Perry DeValpine

Assistant Professor
Ph.D.  University of California, Davis
  

201 Wellman Hall
Berkeley, California 94720
pdevalpine@nature.berkeley.edu
office: 510-642-0676   lab: 510-643-7430   fax:  510-643-5438

Web site         Recent publications      People

My research aims to develop and apply mathematical and statistical modeling methods to evaluate hypotheses from complex ecological data, with a primary focus on population dynamics. Population ecology is an important level for understanding nature because populations comprise individuals, on which natural selection typically acts, and are parts of communities and ecosystems. Understanding population dynamics in real systems, whether they be insects, birds, mammals, plants, microbes, diseases, or whatever, requires synthesis of life history evolution, behavioral ecology, physiology, species interactions, and statistical analysis of typically noisy data in relation to mathematical models. Applications of population dynamics include arthropod population management in agriculture, fishery stock assessment, and population viability analysis in conservation biology. I collaborate with empirical researchers in population ecology as well as other areas to bring together novel data and analysis approaches.

Fitting mathematical models to population dynamics data and making statistical comparisons among hypotheses are challenging problems for several reasons. Typically, data are only rough estimates of a full population; changes through time of a population are highly variable; and our knowledge of population age-structure, species interactions, and other relevant biology may be very limited. For example, field experiments to study factors that affect demographic processes -- such as growth, survival, reproduction, species interactions, movement, and behavior -- often produce noisy estimates of the state of highly variable nonlinear processes. I use applied math and statistics to study methods to fit stochastic, biologically-structured models to such data in order to test hypotheses from ecological field studies. These problems have led me into general work on Monte Carlo computational methods for maximum likelihood estimation (and related calculations) for nonlinear state-space time-series models. Other areas of interest include life-history evolution, species interactions, insect ecology, fishery dynamics, phenotypic plasticity, classification with high-dimensional data (bioinformatics), experimental design, and general statistical theory and computational methods.

Graduate students in my group combine a balance of mathematical modeling, statistics, and field experiments, which varies for each student depending on interests. Prospective graduate students should follow the link to my web site and read more.

Current projects include development of Monte Carlo statistical methods for maximum likelihood analysis of stage-structured population (time-series) models incorporating measurement error and environmental stochasticity; application of new analysis methods to insect population studies from agricultural systems to address questions about stage-structured demography, predator-prey interactions, biological control and insect outbreaks; and evaluation of population time-series analysis methods for fisheries systems to address questions about stock-assessment and nonlinear dynamics.

de Valpine, P. and R. Hilborn. 2005. State-space likelihoods for nonlinear fisheries time-series. Canadian Journal of Fisheries and Aquatic Sciences 62: 1937-1952.

de Valpine, P. 2004. Monte Carlo state-space likelihoods by weighted posterior kernel density estimation. Journal of the American Statistical Association 99:523-535.

de Valpine, P. 2003. Better inferences from population-dynamics experiments using Monte Carlo state-space likelihood methods. Ecology 84:3064-3077.

de Valpine, P. 2002. Review of methods for fitting time-series models with process and observation error, and likelihood calculations for nonlinear, non-Gaussian state-space models. Bulletin of Marine Science 70: 455-471.

de Valpine, P. and A. Hastings. 2002. Fitting population models incorporating process noise and observation error. Ecological Monographs 72:57-76.

de Valpine, P. and J. Harte. 2001. Effects of warming on a montane meadow ecosystem: how species responses comprise the ecosystem response. Ecology 82: 637-648.

de Valpine, P. 2000. A new demographic function maximized by life-history evolution. Proceedings of the Royal Society of London B 267: 357-362.