In search of earthquake-related hydrologic and chemical changes along Hayward Fault
Summary
WIVVI summary (paraphrased): Researchers monitored flow and water chemistry at two springs in Alum Rock Park in eastern San Jose beginning in 1976 and at two shallow wells in eastern Oakland beginning in 1980. All sites were on or near the Hayward Fault, and the two monitoring areas were approximately 55 km apart. Measurements were generally made monthly and included temperature, electrical conductivity, water level or spring flow, together with laboratory chemical and isotopic analyses. Much of the observed variation was seasonal or associated with rainfall and drought. Several short- and long-term changes could not be readily explained by rainfall and occurred before nearby earthquakes, but the reported relationships were tentative and inconsistent. The study therefore documents possible earthquake-related hydrologic and chemical signals while also demonstrating how strongly environmental variability complicates their interpretation.
Source
Applied Geochemistry
Keywords
Hayward Fault, groundwater chemistry, hydrologic monitoring, spring flow, groundwater level, electrical conductivity, rainfall variability, earthquake precursors, chemical anomalies, long-term monitoring
Source Type
research article
Key Findings
Most temperature and chemical variation in the two wells was seasonal and largely associated with rainfall. Flow at one spring declined by approximately 40% after a multiyear California drought began in 1987. Several temporary spring-flow increases were identified, but at least two were plausibly attributable to exceptionally heavy rainfall. In 1980, the wells showed chemical changes not readily explained by rainfall that persisted for several months and were followed by a magnitude 4 earthquake approximately 37 km away. Longer-term chemical variations at one well and the springs were described as possibly correlated with several earthquakes. These temporal coincidences were suggestive but did not establish a consistent or diagnostically specific earthquake precursor.
Limitations
Sampling was generally monthly, which could miss short-lived changes and limits precise determination of onset and duration. The monitoring sites differed in hydrology and were separated by approximately 55 km, while the earthquakes varied in location and magnitude. Rainfall, seasonal recharge, and prolonged drought produced substantial changes in flow, temperature, and chemistry that could resemble or obscure seismic signals. Several proposed associations were identified retrospectively, and the small number of earthquakes and monitoring locations limited statistical validation. No predefined anomaly thresholds, prospective alarm procedure, sensitivity estimate, false-alarm rate, or independent replication was demonstrated. The study therefore cannot establish that the observed changes reliably predicted earthquakes.
Methods Summary
Flow and chemical measurements were collected approximately monthly, with more frequent observations when considered necessary, at two springs in Alum Rock Park in eastern San Jose from 1976 and at two shallow wells in eastern Oakland from 1980. Temperature, electrical conductivity, and water level or spring-flow rate were measured in the field using portable instruments. Water samples were collected for subsequent chemical and isotopic analyses in the laboratory. Time-series changes were compared with rainfall, drought conditions, and the timing and location of regional earthquakes to distinguish likely environmental variation from changes that might be related to seismic activity.