Campaign and site effects in LiDAR prediction models for site-quality assessment of radiata pine plantations in South Australia

Abstract

Site-quality (site-productivity) information underpins many aspects of radiata pine plantation management in South Australia. Site-quality assessment is, in essence, a problem of assessing the spatial variation in stand volume at reference age. Trials examining the use of airborne Light Detection And Ranging (LiDAR) for site-quality assessment were commenced in 2002. LiDAR data was captured using three different LiDAR systems in 2002, 2006 and 2007, generating datasets captured at significantly different flying altitudes and point densities. A total of 169 field plots were measured across nine study sites. The study evaluated how operational LiDAR data-collection parameters (Campaign) and site characteristics (Site) affected LiDAR variables and prediction models for predominant height and stand volume. Significant Campaign effects were detected, which appeared to be due to differences in laser point density and flying height, instrumentation or seasonal effects such as drought. The effects were far more pronounced in models for stand volume than predominant height. Evidence of Site effects was inconclusive, but incorporation of Site variables in prediction models improved model fit under some circumstances. LiDAR-based site-quality assessment proved robust, even at the lowest point densities of LiDAR used in the study. Calibration protocols and modelling strategies that recognize Campaign effects need to be developed for operational deployment.

Acknowledgements

The considerable financial and logistic support of ForestrySA in conducting this study is gratefully acknowledged. Many thanks to ForestrySA technical staff for their professionalism in the field. Thanks to Andrew Robinson and two anonymous reviewers for their valuable suggestions and comments.