The 2023 recipient of the Statistical Society of Canada Award for Impact of Applied and Collaborative Work is Pierre Dutilleul, Professor of Statistics in the Department of Plant Science at McGill University. The award recognizes outstanding contributions by a member of the SSC in collaborative research and applied work, the importance of which derives primarily from its relatively recent impact on a subject area outside of the statistical sciences, on an area of application, or on an organization.

Born and raised in Belgium, Pierre Dutilleul is a graduate of the Université catholique de Louvain (BSc 1983, MSc 1984, DSc 1990), where he wrote his thesis under the supervision of Guy Gérard and José Paris. Living in Montréal since 1990, Pierre is well known and frequently cited for his various adaptations of the celebrated Student’s t test and Fisher’s F test in various correlation and regression contexts (linear, multiple, partial, stepwise, multi-scale, etc.) for data varying over time and space. His seminal paper on this topic, published in Biometrics in 1993, has been cited nearly 1000 times.

In addition to studying the empirical validity and other statistical properties of these new tests, Pierre facilitated their use through software. He also promoted them by showing their relevance and value in several fields of knowledge, including plant and soil sciences, entomology, limnology, and forestry, where spatio-temporal data are abundant. He also contributed in many ways to the development and application of a new statistical approach to the quantification of spatial heterogeneity. This vast body of work, carried out over two decades, culminated in his user-friendly book entitled Spatio-Temporal Heterogeneity: Concepts and Analyses, which was published by Cambridge University Press in 2011.

The development of analytical tools, the study of their properties, their programming, and their use in a variety of contexts are fundamental to statistical research. Pierre has mastered this art and his CV documents many other achievements of this type throughout his career. Examples include his groundbreaking work on the tensor (multilinear) generalization of the Gaussian distribution and his highly cited 1999 paper in the Journal of Statistical Computation and Simulation, proposing an original algorithm for maximum likelihood estimation of the parameters of a matrix normal distribution.

As numerous and impressive as Pierre’s methodological contributions may be, what sets him apart as an exceptionally worthy recipient of the SSC Award for Impact of Applied and Collaborative Work is that he systematically goes beyond the boundaries of his field. As a result, he eventually gained recognition as a plant, soil, and wood science specialist in his own right, through very original and lasting subject-matter contributions for which statistics was essential.

In bold and innovative work in plant science, initially involving his colleague Donald Smith and a PhD student (Foroutan-pour et al., 1999, Appl. Math. Comput.), Pierre revolutionized the study of branching patterns and root systems by characterizing their structural complexity using fractal geometry and developing procedures for 3-D spatial analysis. One seminal contribution of his group was a refinement in 2001 of the famous Beer-Lambert law and its application to plant light interception, to account for the fractal dimension of the branching pattern supporting a leaf canopy. The original Beer-Lambert law, which links the attenuation of a light beam to the properties of the medium through which it passes, is the basis of computerized tomography (CT) scanning technology.

After acquiring a CT scanner, Pierre and an expanded group started documenting and analyzing the growth of live plants non-invasively in space and time. Being the first to use CT images to perform fractal analyses for branching patterns and root systems, Pierre quickly became a leader in the field and a major contributor to the rise of plant, soil, and wood CT scanning. He also developed a very original approach to dendrochronology.

Today, Pierre is regarded as a true “éminence grise” of modern phytometrics. An overview of the current and potential applications of this approach is provided in his 2016 Macmillan book with one of his PhD graduates, Jonathan Lafond (now affiliated to Agriculture and Agri-Food Canada): Branching and Rooting Out with a CT Scanner: The Why, the How, and the Outcomes, Present and Possibly Future.

As early as February 2006, the prestigious magazine Science was hailing Pierre Dutilleul as a bridge builder between “two distinct approaches, the theoretical and the experimental.” Already recognized at the time as a “Statistician in the Plant Science World,” Pierre has continued to cover the green side of his discipline, the Statistical Sciences, ever since.

In parallel, Pierre has built an enviable reputation in geostatistics and, with other collaborative groups, has helped to bridge gaps between plant and insect ecology, geology and geophysics, and by developing statistical methods such as “co-regionalization analysis with a drift” and “multi-frequential periodogram analysis.” In particular, his expertise has led to substantial quantitative gains in the analysis of seismic data. Not surprisingly, the thirty-or-so trainees and mentees he has supervised during his career work in a very broad spectrum of disciplines. Moreover, from 2011 to 2022, he was the co-Editor-in-Chief of the Springer-Nature journal Environmental and Ecological Statistics, whose remit intersects many of his fields of expertise.

On a personal note, Pierre and his wife, Marie-Pierre, have two children and five grandchildren. She loves growing plants, while he appreciates good wine, not necessarily French or Canadian, and good beer, preferably Belgian! Trout fishing in a Québec lake in the summer is a must for Pierre. A great sense of humor, an eagerness to learn, and a Zen attitude are three of his key personal traits.

It is particularly apposite that the SSC has chosen to recognize the breadth and depth of Pierre Dutilleul’s statistical work, and the major impact it had in various fields where space-time data are frequent. 

Christian Genest, McGill University, was primarily responsible for preparing this material.