Sensitivity analysis on distance-adjusted propensity score matching for wildfire effect quantification using national forest inventory data. (October 2021)

• Record Type: Journal Article
• Authors: Woo, Hyeyoung; Eskelson, Bianca N.I.; Monleon, Vicente J.
• Is Part Of: Environmental modelling & software. Volume 144(2021)
• Publication Date: 2021-10
• Subjects: Data availability -- Sample size -- Environmental covariates -- Natural disturbance -- Quasi-experimental method -- Monte Carlo simulation; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2021.105163
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Propensity score matching (PSM) and distance-adjusted PSM enable estimation of causal effects from observational data by selecting controls that are similar to treated observations in terms of environmental covariates and spatial locations. Quantifying effects of natural disturbances such as wildfires often encounters limited availability of observational data due to the scarcity of ecological events or cost of data collection over large areas. Using empirical data of national forest inventory plot measurements, we conducted a sensitivity analysis of distance-adjusted PSM on data availability for wildfire effect quantification. Using Monte Carlo simulations, we assessed the influence of sample size and covariates on the balance of propensity score distributions and the performance of effect estimates. The inclusion of the distance measure in matching compensated for the omission of key covariates. This study provides a practical guide of determining sample size and covariates for matching with spatial information to analyze ecological data. Highlights: Performs sensitivity analysis of matching methods using empirical ecological dataset. Evaluates matching performance depending on sample size and covariates. Highlights contribution of spatial distances in matching in relation to sample size. Provides a practical guide to select sample size and covariates for forest inventory data.