This document discusses methods for calculating the standard error of the b-value parameter in the Gutenberg-Richter magnitude-frequency relationship. It presents two key formulas:
1) For large samples where b can be treated as constant over time, the standard error of b is given by σ(b) = 2.30b2/√n, where n is the sample size and σ2(M) is the sample variance of magnitudes.
2) When b varies slowly over time, its standard error has two components - the time-averaged variance of b at different times, plus the variance of the mean b averaged over the entire time period.
The document also provides tables to determine confidence intervals for
The usual theory of inflation breaks down in eternal inflation. We derive a
dual description of eternal inflation in terms of a deformed Euclidean CFT located at the
threshold of eternal inflation. The partition function gives the amplitude of different geometries
of the threshold surface in the no-boundary state. Its local and global behavior
in dual toy models shows that the amplitude is low for surfaces which are not nearly conformal
to the round three-sphere and essentially zero for surfaces with negative curvature.
Based on this we conjecture that the exit from eternal inflation does not produce an infinite
fractal-like multiverse, but is finite and reasonably smooth
A Study of Anomalous Value of Free-Air Vertical Gradient for Density Determin...Premier Publishers
Underground Gravity Vertical Gradient is an important practice for prospecting underground densities, although in most cases it does not match the densities obtained directly in the laboratory from rock samples representative of the location. The densities of the laboratory samples were systematically lower when compared to the densities calculated by gravimetric determination suggesting some kind of systematic error. Several researchers propose different sources to explain these systematic errors including an anomalous value of the free-air vertical gradient. The anomalous value was admitted for the free-air vertical gradient in this paper to reinterpret the densities determination research made in a mine at Barberton, Ohio, in 1950, by gravimetric measurements and by laboratory rock samples. The results of both approaches reached similar densities agreeing with the free-air vertical gradient proposed.
The usual theory of inflation breaks down in eternal inflation. We derive a
dual description of eternal inflation in terms of a deformed Euclidean CFT located at the
threshold of eternal inflation. The partition function gives the amplitude of different geometries
of the threshold surface in the no-boundary state. Its local and global behavior
in dual toy models shows that the amplitude is low for surfaces which are not nearly conformal
to the round three-sphere and essentially zero for surfaces with negative curvature.
Based on this we conjecture that the exit from eternal inflation does not produce an infinite
fractal-like multiverse, but is finite and reasonably smooth
A Study of Anomalous Value of Free-Air Vertical Gradient for Density Determin...Premier Publishers
Underground Gravity Vertical Gradient is an important practice for prospecting underground densities, although in most cases it does not match the densities obtained directly in the laboratory from rock samples representative of the location. The densities of the laboratory samples were systematically lower when compared to the densities calculated by gravimetric determination suggesting some kind of systematic error. Several researchers propose different sources to explain these systematic errors including an anomalous value of the free-air vertical gradient. The anomalous value was admitted for the free-air vertical gradient in this paper to reinterpret the densities determination research made in a mine at Barberton, Ohio, in 1950, by gravimetric measurements and by laboratory rock samples. The results of both approaches reached similar densities agreeing with the free-air vertical gradient proposed.
My books- Hacking Digital Learning Strategies http://hackingdls.com & Learning to Go https://gum.co/learn2go
Resources at http://shellyterrell.com/classmanagement
The reality for companies that are trying to figure out their blogging or content strategy is that there's a lot of content to write beyond just the "buy now" page.
Design of Low-Pass Digital Differentiators Based on B-splinesCSCJournals
This paper describes a new method for designing low-pass differentiators that could be widely suitable for low-frequency signals with different sampling rates. The method is based on the differential property of convolution and the derivatives of B-spline bias functions. The first order differentiator is just constructed by the first derivative of the B-spline of degree 5 or 4. A high (>2) order low-pass differentiator is constructed by cascading two low order differentiators, of which the coefficients are obtained from the nth derivative of a B-spline of degree n+2 expanded by factor a. In this paper, the properties of the proposed differentiators were presented. In addition, we gave the examples of designing the first to sixth order differentiators, and several simulations, including the effects of the factor a on the results and the anti-noise capability of the proposed differentiators. These properties analysis and simulations indicate that the proposed differentiator can be applied to a wide range of low-frequency signals, and the trade-off between noise- reduction and signal preservation can be made by selecting the maximum allowable value of a.
An Asymptotic Approach of The Crack Extension In Linear PiezoelectricityIRJESJOURNAL
Abstract: As a result of a theoretical technique for elucidating the fracture mechanics of piezoelectric materials, this paper provides, on the basis of the three-dimensional model of thin plates, an asymptotic behavior in the Griffith’s criterion for a weakly anisotropic thin plate with symmetry of order six, through a mathematical analysis of perturbations due to the presence of a crack. It is particularly established, in this work, the effects of both electric field and singularity of the in-plane mechanical displacement on the piezoelectric energy
The Inverse Scattering Series (ISS) is a direct inversion method
for a multidimensional acoustic, elastic and anelastic earth. It
communicates that all inversion processing goals are able to
be achieved directly and without any subsurface information.
This task is reached through a task-specific subseries of the
ISS. Using primaries in the data as subevents of the first-order
internal multiples, the leading-order attenuator can predict the
time of all the first-order internal multiples and is able to attenuate
them.
However, the ISS internal multiple attenuation algorithm can
be a computationally demanding method specially in a complex
earth. By using an approach that is based on two angular
quantities and that was proposed in Terenghi et al. (2012), the
cost of the algorithm can be controlled. The idea is to use the
two angles as key-control parameters, by limiting their variation,
to disregard some calculated contributions of the algorithm
that are negligible. Moreover, the range of integration
can be chosen as a compromise of the required degree of accuracy
and the computational time saving.
This time-saving approach is presented
I am Nathan J. I am a Magnetic Materials Assignment Expert at eduassignmenthelp.com. I hold a Master’s Degree in Electromagnetic, from The University of Queensland, Australia. I have been helping students with their assignments for the past 12 years. I solve assignments related to Magnetic Materials.
Visit eduassignmenthelp.com or email info@eduassignmenthelp.com. You can also call on +1 678 648 4277 for any assistance with Magnetic Materials Assignments.
Finite-difference modeling, accuracy, and boundary conditions- Arthur Weglein...Arthur Weglein
This short report gives a brief review on the finite difference modeling method used in MOSRP
and its boundary conditions as a preparation for the Green’s theorem RTM. The first
part gives the finite difference formulae we used and the second part describes the implemented
boundary conditions. The last part, using two examples, points out some impacts of the accuracy
of source fields on the results of modeling.
I am Baddie K. I am a Magnetic Materials Assignment Expert at eduassignmenthelp.com. I hold a Masters's Degree in Electro-Magnetics, from The University of Malaya, Malaysia. I have been helping students with their assignments for the past 12 years. I solve assignments related to Magnetic Materials.
Visit eduassignmenthelp.com or email info@eduassignmenthelp.com. You can also call on +1 678 648 4277 for any assistance with Magnetic Materials Assignments.
Quantum Minimax Theorem in Statistical Decision Theory (RIMS2014)tanafuyu
This is almost self-contained explanation of our recent result. The contents are based on our talk in the RIMS2014 conference.
Recently, many fundamental and important results in statistical decision theory have been extended to the quantum system. Quantum Hunt-Stein theorem and quantum locally asymptotic normality are typical successful examples.
In our recent preprint, we show quantum minimax theorem, which is also an extension of a well-known result, minimax theorem in statistical decision theory, first shown by Wald and generalized by LeCam. Our assertions hold for every closed convex set of measurements and for general parametric models of density operator. On the other hand, Bayesian analysis based on least favorable priors has been widely used in classical statistics and is expected to play a crucial role in quantum statistics. According to this trend, we also show the existence of least favorable priors, which seems to be new even in classical statistics.
Similar to Öncel Akademi: İstatistiksel Sismoloji (20)
Gravimetri Dersi için aşağıda ki videoları izleyebilirsiniz.
Link 01: https://www.youtube.com/watch?v=HTyjVaVGx0k
Link 02: https://www.youtube.com/watch?v=fUkfgI8XaOE
Geopsy yaygın olarak kullanılan profesyonel bir program. Özellikle, profesyonel program deneyimi yeni mezunlarda çok aranan bir özellik. Bir öğrencim çalışmasında kullanmayı planlıyor.
Acetabularia Information For Class 9 .docxvaibhavrinwa19
Acetabularia acetabulum is a single-celled green alga that in its vegetative state is morphologically differentiated into a basal rhizoid and an axially elongated stalk, which bears whorls of branching hairs. The single diploid nucleus resides in the rhizoid.
Introduction to AI for Nonprofits with Tapp NetworkTechSoup
Dive into the world of AI! Experts Jon Hill and Tareq Monaur will guide you through AI's role in enhancing nonprofit websites and basic marketing strategies, making it easy to understand and apply.
The French Revolution, which began in 1789, was a period of radical social and political upheaval in France. It marked the decline of absolute monarchies, the rise of secular and democratic republics, and the eventual rise of Napoleon Bonaparte. This revolutionary period is crucial in understanding the transition from feudalism to modernity in Europe.
For more information, visit-www.vavaclasses.com
Read| The latest issue of The Challenger is here! We are thrilled to announce that our school paper has qualified for the NATIONAL SCHOOLS PRESS CONFERENCE (NSPC) 2024. Thank you for your unwavering support and trust. Dive into the stories that made us stand out!
2024.06.01 Introducing a competency framework for languag learning materials ...Sandy Millin
http://sandymillin.wordpress.com/iateflwebinar2024
Published classroom materials form the basis of syllabuses, drive teacher professional development, and have a potentially huge influence on learners, teachers and education systems. All teachers also create their own materials, whether a few sentences on a blackboard, a highly-structured fully-realised online course, or anything in between. Despite this, the knowledge and skills needed to create effective language learning materials are rarely part of teacher training, and are mostly learnt by trial and error.
Knowledge and skills frameworks, generally called competency frameworks, for ELT teachers, trainers and managers have existed for a few years now. However, until I created one for my MA dissertation, there wasn’t one drawing together what we need to know and do to be able to effectively produce language learning materials.
This webinar will introduce you to my framework, highlighting the key competencies I identified from my research. It will also show how anybody involved in language teaching (any language, not just English!), teacher training, managing schools or developing language learning materials can benefit from using the framework.
Instructions for Submissions thorugh G- Classroom.pptxJheel Barad
This presentation provides a briefing on how to upload submissions and documents in Google Classroom. It was prepared as part of an orientation for new Sainik School in-service teacher trainees. As a training officer, my goal is to ensure that you are comfortable and proficient with this essential tool for managing assignments and fostering student engagement.
Embracing GenAI - A Strategic ImperativePeter Windle
Artificial Intelligence (AI) technologies such as Generative AI, Image Generators and Large Language Models have had a dramatic impact on teaching, learning and assessment over the past 18 months. The most immediate threat AI posed was to Academic Integrity with Higher Education Institutes (HEIs) focusing their efforts on combating the use of GenAI in assessment. Guidelines were developed for staff and students, policies put in place too. Innovative educators have forged paths in the use of Generative AI for teaching, learning and assessments leading to pockets of transformation springing up across HEIs, often with little or no top-down guidance, support or direction.
This Gasta posits a strategic approach to integrating AI into HEIs to prepare staff, students and the curriculum for an evolving world and workplace. We will highlight the advantages of working with these technologies beyond the realm of teaching, learning and assessment by considering prompt engineering skills, industry impact, curriculum changes, and the need for staff upskilling. In contrast, not engaging strategically with Generative AI poses risks, including falling behind peers, missed opportunities and failing to ensure our graduates remain employable. The rapid evolution of AI technologies necessitates a proactive and strategic approach if we are to remain relevant.
A Strategic Approach: GenAI in EducationPeter Windle
Artificial Intelligence (AI) technologies such as Generative AI, Image Generators and Large Language Models have had a dramatic impact on teaching, learning and assessment over the past 18 months. The most immediate threat AI posed was to Academic Integrity with Higher Education Institutes (HEIs) focusing their efforts on combating the use of GenAI in assessment. Guidelines were developed for staff and students, policies put in place too. Innovative educators have forged paths in the use of Generative AI for teaching, learning and assessments leading to pockets of transformation springing up across HEIs, often with little or no top-down guidance, support or direction.
This Gasta posits a strategic approach to integrating AI into HEIs to prepare staff, students and the curriculum for an evolving world and workplace. We will highlight the advantages of working with these technologies beyond the realm of teaching, learning and assessment by considering prompt engineering skills, industry impact, curriculum changes, and the need for staff upskilling. In contrast, not engaging strategically with Generative AI poses risks, including falling behind peers, missed opportunities and failing to ensure our graduates remain employable. The rapid evolution of AI technologies necessitates a proactive and strategic approach if we are to remain relevant.
Welcome to TechSoup New Member Orientation and Q&A (May 2024).pdfTechSoup
In this webinar you will learn how your organization can access TechSoup's wide variety of product discount and donation programs. From hardware to software, we'll give you a tour of the tools available to help your nonprofit with productivity, collaboration, financial management, donor tracking, security, and more.
Macroeconomics- Movie Location
This will be used as part of your Personal Professional Portfolio once graded.
Objective:
Prepare a presentation or a paper using research, basic comparative analysis, data organization and application of economic information. You will make an informed assessment of an economic climate outside of the United States to accomplish an entertainment industry objective.
Palestine last event orientationfvgnh .pptxRaedMohamed3
An EFL lesson about the current events in Palestine. It is intended to be for intermediate students who wish to increase their listening skills through a short lesson in power point.
1. Bulletin of the Seismological Society of America, Vol. 72, No. 5, pp. 1677-1687, October 1982
THE STANDARD ERROR OF THE MAGNITUDE-FREQUENCY b VALUE
BY YAOLIN SHI AND BRUCE A. BOLT
ABSTRACT
Estimated b values in log N = a - bM are widely used in seismicity comparisons
and risk analysis, but uncertainties have been little explored. In this paper, the
usual F probability density distribution for b is given and compared with an
asymptotic form for temporally varying b. Convenient tables for the standard
error of b are given that allow statistical tests to accompany investigations of
both temporal and spatial variations of b. With large samples and slow temporal
changes in b, the standard error of b is
where
o(l~) = 2.30b2a (?~) ,
n
a2(M ) = T (M,- K.f)2 /n(n- 1).
i=1
In an example from central California, stable estimates of b require a space-
time window containing about 100 earthquakes. From 1952 to 1978, the average
b and 90 per cent confidence limits are 0.95 (+0.94, -0.30). Some fluctuations
of b are statistically significant but some are not. Within 90 per cent confidence
limits, b changes from a low of 0.60 (+0.11, -0.09) in 1955 to a high of 1.39
(+0.25, -0.21) in 1967 and drops to 0.72 (+0.13, -0.10) in 1975. In this
example, no correlation between large earthquakes (M > 5) and b variations
occurred.
INTRODUCTION
The b value in Gutenberg and Richter's relation (1954)
log N = a- bM (1)
has been widely used in research on seismicity and seismic risk analysis. In labora-
tory experiments on rocks, the occurrence of microcracks follows a similar form, and
the b value is inversely related to the stress level (Scholz, 1968). For some earthquake
sequences, the b value is reported to change before the main shock (e.g., Suyehiro,
1966; Gibowicz, 1973; Robinson, 1979; Smith, 1981). This property is consistent with
an initially high stress level followed by a stress drop due to the main shock and it
has been considered useful in earthquake prediction (Scholz, 1968; Li et al., 1978).
For use in prediction and comparative mechanism studies, standard errors of b
must be computed so that meaningful statistical tests can be performed. Because
earthquakes are stochastic processes and the b value is a random variable, the
probability distribution and the variance of b are essential in studying its temporal
and spatial variations.
The b value can be calculated by least-squares regression, but the presence of
even a few large earthquakes influences the resulting b value significantly. In least-
squares regression of equation (1), the variance of one observation, a 2 = Eri2/(n - 1),
is only a measure of the deviation of the data points from the linear curve. The
problem is to derive a formula for an estimation of the variance of b itself.
As an alternative, the maximum likelihood method has been suggested as pref-
1677
2. 1678 YAOLIN SHI AND BRUCE A. BOLT
erable for calculating the b value because it yields a more robust estimate when the
number of the infrequent large earthquakes changes. There will be cases, however,
such as estimating the probability of the largest magnitude of earthquakes from
relation (1), where the least-squares method is more suitable.
Suppose the probability density of magnitude M is exponential with mean 1/fi,
i.e.,
fi exp{-fi(M - M0)}, M > Mo. (2)
The maximum likelihood estimation of fi(b = fi log e) is (Aki, 1965; Utsu, 1965)
1
(3)P = -M - Mo'
where 2~ris the mean magnitude and Mo is the smallest (or "threshold") magnitude
considered. In these seminal papers, only an approximate formula to estimate the
variance of b was given.
First, confidence limits need to be defined. If we form a linear combination Y so
that
n
Y= E Y~, (4)
i=l
where, in terms of the density function, yi = O/Off log f(Mi, fi), it can then be shown
(e.g., Aki, 1965) that for large n, Y has a normal distribution with a zero mean and
variance n / fi 2.
The probability that
-dE < --=fly < dE (5)
is given by
- - exp(-x2/2) dx. (6)
• = ~ d~
Then e = 95 per cent if d~ = 1.96 and • = 90 per cent if dE = 1.65. When defined in
this way, the confidence level is directly related only to Y, and consequently, it is
difficult to assess information about the distribution of b itself.
More directly, the distribution of/~ is
[~ =~2nfi/X~,~ = ~F~,2n. (7)
This chi-squared form is given by Utsu (1966) and rederived by Zhang and Song
(1981). In brief,
n
M - Mo =-I ~ (Mi- Mo)
ni=l
1 n
=- 2 e,/fi,
ni=l
3. THE STANDARD ERROR OF THE MAGNITUDE-FREQUENCY b VALUE 1679
where ei are independent unit exponentials that are distributed like X22/2. Then,
from the additivity property of X2,
- Mo d X~n/2nfi.
It follows from (7) (see Abramowitz and Stegun, 1965, 26.6.3) that
E(fi)=fin/(n-1), n>l (8)
var fi = fi2n2/[(n - 1)2(n - 2)], n > 2. (9)
Zhang and Song (1981) concluded from (8) that the maximum likelihood method
yields a biased estimation of b, and they also investigated, by a Monte Carlo
procedure, the least-squares estimate of the standard error of b. The formula fl/~n
(Aki, 1965) is an approximation to (9).
Once the distribution is derived, confidence intervals can be found by using tables
of F or X2 distributions, e.g., for 0.90 per cent confidence level
F~,2n(0.05) < ~ < F~,2n(0.95) (10)
or
2nb
X22n(0.05) < A < X22n(0.95). (11)
b
Appropriate tables are not readily available in the seismological literature for
n > 50, but graphs giving the cumulative distribution of the ratio (10) are given by
Utsu (1966).
In the above work, b was considered a constant in time. We will discuss the
distribution of b and estimate its standard error for both constant and slowly
changing b. Our results are then checked by comparison with real earthquake
sequences.
NUMERICAL FORMULA
We will now take an alternative approach to the distribution of b and consider
the distribution of Mi without any assumption of the constancy of b. By the central
n
limit theorem, the distribution function of M, M = _1 ~ Mi, approaches a normal
ni=l
distribution for large n if each Mi has finite mean and variance, and the M~ are
independent. Therefore, in the limit, the distribution of 3~t can be expressed as
1
f(3I) - exp(-(Al - E(2~r))2/2 var 214).
~/2~ var 214
From (2),
1
E (M) = 214o+ -
B
1
var M -
B ~
1
E(3YI) = E(M) = Mo + -
1
var AI = - var M = 1~nil 2.
n
(12)
4. 1680 YAOLIN SHI AND BRUCE A. BOLT
Because fi (3)) is a monotonically decreasing function of 3I, there exists a single-
valued inverse fimction
~(j~) = i
-~ + Mo. (13)
P
Then the distribution of b is
-- ^
1 log e
= ~/2~ var3) ~2 exp(-(1/b- 1//~)2log2e/2 var3))
where/~ = log e/[E (3)) - M0]. Typical results are shown in Figure 1.
3 0
(14)
2
<_Q
1
0 4.04.55.0
,i
.51.01.52.02.53.03.5
b value
FIG. 1. Density functionsg(b) calculated from (14) and (15a). Light line for b = 1.6, o_~ = 0.1; heavy
line for b = 1.6, n = 20 (or o114 = 0.05); dashed line for b = 1.0, n = 30.
Here the distribution function of/~, is given for large n; the mean and variance of
3), and then of/~, can be calculated from earthquake catalogs. Statistical tests can,
as a consequence, be carried out conveniently by table reference. Because the value
of b in a region is observed to change with time and location, its mean and variance
might also be expected to change. In other words, b should be regarded, in general,
as a nonstationary stochastic process b (t). When sampling with a small time and
space window, however, b can be usually taken as stationary and with constant
expectation. Then, the above formulas hold and the distribution function g(/~ ) can
be expressed simply as
g(/~) = ~-~ b
exp(-n(b/[~ - 1)2/2). (15a)
5. THE STANDARD ERROR OF THE MAGNITUDE-FREQUENCY b VALUE 1681
For comparison, the exact formula for g(/~ ) for constant b is (Utsu, 1966)
express the sampling weight at different times. Then
g(/~) - n ~ b ~
(n- 1)! /~#Ji exp(-nb/b). (15b)
Sample confidence levels, calculated from (15) are given in Table 1 and may be
TABLE 1
PERCENTAGE POINTS FOR THE DISTRIBUTION OF g(b) AND n
ConfidenceLevel 0.90
bin
30 50 100 200
0.4 -0.10 -0.08 -0.06 -0.04
0.14 0.11 0.07 0.05
0.5 -0.13 -0.10 -0.07 -0.05
0.18 0.13 0.09 0.06
0.6 -0.15 -0.12 -0.09 -0.06
0.21 0.16 0.11 0.07
0.7 -0.18 -0.14 -0.10 -0.08
0.25 0.19 0.13 0.09
0.8 -0.20 -0.16 -0.12 -0.09
0.29 0.21 0.14 0.10
0.9 -0.23 -0.18 -0.13 -0.10
0.32 0.24 0.16 0.11
1.0 -0.25 -0.20 -0.15 -0.11
0.36 0.27 0.18 0.12
1.1 -0.28 -0.22 -0.16 -0.12
0.39 0.29 0.20 0.14
1.2 --0.30 -0.24 --0.18 --0.13
0.43 0.32 0.22 0.15
1.3 -0.33 -0.26 -0.19 -0.14
0.46 0.35 0.24 0.16
1.4 -0.35 -0.28 -0.21 -0.15
0.50 0.37 0.25 0.17
1.5 -0.38 -0.30 -0.22 -0.16
0.54 0.40 0.27 0.19
1.6 -0.40 -0.32 -0.24 -0.17
0.57 0.43 0.29 0.20
1.7 -0.43 --0.35 -0.25 -0.18
0.61 0.45 0.31 0.21
1.8 -0.45 -0.37 -0.27 -0.19
0.64 0.47 0.33 0.22
1.9 -0.48 -0.39 -0.28 -0.21
0.68 0.51 0.34 0.24
2.0 -0.50 -0.41 -0.30 -0.22
0.71 0.53 0.36 0.25
checked against standard X~ tables. Differences between (15a) and (15b) are not
large for large n.
Formula (15b) is important not only because of the assumption that a constant b
within a small sampling window may be acceptable in many cases, but also because
most published b values are given without standard errors and var 2tl. This omission
means that they cannot be used with formula (14) nor for quantitative statistical
tests. Formula (15b) allows, however, the upper limits of the standard error of b
6. 1682 YAOLIN SHI AND BRUCE A. BOLT
values to be estimated, as illustrated later. In this way, b values can be used for
statistical tests.
When b changes with time, a distribution function of b, f(b, t), can be assigned
and
t-
vat b = I (b - E(b))2f(b, t) db dt. (16)
2
If b changes slowly with time, f(b, t) can be decomposed to
f(b, t) = g(b)w(t), (17)
where g(b) is defined earlier. At any time, b has the same form of distribution
function as (14), although its mean and variance may change. Let function w(t)
express the sampling weight at different times. Then
var b = f (b- E(b)}2g(b)w(t) db dt
C
= J ((b -/~t) + (E - E(b))}2g(b)w(t) db dt
= (var bt ) "t- var/~,
where bt is the expectation of b at time t, i.e.,
-- I bg(b)5t db
J
(18)
(19)
and ( ) indicates the time average. Thus, because the cross-product terms vanish,
the variance of b is composed of two parts. The first part is the time average of the
variance of b at different times as calculated previously. The second part is the
variance of 6, the mean of b averaged over the whole time range.
Because var b = (var be), formula (15b) may give an underestimated standard
error when the temporal and spatial window is too large for b to be constant. In this
case, we have to estimate the average 214and var M = ~ (Mi - M)2/n(n - 1), and
j=l
then determine the probability interval of b from Table 2, computed using (14).
In the general case, a large sample distribution of/~ can be derived in the usual
way. Let Xi = Mi - Mo. Then/~ = 1~. IfXi has mean/t and variance a 2, it can be
shown as a special case of (14) that fi is asymptotically normal with mean 1//z and
variance
(llt~e)2a21n = fl4o21n. (20)
n
Here o2 may be estimated, as above, from ~ (Mi - -ff/I)2/(n - 1). Note that log/~
i=l
is also asymptotically normal with mean log fi and variance fl2o2/n. These functions
may also fluctuate with time. It also follows that for large n,
b~
o(/~)= ~ o(M) = 2.3052~(M). (21)
In practice, starting from an earthquake catalog, we can estimate b by maximum
likelihood from M using (3). We can also obtain a probability interval for b from
7. THE STANDARD ERROR OF THE MAGNITUDE-FREQUENCY b VALUE 1683
formula (15), (Table 1) or formula (14), (Table 2) depending on whether the b value
can be regarded as an approximate constant or not. Brillinger (personal communi-
cation) describes the model contemplated as "doubly stochastic." The process
b (M, t) may, of course, depend not only on t, but also on M, and it is an assumption
that the distribution of/~ remains exponential in these cases.
TABLE 2
PERCENTAGE POINTS FOR THE DISTRIBUTION OF g(b) AND
ConfidenceLevel0.90
b/(J)71
0.15 0.10 0.05 0.03
0.4 -0.07 -0.05 -0.03 -0.01
0.12 0.07 0.04 0.02
0.5 -0.11 -0.08 -0.04 -0.02
0.20 0.12 0.05 0.03
0.6 -0.15 -0.11 -0.06 -0.04
0.31 0.18 0.08 0.05
0.7 -0.20 -0.14 -0.08 -0.05
0.46 0.26 0.11 0.06
0.8 -0.25 -0.18 -0.10 -0.06
0.67 0.35 0.15 0.08
0.9 -0.30 -0.23 -0.13 -0.08
0.94 0.47 0.19 0.11
1.0 -0.36 -0.27 -0.16 -0.10
1.32 0.61 0.24 0.13
1.1 -0.42 -0.32 -0.19 -0.12
1.84 0.79 0.29 0.16
1.2 -0.50 -0.38 -0.22 -0.14
3.20 0.99 0.36 0.19
1.3 -0.55 -0.43 -0.25 -0.16
3.67 1.26 0.43 0.23
1.4 -0.62 -0.48 -0.29 -0.19
5.44 1.58 0.51 0.27
1.5 -0.69 -0.54 -0.33 -0.22
8.64 1.98 0.60 0.31
1.6 -0.76 -0.60 -0.37 -0.24
15.97 2.46 0.70 0.36
1.7 -0.84 -0.66 -0.41 -0.27
47.97 3.08 0.81 0.41
1.8 -0.91 -0.73 -0.46 -0.30
--* 3.86 0.93 0.47
1.9 -0.99 -0.79 -0.50 -0.32
-- 4.88 1.07 0.53
2.0 -1.07 -0.86 -0.55 -0.37
-- 6.26 1.22 0.59
* -- > 100.
CENTRAL CALIFORNIASEISMICITY
The seismicity in central California is taken as an example. The catalog is
considered complete for earthquakes with M _->2.5 in this region since 1952 (Bolt
and Miller, 1975). The sample has 2572 earthquakes in 22 yr.
Through a time window containing n = 30, 50, 100, and 200 earthquakes,
respectively, b values were scanned in the time domain. Two distinct b (t) curves for
n = 30 and n = 200 are plotted in Figures 2 and 3. Error bars for the 90 per cent
8. 1684 YAOLIN SI-II AND BRUCE A. BOLT
~7~--~ ur~ (M
r~
r~
o~
r~
r~
r~
cD
CO
r~
~o
~o
~o
o
~o
oo
CD
~o
kn
t--
un
c~
kn
U3
u~
u~
t~
c,-J
anloA q
h~
A~
8~
c~
9. THE STANDARD ERROR OF THE MAGNITUDE-FREQUENCY b VALUE
co
1685
OJ(M
~Ldu u
~d
o
u
~d~d
Lr~ C) I~ 0 LF]
F~
Ld
m~
o
r~
o~
~o
co
r~
Lo
tD
Lrl
CO
m-
tO
CO
~J
~0
~0
C3
~0
Cn
~q
CO
bq
I
m
LA
(.0
Ln
LN
cDLn
LJ
t--
II
8
!
c6
anlDA q
10. 1686 YAOLIN SHI AND BRUCE A. BOLT
confidence level are evaluated from Table 1 (formula 15). For comparison, error
intervals were also evaluated from Table 2 (formula 14). These comparisons show
that although for small time windows (e.g., n = 30), the estimates sometimes differ
by over 100 per cent, for larger time windows (e.g., n = 100), the error intervals
calculated from the two formulas are almost the same. This implies that the variance
of b can be estimated from previous published data of b, even if var M is not given,
provided n is not too small.
For a time window containing n = 30 earthquakes, the b(t) curve has more
fluctuations than for the case n = 200. However, many such fluctuations are not
statistically significant when the chance variation is considered, and this result must
be taken into account if changes of b value are used as a precursor in earthquake
prediction [c.f. Wyss and Lee {1973)].
Over a longer time interval, significant temporal variations of b are obvious in
Figure 3. (Alternatively, log /~ could be graphed in Figures 2 and 3 so that the
3.0
V
2.5
2.0
1.5
1.0
f -- i i i i
• 5 .0 ] .5 2.0 2.5 3.0 ~.5 4.0 4.5 5.0
b vo lue
FIG. 4. Histogram of b distribution for central C~lifornia seismicity, from 397 windows each containing
100 earthquakes, compared with theoretical curve b = 0.95, aM = 0.13.
confidence intervals have constant width. The hypotheses of time invariant b may
then be tested by calculating the proportion of b values that exceed the intervals.)
The times of occurrence of earthquakes larger than magnitude 5 are also plotted in
the figure. The value of b changes from a low level of 0.60 (+0.11, -0.09) in 1955 to
a high value of 1.39 (+0.25, -0.21) in 1967 and drops again to 0.72 (+0.13, -0.10) in
1975. However, we can find no correlation between changes of b and the time of
occurrence of the larger earthquakes. The meaning of this longer term variation of
b is not yet clear.
Finally, the histogram of observed b since 1952 is plotted in Figure 4. The
theoretical distribution curve of/~ = 0.95 and oM = 0.13 is also drawn. Because
within the 22-yr period b changed with time, formula (15) cannot be used. The
theoretical curve is calculated from (14) and a X2 test made with the histogram
grouped into 10 intervals. The computed X~o= 17.5 is close to the expected value of
18.3 for 10 degrees of freedom at the 5 per cent significance level. Therefore, the
11. THE STANDARD ERROR OF THE MAGNITUDE-FREQUENCY b VALUE 1687
evidence does not suggest that the actual distribution differs from the derived
theoretical form at this level.
It should be emphasized that the uncritical use of formula (15) leads to a 90 per
cent confidence level estimate for b of 0.95 (+0.17, -0.14) and a standard error of
0.10. In a similar way, the simple application of the formula of Aki (1965) or Zhang
and Song (1981) yields 0.95 (±0.10) for the standard error. Slow changes of b with
time entail a broader b distribution. From equation (21), the large sample approxi-
mation gives 0.95 (__0.27) for the standard error for the observed distribution of
slowly varying b in central California. We calculate also that the 90 per cent
confidence level of b is more realistically 0.95 (+0.94, -0.30). The result implies that
even large differences between long-term average b values for different regions or
epochs or focal depths (Gutenberg and Richter, 1954) may not be significant.
ACKNOWLEDGMENTS
The authors thank Professor D. Brillinger, Dr. R. Uhrhammer and a reviewer, and A. Lees for helpful
discussions and suggestions. The computer time was provided by the Computer Center of the University
of California, Berkeley.
REFERENCES
Abramowitz, M. and I. A. Stegun (1965). Handbook of Mathematical Functions, U.S. Department of
Commerce, National Bureau of Standards.
Aki, K. (1965). Maximum likelihood estimate of b in the formula log N = a - bM and its confidence
limits, Bull. Earthquake Res. Inst., Tokyo Univ. 43, 237-239.
Bolt, B. A. and R. D. Miller (1975). Catalog of earthquakes in northern California and adjoining areas,
Seismographic Stations, University of California, Berkeley.
Gibowicz, S. J. (1973). Variation of the frequency-magnitude relation during earthquake sequences in
New Zealand, Bull. Seism. Soc. Am. 63, 517-528.
Gutenberg, B. and C. F. Richter (1954). Seismicity of the Earth, 2nd ed., Princeton University, Princeton,
New Jersey.
Li, Q.-L., J.-B. Chen, L. Yu, and B.-L. Hao (1978}. Time and space scanning of the b value--A method
for monitoring the development of catastrophic earthquakes, Acta Geophys. Sin. 21, 101-125.
Robinson, R. (1979). Variation of energy release, rate of occurrence and b value of earthquakes in the
main seismic region, New Zealand, Phys. Earth Planet. Interiors 18, 209-220.
Scholz, C. (1968). The frequency-magnitude relation of microfracturing in rock and its relation to
earthquakes, Bull. Seism. Soc. Am. 58, 399-415.
Smith, W. D. (1981). The b-value as an earthquake precursor, Nature 289, 136-139.
Suyehiro, S. (1966). Differences between aftershocks and foreshocks in the relationship of magnitude to
frequency of occurrence for the great Chilean earthquake of 1960, Bull. Seism. Soc. Am. 56, 185-200.
Utsu, T. (1965). A method for determining the value of b in formula log N = a - bM showing the
magnitude-frequency relation for earthquakes, Geophys. Bull. Hokkaido Univ. 13, 99-103.
Utsu, T. (1966). A statistical significance test of the difference in b-value between two earthquake groups,
J. Phys. Earth 14, 37-40.
Zhang, J.-Z. and L,-Y. Song (1981). On the method of estimating b value and its standard error, Acta
Seismologica Sin. 3, 292-301.
Wyss, M. and W. H. K. Lee {1973). Proc. conf. on tectonic problems on the San Andreas fault system 13,
24-42, Stanford University, Stanford, California.
SEISMOGRAPHICSTATIONS
DEPARTMENT OF GEOLOGYAND GEOPHYSICS
UNIVERSITY OF CALIFORNIA
BERKELEY, CALIFORNIA 94720
Manuscript received 15 March 1982