plantTracker: An R package to translate maps of plant occurrence into demographic data

Abstract

Long‐term demographic data are rare yet invaluable for conservation, management, and basic research on the underlying mechanisms of population and community dynamics. Historical and contemporary mapped datasets of plant location and basal area present a relatively untapped source of demographic records that, in some cases, span over 20 years of sequential data collection. However, these maps do not uniquely mark individual plants, making the process of collecting growth, survival, and recruitment data difficult. Recent efforts to translate historical maps of plant occurrence into shapefiles make it possible to use computer algorithms to track individuals through time and determine individual growth and survival. We summarize the plantTracker R package, which contains user‐friendly functions to extract neighbourhood density, growth, and survival data from repeatedly‐sampled maps of plant location and basal area. These functions can be used with data derived from quadrat maps, aerial photography, and remote sensing, and while designed for use with perennial plants, can be applied to any repeatedly mapped sessile organism. This package contains two primary functions: trackSpp(), which tracks individuals through time and assigns demographic data, as well as getNeighbors(), which calculates both within and between‐species neighbourhood occupancy around each mapped individual. plantTracker also contains functions to estimate plot‐level recruitment, calculate plot‐level population growth rate, and create quadrat maps. We tested the accuracy of the trackSpp() function on two spatial demographic datasets. The function was nearly perfect at assigning individual identities and survival status when tested on maps of tree basal area and perennial forb point locations. In both cases, the function correctly assigned survival and recruitment with 99% accuracy. These accurate and precise functions will expand the amount of data available to investigate demographic processes, which are fundamental drivers of population, community, and ecosystem processes.

Demography with drones: detecting growth and survival of shrubs with unoccupied aerial systems

“…This overall positive relationship points to the potential capacity to improve estimates of growth for outplants with improved segmentation algorithms. Alternately, if accurate estimates of growth are needed from remotely sensed imagery, hand‐digitization and subsequent pairing of digitized crowns is a feasible, albeit time‐consuming, solution (Stears et al 2022). Estimating growth may also be easier for plants with simpler canopy structure than sagebrush shrubs, where irregular, asymmetric, and disjoint canopies are the norm, particularly for older plants.…”

“…Several studies have demonstrated the feasibility of estimating plant growth and survival using remote sensing. The most straightforward approach to estimating plant demography is to overlay hand‐digitized outlines of plant crowns across time periods (Stears et al 2022). However, manual digitization is not feasible across large scales that encompass >100,000 of individual plants.…”