Please use this identifier to cite or link to this item:
Authors: Murru, M.* 
Zhuang, J.* 
Rodolfo, C.* 
Giuseppe, F.* 
Title: Short-term earthquake forecasting experiment before and during the L’Aquila (central Italy) seismic sequence of April 2009
Issue Date: 1-Nov-2014
Series/Report no.: 6/57 (2014)
DOI: 10.4401/ag-6583
Keywords: L’Aquila earthquake
ETAS model
PPE model
Probability forecast
Point process
Probability gain
Subject Classification04. Solid Earth::04.06. Seismology::04.06.02. Earthquake interactions and probability 
Abstract: In this paper, we compare the forecasting performance of several statistical models, which are used to describe the occurrence process of earthquakes in forecasting the short-term earthquake probabilities during the L’Aquila earthquake sequence in central Italy in 2009. These models include the Proximity to Past Earthquakes (PPE) model and two versions of the Epidemic Type Aftershock Sequence (ETAS) model. We used the information gains corresponding to the Poisson and binomial scores to evaluate the performance of these models. It is shown that both ETAS models work better than the PPE model. However, in comparing the two types of ETAS models, the one with the same fixed exponent coefficient (alpha) = 2.3 for both the productivity function and the scaling factor in the spatial response function (ETAS I), performs better in forecasting the active aftershock sequence than the model with different exponent coefficients (ETAS II), when the Poisson score is adopted. ETAS II performs better when a lower magnitude threshold of 2.0 and the binomial score are used. The reason is found to be that the catalog does not have an event of similar magnitude to the L’Aquila mainshock (Mw 6.3) in the training period (April 16, 2005 to March 15, 2009), and the (alpha)-value is underestimated, thus the forecast seismicity is underestimated when the productivity function is extrapolated to high magnitudes. We also investigate the effect of the inclusion of small events in forecasting larger events. These results suggest that the training catalog used for estimating the model parameters should include earthquakes of magnitudes similar to the mainshock when forecasting seismicity during an aftershock sequence.
Appears in Collections:Papers Published / Papers in press

Files in This Item:
File Description SizeFormat 
6583-14798-1-PB (1).pdfArticle2.18 MBAdobe PDFView/Open
Show full item record

Page view(s)

Last Week
Last month
checked on Aug 20, 2018


checked on Aug 20, 2018

Google ScholarTM