Reclamation Methods

Land reclamation is an industrial operation with a definite catalog of techniques. The methods are well understood, their costs are reasonably predictable, and their regimes of application are empirically established across fifty years of PIANC proceedings and major projects built. This document catalogs them: what each method is, what it costs, where it applies, and what the built record actually shows.

The ontology document defines reclamation as the deliberate production of new substrate where none existed. The methods below are how that substrate gets made.

Hydraulic fill: Trailing Suction Hopper Dredger

The trailing suction hopper dredger (TSHD) is the workhorse of large-scale reclamation. The vessel drags suction pipes with dragheads across the seabed while underway, loading a sand-water slurry into an onboard hopper. When full, it transits to the reclamation site and discharges.

Three discharge modes exist. Bottom dumping opens doors in the hopper floor, dropping fill to the seabed in deep water. Pump-ashore via floating pipeline delivers slurry through a flexible hose and land pipe to the placement area; required once the fill surface rises above sea level. Rainbowing shoots the slurry through a monitor nozzle at roughly 30 degrees from horizontal, projecting fill 20 to 80 m from the vessel without anchoring or pipeline. Rainbowing builds land area fast on a calm-water site and is the standard method for core construction of outer seawall berms.

Throughput and cost

The $3 to $8/m³ range assumes sandy seabed, moderate haul distance, and standard environmental conditions. Cost climbs with haul distance, water depth, and mitigation requirements.

The Maasvlakte 2 expansion at Rotterdam is the production benchmark. The PUMA joint venture (Boskalis and Van Oord) deployed up to 12 TSHDs simultaneously during peak work in spring 2010 and set a world record of 3.8 million m³ extracted and placed in a single week (Boskalis project record). Total sand placed: 240 million m³ over a five-year construction program. The project opened May 22, 2013, on schedule and within its €2.9 to €3.0 billion budget.

European TSHDs carry hoppers up to 46,000 m³. The US fleet tops out near 15,000 m³, a roughly four-to-one size gap that directly constrains throughput on domestic projects (Florez policy paper, 2025).

TSHDs operate effectively on sandy to silty seabeds at 5 to 50 m water depth. Rock or dense clay requires pre-loosening by cutterhead before a draghead can load.

Hydraulic fill: Cutter Suction Dredger

The cutter suction dredger (CSD) uses a rotating cutterhead to disaggregate in-situ material while a centrifugal pump sends slurry through a floating pipeline directly to the fill area. The vessel is stationary or spud-mounted. Material moves continuously rather than in batches, which suits enclosed bays and hard substrates that a TSHD draghead cannot penetrate.

Throughput and cost

The higher unit cost reflects pipeline infrastructure and lower transport efficiency relative to TSHDs on long hauls. For short hauls in protected water, the CSD is competitive.

China’s Spratly Islands campaign from 2013 to 2016 is the most intensively studied CSD application in the open literature. The primary vessel was the Tian Jing Hao: 127.5 m length, 4,500 m³/hr capacity, 30 m maximum dredge depth, 6 km pump range (DredgePoint technical database). AIS tracking confirmed the vessel operated at five reefs across a 193-day operational period between September 2013 and June 2014 (IHS Jane’s, as cited in CRS Report R44072, 2015). Approximately 40 to 50 million m³ of carbonate fill was placed across 12.9 km² of new land at seven reef features over roughly 30 months.

China’s successor vessel, the Tian Kun Hao, commissioned June 2018, is the world’s largest self-propelled CSD: 140 m length, 6,000 m³/hr capacity, 35 m maximum dredge depth, 15 km pump range (GlobalSecurity.org). It arrived after the Spratly campaign proper and represents the direction of Chinese industrial capacity.

For the original Chek Lap Kok platform at Hong Kong International Airport, a combined fleet of CSDs and TSHDs from Dutch, Belgian, Australian, Japanese, British, and American contractors placed approximately 250 million m³ of marine fill between December 1992 and January 1996, plus another 90 million m³ of island rock excavation, for a total earthmoving volume of 347 million m³. Peak placement ran approximately 10 tonnes per second (Zhen Hua Engineering project record). The timeline was driven by the 1997 handover political deadline: dredged reclamation was chosen over polder construction specifically because it could be completed faster.

Polder-and-dike

The polder method builds land by encircling a water body with embankments, then pumping the enclosed area dry. It is the Dutch invention and the Netherlands remains its canonical application. No open-ocean site can use this method; it requires an enclosed or semi-enclosed shallow water body.

Construction sequence: erect enclosing dikes in stages using sand-filled cores with clay and riprap armoring; install pumping stations before final dike closure; seal the final gap by high-speed sand sluicing against tidal currents; pump the enclosed area dry over months; drain and leach salts from the marine clay; allow consolidation; begin development.

Dutch record

The Zuiderzee Works converted 1,650 km² of tidal inland sea to dry land across four major polders:

The Afsluitdijk, 32 km of enclosing dam completed 1932 across the Zuiderzee mouth, created the IJsselmeer freshwater lake that made the polders possible. Twenty-six percent of the Netherlands currently lies below mean sea level, sustained by more than 3,000 polders and continuous pumping from approximately 3,300 pumping stations.

Modern polder-and-dike construction for new land runs approximately $50 to $200/m² of finished land, varying strongly with the ratio of dike perimeter to enclosed area. Sand demand is roughly 40% lower than equivalent open hydraulic fill, because the enclosed seabed itself becomes the land surface once drained rather than requiring fill to cover it.

The method is optimal for silty-clay seabeds in shallow, sheltered water. Once drained, marine clay consolidates into productive agricultural or urban substrate, though the process takes years and the land requires permanent drainage infrastructure.

Caisson and quay wall systems

Where new land abuts existing shoreline or sits in protected water, structural retaining walls define the fill perimeter and allow rapid placement. The caisson is the central element.

A gravity caisson is a prefabricated or cast-in-place concrete box floated to position, sunk onto a prepared granular bed, and ballasted with sand. It forms harbor quay faces and reclamation boundaries in water too deep for earthen dikes. Sheet-pile cofferdams use interlocking steel piles driven to form a watertight perimeter cell, backfilled with granular material. Both systems are routine in container port construction.

Caissons are not a fill method. They are boundary-forming structures that allow hydraulic fill to be placed within a defined perimeter at depth. The combination appears at every major port terminal built in the last four decades: gravity caissons or sheet-pile cells form the quay face, and fill is placed behind them to create the yard.

KIX Island I used 69 steel caisson chambers, each 22 m tall and 22 m in diameter, weighing approximately 200 tonnes apiece, sunk to the bay floor to form the outer seawall. Gaps between caissons were filled with rock and 48,000 concrete tetrapods. Three mountains were excavated to provide approximately 21 million m³ of rock for the seawall (Mesri and Funk 2015, ASCE JGGE; Kansai Airports technical page). The seawall was completed in 1989, three years before fill placement began.

Maasvlakte 2’s hard seawall uses a different structural logic: 3.5 km of concrete armor-block dam on the northwest face, where wave exposure is highest. The 20,000 concrete blocks weigh approximately 40 tonnes each and were placed by a custom 50-tonne E-crane to 15 cm positioning accuracy (Boskalis project record). Seven million tonnes of rock, mostly from Norwegian quarries, form the core.

The quay wall system scales from single berth to kilometer-scale port perimeter. Its design standard is governed by the Port of Rotterdam’s Maasvlakte 2 basin at minus 20 m NAP, sized to accommodate the largest post-Panamax container vessels afloat.

Sand motor and mega-nourishment

The sand motor places an oversized volume of fill at a single concentrated location and relies on natural wave and current processes to redistribute it along the coast over years to decades. The logic is economic and ecological: one large placement is cheaper and less disruptive than repeated small ones, and the sand spreads in ways that benefit habitat and recreation as well as coastal protection.

The Zandmotor at Ter Heijde, Netherlands, is the canonical pilot. Construction ran January to October 2011. Final volumes: 17 million m³ in the main peninsula body plus 4.5 million m³ in flanking shoreface nourishments, totaling 21.5 million m³. The peninsula protrudes 1 km offshore and extends 2 km along the coast, reaching up to 5 m above mean sea level. Cost: €70 million (Deltares/EcoShape Sand Motor project page).

The Delfland coast had previously been nourished at conventional rates of roughly 2 to 5 million m³ per five-year cycle. The Zandmotor concentrated the equivalent of 20 years of nourishment in one event. The 10-year evaluation in 2021 declared the project successful for coastal protection, habitat creation, and recreation. In the first 17 months, 1.8 million m³ eroded from the peninsula as sand distributed northward into dune fields and along the coast.

Throughput and cost

Cost per cubic meter at Zandmotor ran approximately €3.25/m³ ($3 to $5/m³ range), competitive with TSHD hydraulic fill. The savings come from logistics: one mobilization, one borrow site, one monitoring program.

The sand motor method requires a wave climate capable of redistributing fill along the coast on a useful time scale. It is a soft method in the engineering-with-nature tradition and does not produce permanent hard perimeters. It is suited to open sandy coasts with reasonably uniform longshore drift. For the sediment budget accounting framework that determines whether placed sand stays in a given drift cell, see coastal-morphodynamics.

Thin-layer placement

Thin-layer placement (TLP) deposits dredged material at 5 to 30 cm thickness over existing wetland, marsh, or tidal flat surfaces. The purpose is not new land creation but elevation maintenance: counteracting subsidence to keep existing coastal territory above rising sea level.

USACE’s beneficial use program integrates TLP into maintenance dredging operations. Rather than disposing material in confined disposal facilities, it is placed on marshes or barrier island interiors that have lost elevation. The USACE target is 70% beneficial reuse of the approximately 210 million cubic yards dredged annually; the current rate is 30 to 40% (Florez, 2025).

The Poplar Island project in the Chesapeake Bay restored 170 acres of wetland using approximately 1 million m³ per year from Baltimore Harbor maintenance dredging. Vegetation tolerates placement at 5 to 30 cm; thicker lifts applied too quickly smother plants.

Cost

TLP is the lowest-cost fill method per cubic meter placed. Its constraint is scale: it is a maintenance and restoration operation, not a land-creation operation, and throughput is limited by the area that can receive material without ecological damage.

Louisiana provides the stakes. The state has lost approximately 4,900 km² of coastal land since 1932. CPRA has placed 193 million cubic yards of sediment through restoration programs since 2005, benefiting 55,807 acres and restoring 71.6 miles of barrier islands (CPRA Master Plan data). Subsidence rates of 5 to 25 mm/yr in Louisiana marshes mean that without continuous TLP and nourishment, restored areas resubmerge within years.

Geotextile tubes and geocells

Geotextile tubes are large cylindrical fabric bags (woven polypropylene, high-strength) hydraulically filled with sand-water slurry. The geotextile retains sand while filtering water. Once dewatered, tubes are rigid, erosion-resistant structures 1 to 5 m in diameter and 10 to 100 m in length.

Applications include breakwaters, groins, revetments, and temporary reclamation perimeters. Tubes are competitive where rock armor is unavailable, expensive to transport, or where a soft lightweight structure is preferred.

Cost

The cost advantage over rock is three to four times on a per-meter basis, though exposed installations degrade faster due to UV aging and abrasion. Lifespan runs 10 to 20 years for buried or submerged applications, shorter for exposed coastal installations.

Performance at moderate wave heights (significant wave height below 2.5 m) is comparable to rock armor of equivalent cross-section. On exposed coasts with larger waves, geotextile tubes are not a primary defense. They serve as perimeter-formers and secondary protection in sheltered conditions, and as dune cores in beach nourishment projects where excavation for rock placement is impractical.

Deep Cement Mixing

Deep cement mixing (DCM) injects cement slurry into soft in-situ soil using rotating mixing tools, creating columns of cement-treated ground 0.8 to 1.5 m in diameter at depths of 20 to 50 m. Overlapping column clusters create a continuous treated zone with improved bearing capacity, reduced settlement, and, critically, containment of contaminated sediment in place.

DCM serves three distinct functions in reclamation:

1. Stabilizing contaminated soft mud in-place without the regulatory and environmental cost of dredging it out.
2. Reducing primary settlement of reclaimed fill over soft marine clay.
3. Providing bearing capacity for heavy structures where PVD drainage alone is insufficient.

HKIA Third Runway System

The Hong Kong International Airport Third Runway System (3RS) executed the largest single DCM application in history. The 650 ha reclamation footprint contained Contaminated Mud Pits (CMPs) from prior reclamation projects carrying heavy metals and persistent organic pollutants. Conventional dredging would have mobilized these contaminants. The EIA (AEIAR-185/2014) mandated DCM as the only acceptable ground improvement method over CMP areas.

Samsung C&T and Build King JV installed over 250,000 DCM cluster units across 6 contract areas using up to 62 DCM rigs simultaneously between 2016 and 2019 (AAHK project records; Yin et al. 2022, ASCE JGGE). Treatment depth was approximately 20 m below seabed. Target unconfined compressive strength at 28 days: 800 kPa. DCM achieved geological stability 0.08 to 1.39 years faster than the PVD-only alternative and reduced primary settlement to 29 to 83% of PVD-only settlement.

The 3RS total budget is HK$141.5 billion at 2015 prices (approximately USD 18.3 billion). The third runway commissioned November 25, 2022; full three-runway operations commenced November 28, 2024. Design capacity: 120 million passengers per year and 10 million tonnes of cargo per year (AAHK official data).

Cost

DCM is the most expensive of the ground improvement methods on a per-cubic-meter basis. Its use is justified when contaminated sediment cannot be removed, when bearing capacity requirements for runway or heavy structure loading exceed what drainage alone provides, or when settlement must be reduced faster than drainage allows.

Design references: Kitazume and Terashi (2013), The Deep Mixing Method, CRC Press; FHWA (2013) US standard; Bergado et al. (1996) soft clay behavior.

Prefabricated Vertical Drains

Prefabricated vertical drains (PVDs) are thin geosynthetic strips, typically 100 mm wide and 4 mm thick, installed by mandrel into soft clay on grids of 1 to 2.5 m spacing at depths of 20 to 40 m. They shorten drainage paths from tens of meters (natural consolidation) to roughly 0.5 m (drain spacing), accelerating primary consolidation by factors of 10 to 100. After installation, a sand blanket and preload surcharge drive settlement; the preload is removed or maintained based on measured settlement progress.

Kansai International Airport

KIX Island I installed approximately 1 million sand drains, each 40 cm in diameter, on a 2.5 m by 2.5 m grid through the Holocene Ma13 clay. The 1 million drain count exceeded the total installed in Japan over the preceding 35 years (Mesri and Funk 2015). Island II used 1.2 million PVDs plus deep mixing and marine plastic drains. The Holocene clay layer settled essentially to completion within one year of fill loading on both islands, as designed.

What PVDs cannot treat is the Pleistocene clay, which extends to 400 m depth beneath Osaka Bay. No surface drain reaches it. Island I has settled 13.66 m cumulatively as of December 2024 against an original optimistic prediction of 5.7 m. Island II, constructed with better geotechnical knowledge, has settled 17.47 m, more than Island I, because its deeper and softer Pleistocene clay continues compressing in ways that do not decelerate as standard consolidation theory predicts (Mesri and Funk 2015; Kansai Airports settlement monitoring data, December 2024). Terminal buildings sit on 900 adjustable jack columns to compensate for ongoing differential settlement.

The total KIX investment reached approximately US$20 billion by 2008 (Airways Magazine). Accumulated debt exceeded 1 trillion yen by 2015, leading to privatization via a 44-year Vinci/Orix concession in 2016.

Cost

PVDs are inexpensive relative to DCM and effective where soft clay is thick but not contaminated, where the loading requirements are met by improved drainage alone, and where the project timeline accommodates months to years of monitored consolidation.

Sand Compaction Piles

Sand compaction piles (SCP) install dense compacted sand columns (0.4 to 0.7 m diameter) into soft clay or loose sand by vibration or static excitation. Unlike PVDs, which provide drainage only, SCPs structurally reinforce the ground through replacement and stress concentration effects. In sandy soils, SCP densifies loose fill to prevent liquefaction. In clay, it creates composite ground stiffness proportional to the area replacement ratio.

SCP is a Japanese method developed by Murayama in 1957 and documented comprehensively by Kitazume (2005), The Sand Compaction Pile Method, Routledge.

Japanese applications

KIX used SCP in combination with sand drains for seawall foundation improvement on the outer margins of both islands. Haneda Airport D-runway used SCP to treat the soft Tokyo Bay clay beneath the reclamation section seawall. The Haneda runway is a hybrid structure: two-thirds conventional hydraulic fill (52 million m³ total, including approximately 620,000 m³ of cement-treated lightweight fill from Tokyo Port dredge spoil) and one-third pile-supported steel-jacket platform over the Tama River mouth, where flood conveyance requirements precluded full fill (Watanabe et al., JGSSP, Vol. 2). The jacket structure comprises 198 prefabricated steel units on 1,165 piles. The runway opened October 21, 2010, in 3.5 years of construction.

Kobe Port Island, built 1966 to 1981, used hydraulic fill of Masado (decomposed granite from the Rokko Mountains) without SCP liquefaction treatment. The 1995 Hanshin earthquake produced pervasive liquefaction: 30 to 40 cm of settlement, lateral quay wall displacements averaging 2 to 3 m, and damage to all but 6 of approximately 187 berths. Total damage was approximately 1 trillion yen. Kobe fell from sixth to seventeenth globally as a container port and never recovered its pre-earthquake competitive position (GEER Association 1995 report; Hamada et al. 1996, ASCE JGGE). Post-1995 Japanese standards mandate explicit liquefaction treatment for offshore reclamation; all subsequent major projects, including Chubu Centrair and Haneda, incorporated updated SCP requirements.

Cost

Cost and throughput summary

Method selection

The choice of technique is not primarily a cost optimization. It is governed by site geology, substrate contamination, water depth, available fill sources, required timeline, and environmental constraints. The cases above make the governing logic clear.

Water depth and fill source access. TSHDs operate at 5 to 50 m water depth on sandy seabeds and are the default choice where offshore borrow areas exist within reasonable haul distance. CSDs handle harder substrates and enclosed sites at shorter haul distances. Where the borrow area is the site itself (as in reef-flat extraction), CSD is the only option.

Substrate contamination. The HKIA 3RS decision is the definitive case. Contaminated mud pits in the reclamation footprint made conventional dredging legally and environmentally indefensible; DCM was mandated by the EIA. Any project with legacy contaminated sediment in its footprint will face a version of this constraint. It is worth knowing before selecting the site.

Geology and settlement. Holocene soft clay is treatable by PVD and SCP; primary consolidation can be completed during construction. Pleistocene clay is not treatable by any current method; it compresses slowly and unpredictably for decades. KIX demonstrates the consequence: a 5.7 m settlement prediction that became 13.66 m of actual settlement and counting. Projects built over thick Pleistocene clay are committing to permanent settlement management.

Seismic exposure. Any reclamation in a seismically active region requires explicit liquefaction analysis. Kobe Port Island established that well-graded coarse fill is not immune to liquefaction under extreme shaking. Post-1995 Japanese standards and modern US practice treat liquefaction mitigation as baseline, not optional.

Timeline constraint. The HKIA original platform was built on dredged fill rather than by polder method specifically because dredging could meet a political deadline. The Maasvlakte 2 team ran 12 TSHDs simultaneously to achieve peak weekly throughput. Timeline drives fleet size, and fleet size drives cost. A project that needs speed pays for it in mobilization and simultaneous vessel deployment.

Enclosed versus exposed coast. Polder-and-dike requires a sheltered, shallow water body with a small perimeter-to-area ratio. Sand motor mega-nourishment requires an open coast with wave energy sufficient to redistribute fill. Caisson and quay wall systems work best in protected harbor settings. These are not interchangeable.

Scale and sediment budget. Any project that materially alters the local littoral drift cell will have downstream effects on adjacent shorelines and inlets. Large fill operations that interrupt longshore drift cause predictable downdrift erosion. Tidal prism reductions destabilize inlet cross-sections per the O’Brien relation. The morphodynamic framework that governs these interactions is in coastal-morphodynamics.

Precedents catalogued here

The global-precedents document traces these methods in their built-out forms: Maasvlakte 2 as the TSHD production reference, KIX as the PVD and settlement reference, Chek Lap Kok and 3RS as the earthmoving volume and DCM references, the Spratly campaign as the CSD throughput reference, and the Zuiderzee Works as the polder lineage. The engineering-with-nature document covers the soft-engineering alternatives, including living shorelines, oyster reef structures, and marsh restoration, which sit alongside thin-layer placement at the lower-energy end of the technique spectrum.

The methods above are what the industry actually builds with. Their unit costs and throughput envelopes are well-established. The governing constraint in most US contexts is not technical capacity but permitting throughput, fleet size relative to European competition, and sediment source access. Those constraints are in industrial-base and institutions-and-permitting.