Canadian Journal of Forest Research · 1998 · 409 citations · 28 references
Natural ResourcesForestryPlant BiologyForest BiometricsForest EcologyForest InventoryCanopy StructureHeight MeasurementLidar-based Forest MeasurementStand MeasurementTree Height EstimationTree MeasurementAirborne Laser ScanningCanopy Height MappingField Trial EvaluationLeaf Area DistributionLeaf Area EstimationPlant Height EstimationStand Height ModelingStand Structure Characterization
Airborne laser scanner returns from Douglas‑fir plots are proportional to the vertical distribution of foliage area, linking pulse return proportions to leaf‑area fractions above reference heights. The study tests whether the laser canopy height quantile that matches the probability of leaf area above a target height provides an unbiased stand‑height estimate. Across 36 plots, canopy‑based quantiles were within 6 % of field stand heights, correlated at r≈0.8, and a geometric model explained the mean difference, confirming the method’s accuracy.
The distribution of canopy heights obtained with an airborne laser scanner over a field trial with Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) was a function of the vertical distribution of foliage area. Over a wide range of canopy structures, the proportion of laser pulses returned from or above a given reference height was proportional to the fraction of leaf area above it. We hypothesized that the quantile of the laser canopy heights matching in probability the fraction of leaf area above a desired height would be an unbiased estimator of same. This was confirmed in 36 (20 × 20 m) plots and 6 older validation plots. Canopy-based quantiles of the laser canopy height data were within 6% (mean 3%) of the field estimates. Laser and field estimates were strongly correlated (r ~ 0.8), and statistical tests supported the null hypotheses of no difference in mean stand height (P > 0.3). A geometric model successfully predicted the mean difference between the laser canopy heights and the mean tree height. Our results explicate why estimation of stand heights from laser scanner data based on the maximum canopy height value in each cell of a fixed area grid has been successful in practice.
28
Practical Nonparametric Statistics
David J. Groggel, W. J. Conover · Technometrics · 2000 · 11.1K citations
Continuous Univariate Distributions.
Yasuhiro Omori, Norman L. Johnson, S. Kotz et al. · Journal of the American Statistical Association · 1995 · 9.3K citations
Noel Cressie · Terra Nova · 1992 · 8.9K citations
Computing, Geography, Geoinformatics +12
Richard A. Brown, Trevor Hastie, Robert Tibshirani · Biometrics · 1991 · 8.3K citations