Editing IPCC:AR6/SROCC/Chapter-4 (section)

=== 4.2.1 Processes of Sea Level Change ===

<div id="section-4-2-1processes-of-sea-level-change-block-1"></div>

Sea level changes have been discussed throughout the various IPCC assessment reports as SLR is a key feature of climate change. Complex interactions between the oceans and ice sheets only recently have been recognised as important drivers of processes that can lead to rapid dynamical changes in the ice sheets. Understanding of basal melt below the ice shelves, ice calving processes and glacial hydrological processes was also limited. Projections of future sea level in the IPCC 4th Assessment Report (AR4; Lemke et al., 2007 <sup>[[#fn:r4|4]]</sup> ) were presented with the caveat that dynamical ice sheet processes were not accounted for, as our physical understanding of these processes was too rudimentary and no literature could be assessed (Bindoff et al., 2007 <sup>[[#fn:r5|5]]</sup> ). In AR5 (Church et al., 2013 <sup>[[#fn:r6|6]]</sup> ), a first attempt was made to quantify the dynamic contribution of the ice sheets, although still with modeling based on limited physcis, relying mainly on an extrapolation of existing observations (Little et al., 2013 <sup>[[#fn:r7|7]]</sup> ) and a single process based case study (Bindschadler et al., 2013 <sup>[[#fn:r8|8]]</sup> ). Here the focus is on sea level changes around coastlines and low-lying islands, updating the GMSL rise by including a new estimate of the dynamic contribution of Antarctica. The mechanism driving past and contemporary sea level changes and episodic extremes of sea level is explained, and confidence in regional projections of future sea level over the 21st century and beyond is assessed.

<div id="section-4-2-1-1ice-sheets-and-ice-shelves"></div>

<span id="ice-sheets-and-ice-shelves"></span>
==== 4.2.1.1 Ice Sheets and Ice Shelves ====

<div id="section-4-2-1-1ice-sheets-and-ice-shelves-block-1"></div>

The ice sheets on Greenland and Antarctica contain most of the fresh water on the Earth’s surface. As a consequence, they have the greatest potential to cause changes in sea level. Figure 4.4 illustrates the size of land ice reservoirs and the most important processes that drive mass changes of ice sheets.

Ice sheets change sea level through the loss or gain of ice above flotation, defined as the ice thickness in exceedance of the smallest thickness that would remain in contact with the sea floor at hydrostatic equilibrium. The GIS is currently losing mass at roughly twice the pace of the AIS (Table 4.1). However, Antarctica contains eight times more ice above flotation than Greenland. Furthermore, a substantial fraction of the AIS rests on bedrock below sea level, making the ice sheet responsive to changes in ocean-driven melt and possibly vulnerable to marine ice sheet instabilities (Cross-Chapter Box 8 in Chapter 3) that can drive rapid mass loss.

Ice sheets gain or lose mass through changes in surface mass balance (SMB), the sum of accumulation and ablation controlled by atmospheric processes, the loss of ice to the ocean though melting of ice shelves, and by calving (breaking off of ice bergs) at marine-terminating ice fronts (see Chapter 3). Ice shelves, the floating extensions of grounded ice flowing into the ocean (Figure 4.4) do not directly contribute to sea level, but they play an important role in ice sheet dynamics by providing resistance to the seaward flow of the grounded ice upstream (Fürst et al., 2016 <sup>[[#fn:r9|9]]</sup> ; Reese et al., 2018b <sup>[[#fn:r10|10]]</sup> ). Ice shelves gain mass through the inflow of ice from the ice sheet, precipitation, and accretion at the ice-ocean interface. They lose mass through a combination of calving and by melting from below, especially where basal ice is in contact with warm water (Paolo et al., 2015, Khazendar et al., 2016). Calving rates at the terminus of marine terminating ice fronts are governed by complex ice-mechanical processes, the internal strength of the ice, and interaction with ocean waves and tides (Benn et al., 2007 <sup>[[#fn:r11|11]]</sup> ; Bassis, 2011 <sup>[[#fn:r12|12]]</sup> ; Massom et al., 2018 <sup>[[#fn:r13|13]]</sup> ). Sub-ice shelf melts rates are controlled by ice-ocean interactions involving the large-scale circulation, more localised heat and fresh water fluxes, and micro (mm)-scale processes in the ice-ocean boundary layer (Gayen et al., 2015 <sup>[[#fn:r14|14]]</sup> ; Dinniman et al., 2016 <sup>[[#fn:r15|15]]</sup> ; Schodlok et al., 2016 <sup>[[#fn:r16|16]]</sup> ). Ice shelves are also impacted by surface processes. Where surface melt rates are high, ice shelves not only lose mass, they can collapse (hydrofracture) from flexural stresses caused by the movement of the meltwater and the deepening of water-filled crevasses (Banwell et al., 2013 <sup>[[#fn:r17|17]]</sup> ; Macayeal and Sergienko, 2013 <sup>[[#fn:r18|18]]</sup> ; Kuipers Munneke et al., 2014 <sup>[[#fn:r19|19]]</sup> ). These complex ice-ocean interactions, calving and hydrofracture processes remain difficult to model, particularly at the scale of ice sheets.Our understanding of ice sheets has progressed substantially since AR5, although deep uncertainty (Cross-Chapter Box 5 in Chapter 1) remains with regard to their potential contribution to future SLR on time scales longer than a century under any given emissions scenario. This is particularly true for Antarctica.

<div id="section-4-2-1-2glaciers"></div>

<span id="glaciers"></span>
==== 4.2.1.2 Glaciers ====

<div id="section-4-2-1-2glaciers-block-1"></div>

Glaciers outside of the GIS and AIS are important contributors to sea level change (Figure 4.4). Because of their specific accumulation and ablation rates, which are often high compared to those of the ice sheets, they are sensitive indicators of climate change and respond quickly to changes in  climate. Over the past century, glaciers have added more mass to the ocean than the GIS and AIS combined (Gregory et al., 2013 <sup>[[#fn:r20|20]]</sup> ). However, the mass of glaciers is small by comparison, equivalent to only 0.32 ± 0.08 m mean SLR if only the fraction of ice above sea level is considered (Farinotti et al., 2019 <sup>[[#fn:r21|21]]</sup> ). Sections 2.2.3, 3.3.2 and Cross-Chapter Box 6 in Chapter 2 provide a detailed discussion of glacier response to climate change.

<div id="section-4-2-1-3ocean-processes"></div>

<span id="ocean-processes"></span>
==== 4.2.1.3 Ocean Processes ====

<div id="section-4-2-1-3ocean-processes-block-1"></div>

In general, increasing temperatures lead to a lower density (‘thermal expansion’) and therefore the larger its volume per unit of mass. Thus, warming leads to a higher sea level even when the ocean mass remains constant. Over at least the last 1500 years changes in sea level were related to global mean temperatures (Kopp et al., 2016 <sup>[[#fn:r22|22]]</sup> ), partly because of ice mass loss, and partly because of thermal expansion. Models and observations indicate that over recent decades, more than 90% of the increase in energy in the climate system has been stored in the ocean. Hence, thermal expansion provides insight into climate sensitivity (Church et al., 2013 <sup>[[#fn:r23|23]]</sup> ). Findings from sea level studies and the energy budget are consistent (Otto et al., 2013 <sup>[[#fn:r24|24]]</sup> ). As thermal expansion per degree is dependent on the temperature itself, heat uptake by a warm region has a larger impact on SLR than heat uptake by a cold region. This contributes to regional changes in sea level, which are also caused by the water temperature and salinity variations (e.g., Lowe and Gregory, 2006; Suzuki and Ishii, 2011 <sup>[[#fn:r25|25]]</sup> ; Bouttes et al., 2014 <sup>[[#fn:r26|26]]</sup> ; Saenko et al., 2015 <sup>[[#fn:r27|27]]</sup> ). Regional patterns in sea level change are also modified from the global average by oceanic and atmospheric (fluid) dynamics (Griffies and Greatbatch, 2012 <sup>[[#fn:r28|28]]</sup> ), including trends in ocean currents, redistribution of temperature and salinity (sea water density), buoyancy, and atmospheric pressure. An analysis of these trends in Coupled Model Intercomparison Project Phase 5 (CMIP5) General Circulation Models (GCMs; Yin, 2012) demonstrates the potential for &gt;15 cm of SLR by 2100 and &gt;30 cm by 2300 (RCP8.5) along the east coast of the USA and Canada from fluid dynamical processes alone. However, Coupled Model Intercomparison Project Phase 6 (CMIP6) GCM simulations are not yet available for an updated analysis of these processes in SROCC.

<div id="section-4-2-1-4terrestrial-reservoirs"></div>

<span id="terrestrial-reservoirs"></span>
==== 4.2.1.4 Terrestrial Reservoirs ====

<div id="section-4-2-1-4terrestrial-reservoirs-block-1"></div>

Global sea level changes are also affected by changes in terrestrial reservoirs of liquid water. Withdrawal of groundwater and storage of fresh water through dam construction (Chao et al., 2008 <sup>[[#fn:r29|29]]</sup> ; Fiedler and Conrad, 2010 <sup>[[#fn:r30|30]]</sup> ) in the earlier parts of the 20th century dominated, leading to sea level fall, but in recent decades, land water depletion due to domestic, agricultural and industrial usage has begun to contribute to sea level change (Wada et al., 2017 <sup>[[#fn:r31|31]]</sup> ). Changes in terrestrial reservoirs may also be related to climate variability: in particular, the El Niño Southern Oscillation (ENSO) has a strong impact on precipitation distribution and temporary storage of water on continents (Boening et al., 2012 <sup>[[#fn:r32|32]]</sup> ; Cazenave et al., 2012 <sup>[[#fn:r33|33]]</sup> ; Fasullo et al., 2013 <sup>[[#fn:r34|34]]</sup> ).

<div id="section-4-2-1-5geodynamic-processes"></div>

<span id="geodynamic-processes"></span>
==== 4.2.1.5 Geodynamic Processes ====

<div id="section-4-2-1-5geodynamic-processes-block-1"></div>

Changing distributions of water mass between land, ice and ocean reservoirs cause nearly instantaneous changes in the Earth’s gravity field and rotation, and elastic deformation of the solid Earth. These processes combine to produce spatially varying patterns of sea level change (Mitrovica et al., 2001 <sup>[[#fn:r35|35]]</sup> ; Mitrovica et al., 2011 <sup>[[#fn:r36|36]]</sup> ). For example, adjacent to an ice sheet losing mass, reduced gravitational attraction between the ice and nearby ocean causes RSL to fall, despite the rise in GMSL from the input of melt water to the ocean. The opposite effect is found far from the ice sheet, where RSL rise can be enhanced as much as 30% relative to the global average.

On time scales longer than the elastic Earth response, redistributions of water and ice cause time-dependent, visco-elastic deformation. This is observed in regions previously covered by ice during the Last Glacial Maximum (LGM), including much of Scandinavia and parts of North America (Lambeck et al., 1998 <sup>[[#fn:r37|37]]</sup> ; Peltier, 2004 <sup>[[#fn:r38|38]]</sup> ), where glacio-isostatic adjustment (GIA) is causing uplift and a lowering of RSL that continues today. In other locations proximal to the previous ice load, and where a glacial forebulge once existed, the relaxing forebulge can contribute to a relative SLR, as currently being experienced along the coastline of the northeast United States. Water being syphoned to high latitudes as the peripheral bulges collapse leads to a widespread RSL fall in equatorial regions, while the overall loading of ocean crust by melt water can cause uplift of land areas near continental margins, far from the location of previous ice loading (Mitrovica and Milne, 2003 <sup>[[#fn:r39|39]]</sup> ; Milne and Mitrovica, 2008 <sup>[[#fn:r40|40]]</sup> ). Rates of modern VLM associated with these post-glacial processes are generally on the order of a few mm yr <sup>–1</sup> or less, but can exceed 1 cm yr <sup>–1</sup> in some places. Because these gravity, rotation, and deformation (GRD) processes control spatial patterns of SLR from melting land ice, they need to be accounted for in regional-to-local sea level assessments. GRD processes are also important for marine-based ice sheets themselves, because they locally reduce RSL at retreating grounding lines which can slow and reduce retreat (Gomez et al., 2015 <sup>[[#fn:r41|41]]</sup> ; see 4.3.3.1.2 and Cross-chapter Box 8 in Chapter 3; Larour et al., 2019 <sup>[[#fn:r42|42]]</sup> ).

VLM from tectonics and dynamic topography associated with viscous mantle processes also affect spatial patterns of relative sea level change. These geological processes are important for reconstructing ancient sea levels based on geological indicators (Austermann and Mitrovica, 2015 <sup>[[#fn:r43|43]]</sup> ; see SM4.1). Along with other natural and anthropogenic processes including volcanism, compaction, and anthropogenic subsidence from ground water extraction (Section 4.2.2.4) these geodynamic processes can be locally important, producing rates of VLM comparable to or greater than recent climate-driven rates of GMSL change (Wöppelmann and Marcos, 2016 <sup>[[#fn:r44|44]]</sup> ). In this chapter, GIA and anthropogenic subsidence are used, and other components of VLM are ignored unless explicitly stated.Changing distributions of water mass between land, ice and ocean reservoirs cause nearly instantaneous changes in the Earth’s gravity field and rotation, and elastic deformation of the solid Earth. These processes combine to produce spatially varying patterns of sea level change (Mitrovica et al., 2001; Mitrovica et al., 2011). For example, adjacent to an ice sheet losing mass, reduced gravitational attraction between the ice and nearby ocean causes RSL to fall, despite the rise in GMSL from the input of melt water to the ocean. The opposite effect is found far from the ice sheet, where RSL rise can be enhanced as much as 30% relative to the global average.

<div id="section-4-2-1-6extreme-sea-level-events"></div>

<span id="extreme-sea-level-events"></span>
==== 4.2.1.6 Extreme Sea Level Events ====

<div id="section-4-2-1-6extreme-sea-level-events-block-1"></div>

Superimposed on gradual changes in RSL, as described in the previous sections, tides, storm surges, waves and other high-frequency processes (Figure 4.4) can be important. Understanding the localised impact of such processes requires detailed knowledge of bathymetry, erosion and sedimentation, as well as a good description of the temporal variability of the wind fields generating waves and storm surges. The potential for compounding effects, like storm surge and high SLR, are of particular concern as they can contribute significantly to flooding risks and extreme events (Little et al., 2015a <sup>[[#fn:r45|45]]</sup> ) . These processes can be captured by hydrodynamical models (see Section 4.2.3.4).

<span id="observed-changes-in-sea-level-past-and-present"></span>