Land, Geotechnical, Seismic & Flood Diligence (Secondary Screen)

Land diligence is the part of site selection that nobody chooses a site for and that everybody can lose a site over. Power, water and fiber (the gates of Chapter 3.1) decide which parcels make the shortlist. The survivors face a secondary screen: can you build the building the workload requires on this dirt, in this seismic and flood environment, with clean enough control of the parcel to finance it? The answer is rarely a hard no. It is usually a yes, but: a cost, a delay or an engineering program the headline land price never included.

The consequences are lopsided. A geotechnical surprise found after you have committed the interconnection deposit does more than add a line item. It stalls the project while capital committed to the accelerators you ordered sits idle and their economic life runs down. So the working rule is: spend cheap money early to avoid expensive surprises late. Desktop screens (FEMA flood layers, USGS seismic hazard, historical-use databases, soil surveys) cost thousands. The field programs they set off cost far more, so put a price on each one before the land and interconnection money is committed. The mistakes they prevent cost tens of millions. The sections below follow the order a developer runs them: buildable land → subsurface → seismic → flood → environmental → title and control.

The first land mistake is treating deed acreage as buildable acreage. They are almost never the same number. Build a gross-to-net worksheet from the things that cannot carry a hall: required perimeter setbacks, wetland and riparian buffers, utility and access easements, floodplain you must keep clear, steep or unstable terrain, and the on-site infrastructure footprint (substation and switchyard, generator farm, fuel storage, cooling yard, water treatment, detention ponds and parking). The developable ratio, buildable area divided by gross site area, tells you whether a listing's headline acreage is real. Lay out the site first, giving buildings, utility yards and constrained land their own allocations, then check how much IT capacity that layout really supports. Gross-acreage benchmarks only serve to cross-check it.

Announced campuses show how widely the land needed per MW varies. Vantage's Frontier campus, announced on 19 August 2025, plans 1.4 GW of GPU compute on 1,200 acres: roughly 0.86 gross acres for every MW of planned IT capacity (as of 12 October 2026). Per Turner (16 July 2026), the Hyperion expansion aims for 5 GW of IT capacity; WRKF put the site at more than 3,200 acres on 14 July 2026. Crusoe's 18 March 2025 announcement counts Abilene's 1.2 GW as total facility power. Before you borrow another campus's ratio, find out what its acreage includes and whether its megawatts are IT or total power.

The ratio also depends on choices made upstream. A liquid-cooled, high-rise hall packs far more MW onto a footprint than a single-story air-cooled shed, and a campus that self-generates (gas turbines, fuel cells; Chapter 3.5) gives a large share of its acreage to the power plant. You either buy the larger parcel and bank the optionality for campus power growth and on-site generation, or buy tight and accept that you have capped the campus at today's design basis. The headroom you do not buy at land close is headroom you cannot manufacture later.

The density-and-cooling decision from Chapter 1.1 reaches all the way down into the soil. Heavier racks concentrate point loads, and that load reaches the slab. The fully loaded HPE GB200 NVL72 weighs about 1.47 tonnes (3,245 lb) with PGW coolant. Dividing that mass by HPE's 600 × 1,068 mm rack envelope gives an envelope-average screen near 470 psf. That figure is not a support reaction, and it excludes the in-row CDU, external piping and busway. The Vera Rubin NVL72 MGX reference rack is heavier still: roughly 4,000 lb (1.8 t), in NVIDIA's description of 16 March 2026. Set the structural basis from the installed mass, footprint and support reactions of the generation and OEM you actually buy (Chapter 6.2). The final check needs more from the OEM: foot and wheel reactions, support geometry, the rack's installed, shipping and rolling states, and the full load path.

Derive the hall's equivalent uniform load from the full equipment schedule and the rack layout, with aisles, CDUs, piping and move-in loads included. Check local support reactions as a separate step; the bay and load-path derivations live in Chapter 6.2. The subsurface either supports that load or it does not, and finding out late is the most expensive way to learn. Size the basis for the roadmap as well as for today's purchase. VR200 NVL72 has shipped since June 2026, and Rubin Ultra Kyber comes next: H2 2027 by NVIDIA's target, 2028 by SemiAnalysis's July 2026 report. A new hall can take its basis from rev 0.7.0 of OCP's Open Data Center design principles (effective 27 August 2026), which set 8,000 lb over a 28 × 48 in footprint as the design target for the heaviest 2027–2028 rack, roughly double a Vera Rubin rack's mass (Chapter 6.1). Few parts of the hall are harder to change at refresh than the slab. Replacing a section of slab-on-grade means moving the racks that sit on it, and structural strengthening brings its own cost and disruption (Chapter 6.2).

The geotechnical investigation answers that question in the field. A grid of soil borings and cone-penetration tests characterizes bearing capacity, settlement behavior, groundwater depth, and any expansive or collapsible soils. The killer condition is liquefaction susceptibility, where saturated loose granular soil loses strength under seismic shaking and behaves like a fluid. The findings drive the foundation fork below. Each rung up that ladder adds cost and, more importantly, schedule: a deep-foundation or aggressive ground-improvement program can add weeks to months to the critical path. Retire that surprise in desktop and early field diligence, not after the interconnection deposit is non-refundable.

The table is a cost ladder, read top-down. The top row is the site you are paying the power premium to find. Every row below it is a tax the subsurface levies on the project, and the bottom two rows (liquefaction, karst) are go/no-go gates rather than line items. The strategic move is to get the geotechnical borings into the diligence period before the land option expires, so the foundation fork is priced into the land deal and not discovered as a change order. Heavier racks raise the load the slab passes to the soil, which leaves less margin and raises the cost of a skipped early boring.

Seismic risk depends on where the site sits on the USGS seismic-hazard map and which Risk Category the facility gets under ASCE 7 / IBC. A standard commercial building is Risk Category II. A data center that is genuinely mission-critical, one whose loss endangers the operator's core obligations, is often designed to Risk Category III or IV by the owner's choice. That raises the seismic importance factor and the design forces the structure must resist, and RC IV also triggers stricter standards for nonstructural components and continued operability. The strategist makes that call, but it has deep engineering consequences: it ripples into the bracing of every rack row, the seismic anchoring of CDUs and switchgear, the flexibility of cooling-pipe joints and the snubbing of raised-access floors.

The sharpest seismic fork is base isolation versus fixed-base. A base-isolated structure sits on bearings or sliders that decouple the building from ground motion, sharply reducing the accelerations the equipment sees in a major event. The price is a meaningful capital premium and a non-trivial design effort (flexible utility connections across the isolation plane, a moat around the building). For a high-seismic site in Japan, Taiwan, coastal California or the Pacific Northwest hosting irreplaceable training infrastructure, isolation can be the difference between a facility that rides out a design-level earthquake and resumes, and one that survives structurally but loses every coolant connection and goes dark. For a low-seismic site, isolation is wasted capital. Make the isolation call at design time. Retrofitting isolation under an existing building is possible, but cutting bearings into the load path and rebuilding the utility crossings makes it a major, disruptive program (GSA, Federal Office Building monograph).

Flood is the diligence item with the most-changed rulebook in this cycle, and the one most likely to be screened lazily. The naive screen checks whether the site is in the FEMA 100-year Special Flood Hazard Area (the 1%-annual-chance floodplain). For a critical facility, that screen falls short. Which flood standard governs depends on the building code the locality has adopted, the risk category, the ASCE 7 edition and its supplements, whether ASCE 24 applies, the mapped and modelled hazards, and the project's own requirements. A 500-year (0.2%-annual-chance) flood plus freeboard, a safety margin of one to several feet above the computed flood elevation, is the prudent screen for a critical facility. Where the adopted code references ASCE 7-22 with Supplement 2, the 500-year line is also the code basis: Supplement 2 requires Risk Category II and higher structures, a typical data center among them, to use the 500-year floodplain for flood loads (ASCE, 2023). Under an earlier edition, the 500-year line is not a code minimum for every private project.

The former federal Flood Risk Management Standard (FFRMS) applied agency by agency to covered federal actions. It offered three methods: a climate-informed approach, base-flood-plus-three-feet for critical actions, or the 0.2%-annual-chance floodplain. It never imposed one 500-year-plus-freeboard rule on private siting. On a private campus, find out which code the jurisdiction has adopted, whether ASCE 24-24 applies, and what flood criteria the owner, lender or insurer impose. For a data center, where a few inches of water across the electrical room is a total loss of the hall, the rational design sits well above the regulatory minimum, not at it.

The downstream consequences are concrete and expensive. Designing to the 500-year line plus freeboard sets the finished-floor elevation, which drives sitework volume (fill, grading, retaining), ramp and dock geometry, and the elevation of every utility penetration. It also interacts with stormwater: a large impervious campus generates runoff that must be detained on-site, consuming developable acreage (back to the ratio above) and triggering its own permits (Chapter 3.9). And it interacts with insurance and lender diligence. Federal law (12 CFR 339.3 and FEMA/NFIP guidance, checked 12 October 2026) makes flood insurance mandatory when a building that secures a covered loan sits in a mapped Special Flood Hazard Area and NFIP coverage is available there. In other cases a lender or insurer can still make flood cover a contract term, and can price the exposure into premiums and loan terms. The cleanest answer is usually to site out of the floodplain entirely. Where the power forces you near it, put the cost of elevating and protecting in the land model from day one.

The greenfield-versus-brownfield fork is usually framed as a real-estate question. It is really an environmental risk-transfer question with a schedule kicker. A brownfield (a former industrial, manufacturing or power-generation site) often comes with what an AI campus needs most: existing grid interconnection or a nearby substation, heavy electrical service, water rights, rail, and a municipality that wants the jobs and offers incentives (Chapter 3.10). A retired coal or gas plant is a particularly clean fit because the grid infrastructure is already in place, and the time-to-power advantage (Chapter 3.2) can be measured in years. The interconnection rights themselves do not automatically convey with the land, though, so the surviving capacity, its transferability and the tariff path need their own diligence.

Greenfield gives you a cleaner slate: no industrial tenancy history and full design freedom. Subsurface conditions and agricultural, fill or aerially deposited contamination still have to be proven by diligence rather than assumed. The cost is building every utility from scratch and the longer permitting path of Chapter 3.9.

The catch on brownfield is the environmental tail. Industrial history means potential soil and groundwater contamination, buried tanks, asbestos, PCBs and, the modern frontier, PFAS. Since 8 July 2024, PFOA and PFOS have been CERCLA hazardous substances, so a Phase I must look for them, and legacy firefighting foam is a common source (Chapter 14.10). Under US CERCLA, an owner can inherit liability for contamination it did not create. The protection is to perform All Appropriate Inquiries (a compliant Phase I ESA) before closing. That is the threshold requirement of the bona-fide-prospective-purchaser defense, which then requires continuing obligations (reasonable steps, regulator cooperation, institutional-control compliance) for as long as you own the site. You either take the brownfield's speed-to-power and incentive upside and absorb the cost and schedule of remediation and the liability tail, or pay the greenfield premium for a clean, predictable site. The answer depends on how much the interconnection head-start is worth against a power queue measured in years, which in 2026 is often a great deal.

Environmental diligence is a staged escalation, and the staging is the point: you spend on the next tier only if the prior tier flags a reason. The Phase I ESA (ASTM E1527-21) is a non-intrusive records-and-reconnaissance review: historical use, regulatory databases, adjacent sites, visual inspection. It looks for recognized environmental conditions (RECs). On K2 Environmental's 2023 price list, the base fee for a single parcel of 20 acres or less is $2,750 rural or $3,000 urban, and the standard report arrives 15 working days after the site visit. A multi-parcel campus needs its own scope and quote. If the inquiry meets the AAI rule on scope, professional qualifications, user obligations and timing, you hold the entry ticket to the CERCLA purchaser defenses. A clean Phase I lets you move on. A REC does not forfeit the bona-fide-prospective-purchaser defense, which EPA confirms is open to buyers who know the land is contaminated. It does close off the innocent-landowner and contiguous-property-owner defenses, which both require that the buyer had no reason to know. Timing runs on two clocks. The inquiry must be completed or updated within a year of acquisition, and these components must be conducted or updated within 180 days before it: interviews with past and present owners, operators and occupants; searches for environmental cleanup liens; reviews of government records; a visual inspection of the property and adjoining properties; and the environmental professional's declaration (40 CFR 312.20(b)). The defenses then run on continuing obligations. If the Phase I flags a REC, it triggers a Phase II ESA: intrusive sampling of soil, groundwater and soil vapor to confirm or refute contamination. That is where cost and schedule open up, and where a brownfield can turn from an asset into a liability.

Beyond contamination, the parcel carries living and cultural constraints that are slower and harder to engineer around than dirt:

• Protected species and habitat. Section 7 consultation is required when a discretionary federal action, a Corps 404 permit or federal funding for example, may affect listed species or their designated critical habitat. Projects without such an action still need to check which ESA take prohibitions and state rules apply. A Section 10 incidental-take permit, backed by a habitat conservation plan, can authorize take that would otherwise be prohibited. Either route can impose seasonal work windows and mitigation measures, delaying grading or reshaping the building's envelope (FWS, checked 12 October 2026). Wetlands delineation interacts with the same Clean Water Act regime that governs the 404/401 permits of Chapter 3.9.
• Cultural and archaeological resources. A cultural-resources survey can surface historic, archaeological or tribal sites that demand consultation and, in the worst case, redesign or avoidance. If grading turns up an unanticipated find, the agreed Section 106 discovery procedure governs where that federal process applies; elsewhere, state and local rules do. Stop work in the affected area as the governing procedure requires (36 CFR 800.13).
• Wetlands, streams and buffers. These subtract directly from the developable ratio and turn into permitting long-poles. A parcel that is 30% wetland is a different deal than its acreage suggests.

None of these typically kills a power-good site. Each is a schedule risk that belongs in the diligence period, not in construction, where it becomes a stop-work order against a running depreciation clock.

The last screen is legal, and it decides whether you can finance and build on land you have otherwise proven out. A clean title commitment is table stakes; the diligence is in the exceptions. Easements (utility corridors, access, drainage, conservation) can bisect a parcel and force the building envelope into a corner, quietly subtracting from the developable ratio you computed above. Mineral rights are the under-appreciated landmine. In many US jurisdictions the mineral estate is severed from the surface estate and is dominant, so a third party can hold the legal right to access the surface to extract oil, gas or minerals beneath your campus. A severed, leasable mineral estate under a planned data hall is a financing and operational risk. Cure it before the project is bankable, by acquiring the minerals, negotiating a surface waiver or accepting drilling setbacks.

The control strategy follows from the campus geometry. Few gigawatt campuses sit on a single deed. Most require assemblage of parcels from multiple owners, which creates a holdout problem: the last seller, knowing they are the last, prices accordingly. The countermeasures are the classic ones. Option agreements lock price and parcel during diligence without committing you to purchase. Purchase-and-sale agreements with diligence contingencies tie your exit to the screens in this chapter (a geotechnical or environmental out). Quiet, staged assembly through nominees is meant to keep the opposition coalition of Chapter 3.11, and the market, from catching on. The tension is real: stealth assembly preserves price and reduces opposition, but early, transparent engagement builds a record that survives appeal and earns the social license. Which way you lean is a judgment call that Chapter 3.11 takes up directly.