Assigned Reading:

  • Pedersen, E. J., Miller, D. L., Simpson, G. L., & Ross, N. (2019). Hierarchical generalized additive models in ecology: an introduction with mgcv. PeerJ, 7, e6876.
  • Lawton, D., Scarth, P., Deveson, E., Piou, C., Spessa, A., Waters, C., & Cease, A. J. (2022). Seeing the locust in the swarm: accounting for spatiotemporal hierarchy improves ecological models of insect populations. Ecography, 2022(2).

Overview

Invasive plants cause immense ecological and economic damage. For example, in forested lands, invasive grasses can change fire regimes, invasive shrubs can reduce native tree regeneration, and invasive trees can reduce timber production. Determining patterns in 1) invasive plant abundance increases/decreases over time and 2) spatial distribution across forested lands could help managers assess potential risks to their lands. For today’s lab, we’re going to be using hierarchical generalized additive models (HGAMs) to help us make these determinations–and add some theoretical flavor to them.

For the abundance increase/decrease over time issue, we will ask:

  • Does functional group (e.g., growth habits such as graminoid, tree, shrub, etc.) or taxonomy (e.g., order) best explain invasive plant abundance changes?

For the spatial distribution issue, we will ask:

  • How do abundance patterns differ by growth habit across space?

The first question will help us understand how to construct HGAMs from our hypotheses, appraise HGAM fits, and conduct model selection with HGAMs. Through the second question, we will learn to do spatial predictions (i.e., making predictive maps) with our model outputs.

‘Cogongrass’ by Nancy Loewenstein and John McGuire.
‘Cogongrass’ by Nancy Loewenstein and John McGuire.

Data description

The data you will be using is from the U.S. Forest Service’s “Forest Inventory and Analysis” (FIA) dataset. The FIA dataset is taken on public and private forested and timber lands throughout the U.S. on specific plots. The plots are revisited approximately every 4 - 5 years, but some plots are never revisited. One of the data collected at each plot is percent cover of various invasive plants. Here, I used the “rFIA” R package to extract invasive species cover data from the Eastern U.S. from 2001 - 2020. I also extracted growth habits and orders of each plant species from the USDA Plants Database. Below is a table with descriptions of columns in each dataset: