Rapid seaward expansion of seaport footprints worldwide

The study addresses how and to what extent major container seaports around the world have expanded their physical footprints seaward over the past three decades, and how this expansion relates to container throughput. Seaports are critical nodes in global trade yet are exposed to increasing climate-driven hazards and sea-level rise, while maritime traffic is projected to grow substantially by mid-century. Prior global and regional risk assessments have typically treated port footprints as static, overlooking dynamic, often seaward, expansion through land reclamation. The authors aim to quantify spatio-temporal patterns of seaward expansion among the world’s top container seaports (1990–2020) and evaluate how footprint growth corresponds to reported container throughput, thereby informing understanding of infrastructure exposure, operational risk, and environmental trade-offs.

Existing research highlights geographically heterogeneous exposure of ports and maritime supply chains to climate-driven natural hazards, with risk hotspots in cyclone corridors and increasing functional risk expected before 2050. Global assessments of port infrastructure risk and international trade networks have grown in sophistication, but generally assume static port footprints and do not incorporate temporal expansion. Reports on coastal land reclamation for port expansion are often geographically focused case studies. Broader literature documents environmental consequences of coastal hardening and ocean sprawl, including impacts on sedimentary habitats, biodiversity, and ecological connectivity, and raises concerns over resource demands (e.g., construction-grade sand) associated with large-scale reclamation and adaptation. Earlier efforts relating port area to handled tonnage suggested linear scaling, but the authors question the applicability of simple scaling across diverse ports and contexts.

• Site selection: Mapped the Lloyd’s List (2021) top 100 container ports by 2020 reported throughput and identified those on open-water coastlines (excluding riverine/inshore ports via OpenStreetMap land-use confirmation). This yielded 89 container seaports on open coasts.
• Remote sensing approach: Applied an open-source method (per Sengupta et al., 2023) in Google Earth Engine to quantify annual coastal land reclamation from 1990–2020 using the 30 m Global Surface Water (JRC-GSW) Yearly Water Classification History (v1.4), considering “no water” and “seasonal” bands. Seaward port expansion is detected as lost permanent water surfaces (lateral shoreline shifts) relative to a 1990 baseline coastline (1990 set to zero expansion; non-zero from 1991 onward).
• Region-of-interest delineation: Manually drew bounding polygons for each seaport by iteratively comparing OpenStreetMap industrial/terminal polygons with visual outputs from the land-reclamation analysis to encompass apparent 2020 extents. Landward polygon edges are clipped to the 1990 shoreline by the analysis.
• Inclusion threshold: Excluded 24 seaports with total seaward expansion <1 km² since 1990 to ensure a conservative sample, leaving 65 seaports for analysis.
• Data handling: No tidal correction applied due to annual composite imagery and reclamation processes designed to minimize tidal effects. No manual post-processing (e.g., pixel correction/smoothing) used for primary results; a parallel manual workflow indicated only ~1% difference in global total expansion for 2020, supporting use of automated, reproducible results. Missing annual data within time series were filled by linear interpolation.
• Variability estimation: For each seaport time series, computed mean sliding standard deviation (km²) using a three-year window after detrending each three-point sub-series; also expressed as percentage of 2020 total reclaimed area. Aggregate of all 65 mean sliding standard deviations is ~42 km² (~4% of 2020 total expansion).
• Throughput data: Compiled annual TEU throughput for 2011–2020 for 43 of the 65 seaports from archived Lloyd’s List reports to explore joint trajectories of expansion and throughput.
• Corroboration: Used Google Earth and Planet Basemap imagery and OpenStreetMap queries to verify that measured seaward growth corresponds to seaport complex expansion. The method does not differentiate among seaport land uses (terminal, storage, industry, etc.), so measured seaward footprints represent a partial gauge of gross port area.

• Extent of expansion: Since 1990, 65 of the world’s top 100 container seaports expanded seaward via reclamation by a combined 978 km². This is ~22% of the estimated current global port terminal area (~4500 km²).
• Trade coverage: In 2020, the top 100 container ports handled 632.2 million TEUs (79% of global 798.9 million TEUs). The 65 seaports analyzed handled 500 million TEUs (79% of top-100 volume; 63% of global volume).
• Regional patterns: 43 of the 65 seaports are in Asia, accounting for 871 km² (89%) of total measured expansion. Of these, 21 Chinese seaports account for 627 km² (63%). Tianjin alone reclaimed >183 km² (~18%), more than triple Singapore, the second largest expander.
• Distribution of magnitudes: About half (32/65) reclaimed <5 km². However, growth is substantial in relative terms: all but 7 at least doubled their seaward area since 1990; 39 quadrupled; 12 expanded by an order of magnitude.
• Temporal dynamics: Time series reveal diverse growth patterns, including early step-changes (asymptotic), late rapid growth (exponential), mid-period sigmoidal growth, and punctuated early-and-late growth, with some near-linear trends. Most ports show substantial growth within the past 10–15 years.
• Regional comparison beyond Asia: In Northern Europe, three seaports (3% volume share) reclaimed ~29 km² since 1990, largely Rotterdam; in the Middle East, eight seaports (5.6% volume share) collectively reclaimed ~49 km².
• Expansion–throughput relationship: Although top expanders are often top throughput ports, overall the relationship between total seaward reclamation (1990–2020) and 2020 throughput is scattered. Rank comparisons show 29 of 65 ports (45%) with outsized reclamation relative to throughput (above a 1:1 reference). For 43 ports with 2011–2020 data, many show increasing throughput with expansion, but numerous exceptions and lags indicate expansion does not guarantee immediate or sustained throughput gains.

The findings reveal a widespread, rapid pattern of seaward expansion among major seaports that transcends national policy contexts, underscoring the global drive to accommodate growing maritime trade and sometimes local land constraints. While a general positive association between expansion and throughput exists, port-specific trajectories vary considerably, reflecting market dynamics, policy, and operational factors; expansion operates as a ratchet that assumes continued or growing market share. The quantified expansion contributes to understanding of ocean sprawl and coastal hardening, with implications for marine habitats, biodiversity, and ecological connectivity. Anticipated climate adaptation needs, such as elevating port infrastructure, imply substantial demands for construction materials (notably sand), potentially intensifying environmental and resource pressures and creating competition for adaptation resources. Geopolitical initiatives (e.g., China’s Belt and Road) may shift expansion hotspots, particularly in regions like Africa where reclamation signatures are already evident. Although the analysis provides a global synthesis, vulnerability profiles and adaptation strategies remain case-specific and require integration with local logistical, policy, ecological, and hazard contexts.

This study systematically quantifies three decades (1990–2020) of seaward expansion across 65 of the world’s top container seaports using global remote sensing, revealing a combined 978 km² increase in gross port area and highly variable relationships between footprint growth and container throughput. The results highlight the ubiquity and pace of port expansion, its role in trade competitiveness, and its integration with broader issues of coastal risk, environmental impact, and resource demand for adaptation. Future research should: (1) extend analyses to other port types; (2) conduct case studies that integrate market dynamics, investment policies, climate adaptation planning, and operational sustainability; and (3) leverage advancing Earth observation capabilities to model seaports as dynamic systems of anthropogenic coastal modification and to inform targeted adaptation and risk management strategies.

• Partial footprint: The method captures only seaward expansion since 1990 and does not account for landward growth or reclamation prior to 1990; thus it provides a partial measure of gross port area and land use is not distinguished (terminal vs. storage vs. industry).
• Sample selection: Analysis limited to top 100 container ports and to 65 with ≥1 km² seaward expansion; smaller expansions and non-container ports are not covered.
• Data and processing: Automated extraction may include artifactual annual values; no manual post-processing applied to core results (though tests suggest ~1% difference in totals). Annual composites preclude tidal correction; pixel-scale uncertainties in JRC-GSW remain. Some time series have missing years (filled via linear interpolation) and nonlinearity with abrupt jumps due to major projects.
• Throughput data coverage: Joint expansion–throughput trajectories available for only 43 ports (2011–2020), limiting generalization of dynamic relationships.
• Interpretive constraints: The approach cannot resolve causal policy/market drivers, immediate operational readiness of reclaimed land (lag effects), or detailed vulnerability and adaptation needs at individual ports.