Файл: [D._Kim_Rossmo]_Geographic_Profiling(BookFi.org).pdf

© 2000 by CRC Press LLC

List of Figures

Figure 4.1 

Case linkage chart

Figure 7.1 

Offender/target/environment Venn diagram

Figure 7.2 

Crime site search geography

Figure 7.3 

Distance-decay function

Figure 8.1 

Raptor target pattern

Figure 8.2 

Stalker target pattern

Figure 9.1 

Confirmed and suspected victim numbers by case

Figure 9.2 

Serial murder rates by state

Figure 9.3 

Serial murder by day of week

Figure 9.4 

Distance to crime site

Figure 9.5 

Mean logarithm of crime trip distance over time

Figure 9.6 

Crime trip distance mean standard deviation over time

Figure 10.1 

Journey-to-crime Venn diagram*

Figure 10.2 

Vancouver robberies — Crime sites

Chapter 10 

Colour Figure 1 Vancouver robberies — Jeopardy surface

Chapter 10 

Colour Figure 2 Vancouver robberies — GeoProfile

Figure 10.3 

GeoProfile confidence intervals**

Figure 10.4 

CGT score distribution

Figure 10.5 

CGT operational performance

Figure 10.6 

CGT model learning curve

Figure 10.7 

Understudy training program

Figure 11.1 

Lafayette South Side Rapist — Crime sites

Chapter 11 

Colour Figure 1 Lafayette South Side Rapist — Jeopardy surface

Chapter 11 

Colour Figure 2 Lafayette South Side Rapist — GeoProfile

*

From Rossmo, D.K. (1995b). Multivariate spatial profiles as a tool in crime investigation. In C.R. Block, 

M. Dabdoub, & S. Fregly (Eds.). 

Crime analysis through computer mapping 

(pp. 65-97).

Washington, 

DC: Police Executive Research Forum. Used with permission.

**

From Rossmo, D.K. (1997). Geographic Profiling. In J.L. Jackson, & D.A. Bekerian (Eds.). (1997b). 

Offender profiling: Theory, research and practice 

(pp. 159-175). Chichester: John Wiley & Sons. Used 

with permission.

© 2000 by CRC Press LLC

Quotation

It is quite a three-pipe problem.

— Sherlock Holmes, in 

The Red-Headed League

,

Sir Arthur Conan Doyle, 1891

© 2000 by CRC Press LLC

Introduction

Interview the subjects, what they’ll tell you is, the thing that was really
appealing to them was the hunt, the hunt and trying to look for the
vulnerable victim.

— FBI Special Agent John Douglas; Mind of a Serial Killer, 1992, p. 3

This book is about geographic profiling. It is also about serial crime and
violent predators. Human hunters are not common, but when they do strike,
the public and the criminal justice system are significantly affected. Beyond
the violence and tragedy of the crimes, these offenders generate tremendous
fear in the community and demand significant resources from police, courts,
and prisons. Most homicides and rapes are solved because there is a connec-
tion between the offender and the victim. Such a nexus is lacking in cases of
stranger crime, and its investigation involves sifting through hundreds of
suspects and thousands of tips. Consequently, police suffer from the problem
of information overload. If we wish to enhance the investigative response to
this type of random violence, it is important to expand our knowledge of
serial predators and their hunting behaviour.

Geographic profiling is an investigative methodology that uses the loca-

tions of a connected series of crime to determine the most probable area of
offender residence. It is applied in cases of serial murder, rape, arson, robbery,
and bombing, though it can be used in single crimes that involve multiple
scenes or other significant geographic characteristics. Developed from
research conducted at Simon Fraser University’s School of Criminology, the
methodology is based on a model that describes the criminal hunt. This book
examines and discusses the spatial patterns produced by the hunting behav-
iour and target locations of serial violent criminals. Hunting behaviour refers
to victim search and attack processes engaged in by an offender, and target

1

© 2000 by CRC Press LLC

locations are the various geographic sites connected to a crime series. Serial
murder, for example, includes victim encounter, attack, murder, and body
dump sites. The patterns and methods of offender hunting activity are ana-
lyzed from a geography of crime perspective. By establishing these patterns,
it is possible to outline, through analyzing the locations of the crimes, the
most probable area of offender residence.

The conceptual basis of this relationship is provided by the crime site

selection model of Brantingham and Brantingham, who observe that indi-
viduals, including criminals, do not move randomly through their environ-
ment. The research led to an algorithm for predicting offender residence
from crime site geography. The resulting computer system produces jeopardy
surfaces — three-dimensional probability surfaces that indicate the most
probable area of offender residence. These are displayed through the produc-
tion of colour isopleth maps that provide a focus for investigative efforts.

Geographic profiling can be used as the basis for several investigative

strategies, including suspect and tip prioritization, address-based searches of
police record systems, patrol saturation and surveillance, neighbourhood
canvasses and searches, DNA screening prioritization, Department of Motor
Vehicle searches, postal or zip code prioritization, and information request
mail-outs. It is important to stress that geographic profiling does not solve
cases, but rather provides a method for managing the large volume of infor-
mation typically generated in major crime investigations. It should be
regarded as one of several tools available to detectives, and is best employed
in conjunction with other police methods. Address information is an element
of most record systems, and geographic profiling can be applied in a variety
of contexts as a powerful decision support tool. Geographic crime patterns
are clues that, when properly decoded, can be used to point in the direction
of the offender.

For example, in the investigation of a series of over 20 rapes from 1988

to 1996 in St. Louis, Missouri, Detective Mark Kennedy employed both
psychological and geographic profiling to prioritize a list of some 90 suspects
for DNA testing. In addition to a residential focus, the geoprofile drew atten-
tion to the St. Louis State Hospital, and to what appeared to be likely com-
muting routes used by the offender. When the Southside Rapist was identified
through DNA testing subsequent to a burglary arrest, it was found that he
moved several times during the crime series. The geoprofile identified his
residential area during his most active rape period — one home was in the
top 2% (0.4 mi

2

), and the other, across the street from the St. Louis State

Hospital, in the top 5.6% (1.2 mi

2

) of the hunting area.

The knowledge gained through research and experience of how and

where criminal predators hunt for victims has both practical and theoretical
implications. Geographic profiling is now an investigative support service

© 2000 by CRC Press LLC

offered to law enforcement agencies in cases of serious violent crime. It has
been used by police departments in North America, Europe, and Australia.
An ongoing study of the spatial patterns of criminal offenders is now occur-
ring in several countries.

This book is designed to be a reference work on geographic profiling for

both police practitioners and academic researchers. It covers investigative
initiatives, crimes, and research studies in Canada, the U.S., and Britain. Its
outline follows an order progressing from problem through to solution.
Chapter 1 is an introduction. Chapter 2 discusses serial murder and the
relevance of geography and distance for child murder and homicide investi-
gation. Research on serial rape and arson is covered in Chapter 3. The general
focus is on those aspects of serial crime related to offender hunting, crime
location, movement, and geography. This discussion provides a context for
the investigative problems and forensic behavioural science-based strategies
brought up in Chapter 4. One of those strategies, criminal profiling, warrants
its own examination in Chapter 5. Chapter 6 moves into the area of behav-
ioural geography, and introduces concepts of importance for understanding
the geography of crime research and theories presented in Chapter 7. Influ-
ences on criminal targets, including the hunt for victims, are discussed in
Chapter 8.

Chapter 9 analyzes the spatial patterns of criminal predators, presenting

original research findings from the study on the geography and targets of
serial murderers conducted at Simon Fraser University. Chapter 10 focuses
on the conceptual basis of geographic profiling, performance measures, rel-
evant considerations, operational procedures, and the understudy training
program. Chapter 11 presents several investigative strategies used with a
geographic profile. Case examples illustrate these tactics, but it should be
stressed that profiling only plays a support role. It does not solve crimes —
that is the responsibility of the assigned investigator. There is also a discussion
of a geographic profile based on the 19

th

-century murders of Jack the Ripper.

Chapter 12, the conclusion, considers future research and where we need to
go from here. The appendices contain research data and coding forms. A
glossary explaining specialized terms and a comprehensive bibliography close
the book.

One final note: depending upon the specification in the original source,

both kilometres and miles are quoted in distance measures.

1

1 

The relevant distance conversions are:  1 kilometre = 0.621 miles, 1 mile = 1.609 kilo-

metres; 1 metre = 1.094 yards, 1 yard = 0.914 metres; 1 centimetre = 0.394 inches, 1 inch
= 2.54 centimetres; 1 hectare = 2.471 acres, 1 acre = 0.405 hectares; 1 square kilometre =
0.39 square miles, 1 square mile = 2.59 square kilometres.