ordSpat: Models of heterogeneity, contextuality and self-interaction in ordered spatial point patterns with applications to animal movement and forest inventory (Sep 2017 – Aug 2021)
Consortium project with Natural Resources Institute Finland
ordSpat is a consortium project funded by the Academy of Finland, conducted at Natural Resources Institute Finland (Luke) and University of Eastern Finland (UEF), and coordinated by Juha Heikkinen (Luke) and Lauri Mehtätalo (UEF).
Novel statistical methods are developed for
- analyzing the effects of memory, learning, and interactions on animal movement and
- conducting stand-level forest inventories by remote sensing.
A new class of statistical models for ordered spatial point patterns, originally developed for eye-movement research serves as a flexible framework for both applications.
The research is based on existing extensive GPS telemetry data and accurately measured forest plots. The expected results include
- widely applicable methodology for spatio-temporal point patterns,
- deeper understanding of animal behavior, and
- cost-efficient forest inventory estimates and their standard errors.
The results can be utilized in the assessment and protection of wildlife populations, in the determination of human-wildlife conflicts, and in small and large scale forest inventories.
Left: GPS locations of alpha male wolf Miki recorded with 30 min intervals in June-July 2009. Right: Display of point pattern indicating locations of trees in a forest plot from the Kiihtelysvaara site. The scale shows the varying density of trees.