Fire Lines Volume 13 Issue 2
Research Brief
A spatially explicit model of tree leaf litter accumulation in fire maintained longleaf pine forests of the southeastern US
Authors: Nuria Sánchez-López, Andrew T. Hudak, Luigi Boschetti, Carlos A. Silva, Kevin Robertson, E Louise Loudermilk, Benjamin C. Bright, Mac A. Callaham Jr., & Melanie K. Taylor
In a nutshell: This paper describes the development and evaluation of a model for estimating tree leaf litter loads at a high spatial resolution. The crux of the model's novel value is its use of airborne laser scanning (ALS) to estimate the spatial distribution of tree litterfall across relatively large areas to model litter accumulation.
Effective management of fire-maintained forests requires an understanding of how tree leaf litter acts to carry surface fire. For example, in Coastal Plain pine-dominated communities of the southeast US, characteristics of the needle cast such as depth and continuity are drivers of fire behavior and burn patterns. Existing methods for measuring or estimating litter do not easily capture the variation of these properties across an area.
This study followed three general steps to develop an integrated approach for estimating litter loads at a high spatial resolution. The model was developed and validated with data from pine-dominated communities at multiple sites in the southeast US.
A model was developed to estimate the amount of tree leaf litter produced annually. ALS data was used to map the distribution of individual tree crowns. A machine learning model was trained to estimate the foliage biomass of each individual tree crown. Tree inventory data (e.g., crown diameter, tree height) and allometric equations were used as reference. The estimated foliage biomass was then combined with leaf turnover rates of the dominant species defined by forest types and dispersal of littercast was simulated to finally create a 5 m resolution map of annual litter production.
The Olson model, an ecological-based model for estimating litter loads based on rates of litter production, decomposition, and time of accumulation, was applied using as input the litter production map at 5 m resolution from the first step. Decomposition rates were derived from an annual evapotranspiration product obtained from remotely sensed data, and the number of years since the last fire from managers and history records.
Finally, the precision and accuracy of the approach were evaluated by comparing the model's estimates of litter loads to the true values collected through field measurements.
The results showed that this novel approach can be used to produce accurate, spatially-detailed estimates of leaf litter accumulation based on relatively easy-to-collect information about the canopy (i.e., tree inventory data) and time since the last fire. This integrated approach is an improvement over existing individual methods such as labor-intensive field measures, and remote sensing techniques that so far are limited in capability to directly measure litter loads (e.g., optical sensors that cannot account for litter depth).
Sánchez-López, N., Hudak, A. T., Boschetti, L., Silva, C. A., Robertson, K., Loudermilk, E. L., Bright, B. C., Callaham, M. A., & Taylor, M. K. (2023). A spatially explicit model of tree leaf litter accumulation in fire maintained longleaf pine forests of the southeastern US. Ecological Modeling, 481, 110369. https://doi.org/10.1016/j.ecolmodel.2023.110369
SFE Updates
In March 2022, the Southern Fire Exchange joined the Integrated Research Management Team (IRMT) at Fort Stewart in southeast Georgia for an interagency, collaborative research burn event. Three prescribed burn units (totaling ~3,000 ac.) were burned over a period of 6 days, allowing time before and after each burn for the setup and retrieval of equipment used for data collection by the fuels, fire behavior, and smoke research teams. At IRMT-hosted research burn events such as this one, the fuels team collects data on wildland fuels before and after the fire at designated plots through the burn units. The fire behavior team positions their sensors directly above the fuel team's plots to collect information on the rate and quantity of energy released by the passing fire. The smoke team collects data on the fire's smoke plume including the concentrations of different particles, how dense it is, which direction it is headed, and its velocity.
The collaborative research burns and the IRMT help these projects harmonize, share data, and prioritize tools to come out of the research to help managers reach their prescribed fire management goals.
