Null Detection in Shear-Wave Splitting Measurements

Null Detection in Shear-Wave Splitting Measurements 1209(9), a difference in fast axis estimate close to 45, that is 37 DU 53. Near-Null measurements can be classified by0 q 0.3 and 32DU 58. Remaining measurementsare to be considered as poor quality (see Fig. 3 for furtherillustration).Real DataWe apply our Null criterion to the shear-wave splittingmeasurements of station LVZ in northern Scandinavia. Theanalyzed earthquakes (M w 6) occurred between December1992 and December 2005. The data were processed by usingthe SplitLab environment (A. Wüstefeld et al., unpublishedmanuscript, 2006). This allows us to analyze eventsefficiently and to calculate simultaneously both the RC andSC technique. We mostly used raw data or, where necessary,applied third-order Butterworth bandpass filters with uppercornerfrequencies down to 0.2 Hz. Most usable events havebackazimuths between 45 and 100. Such sparse backazimuthalcoverage is unfortunately the case for many splittinganalyses, and we aim to extract the maximum informationabout the splitting parameters from these sparse distributions.In total we analyzed 37 SKS phases from a wide rangeof backazimuths (Fig. 4). Many results resemble Null characteristicsby showing low energy on the initial transversecomponent, elongated to linear initial particle motion and atypical energy plot. Such characteristics can be replicated insynthetic seismograms with near-Null parameters, that is,when the fast axis deviates less then 20 from backazimuth(Fig. 1). The average fast axis of the good events, as detectedautomatically and manually, is 14.3 and 14.7 for the SCand RC technique, respectively. Such orientation impliesNulls at backazimuths of approximately 15, 105, 195, and285 and favorable backazimuths for splitting measurementsin between. Indeed, good and fair splitting measurementsare found in backazimuthal ranges between 50 and 70 (Table1 and Fig. 4), where the energy on the transverse componentis expected to reach maximum possible values (seeequation 3) and the splitting can be inverted most reliably.Figure 4. Shear-wave splitting estimates from 33 good and fair measurements fromstation LVZ. The upper panels display fast-axis estimates for RC and SC methods. Notethat many RC estimates are situated near the dotted lines that indicate 45. The lowerpanels display the delay-time estimates. The solid horizontal lines indicate our interpretationof the LVZ with fast axis at 15 and 1.1-sec delay time, based on the meanof the good splitting measurements.