Comparison of a nearest neighbor and other approaches to the detection of space-time clustering

1984 ◽  
Vol 2 (2) ◽  
pp. 125-142 ◽  
Author(s):  
Timothy L. McAuliffe ◽  
A.A. Afifi
Keyword(s):  
Nearest Neighbor ◽  
Space Time

Enahancing the detail on low-level seismicity and swarms in central-southern Italy by template matching

2021 ◽  
Author(s):  
Luca Carbone ◽  
Rita de Nardis ◽  
Giusy Lavecchia ◽  
Laura Peruzza ◽  
Enrico Priolo ◽  
...  
Keyword(s):  
Template Matching ◽  
Nearest Neighbor ◽  
Matched Filter ◽  
Central Italy ◽  
Total Duration ◽  
Space Time ◽  
Risk Area ◽  
Low Level ◽  
Seismic Catalogue ◽  
Completeness Magnitude

<p> </p><p>During the seismic sequence which followed the devastating L’Aquila 2009 earthquake, on 27 May 2009 the OGS (Istituto Nazionale di Oceanografia e di Geofisica Sperimentale) and the GeosisLab (Laboratorio di Geodinamica e Sismogenesi, Chieti-Pescara University) installed a temporary seismometric network around the Sulmona Basin, a high seismic risk area of Central Italy located right at SE of the epicentral one. This area of the central Apennines is generally characterized by low level seismicity organized in low energy clusters, but it experienced destructive earthquakes both in historical and in early instrumental time (Fucino 1915 =XI MCS, Majella 1706 =X-XI MCS, Barrea 1984 =VIII MCS).</p><p>From the 27 May 2009 to 22 November 2011, the temporary network provided a huge amount of continuous seismic recordings, and a seismic catalogue covering the first seven months of network operation (-1.5≤M<sub>L</sub>≤3.7, with a completeness magnitude of 1.1) and a spatial area that stretches from the Sulmona Basin to Marsica-Sora area. Aiming to enhance the detection of microearthquakes reported in this catalogue, we applied the matched-filter technique (MFT) to continuous waveforms properly integrated with data from permanent stations belonging to the national seismic network. Specifically, we used the open-source seismological package PyMPA to detect microseismicity from the cross-correlation of continuous data and templates. As templates we used only the best relocated events of the available seismic catalogue. Starting from 366 well located earthquakes<strong> </strong>we obtain a new seismic catalogue of 6084 new events (-2<M<sub>L</sub><4) lowering the completeness magnitude to 0.2. To these new seismic locations, we applied a declustering method to separate background seismicity from clustered seismicity in the area. All the seismicity shows a bimodal behaviour in term of distribution of the nearest-neighbor distance/time with the presence of two statistically distinct earthquake populations. We focused the attention on two of these clusters (C1 and C2) that numerically represent the 60% of the catalogue. They consist in 2619 and 995 events, respectively, with magnitude -2.0<M<sub>L</sub><3.6 and -0.5<M<sub>L</sub><3.2 occurred in Marsica-Sora area. C1 shows the typical characteristics of a seismic swarm, without a clear mainshock, but with 8 more energetic events (3.0≤M<sub>L</sub>≤3.5); the temporal evolution is very articulated with a total duration of one month with different bursts of seismicity and characteristic time extension of approximately one week. C2 instead has a different space-time evolution and consists of different swarm-like seismic sequences more discontinuous in comparison with C1. These swarms are described in greater detail to investigate the influence of overpressurized fluids and their space-time distribution.</p>

scholarly journals A Uniform Space–Time Gridding Algorithm for Comparison of Satellite Data Products: Characterization and Sensitivity Study

2013 ◽  
Vol 52 (1) ◽  
pp. 255-268 ◽  
Author(s):  
Nadia Smith ◽  
W. Paul Menzel ◽  
Elisabeth Weisz ◽  
Andrew K. Heidinger ◽  
Bryan A. Baum
Keyword(s):  
Nearest Neighbor ◽  
Uniform Space ◽  
Dynamic Environment ◽  
Grid Cell ◽  
Sensitivity Study ◽  
Time Of Day ◽  
Space Time ◽  
Viewing Angle ◽  
Minimum Number ◽  
High Level

AbstractTo overcome the complexities associated with combining or comparing multisensor data, a statistical gridding algorithm is introduced for projecting data from their unique instrument domain to a uniform space–time domain. The algorithm has two components: 1) a spatial gridding phase in which geophysical properties are filtered on the basis of a set of criteria (e.g., time of day or viewing angle) and then aggregated into nearest-neighbor clusters as defined by equal-angle grid cells and 2) a temporal gridding phase in which daily statistics are calculated per grid cell from which longer time-aggregate statistics are derived. The sensitivity of the gridding algorithm is demonstrated using a month (1–31 August 2009) of level 2 Aqua/Moderate Resolution Imaging Spectroradiometer (MODIS) cloud-top pressure (CTP) retrievals as an example. Algorithm sensitivity is tested for grid size, number of days in the definition of a time average, viewing angle, and minimum number of observations per grid cell per day. The average CTP for high-level clouds from a number of different polar-orbiting instruments are compared on a 1° × 1° global grid. With the data projected onto a single grid, differences in CTP retrieval algorithms are highlighted. The authors conclude that this gridding algorithm greatly facilitates the intercomparison of CTP (or any other geophysical parameter) and algorithms in a dynamic environment. Its simplicity lends transparency to understanding the behavior of a given parameter and makes it useful for both research and operational use.

Enhancing the Jacquez k nearest neighbor test for space-time interaction

2012 ◽  
Vol 31 (21) ◽  
pp. 2318-2334 ◽  
Author(s):  
Nicholas Malizia ◽  
Elizabeth A. Mack
Keyword(s):  
Nearest Neighbor ◽  
Space Time ◽  
K Nearest Neighbor ◽  
Time Interaction

scholarly journals Synthesis of space-time optimal systolic algorithms for the Cholesky factorization

2002 ◽  
Vol Vol. 5 ◽  
Author(s):  
Clémentin Tayou Djamegni
Keyword(s):  
Projection Method ◽  
Nearest Neighbor ◽  
Space Time ◽  
Cholesky Factorization ◽  
Systolic Arrays ◽  
Allocation Method ◽  
Time Optimal ◽  
International Audience ◽  
Systolic Algorithms ◽  
Allocation Methods

International audience In this paper we study the synthesis of space-time optimal systolic arrays for the Cholesky Factorization (CF). First, we discuss previous allocation methods and their application to CF. Second, stemming from a new allocation method we derive a space-time optimal array, with nearest neighbor connections, that requires 3N + Θ (1) time steps and N^2/8 + Θ (N) processors, where N is the size of the problem. The number of processors required by this new design improves the best previously known bound, N^2/6 + Θ (N), induced by previous allocation methods. This is the first contribution of the paper. The second contribution stemms from the fact that the paper also introduces a new allocation method that suggests to first perform clever index transformations on the initial dependence graph of a given system of uniform recurrent equations before applying the weakest allocation method, the projection method.

scholarly journals Space-time tradeoffs for approximate nearest neighbor searching

2009 ◽  
Vol 57 (1) ◽  
pp. 1-54 ◽  
Author(s):  
Sunil Arya ◽  
Theocharis Malamatos ◽  
David M. Mount
Keyword(s):  
Nearest Neighbor ◽  
Space Time ◽  
Approximate Nearest Neighbor ◽  
Nearest Neighbor Searching
Sign in / Sign up

Export Citation Format

Share Document