUENCIES: A NATIONAL PERSPECTIVE .................... 35
6.3 BUILDING FOR A MAJOR FLOOD EVENT: A CASE STUDY FOR SOUTH FLORIDA ........................................... 39
7.0 SUMMARY AND NEXT STEPS ................................................................................................ 43
ACKNOWLEDGEMENTS .................................................................................................................... 45
REFERENCES .................................................................................................................................... 47
LIST OF APPENDICES ......................................................................................................................... 55
APPENDIX A. SEA LEVEL RISE AND COASTAL FLOOD HAZARD SCENARIOS AND TOOLS INTERAGENCY
TASK FORCE ........................................................................................................... A-1
APPENDIX B. LOW AND HIGH CLIMATE-RELATED RSL CHANGE CORRESPONDING TO GMSL
SCENARIOS ............................................................................................................ B-1
APPENDIX C. LOW AND HIGH TOTAL RSL CHANGE CORRSPONDING TO GMSL SCENARIOS ............ C-1
APPENDIX D. CMIP5 MODELS USED ............................................................................................. D-1
ACRONYMS .........................................................................................................................................
iv
LIST OF FIGURES
Figure 1. a) Multi-year empirical (smoothed) distributions for daily highest water levels in Norfolk, Va. (Sweet and
Park 2014) for the 1960s and 2010s, showing extent that local RSL rise has increased the flood
probability relative to impact thresholds defined locally by the National Weather Service
(http://water.weather.gov/ahps) for minor (~0.5 m: nuisance level), moderate (~0.9 m) and major (~1.2
m: local level of Hurricane Sandy in 2012) impacts, relative to mean higher high water (MHHW) tidal
datum of the National Tidal Datum Epoch (1983-2001). b) Due to RSL rise, annual flood frequencies
(based upon 5-year averages) in Norfolk for recurrent nuisance tidal floods with minor impacts are
accelerating, as shown by the quadratic trend fit (goodness of fit [R2]=0.84). Flood rates are rapidly
increasing in similar fashions along dozens of coastal cities of the U.S. (e.g., Sweet et al., 2014; Sweet
and Park, 2014; Sweet and Marra, 2016). .................................................................................................... 2
Figure 2. Schematic showing the intersection of scenario approaches with emission-dependent (conditional)
probabilistic projections of sea level rise under the climate modeling community’s Representative
Concentration Pathways (RCP) (van Vuuren et al., 2011). .......................................................................... 4
Figure 3. a) GMSL rise from -500 to 1900 CE from Kopp et al. (2016a)’s geological and tide gauge-based
reconstruction [black line with blue error estimates], from 1900 to 2010 from Hay et al. (2015)’s tide
gauge-based reconstruction [black], and from 1992 to 2015 from the satellite-based reconstruction
updated from Nerem et al. (2010) [magenta] and b) comparisons of GMSL since 1992 from
NOAA/NESDIS/STAR (black line) and the summation (purple line) of global mean ocean mass from
GRACE (blue line) and steric (density) sea level from Argo (red line) with seasonal variations removed
and 60-day smoothing applied (from Leuliette and Nerem, 2016). .............................................................. 8
Figure 4. a) Sea level change rates from 1992-2016 from TOPEX/Poseidon, Jason-1 and Jason-2
(www.star.nesdis.noaa.gov/sod/lsa/SeaLevelRise) and b) relative sea level trends based upon full record
(>30-year period of record in all cases) measured and published for NOAA tide gauges through 2015
(tidesandcurrents.noaa.gov/sltrends). ........................................................................................................... 9
Figure 5. VLM trend estimates (mm/year) derived from GPS platforms used by Hall et al. (2016) obtained from
NASA Jet Propulsion Laboratory (http://sideshow.jpl.nasa.gov/post/series.html accessed March 2016)
and University of Wisconsin (personal communication March 2016, Chuck DeMets of the University of
Wisconsin). The higher-resolution plots (on right) of the U.S. East and West/Alaskan Coast share the
same color scale. Negative values (red colored) occur where VLM is downward, which increases RSL at
the coast. ..................................................................................................................................................... 10
Figure 6. The GMSL rise scenarios of Parris et al. (2012). ........................................................................................ 12
Figure 7. Ratio of RSL to GMSL change for mass loss from specific land-ice sources; these constitute the static-
equilibrium fingerprints of the source. a) Greenland ice sheet (GIS), b) West Antarctic ice sheet (WAIS),
c) East Antarctic ice sheet (EAIS) and d) the median of 18 mountain glaciers after Kopp et al. 2014,
2015. Values more than (less than) 1 indicate RSL rise higher (lower) than GMSL rise. .......................... 17
Figure 8. This study’s six representative GMSL rise scenarios for 2100 (6 colored lines) relative to historical
geological, tide gauge and satellite altimeter GMSL reconstructions from 1800-2015 (black and magenta
lines; as in Figure 3a) and central 90% conditional probability ranges (colored boxes) of RCP-based
GMSL projections of recent studies (Church et al., 2013a; Kopp et al., 2014; 2016a; Slangen et al., 2014;
Grinsted et al., 2015; Mengel et al., 2016). These central 90% probability ranges are augmented (dashed
lines) by the difference between the median Antarctic contribution of Kopp et al. (2014) probabilistic
GMSL/RSL study and the median Antarctic projections of DeConto and Pollard (2016), which have not
yet been incorporated into a probabilistic assessment of future GMSL. .................................................... 22
Figure 9. Climate-related RSL change at 1-degree resolution for 2100 (in meters) relative to the corresponding
(median-value) GMSL rise amount for that scenario. To determine the total climate-related RSL change,
add the GMSL scenario amount to the value shown. ................................................................................. 25
Figure 10. Component contributions (in meters) to the median RSL Intermediate (1.0 m GMSL rise) scenario values
in 2100 from a) the AIS, b) GIC, c) GIS and d) oceanographic processes (global mean thermal expansion
and regional atmosphere/ocean dynamics; note that a different scale is used). .......................................... 26
v
Figure 11. a) Median climate-related RSL amounts under the Intermediate (1-m GMSL rise) scenario in year 2100
of Hall et al. (2016) (with all grid values shown with GMSL amount [1.0 m] removed) and b) difference
between RSL in this study’s (Sweet et al., 2017) Intermediate scenario as shown in Figure 9 and Hall et
al. (2016) results shown in a)...................................................................................................................... 27
Figure 12. Gridded a) background RSL rates and b) GPS VLM rates (with directionality switched as to directly
compare to RSL rate) with trend standard deviations shown in c) and d). In e) is the difference between
the two rates shown in a) and b). In f) is a scatterplot and linear regression of the two sets of gridded rates
shown in a) and b) as black dots and the red dots show background RSL rates as in a) but derived and
compared to VLM estimated for more than 100 specific tide gauge locations used by Hall et al. (2016)
and based upon Zervas et al. (2013). .......................................................................................................... 29
Figure 13. Total RSL change at 1-degree resolution for 2100 (in meters) relative to the corresponding (median-
value) GMSL rise amount for that scenario. To determine the total RSL change, add the GMSL scenario
amount to the value shown. ........................................................................................................................ 31
Figure 14. Average annual RSL for New York City (The Battery), Miami (Virginia Key), Fla., Galveston, Tex. and
San Francisco, Calif. with their respective (median-value) RSL under the six scenarios. The NOAA RSL
observations (tidesandcurrents.noaa.gov/sltrends) are shown relative to the midpoint (year 2000) of the
1991-2009 epoch (1994-2009 at Virginia Key), which is the reference level for the scenarios. ................ 36
Figure 15. Generalized Pareto Distribution and 95% confidence interval (CI: red dash) fit for high-water extremes
for historical data through 2015 (black dots) based upon Sweet et al. (2014) for NOAA tide gauges in a)
New York City (The Battery), N.Y. and b) San Francisco, Calif. In c) is a map for water level heights
(whose magnitudes are shown relative to the 1991-2009 epoch) with a 5-year recurrence interval (20%
annual chance of occurrence) for a set of NOAA tide gauges with more than 20 years of hourly record [as
in a) and b)], and d) shows the height difference between the water levels with a 5-year and a 0.2-year
recurrence interval (happening five or more times per year). ..................................................................... 37
Figure 16. The decade (±5 years about the year shown in the legend) when the flood event with a 5-year recurrence
interval (20% annual chance event) becomes a 0.2-year recurrence interval (or annual probability
increases 25-fold) under location-specific RSL associated with the a) Low b) Intermediate-Low, c)
Intermediate and d) Intermediate-High scenarios. Black dots are locations where the 5-year event does
not transition to a 0.2-year event by 2200. Note: flooding during an event can occur at high tide for
several days (e.g., Sweet et al., 2016). ........................................................................................................ 38
Figure 17. RSL under the Intermediate (1-m), Intermediate-High (1.5-m), and Extreme (2.5-m) GMSL rise
scenarios (solid curves) for Florida Keys region, showing how the water level height with a 1% annual
chance of occurring (dashed lines) and 95% confidence intervals (black error bars) estimated for year
2070 from hourly water levels at the NOAA Key West tide gauge changes in magnitude under each
scenario. All curves have been expressed in terms of the geodetic datum NAVD88 using the tidal datums
at Key West. ............................................................................................................................................... 40
LIST OF TABLES
Table 1. Spatial and temporal scales of geophysical processes affecting water levels. ................................................ 7
Table 2. Key processes contributing to RSL and GMSL change, and sources of information. .................................. 19
Table 3. Constraints used to stratify the projections and number of sample estimates per scenario. Note the
definition of the Low scenario is the continuation of the current rate of GMSL rise (~3 mm/year) through
2100, whereas the others are just defined by 2100 values. ......................................................................... 19
Table 4. Probability of exceeding GMSL (median value) scenarios in 2100 based upon Kopp et al. (2014)............. 22
Table 5. GMSL rise scenario heights in meters for 19-year averages centered on decade through 2200 (showing
only a subset after 2100) initiating in year 2000. Only median values are shown...................................... 23
Table 6. Rise rates (in millimeters per year for 19-year averages centered on decade) associated with the median
GMSL scenario heights this century (as shown in Table 5). ...................................................................... 23
Table 7. RSL with 1% annual chance flood heights and confidence intervals (CI) over the 50-year economic period
of analysis for Virginia Key, Fla. The RSL and extreme sea levels are relative to the geodetic datum
NAVD88, which is 0.27 cm above local mean sea level for the 1983-2001 epoch. ................................... 41
vi
EXECUTIVE SUMMARY
The Sea Level Rise and Coastal Flood Hazard Scenarios and Tools Interagency Task Force, jointly
convened by the U.S. Global Change Research Program (USGCRP) and the National Ocean Council
(NOC), began its work in August 2015. The Task Force has focused its efforts on three primary tasks:
1) updating scenarios of global mean sea level (GMSL) rise, 2) integrating the global scenarios with
regional factors contributing to sea level change for the entire U.S. coastline, and 3) incorporating these
regionally appropriate scenarios within coastal risk management tools and capabilities deployed by
individual agencies in support of the needs of specific stakeholder groups and user communities. This
technical report focuses on the first two of these tasks and reports on the production of gridded relative sea
level (RSL, which includes both ocean-level change and vertical land motion) projections for the United
States associated with an updated set of GMSL scenarios. In addition to supporting the longer-term Task
Force effort, this new product will be an important input into the USGCRP Sustained Assessment process
and upcoming Fourth National Climate Assessment (NCA4) due in 2018. This report also serves as a key
technical input into the in-progress USGCRP Climate Science Special Report (CSSR).
In order to bound the set of GMSL rise scenarios for year 2100, we assessed the most up-to-date scientific
literature on scientifically supported upper-end GMSL projections, including recent observational and
modeling literature related to the potential for rapid ice melt in Greenland and Antarctica. The projections
and results presented in several peer-reviewed publications provide evidence to support a physically
plausible GMSL rise in the range of 2.0 meters (m) to 2.7 m, and recent results regarding Antarctic ice-
sheet instability indicate that such outcomes may be more likely than previously thought. To ensure
consistency with these recent updates to the peer-reviewed scientific literature, we recommend a revised
‘extreme’ upper-bound scenario for GMSL rise of 2.5 m by the year 2100, which is 0.5 m higher than the
upper bound scenario from Parris et al. (2012) employed by the Third NCA (NCA3). In addition, after
consideration of tide gauge and altimeter-based estimates of the rates of GMSL change over the past
quarter-century and of recent modeling of future low-end projections of GMSL rise, we revise Parris et al.
(2012)’s estimate of the lower bound upward by 0.1 m to 0.3 m by the year 2100.
This report articulates the linkages between scenario-based and probabilistic projections of future sea
levels for coastal-risk planning, management of long-lived critical infrastructure, mission readiness, and
other purposes. The probabilistic projections discussed in this report recognize the inherent dependency
(conditionality) of future GMSL rise on future greenhouse-gas emissions and associated ocean-atmosphere
warming. In recognition of the different time horizons of relevance to different decision contexts, as well
as the long-term GMSL rise commitment (lagged GMSL response) from on-going increases in ocean-
atmosphere warming, GMSL rise and associated RSL change are quantified from the year 2000 through
the year 2200 (on a decadal basis to 2100 and with lower temporal frequency between 2100 and 2200).
The 0.3 m-2.5 m GMSL range for 2100 is discretized by 0.5-m increments and aligned with emissions-
based, conditional probabilistic storylines and global model projections into six GMSL rise scenarios: a
Low, Intermediate-Low, Intermediate, Intermediate-High, High and Extreme, which correspond to GMSL
rise of 0.3 m, 0.5 m, 1.0 m, 1.5 m, 2.0 m and 2.5 m, respectively. These GMSL rise scenarios are used to
derive regional RSL responses on a 1-degree grid covering the coastlines of the U.S. mainland, Alaska,
Hawaii, the Caribbean, and the Pacific island territories, as well as at the precise locations of tide gauges
along these coastlines. These scenario-based RSL values fill a major gap in climate information needed to
vii
support a wide range of assessment, planning, and decision-making processes. GMSL was adjusted to
account for key factors important at regional scales, including: 1) shifts in oceanographic factors such as
circulation patterns; 2) changes in the Earth’s gravitational field and rotation, and the flexure of the crust
and upper mantle, due to melting of land-based ice; and 3) vertical land movement (VLM; subsidence or
uplift) due to glacial isostatic adjustment (GIA, which also changes Earth’s gravitational field and
rotation, as well as the overall shape of the ocean basin), sediment compaction, groundwater and fossil
fuel withdrawals, and other nonclimatic factors. Key findings include:
● Along regions of the Northeast Atlantic (Virginia coast and northward) and the western Gulf of
Mexico coasts, RSL rise is projected to be greater than the global average for almost all future
GMSL rise scenarios (e.g., 0.3-0.5 m or more RSL rise by the year 2100 than GMSL rise under
the Intermediate scenario).
● Along much of the Pacific Northwest and Alaska coasts, RSL is projected to be less than the
global average under the Low-to-Intermediate scenarios (e.g., 0.1-1 m or less RSL rise by the
year 2100 than GMSL rise under the Intermediate scenario).
● Along almost all U.S. coasts outside Alaska, RSL is projected to be higher than the global
average under the Intermediate-High, High and Extreme scenarios (e.g., 0.3-1 m or more RSL
rise by the year 2100 than GMSL rise under the High scenario).
Finally, the consequences of rising RSL are presented in terms of how the frequency of moderate-level
flooding associated with a NOAA coastal/lakeshore flood warning of a serious risk to life and property
may change in the future under the sea level scenarios. The elevation threshold used to classify such
events by NOAA on their tide gauges varies along the U.S. coastline, but in general it is about 0.8 m (2.6
feet) above the highest average tide and locally has a 20% annual chance of occurrence. For example,
using the flood-frequency definition, we find at most locations examined (90 cities along the U.S.
coastline outside of Alaska) that with only about 0.35 m (<14 inches) of local RSL rise, annual
frequencies of such disruptive/damaging flooding will increase 25-fold by or about (±5 years) 2080, 2060,
2040 and 2030 under the Low, Intermediate-Low, Intermediate and Intermediate High subset of
scenarios, respectively.
viii
1
1.0 INTRODUCTION
Long-term sea level rise driven by global climate change presents clear and highly consequential risks
to the United States over the coming decades and centuries. Today, millions of people in the United
States already live in areas at risk of coastal flooding, with more moving to the coasts every year
(Melillo et al., 2014). Rising seas will dramatically increase the vulnerability of this growing
population, along with critical infrastructure related to transportation, energy, trade, military readiness,
and coastal ecosystems and the supporting services they provide (Parris et al., 2012; Hall et al., 2016).
One recent study estimates that 0.9 meters (m) of sea level rise would permanently inundate areas
currently home to 2 million Americans; 1.8 meters would inundate areas currently home to 6 million
Americans (Hauer et al., 2016).
Global mean sea level (GMSL) has increased by about 21 centimeters (cm) to 24 cm (8–9 inches [in])
since 1880, with about 8 cm (3 in) occurring since 1993 (Church and White, 2011; Hay et al., 2015;
Nerem et al., 2010). In addition, the rate of GMSL rise since 1900 has been faster than during any
comparable period over at least the last 2800 years (Kopp et al., 2016a). As is discussed in detail in
this report, scientists expect that GMSL will continue to rise throughout the 21st century and beyond,
because of global warming that has already occurred and warming that is yet to occur due to the still-
uncertain level of future emissions. GMSL rise is a certain impact of climate change; the questions are
when, and how much, rather than if. There is also a long-term commitment (persistent trend); even if
society sharply reduces emissions in the coming decades, sea level will most likely continue to rise for
centuries (Golledge et al., 2015; DeConto and Pollard, 2016).
While the long-term, upward shift in sea level is an underlying driver of changes to the nation’s coasts,
impacts are generally expressed through extreme water levels (short-period, lower-probability events
both chronic and acute in nature) occurring against the background of this shifting baseline. Higher sea
levels worsen the impacts of storm surge, high tides, and wave action (e.g., Theuerkauf et al., 2014),
even absent any changes in storm frequency and intensity. Even the relatively small increases in sea
level over the last several decades have led to greater storm impacts at many places along the U.S.
coast (Parris et al., 2012; Miller et al., 2013; Sweet et al., 2013). Similarly, the frequency of
intermittent flooding associated with unusually high tides has increased rapidly (accelerating in many
locations) in response to increases in relative sea level (RSL) as shown in Figure 1. At some locations
in the United States, the frequency of tidal flooding (events typically without a local storm present)
has increased by an order of magnitude over the past several decades, turning it from a rare event into
a recurrent and disruptive problem (Sweet et al., 2014; Sweet and Park, 2014; Sweet et al., 2016).
Significant, direct impacts of long-term RSL rise, including loss of life, damage to infrastructure and
the built environment, permanent loss of land (Weiss et al., 2011), ecological regime shifts in coastal
wetlands and estuary systems (Kirwan et al., 2010), and water quality impairment (Masterson et al.,
2014), also occur when key thresholds in the coastal environment are crossed (Wong et al., 2014).
Some of these impacts have the potential to ‘feedback’ and influence wave impacts and coastal
flooding. For example, there is evidence that wave action and flooding of beaches and marshes can
induce changes in coastal geomorphology, such as sediment build up, that may iteratively modify the
future flood risk profile of communities and ecosystems (Lentz et al., 2016).
2
Figure 1. a) Multi-year empirical (smoothed) distributions for daily highest water levels in Norfolk, Va. (Sweet and
Park 2014) for the 1960s and 2010s, showing extent that local RSL rise has increased the flood probability relative to
impact thresholds defined locally by the National Weather Service (http://water.weather.gov/ahps) for minor (~0.5 m:
nuisance level), moderate (~0.9 m) and major (~1.2 m: local level of Hurricane Sandy in 2012) impacts, relative to mean
higher high water (MHHW) tidal datum of the National Tidal Datum Epoch (1983–2001). b) Due to RSL rise, annual
flood frequencies (based upon 5-year averages) in Norfolk for recurrent nuisance tidal floods with minor impacts are
accelerating, as shown by the quadratic trend fit (goodness of fit [R2]=0.84). Flood rates are rapidly increasing in similar
fashions along dozens of coastal cities of the U.S. (e.g., Sweet et al., 2014; Sweet and Park, 2014; Sweet and Marra, 2016).
In this context, there is a clear need—and a clear call from states and coastal communities (White
House, 2014)—to support preparedness planning with consistent, accessible, authoritative and more
locally appropriate knowledge, data, information, and tools about future changes in sea level and
associated coastal risks. In response to this need, the White House Council on Climate Preparedness
and Resilience in 2015 called for the establishment of the Federal Interagency Sea Level Rise and
Coastal Flood Hazard Scenarios and Tools Task Force,1 a joint task force of the National Ocean
Council (NOC) and the U.S. Global Change Research Program (USGCRP). The Task Force’s charge
is to develop and disseminate, through interagency coordination and collaboration, future RSL and
associated coastal flood hazard scenarios and tools for the entire United States. These scenarios and
tools are intended to serve as a starting point for on-the-ground coastal preparedness planning and risk
management processes, including compliance with the new Federal Flood Risk Management Standard
(FFRMS).2 The Task Force is charged with leveraging the best available science; incorporating
regional science and expertise where appropriate; building this information into user-friendly mapping,
visualization, and analysis tools; and making it easily accessible through established Federal web
portals.3 Part of the motivation for forming the Task Force was to bring together key efforts within
individual agencies, such as the Federal Emergency Management Agency (FEMA), National Oceanic
and Atmospheric Administration (NOAA), U.S. Army Corps of Engineers (USACE), U.S. Geological
Survey (USGS), Department of Defense (DoD), Environmental Protection Agency (EPA) and
National Aeronautics and Space Administration (NASA), that could serve as building blocks of an
overall Federal system of sea level information and decision support, and to provide synthesis and
coverage of the entire United States coastline.
1 See appendix A for Task Force membership.
2 E.O. 13690: Establishing a Federal Flood Risk Management Standard and a Process for Further Soliciting and Considering
Stakeholder Input
(https://www.whitehouse.gov/the-press-office/2015/01/30/executive-order-establishing-federal-flood-risk-management-
standard-and-)
3 e.g., Digital Coast, globalchange.gov, and the Climate Resilience Toolkit
3
This report describes the output from a set of subtasks of the overall Task Force effort—specifically,
developing updated scenarios of GMSL rise, and then regionalizing these global scenarios for the
entire U.S. coastline, to serve as inputs into assessments of potential vulnerabilities and risks in the
coastal environment. In addition to supporting the longer-term Task Force goals, this new set of
products will also be a key input into the USGCRP Sustained Assessment process and the upcoming
Fourth National Climate Assessment (NCA4), due in 2018, including serving as a technical input to
the in-progress USGCRP Climate Science Special Report (CSSR).
This current effort builds upon recent advances in both sea level science and the synthesis and
assessment of this science to support planning and decision-making needs. A Federal interagency
effort described in Parris et al. (2012) defined a set of future GMSL rise scenarios that spanned the
range of published estimates (at the time of development of the report), and could be used to support
assessment and planning. The two-fold purpose of Parris et al. (2012) was to provide a scientific
synthesis across the large range of published future GMSL rise estimates and to complement existing
scientific assessments (e.g., from the Intergovernmental Panel on Climate Change [IPCC]) by
presenting the science from the perspective of scenario analysis within a risk-based context. The report
noted the wide range of estimates for GMSL rise scattered throughout the scientific literature, along
with the lack of any coordinated, interagency effort in the United States to harness this literature in the
development of agreed-upon estimates to support coastal planning, policy, and management. Coastal
managers had correspondingly been left to identify appropriate and relevant scenarios on an ad hoc
basis. Parris et al. (2012) developed a set of four scenarios, spanning a range of 0.2 to 2.0 m by the
year 2100, that described potential future GMSL conditions under varying assumptions about climate
change and the behavior of large ice sheets.
The four Parris et al. (2012) GMSL rise scenarios were designed to help planning, policy, and
decision-making stakeholders analyze vulnerabilities and future risks under conditions of scientific
uncertainty. As discussed in more detail below, a key advance of Parris et al. (2012) was to evaluate
the available science from the perspective of user needs, expanding the range of future conditions
considered to support a diversity of users with potentially very different decision contexts and risk
tolerances in their planning. This includes the need to test plans and policies against extreme cases
with a low probability of occurrence but severe consequences if realized. The Parris et al. scenarios
initially served as input into the Third NCA (NCA3; Melillo et al., 2014), but have since been taken up
in a variety of assessment, planning, and decision-making processes at the Federal, state, and local
level (e.g., USACE, 2014), illustrating the clear demand for such information, even when available
only for GMSL.
Although Parris et al. (2012) represented a significant step forward in developing actionable ranges of
sea level rise and placing them in an applied context, it had several limitations that subsequent efforts
have sought to address. For example, planners would greatly benefit from more locally and regionally
relevant information that accounts for location-specific adjustments to GMSL. In addition, sea level
science has advanced significantly over the last few years, especially improving understanding of the
complex behaviors of the large, land-based ice sheets in Greenland and Antarctica under global
warming, and the correspondingly larger range of possible 21st century rise in GMSL than previously
thought (Oppenheimer and Alley, 2016; DeConto and Pollard, 2016). Two recent efforts representing
significant progress in providing updated future GMSL and RSL rise information are the U.S. DoD
Coastal Assessment Regional Scenario Working Group study, reported in Hall et al. (2016), and Kopp
et al. (2014). Both produced local-scale RSL rise estimates across a range of future GMSL rise values:
at individual U.S. military installations worldwide in Hall et al. (2016) and for a global network of tide
4
gauge locations in Kopp et al. (2014). Hall et al. (2016) used the Parris et al. (2012) GMSL rise
scenarios as the basis for their regional adjustments, while Kopp et al. (2014) constructed probabilistic
projections of the factors driving GMSL rise as the basis for their estimates of the probability of
different levels of relative sea level change (conditional on alternative scenarios of greenhouse gas
emissions) at a global tide gauge network. Since the analysis described in this report draws extensively
from both studies, the approach, findings, and underlying methodology of both are discussed in more
detail in the following sections.
These two efforts provide the foundation for the current development of regional gridded RSL scenarios
for the United States. This report leverages this prior work to generate a new set of sea level products
that represents 1) the most up-to-date science and 2) the most up-to-date set of methodologies for
regionally adjusting a given GMSL rise scenario. In addition, it attempts to better support both
scientific assessment and decision-making by providing a more unified look across both emissions-
dependent probabilistic approaches and discrete scenario-based methods, as conceptualized in Figure 2.
Figure 2. Schematic showing the intersection of scenario approaches with emission-dependent (conditional) probabilistic
projections of sea level rise under the climate modeling community’s Representative Concentration Pathways (RCP)
(van Vuuren et al., 2011).
This report pursues the development of new sea level scenario products from a risk-based perspective,
where the primary motivation is to synthesize the available science and identify appropriate scenarios
to support evaluation and management of future risks associated with rising seas. As described in
Hinkel et al. (2015), the goals of traditional scientific assessment can often diverge from the goals of
packaging science needed to support assessment and management of risk. For example, the IPCC Fifth
Assessment Report (AR5; Church et al., 2013a) stresses the central or ‘likely’ range of 21st century
rise in GMSL based primarily on process-based models. As defined by the IPCC, the ‘likely’ range is
assessed as having at least a 66% probability of containing the true value; as Church, et al. (2013b)
explained, the chapter authors assessed that “there [was] roughly a one-third probability that sea level
5
rise by 2100 may lie outside the ‘likely’ range.” Risk-averse decision-makers, however, may find the
IPCC’s ‘likely’ range of future GMSL rise inadequate for their planning purposes, given the roughly
33% chance that GMSL rise falls outside of this range. For example, the operator and regulator of a
nuclear power plant near the coastal zone might be more interested in the 99th percentile (or 99.9th
percentile, etc.) of the estimated future distribution of GMSL outcomes to robustly manage their risks,
since a disproportionate fraction of total risk will often be associated with low-probability (but high-
consequence) outcomes; many impacts in the coastal environment are highly nonlinear with respect to
the amount of RSL rise (Gutierrez et al., 2009). For example, in the Thames Estuary 2100 project,
planners considered a plausible worst-case RSL rise scenario as a key part of the technical analysis
developed in support of planning new flood-control infrastructure to protect the city of London from
tidal flooding of the Thames River basin over the 21st century (Ranger et al., 2013). Such a risk-based
approach, with consideration of the full range of scientifically plausible outcomes including
potentially consequential outcomes with low probability of occurrence, is consistent with standard
practice in a variety of risk-centered fields, from insurance to environmental toxicology. In very
general terms, synthesis and assessment of the best-available science to most effectively support risk
assessment should not only aim to address the question, “What is most likely to occur?” but also “How
bad could things get?” along with other questions relevant to the current state of knowledge, rate of
change, time of emergence of extreme impacts, when more might be known, and so on.
Accordingly, this report assesses global and regional sea level rise science pertinent to assessing U.S.
coastal risks, including how best to characterize uncertainty to support planning and decision-making
in a consistent manner nationwide. This kind of bounding analysis is often given much less attention
in traditional climate science assessments, so this report focuses more substantially on it with respect
to future sea levels along U.S. coasts.
Therefore, in support of these goals, the focus in this report is on:
● Examining the full range of scientifically plausible future rises in sea level and leveraging
multiple lines of scientific evidence, not just the process-based models that have important
known deficiencies with respect to the representation of ice sheet dynamics (sections 2-5)
● Providing scenario selection guidelines for planning and decision-making applications as a
function of decision context (section 6)
Sections 2-5 are therefore primarily scientific and technical, describing the observational record, the
foundational work upon which this current effort is built, and the results of the analysis, whereas
Section 6 is geared more toward the needs of users for applying these results in planning processes.
Section 7 concludes the report by providing a summary and a look ahead.
6
7
2.0 SEA LEVEL RISE: HISTORIC INSIGHTS AND RECENT
OBSERVATIONS
Updated estimates of possible future GMSL and RSL rise amounts and associated rates depend in part
upon historical sea level rise. This section reviews recent observations of sea level change in the
context of recent reconstructions of past responses.
Water levels vary in response to multiple processes operating over multiple temporal and spatial scales
(Table 1; see Kopp et al. (2015) for a review). Changes in RSL occur in response to changing sea
surface height (SSH) and due to both natural and anthropogenic VLM (where a downward sinking
land motion with a ‘negative’ sign contributes to a RSL rise) and can be expressed as:
1) = −
In a simple view, long-term trends in global SSH arise from increases in global ocean volume (due
primarily thermal expansion) and ocean mass additions (due to primarily to melting of land-based ice,
with changes in storage of water on land). Superimposed are several factors that can cause changes in
regional SSH, including decadal-scale variability in ocean circulation (e.g., Firing et al., 2004), often
associated with large-scale climatic patterns such as the Pacific Decadal Oscillation (PDO; Bromirski
et al., 2011), as well as interannual variability such as occurs during phases of the El Niño Southern
Oscillation (ENSO, Hamlington et al., 2015). RSL change also arises from regional and more-
localized VLM. VLM can be downward (negative rates; i.e., subsidence) or upward (positive rates;
i.e., uplift) and cause a relative rise or drop in sea levels, respectively. VLM results from natural
processes (e.g., GIA or sediment compaction) whose rates are quasi-steady for centuries, as well as
more punctuated natural events (e.g., earthquakes). Human-induced VLM is another contributor,
typically associated with local-to-regional groundwater and fossil-fuel withdrawal that have time
scales associated with the disturbance.
Table 1. Spatial and temporal scales of geophysical processes affecting water levels.
Physical Process
Spatial Scale
Global Regional Local
Temporal
Scale
Potential
Magnitude
(yearly)
Wind Waves (e.g., dynamical effects,
runup)
X
seconds to minutes
<10 m
Tsunami
X
X
minutes to hours
<10s of m
Storm Surge (e.g., tropical storms or
nor’easters)
X
X
minutes to days
<15 m
Tides
X
hours
<15 m
Seasonal Cycles
X
X
months
<0.5 m
Ocean/Atmospheric Variability (e.g.,
ENSO response)
X
X
months to years
<0.5 m
Ocean Eddies, Planetary Waves
X
X
months to years
<0.5 m
Ocean Gyre and Over-turning
Variability (e.g., PDO response)
X
X
years to decades
<0.5 m
Land Ice Melt/Discharge
X
X
X
years to centuries
millimeters to
centimeters
Thermal Expansion
X
X
X
years to centuries
millimeters to
centimeters
Vertical Land Motion
X
X
minutes to centuries
millimeters to
centimeters
8
Changes in SSH are measured by satellite altimeters, which observe subtidal water levels offshore.
Changes in nearly instantaneous water level (for safe navigation, storm-surge preparedness and other
commerce-supporting activities) and in RSL (including VLM effects) are measured by tide gauges at
the land-ocean interface. However, it should be noted that, in the discussion of the consequences of
RSL rise on tidal flood frequencies (see section 6), waves and their high-frequency (seconds to
minutes) dynamical effects are not considered; we refer to ‘still water levels’ instead of a more ‘total
water level’ (Hall et al., 2016) as operationally reported by tide gauges due to the mechanical low-pass
filtering associated with their protective wells, multi-minute averaging scheme and general placement
in protected waters (Sweet et al., 2015). Also, for context, the term ‘sea level’ is distinguished from
‘water level’ in that the former is generally considered to represent monthly and longer-scale
variability; when discussing spatial scales, responses locally refer loosely to scales upwards of about
tens of kilometers and regionally to scales upwards of about hundreds of kilometers.
2.1 Global Mean Sea Level Changes
As shown in Figure 3a, the rate of GMSL rise as measured by altimeter (3.4 ± 0.4 mm/year) since 1993
(https://sealevel.nasa.gov; accessed in December, 2016) is more than double the rate estimated by a
global network of tide gauges during the 20th century (~1.4 mm/year; Church and White, 2011; Hay et
al., 2015), and the 20th century rate is found to be faster than any century in at least 2800 years (Kemp
et al., 2011; Kopp et al., 2016a). The altimeter and tide gauge reconstructions of GMSL (Figure 3a) are
in close agreement during their common record (since 1993). Over the last 30 years (since the mid
1980’s), tide gauge based reconstructions of GMSL report trends of ~3 mm/year. The main drivers for
GMSL rise are atmospheric and ocean warming, which act to increase both the mass of the ocean,
primarily through the melting of land ice (anthropogenic changes in the storage of water on land has
been an additional effect), and the volume of the ocean, primarily through thermal expansion. Over the
last decade, these three components measured by satellite gravimetry measurements of GRACE
(Gravity Recovery and Climate Experiment) (www.nasa.gov/grace) and in situ by the worldwide array
of Argo profiling drifters (www.argo.ucsd.edu) explain the global altimeter trend (Leuliette, 2015;
Merrifield et al., 2015; Chambers et al., 2016; Leuliette and Nerem, 2016) (Figure 3b).
Figure 3. a) GMSL rise from -500 to 1900 CE from Kopp et al. (2016a)’s geological and tide gauge-based reconstruction
[black line with blue error estimates], from 1900 to 2010 from Hay et al. (2015)’s tide gauge-based reconstruction [black],
and from 1992 to 2015 from the satellite-based reconstruction updated from Nerem et al. (2010) [magenta] and b)
comparisons of GMSL since 1992 from NOAA/NESDIS/STAR (black line) and the summation (purple line) of global mean
ocean mass from GRACE (blue line) and steric (density) sea level from Argo (red line) with seasonal variations removed
and 60-day smoothing applied (from Leuliette and Nerem, 2016).
9
2.2 Regional Sea Level Changes
Consistent with scientific understanding, sea levels have not been rising uniformly across the globe
over the last century (Church et al., 2004; Hay et al., 2015; Merrifield et al., 2016) or indeed at any
time for which geological records exist (Khan et al., 2015; Kopp et al., 2016a); sea level has been
spatially quite variable (Figure 4a). One driver of differences is the dynamic redistribution of ocean
mass, which is the result of both episodic and long-term changes in winds, air pressure, air-sea heat
and freshwater fluxes, and ocean currents. There are two regional patterns that impact U.S. coastlines:
● There has been a large west-to-east difference in sea level rise rates across the Pacific Basin
over the last several decades. Trends range from much higher than the global rate (>10
mm/year) within the Western Pacific and at several U.S. Affiliated Pacific Islands (USAPI) to
much less (<1 mm/year) within the Eastern Pacific and regions of the U.S. West Coast (Figure
4a). These zonal rate differences are thought to be primarily associated with multi-decadal
fluctuations in basin-scale forcing driven primarily by altered prevailing trade wind forcing
associated with the PDO (Bromirski et al., 2011; Merrifield et al., 2012), which appears to
have switched phases in the last couple of years (Hamlington et al., 2016; Merrifield et al.,
2016). A PDO-phase switch could signal the start of higher amounts of RSL rise along the
U.S. West Coast in the coming decades, similar to higher rates that have occurred there during
portions of the last century (Zervas, 2009; Church et al., 2004).
● Along the Northeast Atlantic, sea level trends have been higher than the global rate over the
last several decades, capped by a recent multiyear jump in sea level beginning in 2009
(Goddard et al. 2015). The relatively high rates have been attributed to changes in the Gulf
Stream (Yin and Goddard, 2013; Ezer, 2013; Kopp, 2013; Kopp et al., 2015), although there is
some debate as to whether these changes represent natural variability or a long-term trend
(Rahmstorf et al., 2015).
Figure 4. a) Sea level change rates from 1992-2016 from TOPEX/Poseidon, Jason-1 and Jason-2
(www.star.nesdis.noaa.gov/sod/lsa/SeaLevelRise) and b) relative sea level trends based upon full record (>30-year period of
record in all cases) measured and published for NOAA tide gauges through 2015 (tidesandcurrents.noaa.gov/sltrends).
2.3 Relative Sea Levels
Around the U.S., VLM can be a significant factor in the overall rate of RSL trends experienced. As
shown in Figure 4b, the highest RSL rise trends are found in regions of Louisiana (8–10 mm/year),
Texas (4–7 mm/year) and along the Northeast Atlantic from Virginia to New Jersey (3–5 mm/year). In
these regions, natural, nonclimatic processes like GIA and sediment compaction add about 0.5–
10
2 mm/year to RSL change (Kopp et al., 2016a), while artificial groundwater and oil/gas extraction
processes further enhance RSL rise (Galloway et al., 1999; Sella et al., 2007; Boon et al., 2010;
Eggleston and Pope, 2013; Miller et al., 2013). Where RSL-rise trends are high, land subsidence with
rates of 2–5 mm/year or more is not uncommon as estimated by global positioning systems (GPS) for
regions of the Northeast Atlantic and Gulf Coasts (Figure 5). On the other hand, there are regions, such
as parts of southern Alaska, where RSL trends are negative with land uplift >10 mm/year. In Alaska, this
is due to GIA-related uplift from the last ice age, as well as more recent uplift due to glacier retreat over
the last several decades (e.g., Sato et al., 2011) and the averaged effects of tectonics over the
observational record.
Figure 5. VLM trend estimates (mm/year) derived from GPS platforms used by Hall et al. (2016) obtained from NASA Jet
Propulsion Laboratory (http://sideshow.jpl.nasa.gov/post/series.html accessed March 2016) and University of Wisconsin
(personal communication March 2016, Chuck DeMets of the University of Wisconsin). The higher-resolution plots (on
right) of the U.S. East and West/Alaskan Coast share the same color scale. Negative values (red colored) occur where
VLM is downward, which increases RSL at the coast.
11
3.0 FUTURE SEA LEVELS: SCENARIOS AND PROBABILISTIC
PROJECTIONS
As discussed in Kopp et al. (2014) and Hall et al. (2016), development of the scenario products
reported on here starts with estimates of the probability of GMSL change and underlying contributing
processes, conditional upon greenhouse-gas emissions pathways. These emissions pathways are
represented by the Representative Concentration Pathways (RCPs) (van Vuuren et al., 2011). From
these conditional probability distributions, scenarios to support planning and decision-making in the
face of uncertain future sea level rise risks can be defined and selected, depending on the unique
characteristics of a given decision, as will be discussed in more detail in section 6.
Because of the importance of sea level in a coastal-hazards context, and in light of these substantial
global and regional trends, the natural question is how much GMSL might rise over the next few
decades, this century, and beyond. Until recently, it has been commonplace for studies to provide a
single ‘mid-range’ or ‘central’ projection (Tebaldi et al., 2012; NRC, 2012). Such an approach may be
sufficient to address near-term planning needs, because at these timescales, variations in projected sea
level rise—e.g., under current assumptions about future greenhouse gas emissions—will be small
relative to the noise associated with storm surge activity. Such planning decisions could include normal
budgeting cycles for anticipated high-water occurrences for preparedness purposes (e.g., budgeting for
mobilization of personnel, pumps, sandbags and other temporary flood-control measures). However,
mid-range projections are typically insufficient for many decisions. As discussed in section 1, decision-
makers charged with planning for upgrades to existing long-life critical infrastructure (e.g., power
plants, military installations), or building new infrastructure, need to consider the risks across a broad
range of possible outcomes, including those associated with high-consequence, low-probability
situations.
Parris et al. (2012) recognized the need for an interagency effort to define a set of future-possible
GMSL rise scenarios for coastal planning, policy, and management to allow for recognition of trend
changes and adaptive management strategies (USACE, 2013). Because of the large uncertainties
involved in predictions of the land-based ice melt contribution to GMSL rise and the significant
consequences associated with impossible-to-rule-out extreme outcomes, Parris et al. (2012)
recommended a scenario approach covering a broad range (0.2 m to 2.0 m GMSL rise by 2100) of
existing sea level study results (trends, process modeling, semi-empirical approaches, etc.). Their
scenario set, which was developed to support NCA3 (Melillo et al., 2014), was not intended to provide
probabilistic prediction of future changes. Rather, the scenarios were intended to describe plausible
conditions that support decision-making under uncertainty, given specific assumptions about which sea
level rise science to include in one’s risk assessment. Parris et al. (2012) explicitly conceptualized
scenarios as being defined by considerations of use, writing, “Scenarios do not predict future changes,
but describe future potential conditions in a manner that supports decision-making under conditions of
uncertainty. Scenarios are used to develop and test decisions under a variety of plausible futures. This
approach strengthens an organization’s ability to recognize, adapt to, and take advantage of changes
over time” (Parris et al., 2012).
12
Figure 6. The GMSL rise scenarios of Parris et al. (2012).
The Parris et al. (2012) scenarios provided a range of possible future GMSL rise by 2100, bounded by a
low- (0.2 m) and a high-end (2.0 m) member with two intermediate members (0.5 m and 1.2 m)
(Figure 6). Each of these four scenarios exemplifies a specific set of scientific assumptions about 21st
century GMSL. For example, the low scenario represents an amount about 5 cm above the extrapolated
rate of the GMSL rise trend over the 20th century, while the high scenario represents an upper limit
reflecting GMSL rise occurring under more extreme land-ice contributions, as modeled by Pfeffer et al.
(2008). The two intermediate scenarios represent rise obtained from the upper-end of the projections
from the IPCC Fourth Assessment Report, from climate models using the low-emissions B1 scenario
(Intermediate-Low), and several semi-empirical based studies (Rahmstorf et al., 2007; Horton et al.,
2008; Intermediate-High), respectively.
More recently, Hall et al. (2016) provided a set of five GMSL rise scenarios for 2100 (0.2, 0.5, 1.0, 1.5
and 2 m) with a similar, user-focused conceptual framing as Parris et al. (2012), including lower and
upper bounds to provide a plausible range of GMSL-rise related risk of concern for DoD installation
managers. Hall et al. (2016) assessed this GMSL range against the extant scientific literature after the
Parris et al. (2012) report (discussed in the next section). Intermediate scenarios were simply
discretized by 0.5-m increments and aligned with emissions-based, conditional probabilistic storylines
and global model projections. In a significant update to Parris et al. (2012), Hall et al. (2016) also
provided local adjustments to the global scenarios to provide site-specific scenarios, which are further
described in section 4.
3.1 Probabilistic GMSL Rise Projections
Since the Parris et al. (2012) report, there have also been significant advances in developing
probabilistic estimates of future GMSL. Estimates of full probability distributions of possible future
GMSL outcomes (conditional on major assumptions such as future greenhouse gas emissions) provide
decision-makers with the most comprehensive information base, from which they can select the most
relevant individual scenarios to be used in their planning processes (Oppenheimer and Alley, 2016).
Specifically, several recent studies have provided probabilistic projections of the extent of future
13
GMSL rise conditional on forcing from the RCPs (van Vuuren et al., 2011). The RCPs provide a set of
possible future greenhouse gas concentrations through the year 2300 and were used by the IPCC AR5.
Each RCP represents possible underlying (though implicit) socioeconomic conditions and
technological considerations, including a low-end member (RCP2.6) requiring strong mitigation (net-
negative emissions in the last decades of the 21st century), a moderate mitigation member (RCP4.5)
stabilizing emissions through 2050 and declining thereafter, and a high-end, fossil-fuel-intensive,
‘business-as-usual’ emission scenario (RCP8.5). In response, global mean temperatures are modeled
as likely (>66% probability) to increase 1.9–2.3, 2.0–3.6 and 3.2–5.4 degrees Celsius, respectively, for
RCP2.6, RCP4.5 and RCP8.5 over the 2081–2100 period relative to 1850–1900 levels (IPCC, 2013).
The IPCC AR5 (Church et al., 2013a), using an ensemble of process-based models and other sources
of information, projects a median and likely (66% probability) GMSL rise of 0.44 m (0.28–0.61 m),
0.53 m (0.36–0.71 m) and 0.74 m (0.52–0.98 m) by 2100 for RCP2.6, 4.5 and 8.5, respectively. (There
is an additional pathway, RCP6.0, which is not analyzed here since its 21st-century GMSL projections
are nearly identical to those for RCP4.5, and few models ran RCP6.0 projections beyond 2100.)
However, AR5 recognized the challenges of modeling additions due to the collapse of marine-based
sectors of the Antarctic ice sheet. More recent studies of GMSL rise have reported probability ranges
that have focused on resolving ranges spanning lower probabilities and/or providing complete
conditional probability distributions. An assessment of recent probabilistic studies finds GMSL rise by
2100 projected for the 90% probability (5th–95th%) range to fall between 0.25–0.80 m, 0.35–0.95 m
and 0.5–1.3 m, respectively, for RCP2.6, 4.5 and 8.5 (Miller et al., 2013; Kopp et al., 2014, 2016a;
Slangen et al., 2014; Mengel et al., 2016). These projections are consistent with the global
temperature/GMSL relationship found to occur over the last 2800 years (Kopp et al., 2016a). To obtain
GMSL rise estimates whose ranges extend beyond AR5, additional assumptions are included within a
particular probabilistic framework. A few examples include reliance on structured expert elicitation of
potential ice-melt contributions not captured in the process models (Bamber and Aspinall, 2013) or
from geologic evidence comparing past sea levels and atmospheric greenhouse gas concentrations
(Rohling et al., 2013) or temperature (Kopp et al., 2016a). Under such frameworks, estimates of high-
end GMSL rise by 2100 under RCP8.5 include ~1.8 m [95th percentile] (Jevrejeva et al., 2014, Rohling
et al., 2013 and Grinsted et al., 2015), ~2.2 m [99th percentile] (Jackson and Jevrejeva, 2016), and
~2.5 m [99.9th percentile] (Kopp et al., 2014).
3.2 Upper and Lower GMSL Rise Scenario Bounds
Determining a potential upper limit of GMSL rise by 2100 (and beyond) is considered an important
target for critical and long-lived infrastructure decisions and a primary objective of this report. Since the
upper limit established by Pfeffer et al. (2008), which was based primarily on assessment of the
maximum plausible loss rate from Greenland and which was the basis for the 2.0-m high scenario for
2100 of Parris et al. (2012), there has been continued and growing evidence that both Antarctica and
Greenland are losing mass at an accelerated rate based upon gravimetry satellites (GRACE), repeat
altimetry, in-situ GPS monitoring, and input-output calculations (Shepherd et al., 2012; Khan et al.,
2014; Scambos and Shuman, 2016; Seo et al., 2015; Martín-Español et al., 2016). Such evidence
suggests that the collapse of some sectors of the Antarctic ice sheet may be inevitable as surrounding
ocean waters warm (Joughin et al., 2014; Rignot et al., 2014). In addition, recent modeling of physical
feedbacks related to marine ice-cliff instabilities and ice-shelf hydrofracturing (rain and meltwater-
enhanced crevassing and calving) used within the physical process models generating GMSL estimates
are being incorporated into ice-sheet models (Pollard et al., 2015). With such feedbacks modeled for
Antarctica, additional GMSL rise upwards of 0.6–1.1 m to median estimates under RCP8.5 are possible
14
by 2100 (DeConto and Pollard, 2016), potentially raising median GMSL projections for RCP8.5 of
Kopp et al. (2014) as high as 1.9 m by 2100. Meanwhile, in Greenland, there are indications that
processes already underway have the potential to lead to an accelerating high-end melt risk. Important
changes in surface albedo are occurring in response to ice melt and associated unmasking and
concentration of impurities in snow and ice (Tedesco et al. 2016). At the base of the ice sheet, important
changes in ice dynamics are occurring, through interactions with surface runoff and a warming ocean,
which may make the Jakobshavn Isbræ, Kangerdlugssuaq Glacier, and the Northeast Greenland ice
stream vulnerable to marine ice sheet instabilities (Khan et al., 2014).
The growing evidence of accelerated ice loss from Antarctica and Greenland only strengthens an
argument for considering worst-case scenarios in coastal risk management. Miller et al. (2013) and
Kopp et al. (2014) discuss several lines of arguments that support a plausible worst-case GMSL rise
scenario in the range of 2.0 m to 2.7 m by 2100: (1) The Pfeffer et al. (2008) worst-case scenario
assumes a 30-cm GMSL contribution from thermal expansion. However, Sriver et al. (2012) find a
physically plausible upper bound from thermal expansion exceeding 50 cm (an additional ~20-cm
increase). (2) The ~60 cm maximum contribution by 2100 from Antarctica in Pfeffer et al. (2008)
could be exceeded by ~30 cm, assuming the 95th percentile for Antarctic melt rate (~22 mm/year) of
the Bamber and Aspinall (2013) expert elicitation study is achieved by 2100 through a linear growth in
melt rate. (3) The Pfeffer et al. (2008) study did not include the possibility of a net decrease in land-
water storage due to groundwater withdrawal; Church et al. (2013) find a likely land-water storage
contribution to 21st century GMSL rise of -1cm to +11 cm. Thus, to ensure consistency with the
growing number of studies supporting upper GMSL bounds exceeding Pfeffer et al. (2008)’s estimate
of 2.0 m by 2100 (Sriver et al., 2012; Bamber and Aspinall, 2013; Miller et al., 2013; Rohling et al.,
2013; Jevrejeva et al., 2014; Grinsted et al., 2015; Jackson and Jevrejeva, 2016; Kopp et al., 2014) and
the potential for continued acceleration of mass loss and associated additional rise contributions now
being modeled for Antarctica (e.g., DeConto and Pollard, 2016), this report recommends a revised
worst-case (Extreme) GMSL rise scenario of 2.5 m by 2100.
As for the lower-end scenario of 0.2 m by 2100 of Parris et al. (2012), this report now recommends
this value be revised upward to 0.3 m by 2100, for several reasons. The primary reason is that the
GMSL rise rate as measured by satellite altimeters has been tracking >3 mm/year for almost a quarter-
century, mostly fluctuating with interannual variability associated with ENSO (Nerem et al., 2010;
Boening et al., 2012; Fasullo et al., 2013; Cazenave et al., 2014). In addition, over the last 30 years
(since the mid-1980s), tide-gauge-based reconstructions of GMSL (Church and White, 2011; Hay et
al., 2015), which are in close agreement with the altimeter record since 1993, continue to report trends
of ~3 mm/year. Continuation of this rate throughout the 21st century results in a GMSL rise of at least
0.3 m. The altimeter record and most recent 30-year record of tide-gauge-GMSL reconstructions are
longer than the 19-year epoch used by NOAA to update their mean sea level tidal datum (performed to
ensure accurate charting for safe navigation), and the altimeter record is nearing NOAA’s 30-year
requirement to compute a local RSL rate trend (Zervas, 2009). Lastly, even under the lowest of the
RCPs (RCP2.6), which entails net-zero greenhouse gas emissions in the third quarter of this century
and net anthropogenic removal of carbon dioxide from the atmosphere thereafter, multiple
probabilistic studies (Kopp et al., 2014, 2016a; Mengel et al., 2016) indicate that GMSL is very likely
(>90% probability) to still increase by 25–30 cm.
15
4.0 REGIONALIZATION OF THE GMSL RISE SCENARIOS
This section describes regional-scale climatic and nonclimatic factors that cause sea level to change
regionally. Knowledge of these processes is applied to the global scenarios to develop integrated
projections of regional sea level change.
4.1 Processes Affecting Regional RSL Change
As discussed earlier, GMSL rise is primarily driven by thermal expansion of the ocean as it warms and
from increases in ocean mass from melting ice locked in mountain glaciers and polar ice sheets, with a
secondary contribution from net changes in land-water storage. But a range of other factors also affect
local-to-regional RSL (i.e., equation 1), whether by changing SSH, VLM, or both, as recently
reviewed by Kopp et al. (2015):
● Changes in land-ice mass (e.g., melting glaciers and ice sheets) and in land-water storage not
only change ocean mass and thus GMSL, but also 1) alter SSH with regionally distinct
signatures from changes in the Earth’s gravitational field and rotation, and 2) lead to regional
VLM from flexure of the Earth’s lithosphere. Collectively, these processes give rise to static-
equilibrium fingerprints that characterize the spatial patterns of different mass transfers.
● Freshwater additions from land-ice melt or precipitation changes, as well as changes in the
distribution of heat in the ocean and atmosphere, also alter regional SSH from altered oceanic
and atmospheric processes involving the density, circulation, and distribution of water within
the ocean.
● GIA, the ongoing response of the solid Earth to land-ice shrinkage at the end of the last ice
age, drives post-glacial rebound under the core of the old ice sheets and subsidence on the
margins leading to regional VLM, and secondarily also changes SSH regionally due to
associated effects on the Earth’s gravitational field and rotation, as well as on the overall
volume of the ocean basin.
● Tectonics and sediment compaction, which occur from both natural causes and extraction of
fluids like groundwater or hydrocarbons from sedimentary reservoirs, also drive VLM. These
changes can be punctuated (not steady through time, e.g., earthquake-related or discontinued
anthropogenic resource extraction) and more localized in nature.
In terms of future rise, the GMSL scenarios and gridded RSL responses are composed of 19-year
average estimates reported on a decadal basis through 2100 (and subsequently for 2120, 2150 and
2200). In this context, interannual effects (e.g., ENSO) are negligible and multidecadal effects (e.g.,
PDO) are mostly attenuated or unresolved. Regional RSL projections or scenarios that account for
static-equilibrium fingerprints, oceanographic processes, and VLM have been considered in several
recent studies (Mitrovica et al., 2011; Yin, 2012; Perrette et al., 2013; Church et al., 2013a; Kopp et
al., 2014, 2016a; Horton et al. 2015; Slangen et al., 2014; Grinsted et al., 2015; Hall et al., 2016).
Results of the regionalization process are shown in section 5.
4.2 Regionalization Method
Projecting future RSL change requires accounting for a set of processes affecting SSH and VLM and
their different spatial patterns in a fashion that is consistent with the magnitude of the projections of
GMSL rise. Tasked to deliver local RSL rise scenarios, Hall et al. (2016) assessed all DoD
installations worldwide within 20-km of a tidally influenced body of water to provide screening-level
estimates of related vulnerabilities. They formulated site-specific RSL scenarios by Monte-Carlo
16
resampling of gridded climate-related data from Perrette et al. (2013) and Kopp et al. (2014) to weight
their resultant samples per their specific set of GMSL rise targets for 2100. Hall et al. (2016) also
applied VLM adjustments (based on datasets shown in Figure 5) to further localize the scenarios using
rates from a global set of GPS measurements and derived from tide gauge platforms.
The basis for scenario development in this report is similar to the approach used by Hall et al. (2016), but
distinct in that it uses regenerated datasets entirely based upon slight modifications of Kopp et al. (2014)
to quantify RSL responses on a gridded basis (1-degree) for the U.S. coastline. (Projections are also
provided at the precise locations of tide gauges.) To account for regional processes, the scenarios are
based on probabilistic projections constructed for a set of contributory processes, which follow a
framework and use data sources similar to those of Kopp et al. (2014). Key sources of information used
in constructing each of these probability distributions are described in Table 2. Ice sheet mass changes
were projected based on combining the IPCC expert assessment of likely ranges with information about
the broader probability distribution from the expert elicitation of Bamber and Aspinall (2013). Glacier
mass changes were based on a surface mass-balance model driven by an ensemble of Fifth Coupled
Model Intercomparison Project (CMIP5) climate projections (Marzeion et al., 2012). Spatial patterns for
the Greenland Ice Sheet, the West Antarctic Ice Sheet (WAIS), the East Antarctic Ice Sheet (EAIS), and
18 glacial regions’ contributions to sea level rise were calculated following Mitrovica et al. (2011),
assuming uniform melt across the source region (Figure 7). The assumption of uniform melt assumption
may not be entirely realistic for the largest of the ice sources (Greenland, East Antarctica, and West
Antarctica), but such an assumption would be more of a consequence to RSL near the ice source
(Mitrovica et al., 2011) and generally not an issue for most of the U.S. Thermal expansion and ocean
dynamics were based on a distribution constructed from the CMIP5 ensemble of global climate model
projections (Taylor et al., 2012). The global-mean anthropogenic land-water storage contribution, which
is a small contributor, was assumed to be spatially uniform and estimated based on the historical
relationship between population size, retention of water in dams, and groundwater extraction, combined
with United Nations (2014) population projections (Kopp et al., 2014).
To estimate the long-term contribution of nonclimatic processes such as GIA, tectonics, and sediment
compaction to RSL change, we use results from a spatiotemporal statistical model of tide gauge data
based upon methods described in Kopp et al. (2014). In this model, the spatiotemporal field of RSL
change over 1900–2012 is represented as the sum of three signals: 1) a globally uniform sea level
change, 2) a constant-rate average, long-term, regionally varying trend, and 3) temporally and spatially
varying regional sea-level contributions. This model is separately fitted to tide gauge data in several
different regions (of primary relevance to this analysis: the U.S. and Canadian Atlantic Coast, the Gulf
of Mexico, the contiguous U.S. Pacific Coast, Alaska, and the Atlantic and Pacific Islands State and
Territories). The spatial scales of variability of processes 2 and 3, and the temporal scale of variability
of process 3, are learned in each region from the tide gauge data. The globally uniform signal is
assumed to match the GMSL signal estimated by Church and White (2011); the discrepancy among
different estimates of this signal likely contributes ~0.2 mm/year (2 cm/century) uncertainty to
estimates of the long-term background RSL trend, which is considered small enough to neglect. The
nonclimatic background RSL trend used for forward projections is the second process estimated by
this model. This trend is assumed to continue at a constant rate. This assumption is accurate for GIA,
but likely less so for unsteady processes such as those resulting from tectonic processes and/or
anthropogenic disturbances (e.g., subsurface fluid withdrawal), which may increase or decrease over
time. Both the regional degree of spatial variability in the background RSL trend and the density of
nearby tide gauges affects the magnitude of the standard error during trend computation at the center
of each 1-degree grid point.
17
A 1-degree spatial grid was chosen, since this resolution is typically sufficient for many decision-
making needs and is a useful intermediate scale between the large-scale sources of RSL (estimated
from Global Circulation Models [GCMs]) and the inherently (in some cases) local-scale VLM-related
processes. It should be noted that scenario projections of RSL are also generated for specific tide
gauge locations, so where there is location-specific tide gauge data, higher resolution background rates
are obtainable. Gridded RSL results without a background RSL signal are provided to allow for the
use of alternative sources of information about this background rate of change (see equation 2 in
section 5.3). Users interested in substituting local VLM estimates instead (e.g., derived from a specific
GPS dataset) are reminded that, although background rates primarily reflect VLM, they do include a
related SSH contribution as well. Comparative analysis (see section 5.4) between gridded GPS-derived
VLM and background nonclimatic RSL rates reveal broad agreement along U.S. coastlines.
Figure 7. Ratio of RSL to GMSL change for mass loss from specific land-ice sources; these constitute the static-
equilibrium fingerprints of the source. a) Greenland ice sheet (GIS), b) West Antarctic ice sheet (WAIS), c) East Antarctic
ice sheet (EAIS) and d) the median of 18 mountain glaciers after Kopp et al. 2014, 2015. Values more than (less than) 1
indicate RSL rise higher (lower) than GMSL rise.
18
To tie the regional, probabilistic projections to the GMSL rise scenarios, 20,000 Monte Carlo-sampled
time series of GMSL and regional RSL projections were generated for each of RCP2.6, RCP4.5 and
RCP8.5 following the basic approach outlined by Kopp et al. (2014). Notable differences from Kopp et
al. (2014) include 1) the use of the spatiotemporal statistical model described in that paper to estimate
the background rate of change on a grid rather than exclusively at tide gauge sites, and 2) the use of the
full spatial field of SSH projected by the CMIP5 GCMs to estimate the gridded responses. These
projections were then pooled for the three RCPs considered and the results stratified into subsets as
described in Table 3. Note that only the GMSL (and corresponding RSL) Monte Carlo estimates that fell
within the prescribed range per scenario (e.g., 50 ± 2 cm for the 0.5-m scenario) were utilized, and thus
the total number of samples shown in Table 3 do not equal the overall number of Monte Carlo samples.
Finally, the median of the stratified subset of projections was taken to determine the time-evolution of
GMSL for each scenario and the associated projections of RSL change.
To account for uncertainty in the relationship between GMSL and climate-related RSL, the central
(i.e., median scenario value) and the 66th-percentile range corresponding to the 17th percentile (low
scenario value) and 83rd percentile (high scenario value) of each stratified subset are also computed.
To generate low, medium and high total RSL (climatic and nonclimatic contributions combined) sub-
scenarios for each GMSL scenario in a way that allows easy substitution of alternative background
RSL rate estimates, we combine the 17th percentile (low), 50th percentile (median) and 83rd percentile
(high) of climatic RSL with, respectively, the 17th, 50th, and 83rd percentile of the background RSL rate
contribution. The probability range (conditional upon the GMSL scenario) spanned by the total RSL
sub-scenarios is therefore roughly 80%, with the precise range depending on the shape of the climatic
RSL distribution. (If it were normally distributed, which it is not, it would be 84%.)
One additional modification of the Kopp et al. (2014) framework was required for the Low scenario.
Within each RCP, the Kopp et al. (2014) framework incorporates a correlation between global mean
thermal expansion and regional ocean dynamics derived from the CMIP5 archive. This correlation has
the potential to create a significant discontinuity in ocean dynamics and thus in RSL projections across
2100 in the Low scenario. This is because RCP2.6 is dominantly responsible for the Low scenario, yet
the number of CMIP5 simulations for RCP2.6 drops from 21 to six after 2100. This large drop leads to
a discontinuity in the correlation between thermal expansion and ocean dynamics, and thus to a
discontinuity in ocean dynamics, between 2100 and 2110. Since the contributions of thermal
expansion and ocean dynamics are large relative to other terms in the Low scenario, this discontinuity
appears particularly significant. To ameliorate this effect, we assume in the Low scenario that ocean
dynamics are not correlated with global mean thermal expansion, except inasmuch as both terms
depend upon emissions pathway.
19
Table 2. Key processes contributing to RSL and GMSL change, and sources of information.
Process
Affects
SSH
Affects
VLM
Sources of information used
Ice sheet mass changes
X
X
IPCC AR5 assessment of likely [>66% probability]
ranges (Church et al., 2013a), with expert elicitation
(Bamber and Aspinall, 2013) used to inform shapes of
distribution tails; spatial patterns per Mitrovica et al.
(2011)
Glacier mass changes
X
X
Glacier surface mass-balance model (Marzeion et al.,
2012) driven by CMIP5 models; spatial patterns per
Mitrovica et al. (2011)
Oceanographic processes
(thermal expansion and
atmosphere-ocean dynamics)
X
CMIP5 models (Taylor et al., 2012), with 5–95% range
interpreted as likely range per IPCC convention
Land-water storage
X
X
Empirical relationships between population, dam storage,
and groundwater withdrawal; uniform spatial pattern
assumed
Glacial isostatic adjustment
X
X
Long-term background rate of RSL change inferred from
spatiotemporal model of tide gauge observations
Tectonics and sediment
compaction
X
Long-term background rate of RSL change inferred from
spatiotemporal model of tide gauge observations
Table 3. Constraints used to stratify the projections and number of sample estimates per scenario. Note the definition
of the Low scenario is the continuation of the current rate of GMSL rise (~3 mm/year) through 2100, whereas the others are
just defined by 2100 values.
GMSL rise Scenario
Constraints
Number of samples
Low
2100 GMSL of 30 ± 2 cm; 2050 GMSL of 15 ± 2 cm;
2030 GMSL of 9 ± 1 cm
224
Intermediate-Low
2100 GMSL of 50 ± 2 cm
4407
Intermediate
2100 GMSL of 100 ± 2 cm
1040
Intermediate-High
2100 GMSL of 150 ± 5 cm
168
High
2100 GMSL of 200 ± 5 cm
23
Extreme
2100 GMSL of 250 ± 15 cm
21
As will be described in the next section, stratifying the probabilistic projections into six subsets in this
way creates six scenarios (or families of scenarios) for GMSL and/or RSL at a given grid point (and set
of tide gauge locations). It is important to emphasize that these six scenarios were not necessarily chosen
to be relevant for any specific planning or decision-making process, nor to meet the needs of any specific
decision-maker or other user, and the names (Low, High, etc.) are for convenience in usage. Section 6
provides discussion and guidelines for selecting the appropriate scenario from among these six—more
broadly, from the full conditional probability distributions that these scenarios span— for a specific
application and decision context. These particular scenarios do have meaning in a scientific context. The
Low and Extreme scenarios represent the scientifically plausible lower and upper bounds on 21st century
GMSL rise, respectively, as defined in this report; the remaining four scenarios (from Intermediate-Low
to High), while simply placed at 0.5-m intervals in between, can be shown to correspond to different
likelihood levels under RCP2.6, RCP4.5, and RCP8.5, as discussed in section 5.
20
21
5.0 RESULTS
There are a few important distinctions between the GMSL rise scenarios in this report and those of
Parris et al. (2012) and Hall et al. (2016). Namely, the new scenarios: 1) are anchored in year 2000
(i.e., a 1991–2009 epoch), instead of 1992 (i.e., the National Tidal Datum Epoch of 1983–2001) so as
to align with the framework of Kopp et al. (2014) and other sea level rise projections provided in the
scientific literature, 2) provide decadal-scale estimates through 2100, 3) have a broader and higher
range for GMSL rise by 2100 (0.3–2.5 m) as described above, 4) provide estimates through the year
2200, and 5) are downscaled to a 1-degree gridded basis to provide a systematic spatial framework to
more broadly support regional/local decision making.
In terms of (1) above, we recognize that most of the U.S. is currently on a 1983–2001 datum epoch, and
a RSL-change adjustment is needed for many applications. To obtain a 1991–2009 epoch (centered on
year 2000) for a location, we suggest two methods. First, if a local tide gauge exists with RSL data over
1983–2009 exists, the simplest approach is to calculate the difference between the 1983–2001 average
and the 1991–2009 average and then to apply this difference (NOAA, 2003). Second, if no adequate
nearby tide gauge is available, an estimate of 8-years’ worth of background RSL change (see section
5.4) can be combined with (a) an estimate of 8-years’ worth of SSH change over 1992–2009, measured
offshore by satellite altimetry, (b) a tide-gauge-based estimate of GMSL change between 1983–2001
and 1991–2009 (1.8 cm per Hay et al., 2015), or (c) an altimetry-based estimate of 8-years’ worth of
GMSL change at the rate characterizing 1993–2009 (3.0 cm per Nerem et al., 2010).
5.1 Global Mean Sea Level Rise Scenarios
As listed in Table 3, the six representative GMSL rise scenarios range from a low-end (Low) scenario
of 0.3 m to a worst-case (Extreme) scenario of 2.5 m by 2100. The Intermediate-Low (0.5 m),
Intermediate (1.0 m), Intermediate-High (1.5 m) and High (2.0 m) scenarios are the same as those put
forward by Hall et al. (2016). In Figure 8, temporal evolution of the scenarios is illustrated through
year 2100 relative to GMSL reconstructions from 1800 to 1900 based upon the meta-analysis of
geological and tide gauge data of Kopp et al. (2016a), from 1900 to 2010 from the tide gauge analysis
of Hay et al. (2015), and from 1992–2015 based upon satellite altimetry analysis updated from Nerem
et al. (2010). The six GMSL rise scenarios are also shown (Table 4) relative to the probability of
exceedance in 2100 as assessed by the RCP-based probabilistic projections of Kopp et al. (2014). Note
that the GMSL rise scenarios assume that the rate of ice-sheet mass loss increases with a constant
acceleration; however, this might not be the case (DeConto and Pollard, 2016), so it is, for example,
possible to be on the Intermediate scenario early in the century but the High or Extreme scenario late
in the century. Under the methodological assumptions of Kopp et al. (2014), in 2100 the Low scenario
has a 94% to 100% chance of being exceeded under RCP2.6 and RCP8.5, respectively, whereas the
Extreme scenario has a 0.05% to a 0.1% chance of being exceeded. However, as discussed in section
3, new evidence regarding the Antarctic ice sheet, if sustained, may significantly increase the
probability of the Intermediate-High, High, and Extreme scenarios, particularly for RCP8.5 projections
based upon Kopp et al. (2014). These ice-sheet modeling results have not yet been incorporated into a
(conditional) probabilistic analysis of GMSL.
22
Figure 8. This study’s six representative GMSL rise scenarios for 2100 (6 colored lines) relative to historical geological, tide
gauge and satellite altimeter GMSL reconstructions from 1800–2015 (black and magenta lines; as in Figure 3a) and central
90% conditional probability ranges (colored boxes) of RCP-based GMSL projections of recent studies (Church et al., 2013a;
Kopp et al., 2014; 2016a; Slangen et al., 2014; Grinsted et al., 2015; Mengel et al., 2016). These central 90% probability
ranges are augmented (dashed lines) by the difference between the median Antarctic contribution of Kopp et al. (2014)
probabilistic GMSL/RSL study and the median Antarctic projections of DeConto and Pollard (2016), which have not yet
been incorporated into a probabilistic assessment of future GMSL. (A labeling error in the x-axis was corrected on January
30, 2017).
Table 4. Probability of exceeding GMSL (median value) scenarios in 2100 based upon Kopp et al. (2014).
GMSL rise Scenario
RCP2.6
RCP4.5
RCP8.5
Low (0.3 m)
94%
98%
100%
Intermediate-Low (0.5 m)
49%
73%
96%
Intermediate (1.0 m)
2%
3%
17%
Intermediate-High (1.5 m)
0.4%
0.5%
1.3%
High (2.0 m)
0.1%
0.1%
0.3%
Extreme (2.5 m)
0.05%
0.05%
0.1%
5.2 GMSL Rise Rates this Century and Rise Beyond 2100
Though the GMSL rise scenarios are primarily framed for overall changes occurring by 2100, it is
important to recognize that GMSL rise will not stop at 2100; rather, it will continue to rise for
centuries afterwards (Levermann et al., 2013; Kopp et al., 2014). By 2200, the 0.3–2.5 m range
spanned by the six GMSL rise scenarios increases to 0.4–9.7 m, as shown in Table 5. It can be seen
(Figure 8) that deceleration of GMSL occurs under the Low scenario with only slight increases
through 2200. Continued acceleration is modest under the Intermediate-Low scenario and pronounced
under all other scenarios (Table 6). The amount of GMSL rise by 2200 does not necessarily represent
the maximum physically possible contributions from ice-sheet, ice-cliff or ice-shelf feedback
processes, which, as discussed in section 3, may significantly increase contributions to overall GMSL
rise amounts (DeConto and Pollard, 2016).
23
Table 5. GMSL rise scenario heights in meters for 19-year averages centered on decade through 2200 (showing only a subset
after 2100) initiating in year 2000. Only median values are shown.
GMSL
Scenario
(meters)
2010
2020
2030
2040
2050
2060
2070
2080
2090
2100
2120
2150
2200
Low
0.03
0.06
0.09
0.13
0.16
0.19
0.22
0.25
0.28
0.30
0.34
0.37
0.39
Intermediate-
Low 0.04 0.08 0.13 0.18 0.24 0.29 0.35 0.4 0.45 0.50 0.60 0.73 0.95
Intermediate 0.04 0.10 0.16 0.25 0.34 0.45 0.57 0.71 0.85 1.0 1.3 1.8 2.8
Intermediate-
High 0.05 0.10 0.19 0.30 0.44 0.60 0.79 1.0 1.2 1.5 2.0 3.1 5.1
High
0.05
0.11
0.21
0.36
0.54
0.77
1.0
1.3
1.7
2.0
2.8
4.3
7.5
Extreme
0.04
0.11
0.24
0.41
0.63
0.90
1.2
1.6
2.0
2.5
3.6
5.5
9.7
The rates of GMSL associated with the time-dependent GMSL rise heights are shown in Table 6.
Rates range from a near-constant 3 mm/year of the Low scenario (by definition) this century to 5 mm
to 44 mm/year later (2081–2099 average) in the century, with differing amounts of acceleration. As
previously noted, global mean thermal expansion, ocean dynamics, and glacier melt in the Kopp et al.
(2014) framework are derived from the CMIP5 archive (directly for thermal expansion and ocean
dynamics, as an input to a glacier mass-balance model for glacier melt). The reduction in CMIP5
model simulations after 2100 leads to small discontinuities in GMSL and RSL between 2100 and
2110. We therefore advise caution when using projections for 2110 and advise against calculating
rates over periods spanning 2100. Accordingly, for the 22nd century, Table 5 focuses on projections
for 2120, 2150, and 2200, and the rates in Table 6 are calculated only for the 21st century.
As for the actual GMSL scenario heights, it is possible to be on a rate associated with an Intermediate-
Low or Intermediate scenario earlier in the century, but since the rate of ice-sheet mass loss may not
necessarily change linearly (e.g., DeConto and Pollard, 2016), GMSL rise rates could transition to
those associated with higher GMSL rise scenarios later in the century and could exceed all rates
shown. For context, the geological record reveals that during the last deglaciation (~20,000–9000
years before the Common Era), GMSL rise rates exceeded about 10 mm/year, with rates above
40 mm/year during meltwater pulses (Deschamps et al., 2012; Miller et al., 2013).
Table 6. Rise rates (in millimeters per year for 19-year averages centered on decade) associated with the median GMSL
scenario heights this century (as shown in Table 5).
GMSL Scenario
Rates (mm/year)
2010
2020
2030
2040
2050
2060
2070
2080
2090
Low
3
3
3
3
3
3
3
3
3
Intermediate-Low
4
5
5
5
5
5
5
5
5
Intermediate
5
6
7
9
10
12
13
14
15
Intermediate-High
5
7
10
13
15
18
20
22
24
High
6
8
13
16
20
24
28
31
35
Extreme
6
10
15
20
25
30
35
40
44
24
5.3 Regional Climate-related RSL Changes
The amount of RSL change that occurs regionally under each GMSL rise scenario reflects
contributions from both 1) climate-related processes affecting regional SSH and VLM and 2)
nonclimatic background RSL changes, primarily from VLM but also including any related SSH
changes. In other words, for a given scenario,
2) (,) = (,) + () ( − 0)
where , the total RSL change relative to RSL at time 0 in the sea level scenario, is defined for
spatial locations x and times t, and the nonclimatic, background change is assumed to be linear in time.
The climatic and nonclimatic responses are quantified separately to allow supplemental VLM
estimates (e.g., local GPS measurements) to be substituted for the background RSL rate estimates
instead. However, one caution as stated earlier is that using localized GPS-VLM rate estimates in place
of the tide gauge-estimated background rates may lead to the neglect of any SSH effects within the
background RSL rate. These effects can be significant; for instance, in some GIA models (e.g., Peltier,
2004), SSH fall offsets about one-third of the RSL rise caused by land subsidence in New York City.
Figure 9 shows how the median climate-related RSL projections at 1-degree resolution for each
scenario in 2100 differ from the median GMSL rise for that scenario. This difference can be either
positive or negative; i.e., the climatic contribution to RSL change can be either greater or less than the
GMSL rise, with differences between the climate-related RSL change and GMSL rise arising due to
both ocean dynamics (e.g., along East Coast) and the static-equilibrium fingerprint effects of land-ice
mass changes. In addition to the median value, low and high RSL values for each GMSL rise scenario
are shown in appendices B and C, for climate-related (not including background) and total RSL,
respectively. The climate-related RSL projection patterns in Figure 9 are summarized with those of the
nonclimatic background RSL rates later in section 5.5.
25
Figure 9. Climate-related RSL change at 1-degree resolution for 2100 (in meters) relative to the corresponding (median-
value) GMSL rise amount for that scenario. To determine the total climate-related RSL change, add the GMSL scenario
amount to the value shown.
Figure 10 shows several key components that contribute to the median RSL projections using the
Intermediate (1-m GMSL) scenario as an example. Note that this figure shows the total contribution of
a process to RSL change, not just the difference from GMSL change as in Figure 9. Accounting for the
amplification of its contribution by static-equilibrium processes (Figure 7), the Antarctic Ice Sheet
(AIS) contributes 0.1–0.2 m to RSL rise across the U.S. The Greenland Ice Sheet (GIS) contributes
0.1–0.2 m as well across most of the U.S., with higher (0.2–0.3 m) amounts within the Pacific due to
far-field static-equilibrium effects and lower amounts (0–0.1 m) along the Northeast Atlantic and
Northern Alaska due to intermediate-field effects. (Both AIS and GIS progressively contribute to
higher RSL rise along most U.S. shorelines, with similar patterns under the Intermediate-High, High
and Extreme scenarios). RSL contributions from glacier and ice caps (GIC) melt are about 0.2 m from
melt around the U.S., but are substantially lower (sizable RSL fall in some places) due to near-field
and intermediate-field static-equilibrium effects along the Alaskan coast and in the Pacific Northwest.
26
Lastly, oceanographic (OC) processes, including both global mean thermal expansion and regional
atmosphere/ocean dynamics, contribute about 0.4–0.5 m to RSL rise around the U.S. and higher along
the Northeast Atlantic Coast (0.7–0.8 m) related to changes in the Gulf Stream System.
Figure 10. Component contributions (in meters) to the median RSL Intermediate (1.0 m GMSL rise) scenario values in 2100
from a) the AIS, b) GIC, c) GIS and d) oceanographic processes (global mean thermal expansion and regional
atmosphere/ocean dynamics; note that a different scale is used).
Figure 11 compares the median climate-related RSL values shown in Figure 9 for the Intermediate
(1-m GMSL rise) scenario for the year 2100 to those of Hall et al. (2016). For this comparison, the
Hall et al. (2016) 1-m sea level scenario has been gridded (Figure 11a). Similar patterns emerge, such
as higher values along the Hawaiian coastline due to far-field effects from Antarctic ice-mass loss;
lower values (negative) occur along the Alaskan and Northwest U.S. coasts from near-field effects of
ice-mass loss of Alaskan mountain glacier and Greenland melt. Because of differences between the
methods used in this report as compared to those in Hall et al. (2016), some variations are expected.
Notable differences between this study’s results (Figure 11b) include higher values (10–20 cm) along
the U.S. East Coast and generally higher values along the Southern Alaskan Coast (50 cm or more)
than the Hall et al. (2016) values. This may be due to Hall et al. (2016)’s usage of a single fingerprint
for all glaciers combined, or to somewhat different Alaska fingerprint values in the Alaska regions of
Perrette et al. (2013) used in the Hall et al. (2016) study. Perrette et al. (2013) relied on the present-day
distribution of mass loss using the model of Bamber and Riva (2010) to develop fingerprints, which
were assumed to remain the same in the future. Also, Hall et al. (2016) used only a single Antarctic
fingerprint, which did not differentiate contributions from the WAIS (Figure 7). This may account for
their lower values along North America, since the WAIS fingerprint has a higher relative effect than
27
the EAIS along both coasts. Another source of difference could be related to a higher overall
contribution from Greenland in the Hall et al. (2016) study, which, due to near- and intermediate-field
static equilibrium effects (Figure 8a), would cause a reduction of climate-related RSL along the U.S.
East Coast as seen in Figure 11a. On average, this study’s gridded values for the Intermediate scenario
are 8 cm higher along the U.S. coastline, though with similar RSL spatial patterns in general
agreement (linear regression of RSL within grids for the Intermediate (1-m GMSL rise) in Figures 9
and 11a: R2=0.65).
Figure 11. a) Median climate-related RSL amounts under the Intermediate (1-m GMSL rise) scenario in year 2100 of
Hall et al. (2016) (with all grid values shown with GMSL amount [1.0 m] removed) and b) difference between RSL in this
study’s (Sweet et al., 2017) Intermediate scenario as shown in Figure 9 and Hall et al. (2016) results shown in a).
5.4 Nonclimatic Background RSL and GPS VLM Trends
The other factor needed to regionalize the GMSL rise scenarios is background RSL change caused by
nonclimatic processes. These background changes are primarily driven by VLM, but also include SSH
changes associated with GIA. The gridded background RSL trends, which are assumed to measure
linear trends and to be independent of scenario, are shown in Figure 12a. The assumption of
background RSL rate persistence this century is valid for risk-framing purposes (Hall et al., 2016) but
could become invalid if, for example, most of the underlying signal stems from anthropogenic-induced
VLM, and the driving disturbance ceases at some point in the future. The background RSL rates reveal
similar patterns as the GPS-derived VLM estimates, which are shown in Figure 12b as a 1-degree
gridded version of Figure 5 for GPS sites within 50 km of the U.S. coastline. Note that the direction of
the GPS VLM rates have been switched in Figure 12b (and Figure 12e, f) for comparison purposes (i.e.,
negative VLM trends or downward motion of the land are shown as a positive contribution to RSL
change). Common patterns between the tide gauge background RSL and GPS-VLM trend estimates
include slight RSL fall (0 to 1 mm per year RSL fall/land uplift) along the U.S. West Coast. RSL
rise/land subsidence rates of about 1–2 mm/year are found broadly along the U.S. East Coast, though
this is higher (2–4 mm/year or more) along the Mid-Atlantic coast from Virginia to New Jersey due to a
combination of both anthropogenic processes, such as extraction of groundwater, and GIA associated
with the disappearance of the North American (Laurentide) Ice Sheet. Within the central and western
Gulf of Mexico, RSL rise/land subsidence rates of about 3–5 mm/year or more reflect sediment
compaction of the Mississippi River Delta and withdrawal of subsurface groundwater and fossil fuels.
RSL fall/land uplift rates of 5 mm/year or more are found in Southern Alaska due in part to GIA
associated with the post-Little Ice Age retreat of mountain glaciers.
28
Figures 12c and 12d show the standard deviations of the background RSL and VLM rate trends, which
are approximately 1–2 mm/year within the continental U.S. and generally higher along the Alaskan
Coast as assessed by tide gauges. Trend standard deviations from the tide gauges (Figure 12c) are
generally less along the contiguous U.S. as compared to those derived from GPS (Figure 12d), likely
due to much shorter GPS record lengths as compared to the tide gauges. As detailed in section 4.2, the
spatial scale of variability of the background RSL trend is learned from the data, which is why the
errors are much larger farther away from tide gauges in Alaska (where gauges are sparse and the
dominant spatial scale of variability short) than on the U.S. East Coast (where gauges are numerous
and the dominant spatial scale of variability long). Similarly, VLM rates and their uncertainties
assessed by GPS will vary, often substantially over relatively short distances. A reason why the
standard deviations of the gridded VLM trends are relatively low in Alaska (Figures 12d) is due to the
computation process. Unlike the background RSL trend, which is estimated for the specific center of
the grid, the VLM rates and their standard deviations are derived by simply pooling available
measurements, averaging their rates, and statistically reconciling the trend variances.
Figure 12e shows the differences in rates between the background RSL and GPS VLM trends. Larger
discrepancies occur in regions where rates are high and likely influenced by human activities that have
varied through time, such as pumping of groundwater/fossil fuels (e.g., Western Gulf of Mexico);
discrepancies are also high where large spatial-differential VLM rates exist due to processes such as
volcanism, tectonics or recent glacier retreat (e.g., Southern Alaska). However, the trend rates derived
from tide gauge (background RSL) and GPS (VLM) are generally in close agreement (±1 mm/year)
along much of the U.S., which is further exemplified by the high goodness of fit measure in Figure 12f
(black dots: R2 = 0.78). Also shown in Figure 12f is a similar comparison among the background RSL
rates as in Figure 12a, but derived and compared to VLM estimated for more than 100 specific tide
gauge locations used by Hall et al. (2016) and based upon Zervas et al. (2013). In this instance, the two
data sets (shown as red dots) are quite close (R2 = 0.92), with discrepancies partly because Zervas et al.
(2013) use a constant 20th century GMSL rise rate instead of a time-varying rate corresponding to the
specific time periods covered by the regional groups of tide gauges used in their assessment.
29
Figure 12. Gridded a) background RSL rates and b) GPS VLM rates (with directionality switched as to directly
compare to RSL rate) with trend standard deviations shown in c) and d). In e) is the difference between the two rates
shown in a) and b). In f) is a scatterplot and linear regression of the two sets of gridded rates shown in a) and b) as
black dots and the red dots show background RSL rates as in a) but derived and compared to VLM estimated for more
than 100 specific tide gauge (TG) locations used by Hall et al. (2016) and based upon Zervas et al. (2013).
5.5 Scenario Projections of Relative Sea Level (RSL)
Figure 13 shows the total median RSL projections relative to the median GMSL values for each
scenario for 2100 (Figure 8). High and low RSL values for each GMSL scenario are provided for each
scenario in appendix C. The patterns reflect those inherent to Figures 9 and 12a as summarized in
section 5.3. For instance, under the Intermediate-High (1.5-m GMSL rise) scenario, RSL along the
U.S. East Coast is about 0.4–0.7 m higher than GMSL rise, 0.2–1.0 m higher along the Gulf Coast,
0.2–0.3 m higher along the West Coast, 0.3–0.5 m higher within Hawaii and the Pacific Islands and
both slightly higher and much lower (-1.0 m to +0.2 m) along the Alaskan Coast.
Some notable patterns projected to occur this century (Figure 13) and their physical interpretations
(based on Yin, 2012; Perrette et al., 2013; Church et al., 2013a; Kopp et al., 2014, 2016a; Slangen et
al., 2014; Grinsted et al., 2015) are listed below.
30
● Future RSL rise is amplified along the Northeast U.S. coast due to the effects of GIA, the far-
field static equilibrium effects of Antarctic melt (Figure 7), and reduced transport of the larger
Gulf Stream System (Atlantic Meridional Overturning Circulation [AMOC]). Future RSL rise
is partially reduced by intermediate-field static-equilibrium effects associated with relative
proximity to Greenland and many northern glaciers.
● RSL rise is amplified along the western region of the Gulf of Mexico and much of the
Northeast Atlantic coast by withdrawal of groundwater (along the Atlantic Coast) and of both
fossil fuels and groundwater (along the Gulf Coast). Continuation of these practices would
amplify future RSL rise. Far-field static equilibrium effects of Antarctic melt (Figure 7) are
also amplified in these two regions (as in the first bullet describing the Northeast U.S.).
● Future RSL is amplified along the Pacific Coast of the continental U.S. due to far-field static-
equilibrium effects of Antarctic ice sheet melt.
● Future RSL is reduced along the Alaska and the U.S. Pacific Northwest coasts due to
proximity to the Alaskan glaciers from both ongoing GIA to past glacier shrinkage and to the
static-equilibrium effects of projected future losses.
● Future RSL rise is amplified in Hawaii and other Pacific islands by static-equilibrium effects,
because they are in the far field of all sources of melting land ice.
It should be noted that the probabilistic construction of the High and Extreme scenarios draw more
heavily on Antarctic contributions than might be the case in a world with 2.0 m or 2.5 m of GMSL rise
by 2100, respectively. This is because: 1) There is no assumed correlation between Greenland and
Antarctic melt other than that associated with the RCP forcing, 2) the High and Extreme scenarios
require a large Antarctic contribution, and 3) under an assumption of noncorrelation, it is unlikely that
both the Greenland and Antarctic contributions will be extreme. Because Antarctic contributions to
RSL rise are amplified along all U.S. coastlines, while Greenland contributions are dampened across
much of the U.S., RSL is projected to be higher than if driven by a more extreme Greenland
contribution and a somewhat less extreme Antarctic contribution.
31
Figure 13. Total RSL change at 1-degree resolution for 2100 (in meters) relative to the corresponding (median-value) GMSL
rise amount for that scenario. To determine the total RSL change, add the GMSL scenario amount to the value shown.
32
33
6.0 USAGE OF SCENARIOS WITHIN A RISK-BASED CONTEXT
Significant uncertainties exist about the exact trajectory (and impacts) of future climate change,
limiting the value of prediction-based frameworks for long-term, climate-related decision-making
(e.g., see Hallegatte et al. 2012, Weaver et al., 2013). In other words, decision-makers must expect to
be surprised (NRC, 2009). Planning approaches using sea level scenarios (e.g., Parris et al., 2012 and
Hall et al., 2016) can help manage uncertainty for continuity of mission and functionality of critical
system(s). For example, long-term coastal planning at a local or regional/state level (e.g., major
infrastructure upgrades) requires information about possible RSL rise that might affect the project in
response to all factors associated with the warming of the oceans and cryosphere. Parris et al. (2012)
did not explicitly provide such information, and although the Hall et al. (2016) study did, their site-
specific results are not publicly available.
This report provides GMSL rise scenarios and associated 1-degree resolution RSL (amounts and rates)
covering what is plausibly a full range of possible futures given current scientific understanding and
different assumptions about future greenhouse gas emissions. However, further information is needed
for their application, such as how best to identify and select the small number of most relevant
scenarios for a specific management application from the full conditional probability distributions.
Providing such guidelines in a tailored way across the diversity of coastal planning contexts is beyond
the scope of this report. However, drawing upon Hall et al. (2016), Hinkel et al. (2015), Kopp et al.
(2016b), King et al. (2015) and others, it is possible to put forth some general guidelines and provide a
couple of different contextual examples.
6.1 General Guidelines for Scenario Selection
Coastal planners making critical decisions should weigh several factors, such as the type of decision to
be made, expected future performance, planning horizon, and overall risk tolerance, including the
criticality of the asset and/or the size and vulnerability of the exposed population (Hall et al., 2016).
The process of selecting a sea level scenario for a specific setting is not a straightforward task for
planners and engineers, and there are only a few case studies regarding its application in the literature
(e.g., Hall et al., 2016; Ranger et al., 2013). Choosing meaningful sea level scenarios for risk
management purposes begins with an understanding of the specific system, problem, goals, and
preferences. These may include:
● What is the decision type? What is the operational timeframe over which the decision needs to
function in an effective manner (planning horizon)? How flexible is the decision should new
information become available?
● Who and what will be affected if sea level increases, and in what way? What is “at risk”?
● What outcomes are desired to avoid, in the near and long term with respect to these valued
things? What is the tolerance for risk?
● Are there thresholds or tipping points in human or natural systems of concern beyond which
damages would increase dramatically?
● What are the specific details of the coastal system: e.g., elevation, characteristics of the
coastline, the locations of things of value (houses, infrastructure, sensitive ecosystems)?
Key to the decision process is determining the extent to which a given amount of RSL rise may cause
impacts to either newly built or existing infrastructure. This will differ by geographic location due to
differences in topography, land cover, physical forcing (i.e., affecting the width of the typical high
water ‘distribution’ in Figure 1), and existing storm-flood defenses. Thus, it is important to recognize
34
the nature of the surrounding physical forcing regime and how the type of anticipated impact
(e.g., magnitude, frequency or duration of high water event) might vary in the future under the range
of sea level scenarios. A relatively small amount of RSL rise will substantially increase the frequency
and spatial extent of flooding more in some locations than others. For instance, tidal flooding has and
will continue to become more frequent and severe with a given amount of RSL rise along regions with
flat, low-lying coastal zones that typically do not experience regular, severe storm surges (e.g., the
Southeast Atlantic Coast) more so than along regions with steeper topography that experience often-
severe extratropical storms (e.g., New England Coast) (Sweet and Park, 2014). Often, if a location is
prone to impacts from severe storms, some flood defenses may likely be in place, such as for storm
surges along a beachfront (e.g., seawalls, elevated houses along the beachfront). But back bay and
inland regions of harbor cities with infrastructure (e.g., transportation, waste and stormwater systems)
at ground level will continue to be exposed to impacts of more-probable and increasingly-extreme
tidal flooding due to RSL rise.
Early stages of planning need to establish the conceptual linkage between sea level rise, the system(s)
of interest, and the threat to those systems. Once a conceptual linkage is established, the next step is to
“stress test” the system and the plans against different futures to assess potential risks. Scenarios, to be
most useful, should delineate between policies or plans that succeed and those that fail
(Lempert, 2013). For decisions involving long planning horizons and with a limited adaptive
management capacity, the high degree of uncertainty in late-21st century GMSL rise looms large.
Failure to adequately account for low-probability, high-consequence outcomes significantly increases
future risks and exposure (Oppenheimer and Alley, 2016). For many decisions, it is essential to assess
worst-case scenarios, not only those assessed as the scientifically ‘likely’ to happen. For example,
drawing on the references cited above, the following is suggested as a potential initial scenario
selection strategy for decisions and planning processes for which long-term risk management is
paramount:
● Define a scientifically plausible upper-bound (which might be thought of as a worst-case or
extreme scenario) as the amount of sea level rise that, while low probability, cannot be ruled
out over the time horizon being considered. Use this upper-bound scenario as a guide for
overall system risk and long-term adaptation strategies.
● Define a central estimate or mid-range scenario (given assumptions about greenhouse gas
emissions and other major drivers). Use this scenario as baseline for shorter-term planning,
such as setting initial adaptation plans for the next two decades. This scenario and the upper-
bound scenario can together be thought of as providing a general planning envelope.
This approach is consistent with that recommended in Kopp et al. (2016b) and used by the Thames
Estuary 2100 project (Ranger et al., 2013; Hinkel et al., 2015). Continuous monitoring of current sea
level behaviors (trends and variability), along with improved scientific understanding of relevant
climate-system processes and feedbacks, can then help identify the evolution of the system over time
with respect to these mid-range and worst-case scenarios. Systematic assessments can determine
current sea level rise (and risk) trajectories and when to implement more aggressive response options
against the high-end scenario if adaptive management strategies are an option (USACE, 2013;
Hall et al., 2016).
Several other logistical aspects should be considered, such as relevant elevation-frequency thresholds,
probability of water level events (recurrent-to-rare in frequency), availability of spatial/mapping and
tidal-geodetic datum information, rates of change, and other factors. The following sections provide
35
more detailed scenario applications that consider these additional factors, focused on increasing
frequency of recurrent tidal flooding at a national scale and the increasing magnitude of a major flood
event along the South Florida coast.
6.2 Scenario Projections of RSL and Tidal Flood Frequencies: A National
Perspective
Impacts of RSL have been and will continue to be experienced as increasingly deep and frequent tidal
flooding (e.g., Ezer and Atkinson, 2014; Sweet et al., 2014; Sweet and Park, 2014; Sweet et al., 2016).
Tidal flooding is an impact event not generally attributed to a particular storm event, but rather from
the continued effects of RSL rise, which elevate the reach of normal tides and water-level setup from
prevailing winds. Eventually, the cumulative toll from recurrent tidal flooding above a local-specific
physical threshold (i.e., duration, magnitude, or frequency) will eventually degrade sector-specific
functionalities and/or exceed economic or public-tolerance thresholds. In the following example, we
highlight how the frequency of tidal-flood frequencies may change in response to RSL rise (or fall)
under the sea level scenarios. Our focus is on events that have been sufficiently sampled in time as to
provide a robust understanding of their probability of occurrence. A similar assumption is commonly
used when providing maps of future high tide levels associated with rising sea levels (e.g., NOAA Sea
Level Rise Viewer: https://coast.noaa.gov/slr).
Recognizing a location’s current RSL trajectory relative to its set of scenarios and past degree of
interannual variability can assist in decision-making processes over the next several decades.
Figure 14 shows observed annual mean RSL for four U.S. cities relative to future RSL under the six
scenarios. In most circumstances, the range of interannual RSL change/variability since 2000 has been
bounded (to date) by the trajectory of the Intermediate-High (1.5-m) scenario. As noted previously, it
is important to recognize that being close to one scenario early in the century does not imply that real-
world behavior will follow that scenario throughout the century; for example, rapid ice sheet collapse
in Antarctica could conceivably bring the world from the Intermediate scenario early in the century to
the Extreme scenario by the end of the century.
In places like San Francisco (and along the West Coast), interannual RSL variability over the last
several decades has been and may continue to be large compared to the RSL trend itself (see
tidesandcurrents.noaa.gov/sltrends). This can be counterintuitive especially for local decision-makers,
since it suggests that historical trends in RSL may be a poor predictor of future sea level rise. Because
there is not expected to be much difference in RSL under the scenarios over the next couple decades
(Church et al., 2013a; Kopp et al., 2014), where yearly RSL variability effectively spans a large
section of the range of (19-year average) scenarios, near-term decision-making (e.g., annual-to-
decadal) may be effectively scenario-independent. In such circumstances, time-dependent event
probabilities of both minor tidal flooding and more severe events respond primarily to RSL variability
and/or climatic phases affecting regions such as by ENSO, the Atlantic Multidecadal Oscillation
(AMO), the North Atlantic Oscillation (NAO), etc. (Menendez and Woodworth, 2010; Park et al.,
2010; Sweet and Park, 2014; Marcos et al. 2015; Woodworth and Menendez, 2015; Wahl and
Chambers, 2016; Sweet et al., 2016; Hall et al., 2016).
36
Figure 14. Average annual RSL for New York City (The Battery), Miami (Virginia Key), Fla., Galveston, Tex. and
San Francisco, Calif. with their respective (median-value) RSL under the six scenarios. The NOAA RSL observations
(tidesandcurrents.noaa.gov/sltrends) are shown relative to the midpoint (year 2000) of the 1991–2009 epoch (1994–2009 at
Virginia Key), which is the reference level for the scenarios.
When planning under any sea level scenario, both short and/or long-term decisions should recognize
that locations with lower elevation thresholds for impacts, less variability in extreme water levels, or
higher rates of RSL rise have been the most prone to rapid (often-accelerating) increases in event
probabilities (Sweet and Park, 2014) and will continue to be so in the future (Hunter, 2012; Tebaldi
et al., 2012; Kopp et al., 2014, Sweet and Park, 2014; Buchanan et al., 2016). Return level interval
curves of extreme water level events in New York City and San Francisco (Figure 15a, b) illustrate the
differences in the extreme variance between locations (proportional to the gradient of the return level
interval curve). As stated above, the focus here is on the higher probability (more frequent) events,
which are better resolved statistically (i.e., tighter 95% confidence intervals) due to their more frequent
nature than, for example, the extremely rare event such as associated with a hurricane landfall (Hall et
al., 2016). Figure 15c shows the water level heights with a 20% annual chance of occurrence (5-year
recurrence interval) for a set of NOAA tide gauges with more than 20 years of hourly observations. The
20% annual chance flood levels range from about 0.3 m (~1 foot), such as where narrow and deep
continental shelves bathymetrically constrain the magnitude of storm surges (e.g., along the West Coast
and Islands where wave runup and/or dynamic water levels [Stockdon et al., 2006; Sweet et al., 2015]
can be larger than storm surge [Hoeke et al., 2013; Ruggiero, 2013; Serafin and Ruggiero, 2014]) to
0.9 m or more at higher latitudes, where powerful extratropical storms and wide continental shelves
allow larger surges to build (Tebaldi et al., 2012; Hall et al., 2016; Merrifield et al., 2015, 2016). The
median of the 20% annual chance flood in Figure 15c is about 0.8 m (ranging from about 0.3 m to
37
1.8 m) above MHHW, which is nearly the same as the National Weather Service’s empirically derived
elevations used to define ‘moderate’ flooding at dozens of NOAA tide gauges (~0.8 m and ranging
from about 0.5 m to 1.8 m; http://water.weather.gov/ahps). Moderate level flooding is disruptive to
commerce and damaging to private and commercial property; when such coastal flooding is imminent
or occurring, NOAA typically issues a coastal/lakeshore flood warning of a serious risk to life and
property (NOAA, 2014). In contrast, the median nuisance flood threshold around the U.S., which is
associated with minor coastal flooding and issuance of a coastal flood advisory, can be expected to
occur about two times per year (0.5-year recurrence interval) and is about 0.5 m above MHHW (Sweet
et al., 2014).
Figure 15. Generalized Pareto Distribution and 95% confidence interval (CI: red dash) fit for high-water extremes for
historical data through 2015 (black dots) based upon Sweet et al. (2014) for NOAA tide gauges in a) New York City (The
Battery), N.Y. and b) San Francisco, Calif. In c) is a map for water level heights (whose magnitudes are shown relative to the
1991–2009 epoch) with a 5-year recurrence interval (20% annual chance of occurrence) for a set of NOAA tide gauges with
more than 20 years of hourly record [as in a) and b)], and d) shows the height difference between the water levels with a
5-year and a 0.2-year recurrence interval (happening five or more times per year).
Estimates of a location’s high water distribution, which forms in response to high tides and storms, are
needed to assess how and when flood frequencies are likely to change under future RSL rise.
Figure 15d illustrates differences in water level heights associated for events with a 5-year (20%
annual change event) and 0.2-year recurrence interval (happening five or more times per year), with
the former representing disruptive/damaging flooding and the latter associated with localized shallow
(more nuisance-like) tidal flooding. The height separating the two flood types provides a RSL rise-
related ‘time horizon’ for a future when disruptive/damaging tidal flooding may become much more
38
commonplace (under current flood defenses). The median value in Figure 15d is about 0.35 m, with a
range from about 0.1 m to 1.1 m. Thus, for most of U.S. tide gauge locations examined (108 locations;
90 along U.S. coastlines outside Alaska), with an additional 0.35-m rise (<14 in) in RSL, exposure to
disruptive/damaging tidal flooding will become much more commonplace.
Figure 16 illustrates the decade when the 5-year event (20% annual chance) becomes a 0.2-year
event—a 25-fold increase in event probability—from RSL rise under the low-to-moderate subset of
the sea level scenarios (Low, Intermediate-Low, Intermediate and Intermediate-High). It is important
to note that flooding during an event can span several high tides over the course of days (e.g., Sweet et
al., 2016). Locations experiencing the most rapid increase in event probability in the coming decades
in general have higher RSL rise and/or less extreme water-level variance (i.e., a smaller scale
parameter—cf. Hunter, 2012 and Church et al., 2013a). Smaller variance in extreme event magnitudes,
not surprisingly, is found within locations with typically calm condition (e.g., Florida Coast) and/or
small storm surges (e.g., California and Pacific Island Coasts). These two parameters – magnitude of
RSL rise and extreme water-level variance (independent variables) – largely explain the pattern of
dates (dependent variable) shown in Figure 16 through a bivariate linear regression model (not shown;
R2 = 0.5, 0.6, 0.8 and 0.9 under the four scenarios shown, respectively). Considering RSL under the
Low, Intermediate-Low, Intermediate and Intermediate-High scenarios (Figure 16a–d),
disruptive/damaging tidal flooding will occur five or more times a year at most locations (90 cities)
along the U.S. coastline (outside Alaska) by or about (±5 years) 2080, 2060, 2040 and 2030,
respectively.
Figure 16. The decade (±5 years about the year shown in the legend) when the flood event with a 5-year recurrence interval
(20% annual chance event) becomes a 0.2-year recurrence interval (or annual probability increases 25-fold) under location-
specific RSL associated with the a) Low b) Intermediate-Low, c) Intermediate and d) Intermediate-High scenarios. Black dots are
locations where the 5-year event does not transition to a 0.2-year event by 2200. Note: flooding during an event can occur at high
tide for several days (e.g., Sweet et al., 2016).
39
6.3 Building for a Major Flood Event: A Case Study for South Florida
Consider a hypothetical flood risk-reduction structure needed by a community to reduce the flood risks
associated with future RSL rise and its effects on extreme water level events. This example will use
sea level scenarios for the region encompassing Key West, with probabilistic information on the
extreme water levels estimated for its NOAA tide gauge as described for Figure 15. It is not intended
to provide a detailed decision-making blueprint for a specific infrastructure facility; rather, this
example illustrates the application of sea level scenarios and related changes in extreme water levels
for decision-making. It also demonstrates how the previously described concepts of risk-based framing
may be used to assist in determining design parameters for the structure.
The Florida Keys are well known to experience impacts from hurricane storm surges and from tidal
flooding of streets and neighborhoods. Suppose that the region of Key West intends to plan, design,
and construct a seawall using a 50-year period of economic analysis starting in 2020 and, as a
consequence, the structure is expected to provide risk-reduction benefits for the period 2020 through
2070. Further suppose that the seawall is intended to protect against the 1% annual chance flood
(i.e., with a 100-year recurrence interval) over this 50-year period. It is assumed that a community in
the region would like to assume reasonably low levels of risks, and therefore, its practitioners have
opted to use higher sea level rise scenarios as design criteria. In this specific example, two scenarios
will be used as the design range. Using Table 4 as a guide, scenarios that have exceedance
probabilities significantly lower than 50% are chosen, and thus the Intermediate (1.0-m) scenario is
used as the lower bound for the design range. Based on the probabilistic analysis of Kopp et al. (2014),
the exceedance probability for the corresponding GMSL is in the range of 2% to 17%. The
Intermediate-High (1.5-m) scenario is chosen as the high end for the design range. Based on the
probabilistic analysis of Kopp et al. (2014), the exceedance probabilities for these scenarios are in the
range of 0.4% to 1.3% (Table 4). As noted previously, the GMSL exceedance probabilities for the
scenarios may be underestimates due to effects such as Antarctic ice sheet instability. The
corresponding low and high percentile RSL values (appendix C) also provide uncertainty bounds on
RSL change (approximately 10th to 90th percentile) under a particular scenario. The Extreme (2.5-m)
scenario, corresponding to about the 99.9th percentile of GMSL (see Table 4), is also selected to
represent a plausible worst-case scenario (least upper bound; Hinkel et al., 2015). This worst-case
scenario can be used to assess the project performance under conditions of extreme RSL rise that are
physically plausible but judged to have extremely low probabilities. Often, decision makers would like
information on the performance of a structure in the event of what is currently believed to be a low-
probability future, but one whose probability may be revised with additional scientific knowledge.
RSL for the scenarios applicable for the Key West location are shown in Figure 17 (solid curves).
40
Figure 17. RSL under the Intermediate (1-m), Intermediate-High (1.5-m), and Extreme (2.5-m) GMSL rise scenarios (solid
curves) for Florida Keys region, showing how the water level height with a 1% annual chance of occurring (dashed lines) and
95% confidence intervals (black error bars) estimated for year 2070 from hourly water levels at the NOAA Key West tide
gauge changes in magnitude under each scenario. All curves have been expressed in terms of the geodetic datum NAVD88
using the tidal datums at Key West.
Practical use of RSL and extreme value projections, which typically reference NOAA’s tidal datums,
require an appropriate connection to a geodetic or map datum (e.g., NAVD88 or NGVD29). The RSL
values shown in Figure 17 are referenced to NAVD88 (through its relationship to the 1983–2001 mean
sea level datum using its local ~2.4 mm/year RSL trend). As was the approximation in Figure 17, it is
assumed that the extreme value distribution is stationary except for its shifting upwards due to a
change in mean sea level as prescribed by the scenario. The change in the 1% annual chance flood in
response to increasing RSL under the various scenarios is shown in Figure 17 (dashed curves).
The uncertainty bounds on the projected 1% annual chance flood assume that RSL and the historical
extremes are independent. The total uncertainty of the projected extremes is estimated by summing the
computed variances of the individual RSL high and low percentile values (appendix C) and the
estimated return level uncertainties (e.g., Figure 15a, b). The corresponding 95% confidence interval
considering the uncertainty of both the RSL and the estimate of the extreme sea level is shown as error
bars in Figure 17. A summary of results of the above calculations are shown in Table 7.
41
Table 7. RSL with 1% annual chance flood heights and confidence intervals (CI) over the 50-year economic period of
analysis for Virginia Key, Fla. The RSL and extreme sea levels are relative to the geodetic datum NAVD88, which is 0.27
cm above local mean sea level for the 1983–2001 epoch.
Intermediate
Intermediate-High
Extreme
GMSL rise between 2000
and 2100
1.0 m
1.5 m
2.5 m
Median RSL (Key West) in
2070
0.43 m
(0.68 m rise since 2000)
0.76 m
(1.01 m rise since 2000)
1.38 m
(1.63 m rise since 2000)
1% annual chance flood
1.33 m
1.66 m
2.28 m
1% annual chance flood
(5% CI)
1.16 m
1.41 m
1.88 m
1% annual chance flood
(95% CI)
1.51 m
1.91 m
2.69 m
For the Intermediate scenario, the 1% annual chance flood (in meters NAVD88) and the corresponding
95% confidence interval is estimated to be 1.33 m (1.16 m, 1.51m), and the same for the Intermediate-
High scenarios is 1.66 m (1.41 m, 1.91 m). Depending on risk tolerance, the practitioner may choose a
value between these ranges for design. In addition, the last column provides information using the
Extreme scenario (2.5 m GMSL rise by 2100), which will allow the practitioner to assess the
consequences of a low-probability, but high-consequence scenario. For this Extreme scenario, the 1%
annual chance flood could be as high as 2.28 m (1.88 m, 2.69 m). The decision to use the 1% annual
chance flood at the end of the planning horizon (2070 in this case) will provide a higher level of risk-
reduction benefits throughout the life of the project. Alternative approaches for risk-based design
under nonstationarity are available, and they may provide design criteria that are either more or less
stringent and costly (Salas and Obeysekera, 2014; Rootzen and Katz, 2013; Obeysekera and Salas,
2016; Buchanan et al. 2016). It is important to note that that the decision to reduce risks for longer
engineering design periods or higher water levels entails additional cost. However, it is also true that,
even if changes in RSL over the chosen time frame are lower than the optimized level of risk-
reduction benefits selected, a high threshold selection can confer additional benefits, such as added
risk reduction for larger storm surges (whether due to stronger storms with climate change, or natural
variability), or risk-reduction benefits that extend beyond the period of economic analysis (just as
infrastructure often provides benefits well beyond its intended design life).
42
43
7.0 SUMMARY AND NEXT STEPS
This technical report provides results and discussion related to two primary tasks: 1) developing an
updated scenario range for possible 21st century GMSL rise and 2) producing a set of gridded RSL
response along the United States coastline based on discrete scenarios drawn from this updated GMSL
rise range. For the first task, we assessed recent observational and modeling literature on worst-case
GMSL projections. Several studies argue for physically plausible GMSL rise in the range of 2.0–2.7 m,
and recent results regarding Antarctic ice-sheet instability indicate that such outcomes may be more
likely than previously thought. As such, this report recommends a range of GMSL rise of 0.3–2.5 m
possible during the 21st century, which was previously reported as 0.2–2.0 m in Parris et al. (2012). The
upward revision to the low-end GMSL rise scenario is based upon tide gauge and altimeter-based
estimates of the rates of GMSL change over the past quarter-century and of recent modeling of future
low-end projections of GMSL rise.
From this revised range for GMSL rise, RSL is projected on a 1-degree grid (and for precise locations
of tide gauges) covering the U.S. mainland coastline, Alaska, Hawaii, the Caribbean, and the Pacific
island territories for six representative GMSL rise scenarios by 2100: a Low, Intermediate-Low,
Intermediate, Intermediate-High, High, and Extreme, which correspond to GMSL rise of 0.3, 0.5, 1.0,
1.5, 2.0 and 2.5 m, respectively. In recognition of long-term GMSL rise commitment (lagged GMSL
response), GMSL and associated RSL values are quantified from the year 2000 through the year 2200
(on a decadal basis to 2100 and with lower temporal frequency between 2100 and 2200). The GMSL
rise scenarios at each grid cell are adjusted to account for key factors important at regional scales,
including 1) shifts in oceanographic factors such as circulation patterns; 2) changes in the Earth’s
gravitational field and rotation, and the flexure of its crust and upper mantle due to melting of land-
based ice; and 3) VLM—subsidence or uplift—due to GIA (which also changes the shape of the ocean
basin and sea level), sediment compaction, groundwater and fossil fuel withdrawals, and other
nonclimatic factors. Key findings include that:
● For almost all future GMSL rise scenarios, RSL rise is projected to be greater than the global
average along the coasts of the U.S. Northeast and the western Gulf of Mexico.
● Under the Intermediate and Low GMSL rise scenarios, RSL is projected to be less than the
global average along much of the Pacific Northwest and Alaska coasts.
● Under the Intermediate-High, High and Extreme GMSL rise scenarios, RSL is projected to be
higher than the global average along almost all U.S. coasts outside Alaska.
This 1-degree gridded set of scenario-based RSL projections for the U.S. is a new data product
intended to fill a major gap in climate information needed to support a wide range of coastal
assessment, planning, and decision-making processes. The gridded set provides scenario estimates of
future RSL rise (and potential impacts) for locations within an individual grid, though it is recognized
that local VLM may spatially deviate over 1 km to 10 km in some regions. By separating the climatic
from the nonclimatic RSL contributions inherent to the RSL response under the scenarios,
practitioners can modify the nonclimatic background rates accordingly (e.g., if anthropogenic forcing
of VLM is thought to vary over the century). Systematic GMSL and RSL assessments will continue to
refine the scenarios per contemporary scientific understanding of the processes contributing to extreme
and rapid sea level change. Important to estimates of the probability of the higher-end scenarios
(though not factored into the probability estimates included in this report or yet available in the peer-
reviewed literature) are contributions from ice-cliff and ice-shelf feedback processes that may
44
significantly increase ice-sheet contributions to GMSL rise, particularly under high emissions
scenarios (DeConto and Pollard, 2016).
This report draws attention to the consequences of RSL rise, which are already occurring. For example,
tidal-flood frequencies for minor (nuisance) impacts are rapidly increasing and accelerating in dozens
of coastal communities (Ezer and Atkinson, 2014; Sweet et al., 2014; Sweet and Park, 2014). We
provide a broad assessment in terms of how continued RSL rise in the future under the sea level
scenarios may affect the frequency of more significant moderate flooding, which impacts property,
public services and commerce and is often associated with a NOAA coastal/lakeshore flood warning.
The elevation threshold used to classify such events by NOAA on their tide gauges varies along the
U.S. coastline, but in general it is about 0.8 m (2.6 feet) above the highest average tide and locally has a
20% annual chance of occurrence. Using this flood-frequency definition, we find at most locations
examined (90 cities along the U.S. coastline outside of Alaska) that with only about 0.35 m (<14 in) of
local RSL rise, annual frequencies of such disruptive/damaging flooding will increase 25-fold by or
about (±5 years) 2080, 2060, 2040 and 2030 under the Low, Intermediate-Low, Intermediate, and
Intermediate High subset of scenarios, respectively.
The GMSL and RSL products developed here will be a key input into the USGCRP Sustained
Assessment process and inform the upcoming NCA4 due in 2018. They will also support the next
steps of the Interagency Sea Level Rise and Coastal Flood Hazard Scenarios and Tools Task Force.
These next steps will focus on the integration of these global and regional scenario products within the
diversity of coastal risk management tools and capabilities deployed by individual agencies in support
of the needs of specific stakeholder groups and user communities. This deployment of scenarios and
tools will help serve as a starting point for on-the-ground coastal preparedness planning and risk
management processes needed to ensure that U.S. coastal communities (and their economies) remain
vibrant and resilient to ongoing and future changes in sea level.
45
ACKNOWLEDGEMENTS
We thank Ashley Miller of NOAA’s Center for Operational Oceanographic Products and Services
(CO-OPS) and Daniel Bader of the Center for Climate Systems Research at Columbia University for
their help in data processing. We also thank Dr. John Hall, Director, Joint Fire Science Program
(formerly with DoD SERDP/ESTCP), Dr. R. Steven Nerem of the University of Colorado, Lisa
Auermuller of the Jacques Cousteau National Estuarine Research Reserve, Dr. Marissa Liang of the
U.S. Environmental Protection Agency, and Dr. Greg Dusek, Senior Scientist of NOAA CO-OPS for
their reviews* of this manuscript and constructive comments (*a review does not necessarily indicate
agreement on all points of the final version).
We acknowledge the World Climate Research Programme’s Working Group on Coupled Modeling,
which is responsible for CMIP, and we thank the climate modeling groups (listed in appendix D) for
producing and making available their model output. For CMIP, the U.S. Department of Energy’s
Program for Climate Model Diagnosis and Intercomparison provides coordinating support and led
development of software infrastructure in partnership with the Global Organization for Earth System
Science Portals.
46
47
REFERENCES
Bamber, J. and R. Riva (2010). The Sea Level Fingerprint of Recent Ice Mass Fluxes. The Cryosphere
4:621–627.
Bamber, J. L., and W. P. Aspinall (2013). An expert judgement assessment of future sea level rise
from the ice sheets, Nat. Clim. Change, 3, 424–427, doi:10.1038/nclimate1778.
Boening, C., J. K. Willis, F. W. Landerer, R. S. Nerem, and J. Fasullo (2012). The 2011 La Niña: So
strong, the oceans fell. Geophys. Res. Lett., 39, L19602, doi:10.1029/2012GL053055.
Boon, J. D., Brubaker, J. M. & Forrest, D. R. Chesapeake Bay Land Subsidence and Sea Level
Change (2010). An Evaluation of Past and Present Trends and Future Outlook Special Report No.
425 in Applied Marine Science and Ocean Engineering (Virginia Institute of Marine Science,
2010).
Bromirski, P. D., Miller, A. J., Flick, R. E., & Auad, G. (2011). Dynamical suppression of sea level
rise along the Pacific coast of North America: Indications for imminent acceleration. Journal of
Geophysical Research: Oceans, 116(C7).
Buchanan, M.K., R.E. Kopp, M. Oppenheimer, C. Tebaldi (2016). Allowances for evolving coastal
flood risk under uncertain local sea-level rise. Climatic Change. doi:10.1007/s10584-016-1664-7
Cazenave, A., H.-B. Dieng, B. Meyssignac, K. von Schuckmann, B. Descharme, and E. Berthier
(2014). The rate of sea-level rise. Nat. Climate Change, 4, 358–361, doi:10.1038/nclimate2159.
Chambers, D. P., Cazenave, A., Champollion, N., Dieng, H., Llovel, W., Forsberg, R., ... & Wada, Y.
(2016). Evaluation of the Global Mean Sea Level Budget between 1993 and 2014. Surveys in
Geophysics, 1-19.
Church J.A., White N.J., Coleman R., Lambeck K., Mitrovica J.X. (2004) Estimates of the regional
distribution of sea-level rise over the 1950 to 2000 period. J. Clim. 17:2609–2625.
Church, J. A., & White, N. J. (2011). Sea-level rise from the late 19th to the early 21st century.
Surveys in Geophysics, 32(4-5), 585-602.
Church, J. A and coauthors (2013a). Sea level change. Chapter 13 in Climate Change 2013: The
Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA, 1535.
Church, J. A. and coauthors (2013b). Sea-level rise by 2100. Science, 342(6165), 1445.
DeConto, R. M. and Pollard, D. (2016). Contribution of Antarctica to past and future sea-level rise.
Nature, 531(7596), 591-597.
Deschamps, P., N. Durand, E. Bard, B. Hamelin, G. Camoin, A. L. Thomas, G. M. Henderson, J.
Okuno, and Y. Yokoyama (2012). Ice-sheet collapse and sea-level rise at the Bølling warming
14,600 years ago, Nature, 483, 559–564, doi:10.1038/nature10902.
Eggleston, J. and J. Pope (2013). Land subsidence and relative sea-level rise in the southern Chesapeake
Bay region: U.S. Geological Survey Circular 1392, 30 p., http://dx.doi.org/10.3133/cir1392.
Ezer, T. (2013). Sea level rise, spatially uneven and temporally unsteady: Why the US East Coast, the
global tide gauge record, and the global altimeter data show different trends. Geophysical Research
Letters, 40(20), 5439-5444.
48
Ezer, T. and L.P. Atkinson (2014). Accelerated flooding along the U.S. East Coast: on the impact of
sea-level rise, tides, storms, the Gulf Stream, and the North Atlantic Oscillations. Earth’s Future.2,
362-382. doi:10.1002/2014EF000252.
Fasullo, J. T., C. Boening, F. W. Landerer, and R. S. Nerem (2013). Australia’s unique influence on
global sea level in 2010–2011. Geophys. Res. Lett., 40, 4368–4373, doi:10.1002/grl.50834.
Firing, Y., M.A. Merrifield, T.A. Schroeder, and B. Qiu (2004). Interdecadal Sea Level Fluctuations at
Hawaii. J. Phys. Oceanogr., 34, 2514–2524, doi: 10.1175/JPO2636.1.
Galloway, D.L., D.R. Jones, and S.E. Ingebritsen (1999). Land subsidence in the United States: U.S.
Geological Survey Circular 1182, 175 p.
Goddard, P. B., J. Yin, S.M. Griffies, S. Zhang (2015). An extreme event of sea-level rise along the
Northeast coast of North America in 2009–2010. Nat. Commun. 6:6346 doi: 10.1038/ ncomms7346.
Golledge, N.R., D. E.Kowalewski, T.R. Naish, R.H. Levy, C.J. Fogwill, and E.G.W. Gasson, (2015).
The multi-millennial Antarctic commitment to future sea-level rise: Nature, v. 526, no. 7573,
p. 421-425, doi:10.1038/nature15706.
Grinsted, A., S. Jevrejeva, R.E.M. Riva, and D. Dahl-Jensen, (2015). Sea Level Rise Projections for
Northern Europe under RCP 8.5. Climate Research 64:15–23.
Gutierrez, B.T., S.J. Williams, and E.R. Thieler, (2009). Ocean coasts. In: Coastal Sensitivity to Sea-
Level Rise: A Focus on the Mid-Atlantic Region. A report by the U.S. Climate Change Science
Program and the Subcommittee on Global Change Research. [J.G. Titus (coordinating lead author),
K.E. Anderson, D.R. Cahoon, D.B. Gesch, S.K. Gill, B.T. Gutierrez, E.R. Thieler, and S.J.
Williams (lead authors)]. U.S. Environmental Protection Agency, Washington DC, pp. 43-56.
Hall, J.A., S. Gill, J. Obeysekera, W. Sweet, K. Knuuti, and J. Marburger (2016). Regional Sea Level
Scenarios for Coastal Risk Management: Managing the Uncertainty of Future Sea Level Change
and Extreme Water Levels for Department of Defense Coastal Sites Worldwide. U.S. Department
of Defense, Strategic Environmental Research and Development Program. 224 pp.
Hallegatte, S., A. Shah, R. Lempert, C. Brown, and S. Gill. (2012). Investment Decision Making
Under Deep Uncertainty – Application to Climate Change. World Bank Policy Research Working
Paper 6193. World Bank, Washington, DC, USA.
Hamlington, B. D., R. R. Leben, K.-Y. Kim, R. S. Nerem, L. P. Atkinson, and P. R. Thompson (2015).
The effect of the El Niño-Southern Oscillation on U.S. regional and coastal sea level, J. Geophys.
Res. Oceans, 120, 3970–3986, doi:10.1002/2014JC010602.
Hamlington, B. D., R. R. Leben, M. W. Strassburg, R. S. Nerem, and K.-Y. Kim (2013). Contribution
of the Pacific decadal oscillation to global mean sea level trends. Geophys. Res. Lett., 40, 5171–
5175, doi:10.1002/grl.50950.
Hamlington, B. D., S. H. Cheon, P. R. Thompson, M. A. Merrifield, R. S. Nerem, R. R. Leben, and
K.-Y. Kim (2016). An ongoing shift in Pacific Ocean sea level, J. Geophys. Res. Oceans, 121,
5084–5097, doi:10.1002/2016JC011815.
Hauer, M. E., J. M. Evans, and D R. Mishra (2016). Millions projected to be at risk from sea-level rise
in the continental United States. Nature Climate Change.
Hay, C. C., Morrow, E., Kopp, R. E., & Mitrovica, J. X. (2015). Probabilistic reanalysis of twentieth-
century sea-level rise. Nature, 517(7535), 481-484.
49
Hinkel, J., C. Jaeger, R.J. Nicholls, J. Lowe, O. Renn, and S. Peijun, (2015). Sea-level rise scenarios
and coastal risk management. Nature Climate Change, 5, 188-190.
Hoeke, R. K., K. L. McInnes, J. C. Kruger, R.J. McNaught, J. R. Hunter, and S. G. Smithers (2013),
Widespread inundation of Pacific islands triggered by distant-source wind-waves, Glob. Planet.
Chang.,108, 128–138, doi:10.1016/j.gloplacha.2013.06.006.
Horton, R., C. Herweijer, C. Rosenzweig, J. Liu, V. Gornitz and A.C. Ruane (2008). Sea level rise
projections for current generation CGCMs based on the semi-empirical method, Geophys. Res.
Lett., Vol. 35, L02715, doi: 10.1029/2007GL032486.
Horton, R.M., C. Little, V. Gornitz, D. Bader, and M. Oppenheimer (2015). Sea Level Rise and
Coastal Storms. Annals of the New York Academy of Sciences 1336 (1), 36-44.
Hunter, J. (2012). A Simple Technique for Estimating an Allowance for Uncertain Sea-Level Rise.
Clim. Chang. 113:239–252.
Intergovernmental Panel on Climate Change (2013). Summary for policymakers, in Climate Change
2013: The Physical Science Basis, edited by T. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K.
Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, and P. Midgley, pp. 3–29, Cambridge Univ. Press,
Cambridge, U. K.
Jackson, L.P. and S. Jevrejeva. (2016). A probabilistic approach to 21st century regional sea-level
projections using RCP and High-end scenarios. Global and Planetary Change, 146, pp.179-189.
Jevrejeva, S., A. Grinsted, & J.C. Moore. (2014). Upper limit for sea level projections by 2100.
Environ. Res. Lett., 9(10), 104008.
Joughin, I., B.E. Smith, & B. Medley. (2014). Marine ice sheet collapse potentially under way for the
Thwaites Glacier Basin, West Antarctica. Science, 344(6185), 735-738.
Kemp, A. C., B.P Horton, J.P. Donnelly, M.E. Mann, M. Vermeer, & S. Rahmstorf. (2011). Climate
related sea-level variations over the past two millennia. Proceedings of the National Academy of
Sciences, 108(27), 11017-11022.
Khan, N. S., E. Ashe, T.A. Shaw, M. Vacchi, J. Walker, W.R. Peltier, ... & B.P Horton. (2015).
Holocene relative sea-level changes from near-, intermediate-, and far-field locations. Current
Climate Change Reports, 1(4), 247-262.
Khan, S. A., K.H. Kjær, M. Bevis, J.L. Bamber, J. Wahr, K.K. Kjeldsen, ... & L. Liu. (2014).
Sustained mass loss of the northeast Greenland ice sheet triggered by regional warming. Nature
Climate Change, 4(4), 292-299.
King, D., D. Schrag, D. Zhou, Q. Ye, and A. Ghosh (2015). Climate Change: A Risk Assessment.
University of Cambridge Centre for Science and Policy, 154 pp.
Kirwan, M.L., Guntenspergen, G.R., D'Alpaos, A., Morris, J.T., Mudd, S.M., and Temmerman, S.
(2010). Limits on the adaptability of coastal marshes to rising sea level: Geophys. Res. Lett., v. 37,
no. 23, p. L23401.
Kopp, R. E. (2013). Does the mid-Atlantic United States sea level acceleration hot spot reflect ocean
dynamic variability?. Geophys. Res. Lett., 40(15), 3981-3985.
Kopp, R. E., C.C Hay, C.M Little, & J.X. Mitrovica. (2015). Geographic variability of sea-level
change. Current Climate Change Reports, 1(3), 192-204.
50
Kopp, R. E., R.M. Horton, C.M. Little, J.X. Mitrovica, M. Oppenheimer, D.J. Rasmussen, B. Strauss,
C. Tebaldi. (2014). Probabilistic 21st and 22nd century sea-level projections at a global network of
tide-gauge sites. Earth's Future, 2(8), 383-406.
Kopp, R. E., A.C. Kemp, K. Bittermann, B.P Horton, J.P Donnelly, W.R. Gehrels, & S. Rahmstorf.
(2016a). Temperature-driven global sea-level variability in the Common Era. Proceedings of the
National Academy of Sciences, 201517056.
Kopp, R.E., A. Broccoli, B. Horton, D. Kreeger, R. Leichenko, J.A. Miller, J.K. Miller, P. Orton, A.
Parris, D. Robinson, C.P. Weaver, M. Campo, M. Kaplan, M. Buchanan, J. Herb, L. Auermuller
and C. Andrews. (2016b). Assessing New Jersey’s Exposure to Sea-Level Rise and Coastal
Storms: Report of the New Jersey Climate Adaptation Alliance Science and Technical Advisory
Panel. Prepared for the New Jersey Climate Adaptation Alliance. New Brunswick, New Jersey.
Lempert, R. (2013). Scenarios that illuminate vulnerabilities and robust responses. Clim. Chang.,
117:627-646.
Lentz, E., E.R. Thieler, N.G. Plant, S.R. Stippa, R.M. Horton, and D.B. Gesch (2016), Evaluation of
dynamic coastal response to sea-level rise modifies inundation likelihood, Nat. Clim. Chang.
Leuliette, E. W. (2015). The balancing of the sea-level budget, Curr. Clim. Chang. Rep., 1, 3, 185–
191, doi:10.1007/s40641-015-0012-8.
Leuliette, E.W., and R.S. Nerem (2016). Contributions of Greenland and Antarctica to global and
regional sea level change. Oceanography 29(4):154–159, doi:10.5670/oceanog.2016.107
Levermann, A., P.U. Clark, B. Marzeion, G.A. Milne, D. Pollard, V. Radic, & A. Robinson. (2013).
The multimillennial sea-level commitment of global warming. Proceedings of the National
Academy of Sciences, 110(34), 13745-13750.
Marcos, M., F. M. Calafat, A. Berihuete, and S. Dangendorf (2015), Long-term variations in global
sea level extremes, J. Geophys. Res. Oceans, 120, 8115– 8134, doi:10.1002/2015JC011173.
Martín-Español, A., A. Zammit-Mangion, P.J. Clarke, T. Flament, V. Helm, M.A. King, S.B. Luthcke,
E. Petrie, F. Rémy, N. Schön. and B. Wouters (2016). Spatial and temporal Antarctic Ice Sheet
mass trends, glacio-isostatic adjustment, and surface processes from a joint inversion of satellite
altimeter, gravity, and GPS data. J. of Geophys. Res.: Earth Surface.
Marzeion, B., A. H. Jarosch, and M. Hofer (2012), Past and future sea-level change from the surface
mass balance of glaciers, Cryosphere, 6, 1295–1322, doi:10.5194/tc-6-1295-2012.
Masterson, J.P., Fienen, M.N., Thieler, E.R., Gesch, D.B., Gutierrez, B.T., and Plant, N.G. (2014).
Effects of sea-level rise on barrier island groundwater system dynamics – ecohydrological
implications: Ecohydrol., v. 7, no. 3, p. 1064-1071.
Melillo, J. M., T.C. Richmond and G. W. Yohe, Eds. (2014). Climate Change Impacts in the United
States: The Third National Climate Assessment. U.S. Global Change Research Program, 841 pp.
doi:10.7930/J0Z31WJ2.
Menéndez, P.L Woodworth (2010). Changes in extreme high water levels based on a quasi-global
tide-gauge dataset. J. Geophys. Res. 115, C10011.doi: 10.1029/2009JC005997
Mengel, M., A. Levermann, K. Frieler, A. Robinson, B. Marzeion, & R. Winkelmann. (2016). Future
sea level rise constrained by observations and long-term commitment. Proceedings of the National
Academy of Sciences, 113(10), 2597-2602.
51
Merrifield, M. A., P. Thompson, and M. Lander (2012), Multidecadal sea level anomalies and trends
in the western tropical Pacific, Geophys. Res. Lett., 39, L13602, doi:10.1029/2012GL052032
Merrifield, M.A., E. Leuliette, P. Thompson, D. Chambers, B. D. Hamlington, S. Jevrejeva, J. J.
Marra, M. Menéndez, G. T. Mitchum, R. S. Nerem, and W. Sweet (2016). [Global oceans] Sea
level variability and change [in “State of the Climate in 2015”]. Bull. Amer. Meteor. Soc.
Merrifield, M.A., P. Thompson, E. Leuliette, G. T. Mitchum, D. P. Chambers, S. Jevrejeva, R.
S. Nerem, M. Menéndez, W. Sweet, B. Hamlington, J. J. Marra (2015). [Global oceans] Sea level
variability and change [in “State of the Climate in 2014”]. Bull. Amer. Meteor. Soc.
Miller, K. G., R. E. Kopp, B. P. Horton, J. V. Browning, and A. C. Kemp (2013), A geological
perspective on sea-level rise and impacts along the U.S. mid-Atlantic coast, Earth’s Future, 1, 3–
18, doi:10.1002/2013EF000135.
Mitrovica, J. X., N. Gomez, E. Morrow, C. Hay, K. Latychev, and M. E. Tamisiea (2011). On the
robustness of predictions of sea level fingerprints, Geophys. J. Int., 187(2), 729–742,
doi:10.1111/j.1365-246X.2011.05090.x.
Nerem, R. S., D. Chambers, C. Choe, and G. T. Mitchum (2010). Estimating Mean Sea Level Change
from the TOPEX and Jason Altimeter Missions. Mar. Geod., 33, 435–446,
doi:10.1080/01490419.2010.4910.
National Research Council (2009). Informing Decisions in a Changing Climate. Panel on Strategies
and Methods for Climate-Related Decision Support, Committee on the Human Dimensions of
Global Change. Division of Behavioral and Social Sciences and Education. Washington, DC: The
National Academies Press.
National Research Council (2012). Sea-Level Rise for the Coasts of California, Oregon, and
Washington: Past, Present, and Future. Washington, D.C.: The National Academies Press. http://
www.nap.edu/openbook.php?record_id=13389
NOAA (2003). Computational Techniques for Tidal Datums Handbook. NOAA Special Publication
NOS CO-OPS 2. Silver Spring, Maryland: U.S. Department of Commerce, NOAA, National Ocean
Service, Center for Operational Oceanographic Products and Services.
NOAA (2014). National Weather Service Instruction 10-320. Surf Zone Forecast & Coastal/Lakeshore
Hazard Services. http://www.nws.noaa.gov/directives/sym/pd01003020curr.pdf
Obeysekera, J. and J.D. Salas (2016). Frequency of Recurrent Extreme Events under Nonstationarity.
J. Hydrol. Engr., pub. Online Feb. 2016, ASCE, ISSN 1084-0699.
Oppenheimer, M., and R.B. Alley (2016). How high will the seas rise? Science, 354, 1375-1377.
Park J., J. Obeysekera, J. Barnes. (2010). Temporal Energy Partitions of Florida Extreme Sea Level
Events as a function of Atlantic Multidecadal Oscillation, Ocean Sci., 6, 587-593.
Park, J., and W. Sweet. (2015). Accelerated sea level rise and Florida current transport. Ocean Sci. 11,
607–615, doi:10.5194/os-11-607-2015.
Parris, A., P. Bromirski, V. Burkett, D. Cayan, M. Culver, J. Hall, R. Horton, K. Knuuti, R. Moss, J.
Obeysekera, A. Sallenger, and J. Weiss (2012). Global Sea Level Rise Scenarios for the US
National Climate Assessment. NOAA Tech Memo OAR CPO-1. 37 pp.
Peltier, W. R. (2004). Global glacial isostasy and the surface of the ice-age Earth: the ICE-5G (VM2)
model and GRACE. Annu. Rev. Earth Planet. Sci., 32, 111-149.
52
Perrette, M., F. Landerer, R. Riva, K. Frieler, and M. Meinshausen. (2013). A Scaling Approach to
Project Regional Sea Level Rise and Its Uncertainties. Earth System Dynamics 4:11–29.
Pfeffer, W. T., Harper, J. T. and O'Neel, S. (2008). Kinematic constraints on glacier contributions to
21st-century sea-level rise. Science, 321(5894), 1340-1343.
Pollard, D., R.M. DeConto, & R.B. Alley. (2015). Potential Antarctic Ice Sheet retreat driven by
hydrofracturing and ice cliff failure. Earth Planet. Sci. Lett., 412, 112-121.
Rahmstorf, S., J.E. Box, G. Feulner, M.E. Mann, A. Robinson, S. Rutherford, and E.J. Schaffernicht,
(2015). Exceptional twentieth-century slowdown in Atlantic Ocean overturning circulation, Nat.
Clim. Chang., vol. 5, pp. 475-480.
Rahmstorf, S. (2007). A Semi-Empirical Approach to Projecting Future Sea-Level Rise, Science Vol.
315
Ranger, N., T. Reeder, and J. Lowe (2013). Addressing ‘deep’ uncertainty over long-term climate in
major infrastructure projects: four innovations of the Thames Estuary 2100 Project. European
Journal on Decision Processes, 1, 233–262. doi:10.1007/s40070-013-0014-5.
Rignot, E., Mouginot, J., Morlighem, M., Seroussi, H., & Scheuchl, B. (2014). Widespread, rapid
grounding line retreat of Pine Island, Thwaites, Smith, and Kohler glaciers, West Antarctica, from
1992 to 2011. Geophys. Res. Lett., 41(10), 3502-3509.
Rohling, E. J., I. D. Haigh, G. L. Foster, A. P. Roberts, and K. M. Grant (2013). A geological
perspective on potential future sea-level rise, Nat. Sci. Rep., 3, 3461, doi:10.1038/srep03461.
Rootzen, H. and R.W. Katz (2013). Design Life Level: Quantifying Risk in a Changing Climate.
Water Resour. Res., 49, 5964-5972, doi:10.1002/wrcr.20425
Ruggiero, P. (2013). Is the intensifying wave climate of the US Pacific northwest increasing flooding
and erosion risk faster than sea-level rise?. J. Waterway Port Coastal Ocean Eng., 139(2), 88–97.
Salas, J.D. and J. Obeysekera (2014). Revisiting the Concepts of Return Period and Risk for
Nonstationary Hydrologic Extreme Events. ASCE J. Hydrol. Engr., 19(3), 554-568.
Sato, T., C. Larsen, S. Miura, Y. Ohta, H. Fujimoto, W. Sun, R. Motyka, and J. Freymueller (2011).
Reevaluation of the viscoelastic and elastic responses to the past and present‒day ice changes in
Southeast Alaska, Tectonophysics, 511, 79–88, doi:10.1016/j.tecto.2010.05.009.
Scambos, T., and C. Shuman, (2016). Comment on ‘Mass gains of the Antarctic ice sheet exceed
losses’ by HJ Zwally and others. J. Glaciol.: 1-5.
Sella, G.F., S. Stein, T.H. Dixon, M. Craymer, S. James, S. Mazzotti. & R.K. Dokka, (2007).
Observation of glacial isostatic adjustment in ‘stable’ North America with GPS. Geophys. Res.
Lett., 34 (L02306), doi:10.10292006GL027081.
Seo, K.W., C.R. Wilson, T. Scambos, B.M. Kim, D.E. Waliser, B. Tian, B.H. Kim. and J. Eom.
(2015). Surface mass balance contributions to acceleration of Antarctic ice mass loss during 2003–
2013. J. Geophys. Res.: Solid Earth, 120(5), pp.3617-3627.
Serafin, K. A. and P. Ruggiero (2014). Simulating extreme total water levels using a time-dependent,
extreme value approach. J. Geophys. Res. Oceans, 119, doi:10.1002/2014JC010093.
Shepherd, A., E. Ivins, G. Aa, V. Barletta, M. Bentley, S. Bettadpur…H.J. Zwally. (2012). A
reconciled estimate of ice-sheet mass balance. Science, 338, 1183–1189.
53
Slangen, A. B. A., M. Carson, C.A. Katsman, R.S.W. van de Wal, A. Köhl, L.L.A. Vermeersen, &
D. Stammer. (2014). Projecting twenty-first century regional sea-level changes. Climatic Change,
124(1-2), 317-332.
Stockdon, H.F., R.A. Holman, P.A. Howd, and A.H. Sallenger. (2006). Empirical Parameterization of
Setup, Swash, and Runup. Coastal Engineering 53:573–588.
Sriver, R. L., N. M. Urban, R. Olson, and K. Keller. (2012). Toward a physically plausible upper
bound of sea-level rise projections, Clim. Chang., 115, 893–902.
Sweet, W.V., C. Zervas, S. Gill, J. Park (2013). Hurricane Sandy Inundation Probabilities Today and
Tomorrow [In “Explaining Extreme Events of 2012 from a Climatic Perspective”], Bull Amer.
Meteor. Soc. 94 (9), S17–S20.
Sweet, W. V., J. Park, J. J. Marra, C. Zervas, and S. Gill (2014), Sea level rise and nuisance flood
frequency changes around the United States, NOAA Tech. Rep. NOS CO-OPS 73, 53 pp.
Sweet, W.V. and J. Park (2014). From the extreme and the mean: Acceleration and tipping point of
coastal inundation from sea level rise. Earth Futures, 2 579-600. DOI: 10.1002/2014EF000272
Sweet, W.V., J. Park, S. Gill, and J. Marra (2015). New Ways to Measure Waves and Their Effects at
NOAA Tide Gauges: A Hawaiian-Network Perspective. Geophysical Research Letters 42:9355–
9361. doi:10.1002/2015GL066030.
Sweet, W. V. and J. J. Marra, (2016). 2015 State of U.S. Nuisance Tidal Flooding. Supplement to
State of the Climate: National Overview for May 2016, published online June 2016.
http://www.ncdc.noaa.gov/monitoring-content/sotc/national/2016/may/sweet-marra-nuisance-
flooding-2015.pdf
Sweet, W., M. Menendez, A. Genz, J. Obeysekera, J. Park, J. Marra (2016). In Tide’s Way: Southeast
Florida’s September 2015 Sunny-day Flood [in “Explaining Extremes of 2015 from a Climate
Perspective”]. Bull. Amer. Meteor. Soc., 97 (12), S25–S30, doi:10.1175/BAMS-D-16-0149.
Taylor, K. E., R. J. Stouffer, and G. A. Meehl (2012). An overview of CMIP5 and the experiment
design, Bull. Amer. Meteor. Soc., 93(4), 485–498, doi:10.1175/BAMS-D-11-00094.1.
Tebaldi, C., B. H. Strauss, and C. E. Zervas. (2012): Modeling sea level rise impacts on storm surges
along US coasts. Environ. Res. Lett., 7, 014032, doi:10.1088/1748-9326/7/1/014032.
Tedesco, M., Doherty, S., Fettweis, X., Alexander, P., Jeyaratnam, J., & Stroeve, J. (2016). The
darkening of the Greenland ice sheet: trends, drivers, and projections (1981–2100). Cryosphere
(The), 10, 477-496.
Theuerkauf, E. J., A. B. Rodriguez, S. R. Fegley, and R. A. Luettich Jr. (2014), Sea level anomalies
exacerbate beach erosion, Geophys. Res. Lett., 41, 5139–5147, doi:10.1002/2014GL060544.
USACE. (2013). Incorporating Sea Level Change in Civil Works Programs. ER 1100-2-8162.
Washington, D.C.: USACE.
USACE, (2014). Procedures to evaluate sea-level change: impacts, responses, and adaptation, U.S.
Army Corps of Engineers, Technical Letter No. 1100-2-1, 254 p.
van Vuuren, D. P., J. Edmonds, M. Kainuma, K. Riahi, A. Thomson, K. Hibbard…S. Rose. (2011).
The representative concentration pathways: An overview, Clim. Chang., 109, 5–31,
doi:10.1007/s10584-011-0148-z.
54
Wahl, T. and D.P. Chambers. (2016). Climate controls multidecadal variability in U. S. extreme sea
level records. J. Geophys. Res. 121.doi:10.1002/2015JC011057.
Weaver, C.P., R.J. Lempert, C. Brown, J.A. Hall, D. Revell, and D. Sarewitz, 2013: Improving the
contribution of climate model information to decision-making: The value and demands of robust
decision frameworks. Wiley Interdisciplinary Reviews Climate Change, 4, 39-60.
Weiss, J., J. Overpeck, and B. Strauss. (2011). Implications of recent sea level rise science for low-
elevation areas in coastal cities of the conterminous U.S.A: Clim. Chang., v. 105, no. 3, p. 635-645.
White House, (2014). President's State, Local, and Tribal Leaders Task Force on Climate Preparedness
and Resilience: Recommendations to the President. Washington, D.C., USA.
https://www.whitehouse.gov/sites/default/files/docs/task_force_report_0.pdf.
Wong, P.P., I.J. Losada, J.-P. Gattuso, J. Hinkel, A. Khattabi, K.L. McInnes, Y. Saito, and A. Sallenger,
(2014). Coastal systems and low-lying areas. In: Climate Change 2014: Impacts, Adaptation, and
Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Field, C.B., V.R. Barros,
D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C.
Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L.White
(eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp.
361-409
Woodworth, P.L. and M. Menéndez. (2015). Changes in the mesoscale variability and in extreme sea
levels over two decades as observed by satellite altimetry. J. Geophys. Res. 120, 64-77.
doi:10.1002/2014jc010363
Yin, J. and P.B. Goddard. (2013). Oceanic control of sea level rise patterns along the East Coast of the
United States. Geophys. Res. Lett., 40(20), 5514-5520.
Yin, J.J. (2012). Century to Multi-Century Sea Level Projections from CMIP5 Models. Geophys. Res.
Lett. 39:L17709. doi:10.1029/2012GL052947.
Zervas, C. (2009). Sea Level Variations of the United States 1854–2006. NOAA Technical Report
NOS CO-OPS 053, 75p, Appendices A–E.
Zervas, C., S. Gill, and W. V. Sweet (2013). Estimating vertical land motion from long-term tide
gauge records, in NOAA Tech. Rep. NOS CO-OPS 65, 22 pp.
55
LIST OF APPENDICES
Appendix A. Sea Level Rise and Coastal Flood Hazard Scenarios and Tools Interagency Task
Force
Appendix B. Climate-related RSL change for 2100 (in meters) corresponding to the low (17th) and
high (83rd) percentile values (Table 5) relative to the GMSL rise amount for that
scenario (as in Figure 9), respectively
Appendix C. Total RSL change for 2100 (in meters) corresponding low and high percentile values
(Table 5) relative to the GMSL rise amount for that scenario (as in Figure 13),
respectively
Appendix D. CMIP5 models used
56
A-1
APPENDIX A. SEA LEVEL RISE AND COASTAL FLOOD HAZARD
SCENARIOS AND TOOLS INTERAGENCY TASK FORCE
Convening organizations: The U.S. Global Change Research Program (USGCRP) and the National Ocean
Council (NOC) at the request of the White House Council on Climate Preparedness and Resilience
Participating agencies: DoD, EPA, FEMA, NASA, NOAA, USACE, USGS
Membership:
Adrienne Antoine (NOAA)
Mark Crowell (FEMA)
John Haines (USGS, Co-Chair)
John Hall (DoD, former member; now BLM)
Radley Horton (Columbia University, NASA/GISS)
Paul Huang (FEMA)
Marissa Liang (EPA)
Carolyn Lindley (NOAA)
Kris Ludwig (USGS)
Audra Luscher (NOAA)
Doug Marcy (NOAA)
Heidi Moritz (USACE)
Rob Thieler (USGS)
Chris Weaver (EPA, Co-Chair)
Kate White (USACE)
A-2
B-1
APPENDIX B. LOW AND HIGH CLIMATE-RELATED RSL CHANGE
CORRESPONDING TO GMSL SCENARIOS
The maps below show, for each GMSL scenario, the 17th and 83rd percentile of RSL change in 2100
(relative to 2000) among those samples from the probabilistic RSL distribution consistent with each
GMSL scenario. They are presented relative to the median GMSL change for the scenario, as in Figure 9.
B-2
Appendix B: continued
C-1
APPENDIX C. LOW AND HIGH TOTAL RSL CHANGE
CORRSPONDING TO GMSL SCENARIOS
The maps below show, for each GMSL scenario, a low and high representation of total RSL change in
2100 (relative to 2000) among those samples from the probabilistic RSL distribution consistent with each
GMSL scenario. They are presented relative to the median GMSL change for the scenario, as in Figure 9.
They are produced by combining the Low scenarios from appendix B with the 17th percentile estimate of
background rate, and the High scenarios from appendix B with the 83rd percentile of background rate.
They represent an approximate 10th to 90th percentile range.
C-2
Appendix C: continued
D-1
APPENDIX D. CMIP5 MODELS USED
The CMIP5 models indicated in the left set of columns were used for the thermal expansion and ocean
dynamics projections in this report. The CMIP5 models indicated in the right set of columns were used as
input by Marzeion et al. (2012) to a glacier mass-balance model. The output of this model is a key input
for the glacier projections in this report. “21” indicates use for the 21st century projections; “22” indicates
the model was also used for 22nd century projections.
Model
Thermal Expansion
and Ocean Dynamics
Glaciers and Ice Caps
RCP 8.5
RCP 4.5
RCP 2.6
RCP 8.5
RCP 4.5
RCP 2.6
access1-0
21
21
access1-3
21
21
bcc-csm1-1
22
22
22
22
22
22
bcc-csm1-1-m
21
21
21
canesm2
21
22
22
21
22
22
ccsm4
21
21
21
21
21
21
cmcc-cesm
21
cmcc-cm
21
21
cmcc-cms
21
21
cnrm-cm5
22
22
21
22
22
21
csiro-mk3-6-0
21
21
21
22
22
21
gfdl-cm3
21
22
21
21
21
21
gfdl-esm2g
21
21
21
gfdl-esm2m
21
21
21
giss-e2-r
22
22
22
22
22
giss-e2-r-cc
21
21
hadgem2-cc
21
hadgem2-es
21
22
22
22
22
inmcm4
21
21
21
21
ipsl-cm5a-lr
22
22
22
22
22
22
ipsl-cm5a-mr
21
22
21
miroc-esm
21
22
21
21
21
21
miroc-esm-chem
21
21
21
miroc5
21
21
21
mpi-esm-lr
22
22
22
22
22
22
mpi-esm-mr
21
21
21
mri-cgcm3
21
21
21
21
21
noresm1-m
21
22
21
21
22
21
noresm1-me
21
21
21
ACRONYMS
AR5
Fifth Assessment Report
AIS
Antarctic Ice Sheet
AMO
Atlantic Multidecadal Oscillation
cm
centimeters
CMIP5
Fifth Model Intercomparison Project
CSSR
Climate Science Special Project
DoD
Department of Defense
EAIS
Eastern Antarctic ice sheet
ENSO
El Niño Southern Oscillation
EPA
Environmental Protection Agency
FEMA
Federal Emergency Management Agency
FFRMS
Federal Flood Risk Management Standard
GIA
Glacial isostatic adjustment
GIS
Greenland ice sheet
GCM
Global Circulation Models
GIC
Greenland ice sheet
GMSL
Global mean sea level
GRACE
Gravity Recovery and Climate Experiment
inch
in
IPCC
Intergovernmental Panel on Climate Change
km
kilometer
m
meters
MHHW
Mean higher high water
mm
millimeter
NAO
North Atlantic Oscillation
NASA
National Aeronautics and Space Administration
NCA4
Fourth National Climate Assessment
NESDIS
National Environmental Satellite, Data, and Information Service
NOAA
National Oceanic and Atmospheric Administration
NOC
National Ocean Council
PDO
Pacific Decadal Oscillation
RCP
Representative concentration pathways
RSL
Relative sea level
SSH
Sea surface height
USACE
United States Army Corps of Engineers
USAPI
United States Affiliated Pacific Islands
USGS
United States Geological Survey
USGCRP
United States Global Change Research Program
VLM
Vertical land movement
WAIS
Western Antarctic ice sheet