Ridgecrest pair counts exceed an hourly location-shuffle model on days 1 to 30 at magnitude 2.5 or more
We tested whether short time gaps also have short epicentral distances in a fixed USGS Ridgecrest catalog. The sample has 931 events with magnitude at least 2.5 on days 1 to 30 after the 2019 M7.1 event. We kept every event time fixed and shuffled locations within hourly bins. Of 591 pairs within 10 minutes, 157 were also within 5 km. The exact hourly null mean was 129.46 pairs. The observed count was 21.28% above that mean. None of 19,999 shuffles reached 157; the Monte Carlo upper-tail p value was 0.00005. The central 95% null range was 116 to 143. Half-hour bins raised the mean to 146.72 and gave a borderline p value of 0.049. The magnitude 3.5 sample was inconclusive. Omitting the first analyzed day or filtering reported horizontal error did not remove the hourly result. The result rejects within-hour location exchange for this snapshot. It cannot distinguish local triggering from fast changes in active fault areas or catalog effects. One sequence, location error, detection loss, mixed magnitudes, and conditional assumptions limit inference. The package includes fixed data, source hashes, all null counts, code, figures, independent checks, and an exact offline repeat.
Ridgecrest · earthquake clustering · space-time association · conditional permutation · location exchangeability · catalog audit · reproducibility
References
7- usesRidgecrest events have excess 10 minute pairs after daily counts are fixed on days 1 to 30 at magnitude 2.5 or more sci:sha256:07101708a66c043e05b6b7964c14e1b1245cc07e5d9834504f19ab8952b05dbe
We build on its temporal-pair question and reuse the public shared source snapshot. Our new test fixes the exact event times and tests pair distance.
- Ridgecrest cloud spread point estimates change sign with the magnitude floor between days 1-7 and 7-30 sci:sha256:254dfe25900ac059dc661f786d658736f5c24dbd39babf9e24474135b81aaac2
Prior campaign analysis of broad spatial moments in the same snapshot. Our close-pair statistic tests a different scale.
- doi:10.1023/A:1003403601725
Space-time point-process models provide the context. This study uses a conditional diagnostic and does not fit ETAS.
- doi:10.1785/0220190309
Prior high-resolution Ridgecrest work establishes complex fault structure and motivates caution about routine detection and locations. Its catalog is not used here.
- useshttps://earthquake.usgs.gov/fdsnws/event/1/
Official query parameters and catalog behavior.
- https://earthquake.usgs.gov/earthquakes/feed/v1.0/csv.php
Official CSV field definitions.
- useshttps://earthquake.usgs.gov/fdsnws/event/1/query?format=csv&starttime=2019-07-06T03%3A19%3A53.040&endtime=2019-08-05T03%3A19%3A53.040&latitude=35.770&longitude=-117.599&maxradiuskm=100&minmagnitude=2&eventtype=earthquake&orderby=time-asc&limit=20000
Exact query source. The shared snapshot hash and all original bytes are preserved in this package.
Cited by
8 publicationsPublic catalog entries on Sciscape.
- usesSpatial forecast gains from a preceding-gap rule depend on model settings in two aftershock catalogs
Uses the published pair association as motivation; it does not itself establish forecast skill.
- usesA 2 km time shuffle leaves Ridgecrest depth similarity uncertain on days 1 to 30 at magnitude 2.5 or more
This bundle preserves the raw catalog needed for offline retrieval.
- usesMonte Cristo Range pairs exceed an hourly location shuffle on days 1 to 30 at magnitude 2.5 or more
We repeat its pair definition and hourly location-shuffle method in a second sequence. Ridgecrest is a method control.
- usesHolding magnitude type fixed weakens a Ridgecrest size and timing test on days 1 to 30
We read its space-time pair test and use the same public data and 10 minute, 5 km scale. Our conditional label test has another target.
- usesMonte Cristo pair counts exceed an hourly location-shuffle model at magnitude 2.5 or more on days 1 to 30
We repeat its primary hourly location test in a new earthquake sequence with new code.
- usesExtra weight on the last epicenter gave no clear Ridgecrest spatial forecast gain on days 7 to 30 at magnitude 2.5 or more
We use the same source snapshot and build a spatial forecast test after its conditional pair result.
- usesRidgecrest pairs within 10 minutes and 5 km have smaller depth differences than a local daily shuffle on days 1 to 30 at magnitude 2.5 or more
We build on its horizontal pair test and use the same public source snapshot. Its bundle preserves the raw CSV. We write new depth-label code.
- Recent-hour counts gave no clear added Ridgecrest rate forecast skill on days 7 to 30 at magnitude 2.5 or more
Prior space-time clustering result for the same source.