• Spatial and spatio-temporal analysis
  • Nonstationary spatial modeling of heterogeneous materials
  • Bayesian spatio-temporal point process models
  • Bayesian hierarchical models
  • Bayesian functional data analysis

Applications of Interest

  • Atmospheric Sciences
  • Material Sciences


Tucker J.D., L. Shand, and K. Chowdhary (2019). Bayesian Function Registration. In preparation.

Harris, T., Tucker J.D., B. Li, and L. Shand (2019). Identifying Shape Outliers in Functional Data with Elastic Depth In preparation.

Tucker, J. D., L. Shand, J. R. Lewis (2019). Handling Missing Data in self-exciting point process models. Spatial Statistics. 29.160-176.

Guo S., M. A. Cooper., and L. Shand (2018). A statistical representation of pyrotechnic research igniter output. AIP Conference Proceedings. 1979 (1). doi: 10.1063/1.5044973

Lyndsay Shand and Bo Li (2018). Spatially Varying Autoregressive Models for Prediction of New HIV Diagnoses. Journal of the Royal Statistical Society: Series C.

Lyndsay Shand and Bo Li (2017). Modeling Nonstationarity in Space and Time. Biometrics.

L. Shand; W. M. Brown; L. F. Chaves; T. L. Goldberg; G. L. Hamer; L. Haramis; U. Kitron; E. D. Walker; M.O. Ruiz (2016). Predicting West Nile Virus Infection Risk From the Synergistic Effects of Rainfall and Temperature. Journal of Medical Entomology. 1-10. doi: 10.1093/jme/tjw042.


Data-driven inferences for aerosol and marine low-cloud interactions using ship-based databases and open-source satellite imagery

It is well-known through the Twomey effect, or aerosol indirect effect, that aerosol emissions impact marine boundary-layer clouds. Anthropogenic aerosols emitted from ships cause perturbations to clouds’ lifetime, albedo, and precipitation efficiency, which have an impact on optical and thermal properties of the cloud. The aerosols’ impact on clouds is highly dependent on the emissions source as well as the local environmental properties such as humidity, boundary layer height, temperature profile, and wind profile. The emissions from ships can sometimes form ship tracks, but the conditions under which they form are poorly understood, and the entire extent of aerosol effects on clouds remains one of the largest sources of uncertainty in climate forcing calculations. To improve our knowledge of aerosol-cloud interactions for ship track formation in marine boundary-layer clouds, we present work using a data-driven approach. Using space-based observational platforms such as GOES-17, we identify and correlate visible ship tracks with AIS ship positioning datasets to confirm and attribute a given ship track with ship type, emission source, and location. Trajectories of the ships that do produce tracks are linked with environmental conditions to ascertain properties such as how long the tracks exist and how the movement of tracks impacts overall cloud properties, thus improving our understanding of aerosol-cloud interactions. This work is possible for the first time given the high-resolution imagery from GOES-17. Results of this work will also be informative for improved understanding of any proposed future marine low-cloud brightening geoengineering efforts and parameterization development for global climate models.


Spatially-varying autoregressive prediction models for HIV

A major challenge with modeling HIV data lies in the rarity of the disease, leading to few to no incidents across much of the domain. This also makes it necessary to borrow strength from neighboring regions in both space and time in order to more accurately estimate the underlying disease risk. Motivated by making prediction for HIV new diagnosis rate based on a HIV dataset that is abundant in space but has short length in time, we have built a class of spatially varying autoregressive models (SVAR), in a Bayesian hierarchical setting and employ copula methods to describe complex spatial dependencies. These models are applied to three regional datasets of the United States.

Modeling Nonstationary Space-Time Data with Dimension Expansion

The Dimension Expansion technique to model nonstationary saptio-temporal processes allows us to take advantage of the abundance of stationary models available while simultaneously retainin the nonstationary properties in the data. Both simulations and real data application to noisy streamflow and wind datasets show that modeling nonstationarity in both space and time, can improve the predictive performance over stationary covariance models or the models that are nonstationary in space but stationary in time.

R package: rdimexp