Newsletter

Volume 7, No. 2, November 2001

Editors: Teresa Alpuim and Bronwyn Harch

In This Issue:

1. A Message from the President

2. TIES News

3. Environmetrics Conferences.

4. Environmetrics Forum

5. Research Projects and Programmes

6. Recently Published Books

7. Book Reviews

8. TIES Board of Directors
 
 

1. A Message from the President,

Sylvia R. Esterby

The success of TIES 2001 conference is recorded in this issue of the Newsletter through the collection of photographs the editors have gathered, the conference reports and the list of new members, many of whom joined at the conference. I would like to both thank and congratulate Tony Olsen and colleagues who took care of the local arrangements and communications and the

Scientific Program Committee for the very successfulconference in Portland. The support of the many sponsoring organizations and societies for TIES 2001 was also very important for a successful conference. In particular, we acknowledge the substantial support of the University of Washington National Research Center for Statistics and Environment, the US Environmental Protection Agency NHEERL Western Ecology Division, the US Geological Survey National Water Quality Assessment Program and the Oregon State University Department of Statistics. We also thank Western Ecosystems Technology Inc., Palisade Corporation, John Wiley & Sons and CRC Press for their valued contributions in a variety of forms. Congratulations to the Student Award winners, Samantha Bates and Eduardo Severino, and to all of the students who participated in the competition. It bodes well for environmetrics to have so many students working on environmental problems and presenting their work at the conference. I would also like to acknowledge the conference participants who, in the end, made the meeting a success, and I hope that many of you will be able to attend TIES 2002 in Genova.

Two new features at this conference were the short course given by Brian Manly and the President's Invited Lecture by Agnes Herzberg. A summary of the Hunter's Lecture delivered by Paul Switzer is included in this issue and you will be able to read the full papers from both lectures in the Environmetrics journal in the near future. These lectures gave us much to think about and much to do, if we take up the challenges still to be met in developing environmetrical methods and the means of influencing policy decision. The sessions on microbiological problems in water quality assessment were in honour of Wesley Pipes, a founding member of the TIES Board of Directors, for his contributions to this field.

Welcome to the new TIES members. We hope that you will continue to find ways to participate in Society activities.

Turning to ongoing activities, some of these are motivated by the recognition that benefits of the Society could be more fully extended to students and internationally with a larger membership base. The Board of Directors has undertaken several initiatives to work towards achieving this goal. We have added more payment options for regular members, and added to the options for individual membership entitlement provided with an institutional membership (Section 2.3 below). I would like to bring the latter to the attention of Universities since a number of students could be provided with memberships in addition to one regular member who would receive Environmetrics. The preferential conference fees for TIES members, put into effect for the 2001 conference, are being adopted for future conferences and documented in the Conference Handbook.

A membership committee has been established (Section 2.4) and, to extend collaboration with other societies, a Liaison and Outreach Committee has also been established. To increase visibility, the brochure, which was initiated by Eric Smith, is being updated. It was recognized that to have a current and more informative web page a substantial commitment of time would be needed. Paul Sampson has agreed to take on this task, and the page, which has been hosted by the National Water Research Institute, Canada, for about 6 years, has been moved to the University of Washington. I wish to express appreciation for this support from the National Water Research Institute. The TIES web page will now be hosted by the University of Washington National Research Center for Statistics and Environment at

Ali Hadi continues to maintain TIES-L through its Cornell location.

I introduced a draft of a TIES handbook for the Board of Directors to aid in smooth transition between boards and to ensure that procedures, policies, etc were adequately documented. The Board will continue working on this, as well as the Handbook for Conferences initiated by Ray Correll, with Bronwyn Harch.

The announcement for TIES 2002 in Genova is in this issue and the invitation to participate has been extended. Note the web page address given in the announcement since the page will be regularly updated as information becomes available.

Finally, may 2002 bring peace for all of us.

Sylvia Esterby (sresterby@okanagan.bc.ca)
 
 

2. TIES News

2.1. New Members,

Francis Philbert

Welcome to the 71 new members who have joined since May 2001. 59 of these new members joined through the registration process for, or during, the 2001 TIES conference in Portland, Oregon. Conferences, the web page and promotion of TIES by current members continue to be the major means by which individuals are learning about the Society.
 

2.2. Member's News

The following three TIES members were recently granted Fellow Awards from the American Statistical Association:

Peter Guttorp, Professor of Statistics, University of Washington: For major contributions to the growth of environmetrics; for research on spatial modeling under nonstationary spatial covariance; for administration of interdisciplinary research groups, especially as Director of the National Research Center for Statistics and the Environment; for service to the profession.

Eric P. Smith, Professor of Statistics, Virginia Polytechnic Institute and State University: For innovative statistical applications in ecology and environmental statistics; for contributions to statistical education and consulting; for editorial service.

Dale L. Zimmerman, Professor of Statistics, University of Iowa: For outstanding research in spatial statistics, antedependence error structures and associated computational procedures; for teaching; for editorial service.
 
 

David Fox: Environmental Statistician, cat herder, and aviator

Congratulations to David Fox (Perth, Australia), a member of TIES who was recently appointed as a Program Leader of the Sustainable Catchment and Groundwater Management Program in the CSIRO division of Land & Water,

A significant thrust of this group's research effort is devoted to solving Australia's biggest environmental threat - dryland salinity,

David's interest in Environmetrics goes back 10 years after he returned from the United States to join CSIRO in Australia. David refocussed and expanded the small biometrics group so that it became much more customer focussed and responsive to a wider clientele both within CSIRO and other agencies. In 1994 David was invited to join the scientific technical group of an ambitious CSIRO project known as the Port Phillip Bay Environmental Study. This AUS$12 million, 4 year study has been hailed as a landmark environmental project in Australia for which David and the rest of the scientific team were awarded CSIRO's most prestigious award - The Chairman's Gold Meda

The Port Phillip Bay Study launched David's environmetrics career and shortly after its completion he found himself working in the remote highlands of Papua New Guinea initially as the statistician on an international scientific team engaged to audit environmental performance for a mining venture and later as independent statistical consultant for Ok Tedi Mining Limited.

Around this time CSIRO was restructuring and David used the opportunity to establish the Environmetrics Project in the newly formed division of Mathematical and Information Sciences. After about 2 years as Project Leader of the Environmetrics group, David was asked to manage CSIRO's Environmental Projects Office (EPO). The EPO is a 'shopfront' to the combined skills of CSIRO's environmental scientists whose aim is to seek, broker, and manage major, multidisciplinary studies for the organisation.

More recently David has played a pivotal role in providing technical advice to Environment Australia in developing the nation's new National Water Quality Management Strategy. David was responsible for assessing the statistical properties of new water quality protocols and also wrote a chapter on statistical methods for the companion 'Monitoring and Reporting Guidelines' document

(http://www.ea.gov.au/water/quality/nwqms/pubs/mg-ch6.pdf).

When he's not travelling on commercial aircraft, David can be found flying his own Cessna over the Western Australian landscape. Here he is after successfully gaining his Private Pilot's license last year (David is the one on the right!).

2.3. Society News

The membership committee for TIES was recently formed and charged with developing strategies for membership growth. The 2001-2002 committee consists of Peter Guttorp, USA (president-elect of TIES), Eric Smith, USA (representing the North American region), Thomas Polfeldt, Sweden (representing the European region) and Bronwyn Harch, Australia (representing Other Countries).

Among the specific tasks for this year are to try to get more institutional members, and preparing a proposal to the next annual meeting on whether the TIES membership list should be made available on the web. Suggestions with respect to these or other membership issues can be sent to Peter Guttorp at e-address:

Additional TIES Membership Classification Introduced, by Francis Philbert

The TIES Board of Directors, at its meeting in Portland, Oregon, in August, decided to add the following new individual membership classifications within the existing categories, effective in the 2002 membership year. The Institutional membership category has also been modified as outlined below.

3. Environmetrics Conferences

3.1. TIES 2001 Conference Report

Scientific Program,

Mohammed El-Saidi, Ferris State University, USA

The International Environmetrics Society (TIES) held its annual conference in Portland, Oregon, USA, August 13-17, 2001. The theme of the conference was "Environmetrics for Decision Making". A total of 136 registered and attended this international conference. Attendees came from 16 countries and many disciplines, with 40 attendees from outside the USA. It was great to see 15 students among the attendees. Many of the first time attendees chose to become members of TIES.

The conference topics included, but were not limited to, contributed and invited sessions on Meteorological Environmetrics, Modeling Environmental Systems, Air Quality Monitoring and Assessment, Environmental Standards, Environmental Survey Design and Analysis, Chemometrics, Human Health Environmetrics, Geostatistics and Spatial Data, Normalization of Air Pollution Data, Change-Point Methods for Environmental Data Analysis, Space-Time Modeling of Disease and Exposure Data, Aquatic Environmetrics, Change Detection and Signal Extraction, Stochastic Precipitation Models, Landscape and Ecological Applications, Water Quality and Fish Monitoring, Sampling for Recreational Water Quality Assessment, and Environmetricians as Frontline Researchers.

Paul Switzer gave the J. Stuart Hunter Lecture, and Agnes Herzberg was the Inaugural President's Invited Lecturer - a new lectureship this year. Sixteen participants attended the Computer Intensive Statistics short course by Bryan Manly. Forty-one joined the Columbia Gorge tour and 12 participated in the Willamette Valley Wine Tour. Ninety-three attended the conference banquet held on the Portland Spirit cruise ship on the Willamette River. At the banquet past J. Stuart Hunter Lecturers were presented with a plaque.

Best Student Presentation awards were given to Samantha Bates, University of Washington, for her paper "Bayesian uncertainty assessment in multi-compartment deterministic simulation models for environmental risk assessment" and to Eduardo Severino, Universidade de Lisboa, Portugal, for his paper "Area rainfall estimates through the use of space-time statistical models". Each received a certificate and a cash prize of US$250. The awards were sponsored by Western EcoSystems Technology, Inc, of Cheyenne, Wyoming.

Samantha Bates also won the award for the Best Risk Assessment Paper at the conference. The award was sponsored by the Palisades Corporation and consisted of a copy of their risk assessment software program @Risk^TM.

A random drawing of tickets was held during the conference banquet to give out prizes donated by conference sponsors: 10 t-shirts, 2 books, and two $35 Amazon.com gift certificates by Insightful Corp - Makers of S-PLUS; 10 books by John Wiley & Sons; two books by CRC Press. John Wiley & Sons also donated a copy of the new Encyclopedia of Environmetrics, which was won by Lotta Hallberg from Sweden.

Social Program

Columbia Gorge Tour,

Sarah L. Hession,

The Columbia Gorge Tour was excellent! We started with a hike up, up, and UP a forested hillside on a trail that was chipped out of rock during the depression years by the Civilian Conservation Corps. Portions of the trail skirted along cliffs overlooking the Columbia River Valley. After passing two waterfalls (and walking under one!), we hiked down and entered the Oneonta Gorge. Several hikers hiked over challenging terrain deep into the Gorge and had some fun in the water. We had lunch on board the Columbia Sternwheeler while cruising on the Columbia River. We listened to tales of the early settlers attempting to cross the river, sometimes unsuccessfully. This was quite a different picture than what we saw later on that day - a recreational activity called "kite-boarding" in which the kite-boarders sailed across the river and into the air propelled by their kites!

After our cruise, we stopped at the Bonneville Hatchery and Dam for a guided tour. We learned about the large system of dams and the generation of electricity. We saw many fish in the hatching and rearing ponds, including one sturgeon that must have been 8 feet long! All in all, it was a wonderful day. The activities were great and the weather was perfect! Most of all, it provided us with the opportunity to talk in a casual setting and get to know one another. The only bad part of the day was that it had to come to an end. A special thanks again to Don Stevens for coordinating this exceptional day.
 
 

3.2. Forthcoming TIES Conferences

TIES 2002, Genoa, Italy,

Daniela Cocchi

The TIES 2002 conference will be held in Genoa, Italy, June 18 - 22, 2002. All conference activities will be at the Faculty of Engineering of the Genoa University. Genoa is the provincial capital of the Liguria region, in the north-west of Italy; it is located at the farthest inmost part of the Gulf of Genoa, along the foothills of the Apennines. Artistically speaking, the liveliest periods for Genoa were the Middle Ages. The oldest part of the urban centre is a distinctive maze of crowded and narrow streets leading to the old port.

Sessions will include (but will not be restricted to):

Abstracts for contributed papers should be submitted by the corresponding authors (or co-authors) who wish to deliver an oral (or poster) presentation. Each abstract should not be longer than 300 words. Abstracts may be submitted by completing the Abstract form on the conference web page (http://www2.stat.unibo.it/ties2002). The deadline for submission of Abstracts by participants is April 30, 2002. The Scientific Program Committee reserves the right to considers abstracts submitted after April 30, 2002, but only if received by June 1, 2002.

The STUDENT PAPER COMPETITION committee will accept student papers until April 30, 2002. This competition includes a cash award of US$500 for the winner. To be eligible the paper must represent work accomplished as a student and be presented at the TIES 2002 conference by the student. The research must have relevance to environmetrics. The paper may have co-authors, but the student should be the principal author. Papers will be judged during the conference and winner announced at the end of the conference. Please encourage students to submit papers following the instructions on the conference web site.

A short course will be given during the conference, on Sunday - Monday June 16 - 17, 2002. The main topic of the course is "The Estimation of Human Impact on the Environment. The course will be organized by Anders Grimvall (Linköping University, Sweden), coordinator of the EU project IMPACT (Estimation of human impact in the presence of natural fluctuations). The course presenters will be: Peter Guttorp (University of Washington, Seattle, USA), Hans Wackernagel (Ecoles de Mines de Paris, Fontainebleau, France), Andrea Saltelli (Joint Research Centre, Ispra, Italy) and Anders Grimvall (Linköping University, Sweden).

Conference registration and accommodation information can be found at
 

Registration fees are:

Before March 31  After March 31

TIES members 300  Euro 380 Euro

Other Participants 350  Euro 430 Euro

New TIES Members
with Environmetrics J. 425   Euro 505 Euro

New TIES Members (students)
with Environmetrics J. 225 Euro 305 Euro

New TIES Members
without Environmetrics J. 325 Euro 405 Euro

New TIES Members (students)
with Environmetrics J. 115 Euro 195 Euro

Members of the Organizing Committee are:

3.3. Other Forthcoming Conferences

The first Biennial meeting of the International Environmental Modelling and Software Society will be held in Lugano, Switzerland from June 24-27, 2002, http://www.iemss.org/iemss2002. The Society (iEMSs) is an umbrella group linking various interests including those working on modelling, software systems, multiobjective decision support and innovative methods such as artificial intelligence. The theme of the Lugano meeting is "Integrated Assessment and Decision Support".

The topics covered by the conference include:

The Workshop on Statistical Monitoring for Environmental Engineering will be held 14 January 2002 at the Bergamo University, Italy.

This one-day long conference is organized by the Statistics Section of the Engineering Department and will focus on environmental data analysis and monitoring. Statistical methods and applications related to air and water quality problems will be considered together with network monitoring systems and engineering. Talks are planned both from local and European universities and from environmental agencies and institutions. Case studies and models will be concerned mainly with local scale problems and a part of them will focus on Bergamo County.

Topics: Environmental time series, Environmental risk assessment, Information dissemination through internet, Normalization of air quality data, Sensitivity Analysis, Space-time models, Statistical mapping, Statistical modelling and forecasting, Tropospheric ozone.

Languages: Italian and English.

Further information are available on http://dipinge.unibg.it/seminari or Contact: Prof. Alessandro Fassò (organizer), email: fasso@unibg.it.
 
 

The International Workshop, Perspectives in Modern Statistical Inference II, will be held from 14 to 17 August in Brno, the center of Moravian Region. It is organized jointly by the Department of Probability and Mathematical Statistics of Charles University in Prague, the Department of Applied Mathematics of Masaryk University in Brno and the Union of Czech Mathematicians and Physicists.

The workshop is a satellite to the 24^th European Meeting of Statisticians, taking place from 19 to 23 August in Prague and follows the successful First Workshop on Perspectives in Modern Statistical Inference: Parametrics, Semiparametrics and Non parametrics, organized in Prague in 1998.

The workshop will focus on recent results in parametric, semiparametric and nonparametric statistical inference presented by outstanding specialists in the area, namely, P. Chaudri (India), M. Csörgo (Canada), P. Deheuvels (France), P. Greenwood (Canada and USA.), L. Horváth (USA), A.M. Kagan (USA), C.A.J. Klaassen (The Netherlands), R. Koenker (USA), B. Levit (Canada), D.M. Mason (USA), H. Oja (Finland), S. Portnoy (USA), P.K. Sen (USA) and J.A. Wellner (USA).

For more informations about the workshop contact the local organizer: Ivana Horová, Department of Applied Mathematics, Janáckovo nám. 2a, CZ - 662 95 Brno, Ccech Republic. E-mail: horova@math.muni.cz. Fax: +420 5 4121 0337. Phone: +420 5 4132 1251.
 
 

The Seventh Valencia International Meeting on Bayesian Statistics, sponsored by the Universitat de Valencia and the International Society for Bayesian Analysis (ISBA) will take place from June 2nd to June 6th, 2002, inclusive.

As on previous occasions it will be a residential conference in a coastal resort: Mare Nostrum at Playa de las Americas, on the South Coast of Tenerife, one of the Spanish Canary Islands; you can see the resort virtually at

The V7 Programme Committee consists of Susie Bayarri, Jim Berger, Phil Dawid, David Heckerman, Adrian Smith, Mike West and Jose Bernardo, who also acts as Local Organizer. The program will include one stream of 25 invited lectures followed by invited discussions, a second stream of 50 contributed talks, and (it is expected) over 150 contributed papers in poster form. As in previous Valencia meetings, contributed papers will mainly be presented in the very popular plenary evening poster sessions. The scientific program of the conference will also include, Selected Contributed Papers, to be presented in oral sessions. The Conference will be preceded by a one-day postgraduate tutorial seminar on Basic Concepts in Bayesian Statistics. This will consist of five 90 minute lectures given by members of the Organizing Committee.

For further information, including the complete invited program, costs and travel details, see the conference web site, http://www.uv.es/valencia7, or its US mirror site,

If you may be interested in attending Valencia 7, or if you just want to be included in the conference mailing list and did not attend Valencia 6 (or have moved since Valencia 6), please send to the local organizer, jose.m.bernardo@uv.es, the following information: Name; Affiliation; Postal address; Telephone; Fax; E-mail address; Web page; Area(s) of interest within Bayesian Statistics.
 
 

3.4. Reports on Related Conferences

International Conference on Statistical Challenges in Environmental Health Problems,

Abdel El-Shaarawi, CCIW, Canada

The conference was held August 30-September 1, 2001 in Fukuoka Soft Research Park, Fukuoka, Japan. It was organized by the Biometric Society of Japan as a Satellite to the International Statistical Institute, 53^rd Biennial Session in Seoul, Korea, August 22-29. The conference was sponsored by The International Environmetrics Society (TIES) and its proceedings will be published as a special issue of Environmetrics. The president of the conference was Professor Yoshimura, Isao, the International Organizing Committee was chaired by Professor Yanagawa and the International Program Committee was chaired by Professor Walter W. Piegorsch, a long time member and supporter of TIES.

Approximately 70 researchers attended the conference. The scientific program consisted of 31 presentations, 16 of which were invited talks by well-known researchers in the field including John Bailer, P.K. Sen and Walter Piegrosch from the United States; Sylvia Richardson from United Kingdom; and L. Edler from Germany. Several invited speakers from Japan gave a nice sample of quantitative environmental research in their country presented in a single session-stream. The topics covered ranged from defining and evaluating the impact of exposure to environmental pollutants to methodological challenges in dealing with the design and analysis of environmental studies particularly the issue of setting safety standards and regulations. Many speakers emphasized the use of Baysian hierarchical models for integrating spatio-temporal models and data particularly in ecological and health studies. There was ample time for discussions and interactions during both the scientific and social components of the conference.
 
 

4. Environmetrics Forum

Introduction

Reflecting back on my several decades of more or less continuous involvement with environmental statistics, in this newsletter summary I will focus on examples from three topic areas that, I believe, have not received sufficient attention from the regulatory community and that reflect some of my personal interests. They are: sampling frequency and variability in relation to environmental standards, combining multi-site information for estimation of environmental trends, and variability of human exposure to pollutants and its implications for modeling short term health effects of ambient air pollution.

Sampling Frequency and Variability in Relation to Environmental Standards

As a first example, consider an emission source that is not constant. On some days the combination of high emissions and unfavorable weather will cause an ambient threshold to be exceeded. Both the variability of emissions and the variability of weather contribute to the frequency of exceedances. One can view this combination of daily fluctuations via a hierarchical model. On day t the ambient regulatory threshold will be exceeded if emissions exceed C_t. C_t varies from day to day because of weather variability, but its value is uncertain. The uncertainty in C_t is modeled by a gamma distribution, say, whose mean and variance are denoted m_t and am_t, respectively. m_t may be interpreted as ouput of a dispersion model with model uncertainty represented by the parameter a. The temporal variation of m_t, induced by temporal weather variation, is represented by another gamma distribution, say, with mean and variance, m and b_m, respectively. The parameter b derives from the day to day weather variability. Finally, the day to day source emissions variability is represented, at high levels, by an exponential tail distribution. In this way the frequency of ambient threshold exceedances can be related to parameters that represent modeled weather induced dispersion, dispersion model uncertainty, and emissions variability.

Some Statistically Challenging Problems In Environmental Regulation, Paul Switzer (continued from page 9)

The second example deals with sampling frequency. The more frequently we measure ambient pollution, the more likely it is that one of our measurements will exceed a fixed threshold value. To compensate for the frequency of sampling, one could make the threshold depend on the sampling frequency. To illustrate a calculation, suppose that D_k denotes the difference between the annual maximum of daily values and the annual maximum of values measured every k-th day. Using the statistical theory of extreme values, we can compute the probability distribution of D_k and compare annual exceedance probabilities as a function of monitoring frequency. For example, let d be the difference between the annual maximum of daily values and the annual maximum for k=6, then d will exceed the standard deviation of the daily values with 80% probability.

One of the problems with calculations of the foregoing kind is that they do not explicitly account for seasonality. Indeed, statistical models for annual extremes of seasonally varying pollutants require more subtle, but nonetheless important, considerations. If the seasonality is ignored and one works with pollutant distribution functions obtained from data across seasons, then exceedance rates will be underestimated. An alternative is to use so-called Lehmann distribution function models of the type F(x)^b(t) where x is concentration, F is a 'core' distribution function, and b(t) is a time-varying modulating exponent that could reflect seasonal patterns. Such models are relatively easy to manipulate in extreme value statistics.

In a vein similar to accounting for increased sampling frequency, adding more monitor stations in an air basin can only increase the probability that some station will record values that exceed a given threshold. Hence, there is a need to account for the density and placement of monitoring stations in relation to basin-wide monitoring thresholds. I hope and expect that statisticians will take up the challenge. Spatial autocorrelation of monitoring sites will play a role, as well as spatial nonstationarity that reflects geographic patterns of pollutant sources and air circulation.

Time Trends

The ability to discern and describe time trends in environmental quality is very much influenced by both sampling design and modeling choices. The first example here involves the combination of information from multiple monitoring sites. If there are sufficient data one might build a time-space variation model. A simpler approach is to estimate a linear time trend separately at each monitoring site and treat the estimated trend parameter as a spatial datum. Then we build a spatial model for the geographically varying time trend. The spatial model of the time trend parameter is then used to estimate a regional average time trend together with a measure of its uncertainty. There are several limitations to this approach: imposing a linear trend model at each site might be too restrictive, and the number of monitoring sites might be too few to build a spatial model for the trend coefficient.

An alternative is to look for a synthetic time series that linearly combines the information from each of the monitoring sites in a way that maximizes the temporal coherence of the synthetic series. An example of this basically nonparametric approach looks for a combined times series that maximizes temporal autocorrelation. Maximizing temporal autocorrelation is equivalent under certain models to maximizing a signal to noise ratio. The synthetic series will be smoother than its constituents and should extract the temporally coherent trend.

A simpler problem involving regional time trends is to estimate the change in average environmental quality by comparing survey data from the region obtained at two different times. But care must be exercised if the measurement locations are not the same for the two surveys. For example, if a resurvey of a contaminated region uses sampling locations that selectively correspond to the high contamination sites of the first survey, then an unadjusted estimate of regional change will be exaggerated because of regression to the mean. Care should be taken in the design of a resurvey to avoid this kind of selection bias or else to compensate in the analysis using an adjustment for site selection.

Human Exposure

While we monitor and regulate ambient air quality, the basis of such regulation lies in studies of human health effects. Ambient monitoring data are used as a surrogate for human exposure in air pollution epidemiology. Finally, the community health consequences of changes in ambient air pollution need to be estimated from the epidemiological models.

Epidemiological studies could, in principle, have greater power and less systematic error if we could get more specific surrogates for individual exposures. To this end, air quality data specific to indoor, vehicle, workplace, and outdoor microenvironments for example, are being combined with community time-activity survey data and time-averaging models to obtain estimates of community exposure distributions. For some pollutants changes in ambient air quality are reflected differently in different microenvironments, and we will see non-proportional effects of ambient pollution on human exposure.

Even if ambient data were a reasonable proxy for mean community exposure, the implied exposure-response function would have systematic error if the health effects on the individual level were not proportional to exposure. Finally, if individual susceptibility to health effects is correlated with individual exposure, then once again systematic errors are introduced in exposure-response analyses based on ambient pollution data. Further research is needed to characterize the relation between exposure and ambient pollution measurements.

A final example deals with the relation between the variability of ambient pollutant concentrations and the corresponding variability of indoor concentrations of the same pollutant. Such indoor-outdoor relationships are important for human exposure studies because people spend almost all their time indoors. To illustrate, we can have a parsimonious model for short-term indoor variability that depends only on the variability and temporal autocorrelation of outdoor concentrations and the indoor-outdoor air exchange rate. Such models, based on rapid indoor mixing, and are derived from simple mass balance considerations. These models can also be used to infer how variability of pollutant concentrations both indoors and outdoors will depend on the averaging time, autocorrelation statistics, and air exchange rates. Such calculations are important for comparing the stringency of air quality standards based on different averaging times. They are also useful in exposure simulation studies for assigning exposure concentrations whose variability depends on exposure duration.
 
 

The editors apologize for publishing in error an incomplete list of references for the article "Open Problems in Stochastic Modelling of Wind", by E. Kestens and J.L. Teugels, which appeared in the last May issue of the TIES Newsletter, vol. 7, no. 1. Transcribed below is the complete list of references.

References

 5. Research Projects and Programmes,

The United States Environmental Protection Agency recently funded two substantial Programs in environmental statistics oriented toward aquatic resources, one based at Oregon State University and one at Colorado State University. The programs have similar goals, but take different methodological approaches. Oregon State University's Program is oriented toward design-based/model-assisted approaches, while the Program at Colorado State University is oriented toward model-based approaches. The two Principal Investigators (Scott Urquhart at CSU and Don Stevens at OSU) are each partially funded on the other Program, and will ensure close coordination between the two Programs. In addition, joint annual meetings are planned to share results. Some details of the programs are described in the accompanying articles.
 
 

United States National Research Program on Design-Based/Model-Assisted Survey Methodology for Aquatic Resources, Don Stevens, Oregon State University, USA

The US Environmental Protection Agency has funded a research program on survey design and analysis methodology to enhance the use of auxiliary information and to develop design-based model-assisted methodology compatible with multi-tiered studies. This Program will cooperate closely with a companion Program on model-based approaches at Colorado State University. A major goal is to provide expert survey design and analysis assistance to States and Tribes; and to transfer that statistical expertise to States and Tribes through a combination of supervised application of statistical tools and structured distance learning techniques. Several doctoral students will be funded by the Program, so applicants interested in careers environmental statistics are encouraged to inquire about opportunities.

Three state agencies (Oregon Department of Fish and Wildlife (ODFW), California State Water Resources Control Board (CSWRCB), and the San Francisco Estuary Institute(SFEI)) have a special role. The three programs represent a microcosm that spans almost all of the circumstances that state or local aquatic monitoring efforts are likely to encounter. These agencies will identify areas where our statistical expertise will be especially beneficial and act as "laboratories" for testing distance learning methods. The interaction between these three agencies and this Program has the potential to develop into model archetypes for state and local level monitoring programs. ODFW initiated a survey of Coho salmon in Oregon streams in 1998 using a rotating panel design and a spatially-balanced sample selection procedure. The sampling is expected to continue indefinitely using the same design, with the objective of estimating both annual escapement of coho salmon, and trends in the escapement. The sample has nested subsamples of juvenile counts and habitat quality, with landscape level information available on several variables, such as vegetation cover, stream order, and basin size. Both spatial and temporal correlation are present. In addition, there is substantial non-response, which may be related to such information as landholder type and size of holding. Thus, the data set provides fertile ground for developing model-assisted, design-based estimation techniques, such as trend estimation, imputation of missing data, small-area estimation, and a variety of model-assisted status estimation techniques. This is a critical, high-visibility issue in the Pacific Northwest because it involves a commercially and recreationally important species that is listed as threatened by the federal government.

The SFEI is presently concluding a re-design of the San Francisco Estuary Regional Monitoring Program for Trace Substances, and is initiating the design of a Bay Area Wetlands Regional Monitoring Program. This design effort will provide us with the opportunity to investigate some model-assisted design approaches on a relatively small scale. The regional monitoring conducted by SFEI is an example of a state-supported estuarine monitoring program encompassing the estuary itself and its associated wetlands.

The third archetype program is the Surface Waters Ambient Monitoring Program (SWAMP) being initiated by CSWRCB. This will be a state-wide program with multiple objectives, which include an estimate of the proportion of the resource in nominal or degraded condition. This is the first state-funded project of this scale that may use a rigorous, probability-based sampling design to assess surface water quality. Its success may encourage other states to follow.

The research will be carried out at Oregon State University (OSU), Colorado State University (CSU); University of Washington (UW), and Iowa State University (ISU). The Principal Investigator is Don L. Stevens, Jr. (OSU). Other investigators include Virginia Lesser (OSU), Alix Gitelman (OSU), Paul Murtaugh (OSU), Robert Hughes (OSU), Alan Herlihy (OSU), Jean Sifneos (OSU), Breda Munoz-Hernández (OSU), Rubén Smith (OSU), N. Scott Urquhart (CSU), F. Jay Breidt (CSU), Jean Opsomer (ISU), Loveday Conquest (UW), and Jean-Yves Courbois (UW). For further information, contact Don L. Stevens , Jr. (stevens@stat.orst.edu) or N. Scott Urquhart (nsu@stat.colostate.edu).
 
 

United States National Research Program on Applying Spatial and Temporal Modeling of Statistical Surveys to Aquatic Resources, N. Scott Urquhart, Department of Statistics, Colorado State University, Fort Collins, CO, USA

The US Environmental Protection Agency (EPA) STAR program has funded a new research effort to develop model-based methodologies, including enhanced the use of landscape information, for the prediction of the quality of aquatic resources. This new research effort, called Spatial and Temporal Modeling of Aquatic Resources (STMAR) will be housed at the Department of Statistics at Colorado State University; it will cooperate closely with a companion program on design-based/model-assisted approaches at Oregon State University.

STMAR has three major goals: to conduct research and development on statistical modeling tools appropriate for the setting, to train future generations of environmental statisticians, and to engage in outreach to the States and Tribes. The research and development effort will be conducted under the auspices of three Projects: Combining Environmental Data Sets; Local inferences from Aquatic Studies; and Development and Evaluation of Aquatic Indicators. The third project, to be pursued mainly during the first two years of the Program using landscape ecology and associated tools, will be oriented mainly toward the development of variables to be used in the first two projects; a major evaluation criterion will be how well the variables perform as predictors in the first two Projects. The first two Projects will make use of a wide array of modeling techniques, including spatial statistical models and Bayesian methods.

Most inland aquatic resources (lakes, streams, rivers, wetlands, estuaries) share an important feature: The size distribution of each resource concentrates heavily on small objects, but societal interest spreads across all sizes. Straight random sampling would concentrate most of the samples on small objects. Consequently, probability-based monitoring programs such as EPA's Environmental Monitoring and Assessment Program (EMAP) often sample aquatic resource classes with probabilities very different from their frequency in the population. Remote sensing platforms such as satellites can provide complete coverage of many attributes of landscapes. This Program is designed to provide and develop statistical models that combine complete coverage data with data gathered from probability-based surveys of aquatic resources.

The provision of expert survey design and analysis assistance to States and Tribes will be a substantial focus of the Program's last two years. The plan is to transfer the needed statistical expertise through a combination of supervised application of statistical tools and structured distance learning techniques, especially using web-based-type learning materials.

STMAR will fund several postdoctoral fellows and several doctoral students each year. Applicants interested in careers environmental statistics are especially invited inquire about opportunities.

The research and development work will be carried out at Colorado State University (CSU), Oregon State University (OSU), the Southern California Coastal Water Research Project (SCCWRP, Westminster, CA) and by Water Quality Technology, Inc. (WQTI, Fort Collins, CO). N. Scott Urquhart (CSU) serves as Program Director. Other investigators include F. Jay Breidt (CSU), Richard A. Davis (CSU), Jennifer A. Hoeting (CSU), Hariharan Iyer (CSU), Jim C. Loftis (CSU), Robin Reicht (CSU), David Theobald (CSU), Alix Gitelman (OSU), Alan Herlihy (OSU), Don L. Stevens (OSU), Kerry J. Ritter (SCCWRP), Steven B. Weisberg (SCCWRP), Steve W. Johnson (WQTI), For further information, contact N. Scott Urquhart (nsu@stat.colostate.edu) or Don L. Stevens, Jr. (stevens@stat.orst.edu).

6. Recently Published Books

In this section of TIES Newsletter members are invited to provide the editors with details of any newly published books that they deem relevant for the readership.

7. Book Reviews

Liliana Gonzalez, Editor

I am very pleased we are able to include two book reviews in this issue of the Newsletter. Thank you Professor Joe Gani and Professor Eric Smith for your continuing support of the Society, the Newsletter and this section. I also wish to thank the book review managers Ms Doris Drechsler. of Springer Verlag at Heidelberg and Mr Robert Kennedy of Kluwer at Dordrecht for providing complementary copies of books to our reviewers.

The main objective of this section is to provide the members of our Society with book reviews that are relevant and of interest to all of us. Therefore, I invite you to contact me (liliana@cs.uri.edu) and suggest books you would like to review. Unfortunately in some cases it has been difficult to obtain complementary copies of books some of you have requested but I can assure you every effort had been and will be made to contact the editorial houses to obtain complementary copies for reviewing.
 
 

Encyclopedia Of Environmental Science

by David E. Alexander and Rhodes W. Fairbridge.

Editors: Kluwer Academic Publishers, Dordrecht, 1999. pp. xxx + 741, US$ 480, ISBN 0-412-74050-8

Reviewer: Joe Gani, School of Mathematical Sciences, Australian National University, Canberra ACT 0200, Australia.

This is a large and informative volume which covers the most relevant topics in Environmental Science. It consists of 374 contributions, written by 225 experts in various aspects of the field. Browsing through the work, one cannot avoid being struck by the richness of this multidisciplinary area, and the marked influence of the environment on practically every aspect of human life.

If one were to attempt a classification of the articles, one might perhaps agree that they consisted of:

The hydrological cycle is described, with its global balance between precipitation, surface retention, subsurface storage and atmospheric evaporation. Formulae expressing the balance are given. The uses of fresh water are described: by 1992, the world's average annual per capita consumption of renewable water resources was 7420 cubic meters, with 73% used for irrigation, 21% for industry and only 6% for domestic purposes. In some countries, renewable ground water from aquifers is being pumped to the surface, but this supply is expected to run out before very long.

Changes due to human interference have altered the quantity, quality and location of water resources. Over 150,000 square kilometers of wetlands have been drained worldwide, and the land reclaimed; canals have been constructed, and there are 2360 reservoirs in the world, storing 5525 cubic kilometers of water. Some of this is used to generate hydroelectric power; in 1982, 23% of all electric power was produced in this way. The rerouting of water, as for example, through the Farakka Dam in India, has often had unexpected detrimental results. Irrigation has also had negative consequences, as with the Aral Sea. The increasing dependence on groundwater reserves can also be detrimental, causing salt leaching, for example; however, such problems can be controlled by a careful approach to water use.

Health risks associated with water supplies are considered. There is a link between inadequate sanitation and infectious water-borne diseases such as cholera, malaria, onchocerciasis, schistosomiasis and typhoid. This is still a serious threat in developing countries, although such diseases have been largely brought under control in the developed countries.

Finally, the author discusses the management of water resources, and water conservation. The legal rights and implications connected with streams and groundwater resources are considered, as for example in the Tigris-Euphrates basin which borders Iraq, Turkey and Syria. Water shortage has led to renewed interest in the conservation of water. Metering of water usage, conservative methods of irrigation, and water harvesting all help to reduce water usage, thus ensuring that water can be shared more equitably among the peoples of the earth.

b) Agricultural impact on the environment, by H. Berend Tirion, pp. 10-12

In this article, the author concentrates on the negative impact of agriculture on the environment. He discusses the bringing of land into cultivation, the use and management of water, and the intensification, mechanization and industrialization of agriculture, as productivity is increased under the pressure of world population growth. The slash-and-burn technique of farming has led to the loss of forests in many developing countries, with eventual soil erosion and aridity. In the USA alone, roughly 90 million hectares have been affected by desertification.

Groundwater is often used to increase crop growth, but causes the water table to fall, the soil to compact, and subsidence to occur. This has happened in the San Joaquin Valley in California, where more than 13,000 square kilometers of land have been affected by subsidence. Salination is another problem which may occur in such circumstances.

Intensification, mechanization and industrialization of the farming process has diminished the use of farm waste, and increased the spread of fertilizers and pesticides. These developments have all had negative effects on the environment. They include the increased vulnerability of monocultures to pests and diseases, and excessive surpluses of manure which pollute groundwater, and are the source of ammonia emissions. The result of heavy ammonia emissions in certain parts of the Netherlands has been serious damage to local natural ecosystems. The author points out that other farm waste, such as chemicals and plastic packaging materials, may also enter the environment and accumulate in foodstuffs. With farming intensification has come a substantial increase in environmental pollution.

c) Pollution, Nature of, by John T. Tancredi, pp. 482-485

Pollution may be defined as the process of polluting, especially the contamination of soil, water or the atmosphere by the discharge of harmful substances. There are natural sources of pollution, such as volcanic activity, but the author concentrates mainly on pollution due to human beings. He mentions the pollution of lakes, the obscuring of natural landscape by tall buildings, and the noise pollution over the Grand Canyon. Pollution from plastics kills 2 million birds and 100,000 marine animals each year, while the depletion of the ozone layer over the Antarctic has caused an increase in the incidence of skin cancer, among other phenomena.

The introduction of exotic or foreign organisms into ecosystems has also caused pollution; an example is the introduction in the USA of phragmites, a plant used for thatching in English homes, which has now crowded out native American species. The use of chemicals to control such invasions only adds further pollutants to the system. The author states that most river systems suffer some degree of pollution: the Mississippi, for example, has several sources of chemical pollution from industrial plants along its length. But perhaps the worst case of chemical pollution was the methyl-mercury which caused Minamata disease, a result of its pollution of Japan's Minamata Bay in 1991.

Other cases of pollution mentioned are the spread of ash from Mt Pinatubo in the Philippines in 1991, the release of 80 million barrels of crude oil into the Persian Gulf during the Gulf War of 1991, the discharge of waste water into the Mediterranean Sea, and the damage to plant life from acid rain in Europe, China and India. Often, it takes a disaster to draw attention to the need for stricter regulations; such was the case at Bhopal in India, where 2300 people died and 50,000 were injured as a result of an accident in the Union Carbide Company's pesticide factory. Other incidents such as the reactor meltdown at Chernobyl, the oil spill from the Exxon Valdez, the debris in space of more than 3000 tons of material (one speck of which penetrated the Challenger front window in 1983), and the high concentration of PCBs in Lake Michigan are mentioned. All indicate that high pollution levels exist on a global scale; human beings must learn to control these in order to protect their support systems.

d) Sir Peter Markham Scott (1909-1989), by David F. Alexander, pp. 547-548

Peter Scott, artist and ornithologist, was the son of the famous Arctic explorer Captain R.F. Scott. He specialized in the painting of birds, having illustrated the "Handbook of British Birds", among other books. Bird watching and exploring were his main recreations, and sailing one of his passions - he won a bronze medal for sailing in the 1936 Olympics.

During the Second World War, he served with distinction on destroyers, and was awarded an MBE in 1942 and a DSC and Bar in 1943. He made 3 expeditions to the Antarctic and while filming at the South Pole, wrote that his father would have been greatly pleased at the enormous outgrowth of scientific endeavour in Antarctica. He was awarded the CBE in 1953 and knighted in 1973, becoming Honorary Chairman of the World Wildlife Fund International from 1985 until his death. He was the founder of the International Union for the Conservation of Nature and Natural Resources.

Despite his eminence, Scott was a man of the people. He supported and chaired numerous youth associations and sports clubs, and enjoyed explaining the wonders of nature on children's television programs. A list of over a dozen of his books concludes the article.

The Encyclopedia makes an extremely valuable contribution to environmental awareness; the authors have presented their material clearly and understandably, while the editors have made an excellent choice of topics for inclusion. I strongly recommend this book: it should be on the shelves of all libraries, and will satisfy all those who are curious about the scope of Environmental Science.
 
 

Multivariate Statistics for Wildlife and Ecology Research, by Kevin McGarigal, Sam Cushman and Susan Stafford.

Editors: Springer-Verlag, New York, 2000. XIII, pp 283, fig 57. US$49.95, ISBN 0-387-98642-1.

Reviewer: Eric Smith, Department of Statistics, Virginia Polytechnic Institute, Blacksburg, Virginia, 24061, USA.

Wildlife ecologists often collect large amounts of data that must be interpreted. Often this information is in the form of measurements on animals or the land that animals use. Students and researchers sometimes face a daunting task in the analysis of this type of information as there are not many texts focusing on the modeling and analysis data specific to wildlife studies. This text has as objectives the presentation of a practical guide to multivariate analysis, to provide wildlife students with a text for multivariate analysis and to provide a readable background on the methods of multivariate analysis. The material in the book is not new but represents information collected from other texts with a focus on applications to wildlife and ecological data. The level is oriented towards graduate students in wildlife and would require a basic knowledge of statistics.

The book is divided into six chapters. The first chapter provides an overview of multivariate analysis. The primary material is in the next four chapters that focus on the main multivariate techniques discussed: ordination methods, cluster analysis, discriminant analysis and canonical correlation analysis. The last chapter summarizes and compares the methods. In addition, there is a useful glossary of terms and acronyms. The book is well organized. Each of the four topic chapters consists of objectives and an overview of the topic. Then details of the method are provided followed by an example. Each chapter contains a list of references and a bibliography of papers that provide additional readings. The examples are drawn from real applications. Attention is paid to the assumptions required for the analysis and common methods for evaluation of the assumptions of the analysis.

The examples in the text are produced by the Statistical Analysis System (SAS). At the end of each chapter is a listing of the code used to generate the analysis and some comments. The usefulness of the code is limited by the lack of the data used for the analysis. It would be helpful if the authors had set up a web site containing the data along with the SAS code. Access to the data would be valuable for students who want to reproduce the results in the text. There is a description of what the code does but no annotation of the individual lines. Thus the material cannot be used by someone unfamiliar with SAS and would require some of the SAS manuals for actual implementation. A web site would also be valuable for providing updates as versions of SAS change.

While the level of the book is good for a student with a basic knowledge of statistics, the authors should have been more careful with the explanations and suggestions. Statements such as "cluster analysis is sensitive to outliers" (page 123) are misleading and require careful explanation to avoid confusion. Generally, outliers do not lead to problems with hierarchical methods as the outlier falls to the outside of the dendrogram. Problems might occur if the data are standardized prior to analysis (since the outlier affects mean and variance) or if the number of clusters is chosen prior to analysis. In the section on principal components analysis, the authors only discuss standardized scores. As there are cases where standardization is not used, this might lead to problems with understanding results.

While the material in the text covers enough information for a course in multivariate analysis there is additional material that should be included to supplement the text. There are numerous macro programs written in SAS that are helpful and would improve analyses (see for example, Michael Friendly's home page at

For example, for evaluating outliers, the authors suggest a univariate approach in which data are standardized and values outside 2.5 standard deviations flagged. There are improvements to this approach that involve observation deletion methods. The material on ordination primarily covers principal components analysis and factor analysis. While other methods are discussed, the depth of discussion is not adequate. To summarize, I would recommend the book as an addition to one's library but would be concerned about using it as the sole text in a course.
 
 

8. TIES Board of Directors

The following are the names of the elected members of TIES Board of Directors. All terms are from September 1, 2000, to August 31, 2002, except the 4-year terms of the regional directors.

President: Sylvia R. Esterby
(SREsterby@ouc.bc.ca)

President-Elect: Peter Guttorp
(peter@stat.washington.edu)

Secretary: Jari Walden
(jari.walden@fmi.fi)

Treasurer: Francis J. Philbert
(francis.philbert@ec.gc.ca)

Publications Officer: Teresa Alpuim
(talpuim@fc.ul.pt)

Regional Representatives:

North America:

Anthony R. Olsen
(Olsen.Tony@epamail.epa.gov)
(01/09/00 - 31/08/04)

Richard W. Katz
(rwk@ucar.edu)
(01/09/98 - 31/08/02)

Europe:

Daniela Cocchi
(cocchi@stat.unibo.it)
(01/09/00 - 31/08/04)

Anders Grimvall
(angri@mai.liu.se)
(01/09/98 - 31/08/02)

Other Regions:

Raymond L. Correll
(Ray.Correll@CMIS.CSIRO.AU)
(01/09/00 - 31/08/04)

Carlos Alberto de Bragança Pereira
(cpereira@ime.usp.br)
(01/09/98 - 31/08/02)
 

TIES Newsletter is a publication of the International Environmetrics Society (TIES). It is published semiannually, or whenever the need arises, by The International Environmetrics Society and distributed to TIES members as part of their annual dues. Contact:

Francis Philbert, francis.philbert@ec.gc.ca,

for questions regarding membership and other benefits.

The Editors would like to encourage TIES members to submit items for publication in the Newsletter. We would like to have a very comprehensive publication that is of interest to our members by including items such as members' and regional news, Environmetrics and related conferences, research projects and programmes, book reviews, letters to the editor and articles of general interest.

We would like to thank the members who responded to our call and contributed to this issue. It is our hope that the Newsletter will be a valuable platform for discussion and exchange of ideas among us. We will be happy to hear your views about the contents and style of this issue. We hope that you will be a reader as well as a contributor.