Local desk · Ground failure literacy
When strong shaking hits loose, water-logged sand, the ground can briefly behave more like a fluid than a solid. That is liquefaction — a well-documented earthquake hazard along coasts, river valleys, and young fill. This page follows USGS framing, classic case histories, and the northern Santa Clara Valley maps that include Campbell — literacy, not a parcel design study.
The U.S. Geological Survey puts it plainly: liquefaction takes place when loosely packed, water-logged sediments at or near the ground surface lose their strength in response to strong ground shaking. Beneath buildings and lifelines, that loss of strength can mean settlement, tilting, cracked pavement, and broken pipes.
Three ingredients usually line up: loose granular sediment (sand and silt), saturation from a shallow water table, and strong enough shaking for pore-water pressure to rise until grains lose contact. Then the mass can settle, flow on gentle slopes, or shove intact surface blocks sideways.
Sand boils (sand blows) erupt sand and water through cracks, leaving craterlets. Settlement follows as the layer densifies. Lateral spreading moves coherent ground toward a free face — creek bank, channel, or bay margin — over a liquefied layer. USGS FAQ and glossary materials treat these as ordinary hazard vocabulary, not scare terms.
Liquefaction is not a new discovery. Field observations and later USGS compilations tie the same pattern — sand boils, settlement, lateral spread — to cities on young alluvium and fill after large quakes.
The USGS FAQ cites the 1964 Niigata earthquake as a textbook illustration: widespread liquefaction destroyed many buildings, including the famous apartment blocks that tilted when bearing soils lost strength. Engineers and geologists still use Niigata as a reference for how loose, saturated sands under strong shaking can turn “solid” foundations into a soft failure mode.
The 1906 San Francisco earthquake produced extensive ground failure in lowland deposits across the Bay region. USGS Professional Paper 993 (Youd and Hoose, 1978) maps landslides, lateral spreads, settlement, and cracks, noting that liquefaction drove most lowland failures — especially Holocene fluvial and deltaic deposits and poorly compacted fills. Craterlets and sand boils were noted in alluvial settings regionally. Treat in-city lore carefully; the USGS record is strongest for mapped locations and geology, not every street anecdote.
The 1971 San Fernando earthquake pushed liquefaction into dam-safety practice: hydraulic fill near the Lower San Fernando Dam liquefied and helped drive a major upstream slide. The October 15, 1979 Imperial Valley earthquake (~M 6.6) produced detailed USGS sand-boil studies. Bennett, Youd, Harp, and Wieczorek (OFR 81-502) documented sand boils, cracks, and lateral spreading at Heber Road and River Park, tying ejecta to loose channel and floodplain sands via cone and SPT work — a calibration of how density, stratigraphy, and water-table depth control surface effects.
The clearest Bay Area case history is San Francisco’s Marina District in the October 17, 1989 Loma Prieta earthquake. The USGS FAQ notes that liquefaction of lagoon fill caused major subsidence, fracturing, and horizontal sliding. USGS Professional Paper 1551-F details natural sands, loose dumped fill, and 1912 hydraulic fill pumped into an artificial lagoon for the Panama-Pacific Exposition. Sand boils were most conspicuous over hydraulic fill; water, gas, and wastewater repairs concentrated in a small footprint; soft soils also amplified shaking. A teaching example of how fill history concentrates damage — not a forecast for every shoreline block.
Outside California, the 2010–2011 Canterbury sequence produced some of the best-documented urban liquefaction in recent decades. The February 2011 Christchurch earthquake drove widespread sand ejecta and lateral spreading in eastern suburbs on native soils with a shallow water table. Reminder: the hazard is global wherever loose saturated sands meet strong shaking.
Today liquefaction appears on planning maps for coastal and riverine cities on young fill, Holocene sands, and shallow groundwater. Regional scenario maps help cities and utilities screen where sand boils, cracking, or lateral spread are more likely under a given earthquake — screening tools, not designs.
In other words: use the maps to ask better questions. Do not treat a color on a PDF as a foundation report.
Marina District 1989 (PP 1551-F) remains the Bay Area’s most complete liquefaction case history: damage concentrated where hydraulic and dumped fills, a high water table, strong shaking, and soft-soil amplification lined up. For the South Bay the transferable lesson is geologic — young fill and creek-margin deposits deserve attention; denser hillsides generally do not behave the same way.
For the shaking history that produced those case studies — 1906, Loma Prieta, and the South Bay Calaveras events — see the companion Bay Area earthquakes desk.
USGS Open-File Report 2008-1270 (Holzer, Noce, and Bennett) published liquefaction hazard maps for three earthquake scenarios covering San Jose, Campbell, Cupertino, Los Altos, Los Gatos, Milpitas, Mountain View, Palo Alto, Santa Clara, Saratoga, and Sunnyvale. The companion USGS page on Northern Santa Clara Valley liquefaction hazard maps presents the same scenarios visually and repeats the regional-scale disclaimer.
Probabilities are highest for an M 7.8 San Andreas–like event comparable to 1906. For young Holocene levee deposits along major creeks, published probabilities of surface manifestations are about 0.33–0.37 with a 1.5 m water table, and about 0.10–0.14 with a 5 m water table; other units are generally below 0.05. Maps also cover M 6.7 Hayward-like and M 6.9 Calaveras scenarios — liquefaction still predicted along those levees, at lower overall probabilities. Figures from the OFR abstract; cite USGS.
Campbell sits in that mapped valley context: creek and levee deposits matter more than hillsides. Holocene corridor materials along major creeks are what the scenario maps flag; other units carry much lower published probabilities under the same assumptions. Regional context for reading maps — not a statement about any address.
Disclaimer. This essay is for general education. It is not a site assessment, geologic report, or engineering advice, and it does not replace USGS or California Geological Survey maps, local hazard disclosures, or a licensed geotechnical engineer. For property, design, or retrofit decisions, check current USGS / CGS products and hire qualified professionals. Scenario probabilities quoted here are regional published values under stated water-table assumptions — not predictions for a parcel.