|Dr Pierre R. L. Dutilleul
Professor, Department of Plant Science, Interim chair
Associate Member, Department of Mathematics and Statistics
Associate Member, McGill School of Environment
pierre [dot] dutilleul [at] mcgill [dot] ca (Email)
My background is in mathematics and statistics, and I am a Faculty member of McGill University, Macdonald Campus, Department of Plant Science, since December 1992. (The presence of a mathematician and statistician among plant scientists is not common, and motivated an interview to the Science Magazine in 2006, titled “The Greener Side of Math: A Statistician in the Plant Science World”.) Since my doctoral studies at Université Catholique de Louvain (Belgium), my research interests have been mostly in the statistical sciences, methods and applications, in time and space. The applications include the use of a computed tomography (CT) scanner with plants, but also soil and wood, and the statistical analysis of CT scan data which are multi-dimensional by nature (see “Modern Phytometry” in the next section). Some of my main findings in research have been summarized in the book “Spatio-Temporal Heterogeneity: Concepts and Analyses”, published by Cambridge University Press in 2011. I generally do research with graduate students and colleagues at McGill and through collaborations in Canada, the United States and Europe.
I am teaching or am involved in the teaching of undergraduate/graduate applied statistics courses: AEMA 411/611 (Experimental Designs), AEMA 414/AEMA 614 (Temporal and Spatial Statistics), AEMA 310 (Statistical Methods 1), and AEMA 403 (Environmetrics Stage). I am the mentor for the Environmetrics Domain program at the McGill School of Environment.
Since January 2011, I have the status of Editor-in-Chief for the journal Environmental and Ecological Statistics with Springer (http://www.springer.com/life+sciences/ecology/journal/10651). Since January 2009, I am a member of the Board of Directors of Centre SÈVE (http://www.centreseve.org/), a multi-institutional research centre in plant science that is funded by the Regroupements stratégiques program of the Fonds de recherche du Québec – Nature et technologies (FRQNT).
Two main axes of research
Spatio-temporal heterogeneity analysis:
In statistical sense and simply put, heterogeneity may concern the mean or variance parameter of the distribution of a random variable, or be related to the autocorrelation function of a stochastic process. When the value of the mean or the variance is susceptible to change, or variability is measured from observations that are partially dependent on each other because they are autocorrelated, in time or space, there is potential for a heterogeneity analysis, starting with the experimental design (Dutilleul, 1993a, Ecology). This opens the door to a lot of interesting situations and problems! A modified t-test (Dutilleul, 1993b, Biometrics) provides a solution to the problem of assessing validly the correlation between two autocorrelated spatial processes, and was followed by a modified F-test and other modified t-tests in the contexts of multivariate and multi-scale analyses (e.g., Dutilleul et al., 2008a; Dutilleul and Pelletier, 2011). Concerning efficient estimation and the decomposition of the variability contained in multivariate spatial datasets, the series of geostatistical articles including Pelletier et al. (2004, 2009a, 2009b) and Larocque et al. (2007) provide solutions based on the fitting of the linear model of coregionalization by estimated generalized least squares and the development of the method of coregionalization analysis with a drift (CRAD) eventually. In a spectral instead of geostatistical approach, the method of multi-frequential periodogram analysis (MFPA; Dutilleul, 2001) allows the decomposition of a time series, univariate or multivariate, into a number of periodic components, the number of periodic components as well as the period values being estimated in a stepwise procedure. I also have rising research interests in point pattern analysis (Dutilleul et al., 2009; Bonnell et al., 2013) and long-term research interests in multi-dimensional statistics (Dutilleul and Pinel-Alloul, 1996; Dutilleul, 1999), actually back to my doctoral studies.
My research work in this area has started before that, via the search and finding of an improved quantification of the structural complexity of crop canopies (e.g. Foroutan-pour et al., 2001), but it was really boosted with the creation of the CT Scanning Laboratory for agricultural and environmental research at Macdonald Campus of McGill, thanks to an NSERC Major Equipment grant (PI: Dutilleul) and the portion of a CFI grant (PI: Fortin) for the equipment of a computer room, both in 2000. Since the official opening of the facility in Fall 2003, our research group developed new procedures for the graphical, quantitative and statistical analyses of CT scan data in a broad range of applications other than the medical one for which the CT scanning equipment was designed originally. This includes fractal analysis of branching patterns of conifers, with McGill collaborators (Dutilleul et al., 2008b), and multi-fractal analysis of soil macropore networks, with U. Laval collaborators (Lafond et al., 2012).
Examples of current and recently completed research projects
Following the two main axes of research described above, I am leading or recently led, or am involved or was recently involved, in the following projects. On the heterogeneity analysis side, I worked with Prof. Yves Carrière (University of Arizona) and entomologist colleagues in the frame of a USDA/RAMP-funded project, to design and implement field and landscape level reduced-risk management strategies for Lygus bug in Western cropping systems; I collaborated with Dr. Tim Haltigin (Canadian Space Agency) on pattern recognition analysis and a quantitative characterization of terrain landforms on Mars and at analog terrestrial sites; with my former Ph.D. student Ameur Manceur, we developed unbiased modified likelihood ratio tests for simple and double separability of a variance-covariance structure, possibly spatio-temporal; and recently, with Prof. Roland Bürgmann (UC Berkeley), we started an application of the MFPA for a spatio-temporal analysis of California earthquake frequencies . On the phytometry side, collaborations with Centre SÈVE members occupy an important place; for example, with Prof. Carole Beaulieu (U. de Sherbrooke) and McGill collaborators, we completed the first phytopathological application of CT scanning (Han et al., 2008); I have accepted the invitation to guest edit a Research Topic titled “Branching and Rooting Out with a CT Scanner” for Frontiers in Plant Science, to be published in 2015; and recently, I started wood CT scanning applications with Dr. Jean Beaulieu (Ressources Naturelles Canada), including a multi-dimensional modeling of wood density (Manceur et al., 2012) and an alternative approach to classical dendrochronology (Dutilleul et al., 2014).
Dutilleul, P. 1993a. Modifying the t test for assessing the correlation between two spatial processes. Biometrics 49:305-314.
Dutilleul, P. 1993b. Spatial heterogeneity and the design of ecological field experiments. Ecology 74:1646-1658.
Dutilleul, P. and Pinel-Alloul, B. 1996. A doubly multivariate model for statistical analysis of spatio-temporal environmental data. Environmetrics 7:551-566.
Dutilleul, P. 1999. The MLE algorithm for the matrix normal distribution. Journal of Statistical Computation and Simulation 64:105-123.
Dutilleul, P. 2001. Multi-frequential periodogram analysis and the detection of periodic components in time series. Communications in Statistics - Theory and Methods 30:1063-1098.
Foroutan-pour, K., Dutilleul, P., and Smith, D. L. 2001. Inclusion of the fractal dimension of leafless plant structure in the Beer-Lambert law. Agronomy Journal 93:333-338.
Pelletier, B., Dutilleul, P., Larocque, G., and Fyles, J. W. 2004. Fitting the linear model of coregionalization by generalized least squares. Mathematical Geology 36:323-343.
Larocque, G., Dutilleul, P., Pelletier, B., and Fyles, J. W. 2007. Characterization and quantification of uncertainty in coregionalization analysis. Mathematical Geology 39:263-288.
Dutilleul, P., Pelletier, B., and Alpargu, G. 2008. Modified F-tests for assessing the multiple correlation between one spatial process and several others. Journal of Statistical Planning and Inference 138:1402-1415.
Dutilleul, P., Han, L., and Smith, D. L. 2008. Plant light interception can be explained via computed tomography scanning: Demonstration with pyramidal cedar (Thuja occidentalis, Fastigiata). Annals of Botany 101:19-23.
Han, L., Dutilleul, P., Prasher, S. O., Beaulieu, C., and Smith, D. L. 2008. Assessment of common scab-inducing pathogen effects on potato underground organs via computed tomography scanning. Phytopathology 98:1118-1125.
Dutilleul, P., Haltigin, T. W., and Pollard, W. H. 2009. Analysis of polygonal terrain landforms on Earth and Mars through spatial point patterns. Environmetrics 20:206-220.
Pelletier, B., Dutilleul, P., Larocque, G., and Fyles, J. W. 2009a. Coregionalization analysis with a drift for multi-scale assessment of spatial relationships between ecological variables 1. Estimation of drift and random components. Environmental and Ecological Statistics 16:439-466.
Pelletier, B., Dutilleul, P., Larocque, G., and Fyles, J. W. 2009b. Coregionalization analysis with a drift for multi-scale assessment of spatial relationships between ecological variables 2. Estimation of correlations and coefficients of determination. Environmental and Ecological Statistics 16:467-494.
Dutilleul, P. and Pelletier, B. 2011. Tests of significance for structural correlations in the linear model of coregionalization. Mathematical Geosciences 43:819-846.
Lafond, J. A., Han, L., Allaire, S. E., and Dutilleul, P. 2012. Multifractal properties of porosity as calculated from computed tomography (CT) images of a sandy soil, in relation to the relative soil gas diffusion coefficient. European Journal of Soil Science 63:861-873.
Manceur, A. M., Beaulieu, J., Han, L., and Dutilleul, P. 2012. A multidimensional statistical model for wood data analysis, with density estimated from CT scanning data as an example. Canadian Journal of Forest Research 42:1038-1049.
Bonnell, T. R., Dutilleul, P., Chapman, C. A., Reyna-Hurtado, R., Sengupta, R., and Sarabia, U. 2013. Analysing small-scale aggregation in animal visits in space and time: the ST-BBD method. Animal Behaviour 85:483-492.
Dutilleul, P., Han, L., and Beaulieu, J. 2014. How do trees grow? Response from the graphical and quantitative analyses of computed tomography scanning data collected on stem sections. Comptes Rendus Biologies (to appear).