Estimating forest attributes in airborne laser scanning based inventory using calibrated predictions from external models

Abstract

Forest management inventories assisted by airborne laser scanner data rely on predictive models traditionally constructed and applied based on data from the same area of interest. However, forest attributes can also be predicted using models constructed with data external to where the model is applied, both temporal and geographically. When external models are used, many factors influence the predictions’ accuracy and may cause systematic errors. In this study, volume, stem number, and dominant height were estimated using external model predictions calibrated using a reduced number of up-to-date local field plots or using predictions from reparametrized models. We assessed and compared the performance of three different calibration approaches for both temporally and spatially external models. Each of the three approaches was applied with different numbers of calibration plots in a simulation, and the accuracy was assessed using independent validation data. The primary findings were that local calibration reduced the relative mean difference in 89% of the cases, and the relative root mean squared error in 56% of the cases. Differences between application of temporally or spatially external models were minor, and when the number of local plots was small, calibration approaches based on the observed prediction errors on the up-to-date local field plots were better than using the reparametrized models. The results showed that the estimates resulting from calibrating external models with 20 plots were at the same level of accuracy as those resulting from a new inventory.

Prediction of forest attributes using airborne laser scanning-based models without new in-situ field measurements

“…The RMSE distributions in study I showed that with suitable sample plot combinations, scenario SN would provide even smaller RMSE values than the regional models (R). The observed benefit of the calibration plots is in line with findings of de Lera Garrido et al (2022). They concluded that 20 calibration plots would result in predictions comparable to the local models when spatially or temporally transferred models are applied to the target area.…”

“…They reported that model transfer to the target area performed best when the models were calibrated with a small number of sample plots from the target area. Use of calibration plots in the context of model transferability was also studied by de Lera Garrido et al (2022) in Norway. They calibrated temporally and spatially transferred models for volume, stem number and dominant height with a different number of local sample plots and concluded that calibration reduces the systematic errors.…”

“…The first return data is considered more stable for different flying altitudes and canopy conditions than the last return data (Naesset 2004a;Naesset 2005). The large scan angles (up to 30°) seem to have only minimal effects on ALS metrics and on ALS-based forest attribute predictions (van Lier et al 2022) but conversely has clear effects on foliage profile estimation (Qin et al 2017).…”