ESA STATISTICAL ECOLOGY SECTION

Spring 1994 Newsletter

Editor: Don Edwards, Department of Statistics, University of South Carolina, Columbia SC 29201. E-mail: edwards@mistered.stat.scarolina.edu on the INTERNET. Phone: (803)777-5073. Please forward relevant items of information and suggestions, preferably by email. These could include conference and workshop reports, announcements of upcoming meetings of interest to section members, software evaluations, etc.

From the Section Chair: This year's ESA symposium, planning next year's one, and teaching statistics.

As I indicated in the last section newsletter, sponsoring and organizing symposia is one of the major activities of the Statistical Ecology section. At this year's ESA Knoxville meeting, the section is sponsoring a symposium on "Incorporating spatial information into ecological studies". It will be held Tuesday afternoon, Aug. 9, from 1-5 pm. The first half of the symposium focuses on sampling, the second half on experimental design and analysis. The speakers and their titles are:

Bruce Milne, The necessity for sampling at multiple scales.

Tom Edwards, Gretchen Moisem, and Richard Cutler, Use of cluster sampling to assess map uncertainty in remotely-sensed vegetation maps: incorporation of a cost function.

John Porter and James Dooley, Spatial sampling designs for detecting animal movements.

Richard Schmoyer, Nonparametric tests for spatial autocorrelation in regression.

Pierre Dutilleul, Modified correlation analysis with autocorrelated and heteroscedastic sample data.

Philip Dixon, Using nearest-neighbor designs and analyses in ecological experiments.

Now that the organization for this year's symposium is finished, it is time to start thinking about next year's one. It is easier to organize a symposium if there is ample lead time. Last newsletter, I asked for suggestions for symposium topics. Marie-Josee Fortin suggested permutation and randomization tests, Aaron Ellison suggested Bayesian statistics, and John Endler suggested causal modeling/path analysis. In addition to these, at last year's business meeting there were other suggested topics: nonlinear dynamics in time series, unbalanced mixed model methods, and generalized linear models (GLiM). The enclosed ballot asks you to vote for up to three of these topics. After we receive the ballots, I will ask the person suggesting the symposium (or an appropriate designee) to prepare a tentative list of speakers and topics that we can discuss at the section business meeting.

The section business meeting is a mixer/business meeting scheduled on Tuesday from 5 to 7 pm, right after the symposium. The main items of business, as of now, are a discussion of next year's symposium, introduction of the new secretary, and a discussion of current activities of the ESA council.

Finally, I want to raise an issue that might prompt considerable discussion: are graduate students (and perhaps others) receiving a good statistical education? If not, what might be done to improve graduate training in statistics for non-majors?

I am raising this issue because of some recent experiences in PhD admission to candidacy exams. I serve on a number of graduate students' committees, where I usually represent statistics or quantitative ecology. Students seem to have a good knowledge of techniques (e.g. I need to fit a line, so I'll use regression), but seem to lack a real understanding of statistical basics (e.g. concepts of sampling distributions, the meaning of confidence intervals, and confusion between "non-significant" and "no difference"). I realize that there is considerable debate among statisticians and philosophers on the logical structure of statistical inference, but I believe that understanding the basics provides a structure to all statistical methods.

This understanding may be too much to expect from graduate students in ecology, this may just represent current teaching at one school, or this may be one example of a widespread problem. So, I would appreciate any and all feedback, especially if anyone knows of a successful program. What makes it successful? The textbook? The teacher? The examples?

I can be reached by email at dixon@srel.edu, or by regular mail at Savannah River Ecology Lab., Drawer E, Aiken, SC 29802.

Philip Dixon

Section Chair

OFFICIAL BUSINESS

1. 1994 Business meeting

The annual business meeting of the Statistical Ecology Section of ESA will be held in conjunction with the AIBS annual meeting in Knoxville, Tennessee, August 7-11. The mixer and meeting will be held Tuesday evening immediately following the section's symposium on "Incorporating spatial information into ecological experiments". Please try to attend the meeting! We need your input.

2. Activities during the 1994 AIBS annual meeting

This year's meeting is rich with activities of interest to statistical ecologists. Besides the aforementioned Tuesday afternoon symposium, there will be a contributed papers session featuring three student papers supported by $200 section travel awards. Beyond these, the ASPT is sponsoring a symposium on "Maximum likelihood inference in plant systematics and population ecology", organized by Paul Lewis, and the BSA is sponsoring one on "Comparative diversity and dynamics of depressional wetlands" which, though it may not be specifically focused on statistical methods, is bound to be a good rigorous symposium, since its organizer is again our own (very busy) Philip Dixon.

3. Election of Section Secretary

The two year term of current section secretary Don Edwards will expire as of the ESA annual meeting August 1994. According to the bylaws, candidates are to be nominated by the executive committee (the current chair, vice-chair, and secretary). The committee has found one willing candidate, Sam Scheiner of Northern Illinois University. Additional names could have been placed on the ballot by petition of ten or more members; a call for these nominations was included in the Fall newsletter. No other nominations were received. In order to ensure that the section members' voices be heard, a write-in blank is provided on the ballot along with Professor Scheiner's name. Biographical data for him is given below.

Biographical sketch: Samuel M. Scheiner

Dr. Scheiner is an associate professor in the Department of Biological Sciences at Northern Illinois University. His research activities include the analysis of plant diversity at landscape and global scales, the development of measures of diversity (especially pattern diversity), plant population biology, and the evolution of phenotypic plasticity. Recently he co-edited with Jessica Gurevitch a book on experimental design and analysis for ecologists. His most recent section activity has been the development of a recommended addition to the Instructions for Authors for the ESA journals on the use and reporting of statistics. He has also served on the Buell/Braun award committee at the past several meetings.

Items of interest to Statistical Ecologists

1. Report on the conference on Statistics in Ecology and Environmental Monitoring, Dunedin, New Zealand 13-17 December 1993.

The aim of this conference was to bring together those interested in the statistical aspects of ecology and environmental monitoring. An interesting mix of ecologists and statisticians came, mostly from New Zealand, Australia, and the United States. The range of talks presented can be illustrated by those presented by the invited speakers:

Mark Boyce (University of Wisconsin). Population viability analysis exemplified by models for the northern spotted owl.

Richard Cormack (University of St. Andrews). 1. Statistical thoughts on the UK Environmental Change Network. 2. Unification of mark-recapture analyses by loglinear modeling.

Roger Green (University of Western Ontario). Aspects of power analysis in environmental monitoring.

Lyman McDonald (Western Ecosystems Technology, Wyoming). Testing for bioequivalence in field studies: has a disturbed site been adequately reclaimed?

Tony Underwood (University of Sydney). What environmental scientists (and statisticians) need to know to receive (and give) better statistical advice.

A limited edition of the proceedings (approximately 300 pages) will be published in July 1994 by Otago University Press. The cost will be NZ$80 including surface postage. Visa and Mastercard can be accepted (give your card number and expiration date when ordering). A full list of the papers to be included in the proceedings is available on request. To order a copy, please write: SEEM Conference proceedings, Centre for Applications of Statistics and Mathematics, University of Otago, PO Box 56, Dunedin, New Zealand.

Bryan F.J. Manly

Upcoming Meetings

1994

June 21. Workshop: analyzing population data: what's new and what's needed? Missoula, MT. In conjunction with the American Ornithologists' Union, Cooper Ornithological Society, and Wilson Ornithological Society. Contact Nadav Nur, Point Reyes Bird Observatory, 4990 Shoreline Highway, Stinson Beach, CA 94970. (415)868-1221.

July 25-29. SIAM Annual Meeting, San Diego, CA. Themes for the meeting include environmental and geophysical modeling. For information, contact the SIAM Conference department at (215)382-9800; email: meetings@siam.org

Aug. 7-11. Ecological Society of America annual meeting, Knoxville, Tennessee. Joint with AIBS. See discussion, this issue, for activities.

Aug. 8-12. 17th International Biometric conference (IBC94), Hamilton, Ontario, Canada. Sponsored by the International Biometric Society. To be held at McMaster University. For further information, contact: IBC94 local organizing committee, Department of Mathematics and Statistics, McMaster University, Hamilton, Ontario, Canada L8S 4K1. Telephone: (416)529-7070, ext. 3423; FAX: (416)522-0935; e-mail: ibc94@mcmail.cis.mcmaster.ca

Aug. 10-19. Third Latin-American Meeting on Mathematical Ecology, Buenos Aires, Argentina. Contact Fernando Momo, Universidad Nacional de Lujan-Ecologia, C.C. 221, (6700); email: ecologia@unlu01.edu.ar

Aug. 12-15 Fifth International Conference on Statistical Methods for the Environmental Sciences, Burlington, Ontario, Canada, Sponsored by the International Environmetric Society. The time and location make it convenient to attend in conjunction with IBC94 and/or the Joint Statistical Meetings. Information from Dr. A.H. El-Shaarawi, National Water Research Institute, P.O. Box 5050, Burlington, Ontario L7R 4A6, Canada. Telephone: (904)846-5467; email: u101@cs.cciw.ca

Aug. 14-18. Joint Statistical Meetings, Toronto, Canada. Co-sponsored by ASA, ENAR, and WNAR. Information from ASA, 1429 Duke St., Alexandria, Va 22314-3402; (703) 684-1221; FAX (703) 684-2037.

Aug. 20-24. Silver Jubilee of Statistical Ecology, in conjunction with the Sixth International Congress of Ecology, in Manchester, UK. A Silver Jubilee of Statistical Ecology is planned to celebrate the 25th anniversary of the First International Symposium on Statistical Ecology held in Yale in 1969. For information, contact G.P.Patil, Department of Statistics, Penn State University, University Park, PA 16802. Telephone: (814)865-9442; FAX: (814)863-7114; e-mail: GPP@PSUVM.bitnet.

Oct. 6-8. Workshop on Exploratory Spatial Data Analysis, Augsberg, Germany. Contact Antony Unwin, Chair of Computer Oriented Statistics and Data Analysis, Mathematics Institute, University of Augsberg, 86135 Augsberg, Germany. Phone: +49 (821)598-2218, FAX: +49 (821)598-2200. email: unwin@uni-augsberg.de

Nov. 16-18. 15th Franco-Belgian Meeting of Statisticians: Wavelets and Statistics. Grenoble, France. Wavelets are finding increasing application in time series and random fields and image analysis. Contact Prof. Anestis Antoniadis, RFBS, IMAG-LMC, Equipe de Statistique, Universite Joseph Fourier, B.P.53, 38041 Grenoble Cedex 09, France. Phone: (33)76514306. FAX: (33)76631263. email: antonia@imag.fr

1995

July 31-Aug. 3. Ecological Society of America, 80th Annual Meeting, Snowbird, Utah. Theme: the Transdisciplinary Nature of Ecology.

Aug. 13-17. Joint Statistical Meetings, Orlando, Florida. To be held at the Walt Disney World Dolphin and Swan hotels. Sponsored by ASA, ENAR, and WNAR. Information from ASA, 1429 Duke St., Alexandria, Va 22314-3402; (703) 684-1221.

Dec. 6-10. Sixth International Conference on Statistical Methods for the Environmental Sciences, Universiti Kebangsaan, Malaysia. Sponsored by the International Environmetric Society. Information from Dr. A.H. El-Shaarawi, National Water Research Institute, P.O. Box 5050, Burlington, Ontario L7R 4A6, Canada. Telephone: (904)846-5467; email: u101@cs.cciw.ca