Predicting Carbon Accumulation in Temperate Forests of Ontario Using a LiDAR-Initialized Growth-and-Yield Model

Climate warming has led to a need for improved estimates of aboveground carbon accumulation (CA); in Ontario, this is particularly true for uneven-aged, mixedwood temperate forests, where there remains high uncertainty. This study investigated the feasibility of using LiDAR-derived tree lists to initialize a Growth and Yield (G&Y) model at the Petawawa Research Forest (PRF) in eastern Ontario, Canada. We aimed to see if models based on LiDAR-derived estimates of tree attributes provide comparable predictions of aboveground CA in complex temperate forests to those from inventory measurements. Applying a local G&Y model (i.e., FVSOntario), we predicted aboveground carbon stock (CS; tons/ha) and CA (tons/ha/yr) using recurring plot measurements from 2012-2016, FVS1. We then used a suite of statistical predictors derived from LiDAR to predict stem density (SD), stem diameter distribution (SDD), and basal area distribution (BA_dist) using an algorithm. These data, along with measured species abundance, were used to construct tree lists and initialize a second G&Y model (i.e., FVS2). Another G&Y model was tested using LiDAR-initialized tree lists and Forest Resource Inventory (FRI) estimates of species abundance (i.e., FVS3). Models were validated using inventory data at years zero (2012) and four (2016). The inventory-based model predicted equivalent CS at year four compared to validation data at all size-class levels, while LiDAR-based models did not; however, models were statistically dissimilar to validation data for CA. There was no significant difference between LiDAR-based models using inventory and FRI species through a 9-year period at the plot level (p<0.05), although forest type groupings were not always equivalent. We found equivalency of inventory and photo-interpreted models for size classes ≥17 cm. Our results demonstrate the importance of precise tree size information when initializing G&Y models using LiDAR-derived estimates, the possibility of circumventing precise species ...