8.4 Sea Level Rise

Global mean sea level has risen approximately 20 cm since 1900 and is accelerating. The current rate of ~3.6 mm/yr is driven by thermal expansion and the melting of glaciers and ice sheets. Sea level rise is not uniform due to gravitational fingerprints, ocean dynamics, and glacial isostatic adjustment (GIA).

Components of Sea Level Rise

The global mean sea level (GMSL) budget is the sum of thermosteric, barystatic, and other contributions:

$$\frac{d\eta_{\text{GMSL}}}{dt} = \underbrace{\frac{\alpha \Delta Q}{\rho c_p}}_{\text{thermal}} + \underbrace{\frac{\Delta M_{\text{ice}}}{\rho_w A_o}}_{\text{barystatic}} + \underbrace{\Delta\eta_{\text{GIA}}}_{\text{GIA}} + \underbrace{\Delta\eta_{\text{TWS}}}_{\text{land water}}$$

Thermal Expansion (Thermosteric)

~1.3 mm/yr. Water expands as it warms. Accounts for ~35% of current GMSL rise. Dominated by upper 700 m warming. Thermal expansion coefficient $\alpha$ increases with temperature and pressure.

Glacier and Ice Cap Melt

~0.7 mm/yr. ~200,000 glaciers worldwide. Total SLR potential ~0.4 m. Accelerating losses from Arctic, Alps, Andes, Himalayas. Dominated by Alaskan and Patagonian glaciers.

Greenland Ice Sheet

~0.7 mm/yr (accelerating). Contains 7.4 m GMSL equivalent. Losses from surface melting and marine-terminating glacier calving. Rate roughly quadrupled since the 1990s.

Antarctic Ice Sheet

~0.4 mm/yr. Contains 58 m GMSL equivalent. West Antarctic Ice Sheet (WAIS) losing mass through marine ice sheet instability (MISI). East Antarctic relatively stable but contains 53 m equivalent.

Observational Record

Tide Gauges (since ~1700s)

Coastal point measurements of relative sea level (includes land motion). Global reconstruction from ~1000 stations yields ~1.5 mm/yr for the 20th century average. Key historical stations: Amsterdam (since 1700), Stockholm, New York.

Satellite Altimetry (since 1993)

TOPEX/Poseidon, Jason-1/2/3, Sentinel-6 Michael Freilich. Near-global coverage (66°S–66°N). Accuracy: 2–3 cm (absolute), <1 mm/yr trend. Current rate: ~3.6 mm/yr with acceleration ~0.1 mm/yr².

$$\text{Altimetric SLR rate} \approx 3.6 \pm 0.3 \text{ mm/yr (2006-2023)}$$

Rate has approximately doubled compared to the 1900–1990 average of ~1.5 mm/yr

GMSL Projections

IPCC AR6 projects GMSL rise by 2100 relative to 1995–2014 under different Shared Socioeconomic Pathways (SSPs):

SSP1-2.6

0.32–0.62 m
(median ~0.44 m)

SSP2-4.5

0.44–0.76 m
(median ~0.56 m)

SSP5-8.5

0.63–1.01 m
(median ~0.77 m)

Low-likelihood, high-impact scenarios involving Antarctic ice sheet collapse could yield >2 m by 2100. Beyond 2100, multi-meter rise is likely under high-emission scenarios.

Sea Level Fingerprints

When an ice sheet loses mass, the gravitational attraction it exerts on surrounding ocean water decreases, causing sea level to fall near the ice sheet and rise more than the global average far away. Each ice source has a unique spatial "fingerprint":

$$\Delta\eta(\theta,\phi) = \sum_i S_i(\theta,\phi) \cdot \dot{M}_i$$

$S_i$ is the sea level fingerprint for ice source i, $\dot{M}_i$ is its mass loss rate

Greenland Melt

Sea level falls within ~2,000 km of Greenland. Maximum rise in Southern Hemisphere (~130% of global mean). US East Coast receives ~110% of GMSL rise from Greenland.

Antarctic Melt

Sea level falls near Antarctica. Maximum rise in Northern Hemisphere. US coasts receive ~110–120% of GMSL from WAIS collapse.

Glacial Isostatic Adjustment (GIA)

GIA is the ongoing response of the solid Earth to the removal of ice sheets at the end of the last ice age (~20,000 years ago). Formerly glaciated regions (Scandinavia, Hudson Bay) are still rebounding at rates up to 10 mm/yr, while peripheral regions subside:

$$\eta_{\text{relative}} = \eta_{\text{absolute}} - \dot{u}_{\text{land}} \cdot \Delta t$$

Relative sea level = absolute sea level minus land uplift

GIA must be corrected in both tide gauge records (which measure relative SL) and satellite altimetry (which measures geocentric SL but the geoid also changes). GIA corrections are ~−0.3 mm/yr for the global altimetric trend.

GPS at Tide Gauges

Co-located GPS receivers at tide gauges measure vertical land motion directly, allowing conversion from relative to absolute sea level. The Global Sea Level Observing System (GLOSS) maintains ~300 core stations. The combination of tide gauge + GPS + altimetry enables long-term (200+ yr) sea level records.

GRACE Satellite Gravimetry

GRACE (2002–2017) and GRACE-FO (2018–present) measure time-variable gravity, providing direct estimates of ice sheet mass loss. Combined with altimetry and Argo, GRACE allows the sea level budget to be closed independently for the first time.

Python: Sea Level Budget, Projections & Tide Gauge Analysis

Python

!/usr/bin/env python3

script.py106 lines

#!/usr/bin/env python3
"""sea_level_rise.py - SL budget, projections, tide gauge analysis"""
import numpy as np
import matplotlib.pyplot as plt
from numpy.polynomial import polynomial as P

# --- 1. Sea level budget ---
fig, axes = plt.subplots(2, 2, figsize=(14, 10))

years = np.arange(1993, 2024)
# Approximate contributions (mm/yr, increasing)
thermal = 1.0 + 0.02 * (years - 1993)
glaciers = 0.5 + 0.01 * (years - 1993)
greenland = 0.3 + 0.03 * (years - 1993)
antarctica = 0.1 + 0.02 * (years - 1993)
land_water = -0.1 * np.ones(len(years))

# Cumulative (integrate rates)
dt = 1.0  # year
cum_thermal = np.cumsum(thermal) * dt
cum_glaciers = np.cumsum(glaciers) * dt
cum_greenland = np.cumsum(greenland) * dt
cum_antarctica = np.cumsum(antarctica) * dt
cum_total = cum_thermal + cum_glaciers + cum_greenland + cum_antarctica

ax1 = axes[0, 0]
ax1.fill_between(years, 0, cum_thermal, alpha=0.7, label='Thermal expansion')
ax1.fill_between(years, cum_thermal, cum_thermal+cum_glaciers,
                 alpha=0.7, label='Glaciers')
ax1.fill_between(years, cum_thermal+cum_glaciers,
                 cum_thermal+cum_glaciers+cum_greenland,
                 alpha=0.7, label='Greenland')
ax1.fill_between(years, cum_thermal+cum_glaciers+cum_greenland,
                 cum_total, alpha=0.7, label='Antarctica')
ax1.plot(years, cum_total, 'k-', lw=2, label='Total')
ax1.set_xlabel('Year'); ax1.set_ylabel('Cumulative SLR (mm)')
ax1.set_title('Sea Level Rise Budget')
ax1.legend(fontsize=7, loc='upper left'); ax1.grid(True, alpha=0.3)

# --- 2. Future projections ---
ax2 = axes[0, 1]
years_proj = np.arange(2000, 2151)
t = years_proj - 2000

scenarios = {
    'SSP1-2.6': {'rate0': 3.0, 'accel': 0.01, 'color': 'green'},
    'SSP2-4.5': {'rate0': 3.2, 'accel': 0.03, 'color': 'orange'},
    'SSP5-8.5': {'rate0': 3.5, 'accel': 0.06, 'color': 'red'},
}

for name, params in scenarios.items():
    slr = params['rate0'] * t + 0.5 * params['accel'] * t**2
    slr_lo = slr * 0.7; slr_hi = slr * 1.3
    ax2.plot(years_proj, slr, color=params['color'], lw=2, label=name)
    ax2.fill_between(years_proj, slr_lo, slr_hi,
                     color=params['color'], alpha=0.15)

ax2.set_xlabel('Year'); ax2.set_ylabel('GMSL Rise (mm)')
ax2.set_title('Sea Level Rise Projections')
ax2.legend(); ax2.grid(True, alpha=0.3)
ax2.set_xlim(2000, 2150)

# --- 3. Tide gauge analysis ---
ax3 = axes[1, 0]
# Synthetic tide gauge record (monthly, 100 years)
np.random.seed(123)
t_months = np.arange(0, 100*12)
t_years_tg = t_months / 12
trend = 1.8 * t_years_tg  # mm
seasonal = 50 * np.sin(2*np.pi*t_years_tg)  # mm
nodal = 10 * np.sin(2*np.pi*t_years_tg/18.6)  # 18.6 yr nodal
noise = 30 * np.random.randn(len(t_months))
sl_tg = trend + seasonal + nodal + noise

# Deseason and fit trend
sl_annual = np.array([np.mean(sl_tg[i*12:(i+1)*12])
                      for i in range(100)])
years_tg = np.arange(100)
coeffs = np.polyfit(years_tg, sl_annual, 2)
trend_fit = np.polyval(coeffs, years_tg)

ax3.plot(years_tg + 1920, sl_annual, 'b-', alpha=0.5, lw=0.8)
ax3.plot(years_tg + 1920, trend_fit, 'r-', lw=2,
         label=f'Trend: {coeffs[1]:.2f} mm/yr')
ax3.set_xlabel('Year'); ax3.set_ylabel('Sea level (mm)')
ax3.set_title('Synthetic Tide Gauge Record')
ax3.legend(); ax3.grid(True, alpha=0.3)

# --- 4. Fingerprint illustration ---
ax4 = axes[1, 1]
lat = np.linspace(-90, 90, 180)
# Greenland fingerprint (simplified)
fp_green = 1.3 - 1.5 * np.exp(-((lat - 70)/20)**2)
# Antarctic fingerprint (simplified)
fp_ant = 1.2 - 1.5 * np.exp(-((lat + 80)/20)**2)

ax4.plot(lat, fp_green, 'b-', lw=2, label='Greenland melt')
ax4.plot(lat, fp_ant, 'r-', lw=2, label='Antarctic melt')
ax4.axhline(1.0, color='gray', ls='--', label='Global mean')
ax4.set_xlabel('Latitude'); ax4.set_ylabel('Normalized SLR')
ax4.set_title('Sea Level Fingerprints')
ax4.legend(); ax4.grid(True, alpha=0.3)

plt.tight_layout()
plt.savefig('sea_level_rise.png', dpi=150)
plt.show()

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Fortran: GIA Model & Ice Sheet Contribution

Fortran

Simple GIA (post-glacial rebound) and ice sheet SLR model

program.f9086 lines

program sea_level_model
  ! Simple GIA (post-glacial rebound) and ice sheet SLR model
  implicit none
  integer, parameter :: dp = selected_real_kind(15)
  real(dp), parameter :: pi = 3.141592653589793_dp
  integer, parameter :: nlat = 180
  real(dp) :: lat(nlat), sl_thermal(nlat), sl_greenland(nlat)
  real(dp) :: sl_antarctica(nlat), sl_gia(nlat), sl_total(nlat)
  real(dp) :: gmsl_rate, dt_projection
  real(dp) :: rate_thermal, rate_glacier, rate_gis, rate_ais
  real(dp) :: fp_gis, fp_ais, gia_rate
  real(dp) :: rho_ice, rho_water, A_ocean
  integer :: i, year

rho_ice = 917.0_dp; rho_water = 1025.0_dp
  A_ocean = 3.61e14_dp  ! m^2

! Setup latitude grid
  do i = 1, nlat
    lat(i) = -89.5_dp + real(i-1, dp)
  end do

! Current contribution rates (mm/yr)
  rate_thermal = 1.3_dp
  rate_glacier = 0.7_dp
  rate_gis = 0.7_dp      ! Greenland
  rate_ais = 0.4_dp      ! Antarctica

gmsl_rate = rate_thermal + rate_glacier + rate_gis + rate_ais

print *, '=== Sea Level Rise Model ==='
  print '(A,F6.2,A)', ' Current GMSL rate: ', gmsl_rate, ' mm/yr'
  print *, ''

! Compute spatial pattern for 100 year projection
  dt_projection = 100.0_dp  ! years

do i = 1, nlat
    ! Thermal expansion: somewhat latitude-dependent
    sl_thermal(i) = rate_thermal * dt_projection * &
                    (1.0_dp + 0.3_dp * cos(lat(i) * pi / 180.0_dp))

! Greenland fingerprint: reduced near GIS, enhanced far away
    fp_gis = 1.3_dp - 1.8_dp * exp(-((lat(i) - 70.0_dp) / 20.0_dp)**2)
    sl_greenland(i) = rate_gis * dt_projection * max(fp_gis, -0.5_dp)

! Antarctic fingerprint: reduced near AIS, enhanced far away
    fp_ais = 1.2_dp - 1.6_dp * exp(-((lat(i) + 80.0_dp) / 20.0_dp)**2)
    sl_antarctica(i) = rate_ais * dt_projection * max(fp_ais, -0.5_dp)

! GIA: uplift in formerly glaciated regions
    gia_rate = 0.0_dp
    ! Fennoscandia rebound
    if (abs(lat(i) - 65.0_dp) < 15.0_dp) then
      gia_rate = -5.0_dp * exp(-((lat(i) - 65.0_dp) / 8.0_dp)**2)
    end if
    ! Hudson Bay rebound
    if (abs(lat(i) - 55.0_dp) < 10.0_dp) then
      gia_rate = gia_rate - 8.0_dp * exp(-((lat(i)-55.0_dp)/6.0_dp)**2)
    end if
    ! Peripheral subsidence
    if (abs(lat(i) - 40.0_dp) < 10.0_dp) then
      gia_rate = gia_rate + 1.0_dp * exp(-((lat(i)-40.0_dp)/8.0_dp)**2)
    end if
    sl_gia(i) = gia_rate * dt_projection

sl_total(i) = sl_thermal(i) + sl_greenland(i) + &
                  sl_antarctica(i) + rate_glacier * dt_projection + &
                  sl_gia(i)
  end do

print *, 'Latitude  Thermal  Greenland  Antarctic  GIA     Total (mm)'
  do i = 1, nlat, 10
    print '(F7.1, 5F10.1)', lat(i), sl_thermal(i), sl_greenland(i), &
          sl_antarctica(i), sl_gia(i), sl_total(i)
  end do

print *, ''
  print '(A,F8.1,A)', ' Global mean SLR (100yr): ', &
        gmsl_rate * dt_projection, ' mm'
  print '(A,F8.1,A)', ' Maximum regional SLR:    ', &
        maxval(sl_total), ' mm'
  print '(A,F8.1,A)', ' Minimum regional SLR:    ', &
        minval(sl_total), ' mm'

end program sea_level_model

Click Run to execute the Fortran code

Code will be compiled with gfortran and executed on the server

Coastal Impacts and Adaptation

Sea level rise poses existential threats to low-lying coastal regions, small island states, and densely populated river deltas. Key impacts and responses include:

Chronic Flooding

High-tide flooding frequency is increasing exponentially at many US tide gauges. A 0.3 m rise converts rare (annual) flood events into monthly occurrences. Saltwater intrusion threatens freshwater aquifers and agricultural land.

Storm Surge Amplification

Higher baseline sea level increases storm surge heights. A 0.5 m SLR can double the frequency of current 100-year flood levels. Critical for tropical cyclone-prone coasts (Bangladesh, Gulf Coast, Philippines).

Coastal Erosion

The Bruun Rule estimates shoreline retreat as $\Delta x \approx 50-100 \times \Delta\eta$. A 1 m rise could erode beaches by 50–100 m, threatening infrastructure and ecosystems.

Adaptation Strategies

Hard protection (seawalls, levees), soft protection (beach nourishment, dune restoration), accommodation (building codes, insurance), and managed retreat. Nature-based solutions (mangroves, wetlands) increasingly favored.

Ice Sheet Instability Mechanisms

The potential for rapid, nonlinear ice sheet collapse represents the largest uncertainty in sea level projections. Two key instability mechanisms have been identified:

Marine Ice Sheet Instability (MISI)

For marine-terminating glaciers resting on beds that slope downward inland (retrograde slope), a retreat of the grounding line leads to thicker ice at the new grounding line position, increasing ice discharge and causing further retreat in a positive feedback loop:

$\text{Flux} \propto h_g^{n+1} \cdot |\nabla b|^{n-1}, \quad \text{where } h_g = \text{ice thickness at grounding line}$

MISI is believed to be already underway in the Amundsen Sea sector of West Antarctica (Thwaites and Pine Island Glaciers), where the bed slopes inland to depths >2000 m below sea level.

Marine Ice Cliff Instability (MICI)

If ice shelves (floating extensions) are lost, tall ice cliffs at the calving front may be structurally unstable if they exceed ~100 m above the waterline. The cliff fails under its own weight, exposing a taller cliff behind it in a cascading retreat. MICI could potentially contribute >1 m of SLR by 2100 from Antarctica alone, though the mechanism remains debated.

Ice shelf buttressing is the critical factor. Warm Circumpolar Deep Water (CDW, ~1°C) reaching sub-ice-shelf cavities in the Amundsen Sea drives basal melt rates of 10–40 m/yr, thinning the shelves and reducing their buttressing effect. The GRACE/GRACE-FO satellite gravity missions detect Antarctic mass loss of ~150 Gt/yr (2012–2020), equivalent to ~0.4 mm/yr of GMSL rise.

Key Concepts Summary

Current Rate

~3.6 mm/yr (satellite era average), accelerating at ~0.1 mm/yr². Has approximately doubled from the 20th century average of ~1.5 mm/yr.

Budget Closure

Thermal expansion (~35%), glaciers (~20%), Greenland (~20%), Antarctica (~10%), land water storage (~5%). Budget is now closed within uncertainties thanks to Argo and GRACE.

Fingerprints

Sea level rise is not uniform. Gravitational fingerprints cause regions far from melting ice to experience above-average rise. Ocean dynamics add further regional variability.

Projections

0.3–1.0 m by 2100 (likely range across SSP scenarios). Antarctic ice sheet instability could add more. Multi-meter rise possible beyond 2100 under high emissions.

← 8.3 Ocean Heat Content 8.5 Paleoceanography →