Main content area

Aboveground forest biomass derived using multiple dates of WorldView-2 stereo-imagery: quantifying the improvement in estimation accuracy

Vastaranta, Mikko, Yu, Xiaowei, Luoma, Ville, Karjalainen, Mika, Saarinen, Ninni, Wulder, Michael A., White, Joanne C., Persson, Henrik J., Hollaus, Markus, Yrttimaa, Tuomas, Holopainen, Markus, Hyyppä, Juha
International journal of remote sensing 2018 v.39 no.23 pp. 8766-8783
aboveground biomass, canopy, data collection, digital elevation models, forest growth, forests, growth models, lidar, prediction, remote sensing
The aim of this study was to investigate the capabilities of two date satellite-derived image-based point clouds (IPCs) to estimate forest aboveground biomass (AGB). The data sets used include panchromatic WorldView-2 stereo-imagery with 0.46 m spatial resolution representing 2014 and 2016 and a detailed digital elevation model derived from airborne laser scanning data. Altogether, 332 field sample plots with an area of 256 m² were used for model development and validation. Predictors describing forest height, density, and variation in height were extracted from the IPC 2014 and 2016 and used in k-nearest neighbour imputation models developed with sample plot data for predicting AGB. AGB predictions for 2014 (AGB₂₀₁₄) were projected to 2016 using growth models (AGBPᵣₒⱼₑcₜₑd_₂₀₁₆) and combined with the AGB estimates derived from the 2016 data (AGB₂₀₁₆). AGB prediction model developed with 2014 data was also applied to 2016 data (AGB₂₀₁₆_ₚᵣₑd₂₀₁₄). Based on our results, the change in the 90ᵗʰ percentile of height derived from the WorldView-2 IPC was able to characterize forest height growth between 2014 and 2016 with an average growth of 0.9 m. Features describing canopy cover and variation in height derived from the IPC were not as consistent. The AGB₂₀₁₆ had a bias of −7.5% (−10.6 Mg ha⁻¹) and root mean square error (RMSE) of 26.0% (36.7 Mg ha⁻¹) as the respective values for AGBPᵣₒⱼₑcₜₑd_₂₀₁₆ were 7.0% (9.9 Mg ha⁻¹) and 21.5% (30.8 Mg ha⁻¹). AGB₂₀₁₆_ₚᵣₑd₂₀₁₄ had a bias of −19.6% (−27.7 Mg ha⁻¹) and RMSE of 33.2% (46.9 Mg ha⁻¹). By combining predictions of AGB₂₀₁₆ and AGBPᵣₒⱼₑcₜₑd_₂₀₁₆ at sample plot level as a weighted average, we were able to decrease the bias notably compared to estimates made on any single date. The lowest bias of −0.25% (−0.4 Mg ha⁻¹) was obtained when equal weights of 0.5 were given to the AGBPᵣₒⱼₑcₜₑd_₂₀₁₆ and AGB₂₀₁₆ estimates. Respectively, RMSE of 20.9% (29.5 Mg ha⁻¹) was obtained using equal weights. Thus, we conclude that combination of two date WorldView-2 stereo-imagery improved the reliability of AGB estimates on sample plots where forest growth was the only change between the two dates.