Introduction

Background and Rationale

Changing climates over the next century may result in significant losses to forest productivity and resilience. Genetic variation within tree species can lead to significantly different climate tolerances among populations, and as climates shift, many forest populations may become maladapted to their local climates (O'Neill et al. 2008). In order to address the issues of climate change effects on forest populations, the relocation of tree populations via planting trees in locations that are projected to better suit their climate tolerances in the future, otherwise known as assisted migration, has become a popular strategy (Williams and Dumroese 2013).

The DIVERSE project is a pan-Canadian research initiative that aims to enhance Canadian forestry practices to better adapt to the stressors associated with global change, including strategies such as assisted migration of forest trees. In order to inform assisted migration practices, the DIVERSE project is developing a tool for foresters which will allow for matching projected climatic conditions of a prospective forestry project with current ecozone climates, identifying where to source trees that will be optimal for current and projected conditions. More information about DIVERSE can be found here: https://diverseproject.uqo.ca/

The provenance selection tool uses ecozones and interpolated weather station data to generate the provenance recommendations. Interpolated weather station data, such as that generated by the software ClimateNA, is a widely accepted approach for modeling continuous climate data from discrete observations (Wu et al., 2020), and is employed by other similar tools (https://seedlotselectiontool.org/sst/). However, there are potential issues that may compromise the accuracy of the provenance selection tool. These may include ecozones being too large to accurately represent a given point within them, ecozones not accurately representing elevational gradients, or inaccuracies within the interpolated climate data. 

To increase the accuracy of the assisted migration tool both the species' distribution maps and ecozone classifications must be updated to be harmonious across North America, and suitable variables that encapsulate the climatic conditions of said ecozones must be identified. While it is possible to calculate the point climate estimate for a given location, doing this for every possible location across North America and having all that information in the database for the assisted migration tool poses a computational problem. Therefore, we aim to simplify the database by using a multifactor climatic variable to represent the climate of each ecozone. In order to do this, we must first verify the suitability of ecozone to approximate local climate.