Multi-model estimation of understorey shrub, herb and moss cover in temperate forest stands by laser scanner data

Latifi, Hooman; Hill, Steven; Schumann, Bastian; Heurich, Marco; Dech, Stefan

doi:10.1093/forestry/cpw066

Cited by 10 publications

(18 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their shrub layer regression model had a relatively low R 2 value of 37%. In a later study, Latifi et al (18) showed an R 2 of 80% for the shrub layer based on thinned LiDAR point clouds and new analytical methods. Campbell et al (20) also compared field and LiDAR measures of understory directly in mixedwood forests and generated an R 2 of 0.44 based on a relative point density similar to metrics that we used here.…”

Section: Discussionmentioning

confidence: 98%

“…We also computed the symmetric mean absolute percentage error (SMAPE), based on 10-fold cross-validation (29) for the top-ranked models, and calculated SMAPE values for each of the 3 vertical strata separately. Parameters of the mixed effects models were estimated by maximum likelihood in R with the nlme package(18, 26, 30).…”

Section: Methodsmentioning

confidence: 99%

“…We explored two statistical approaches for modelling understory vegetation structure as a function of LiDAR data: machine learning and mixed effects regression models. Machine learning, specifically random forest, has been used to model forest inventory variables with a large suite of LiDAR derived predictors (18, 21). Machine learning in this context strives to produce the best prediction of the forest inventory variables.…”

Section: Introductionmentioning

confidence: 99%

“…Occlusion has been discussed in the literature as a possible issue limiting LiDAR effectiveness for prediction of understory structure (22, 23), but more recent studies have shown that the potential occlusion may not interfere with generating predictions. Latifi et al (18) demonstrated that artificially reducing the density of the LiDAR point cloud did not have an appreciable effect on variance explained in models predicting understory structure. In another study, prediction errors of understory vegetation cover were not related with canopy cover (17).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Modelling vegetation understory cover using LiDAR metrics

Venier

Swystun

Mazerolle

et al. 2019

Preprint

View full text Add to dashboard Cite

21 42 model with the voxel-based understory LiDAR metric along with vertical stratum, but that other 43 understory LiDAR metrics (fractional cover, normalized cover and leaf area density) would still 44 be effective in mixed-effects and Random Forest modelling approaches. 3 45 Supporting information 572 S1Table. Definitions of all variables included in at least one of the mixed-effects or Random 573 Forest models. 574 S2 Table. Frequency of explanatory variables among the 18 random forest models run with 575 341 to 7 variables. 576 S3 Data. Data used in analyses for manuscript. Variable definitions are found in S1.

show abstract

Section: Discussionmentioning

confidence: 98%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Modelling vegetation understory cover using LiDAR metrics

Venier

Swystun

Mazerolle

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Qualification of these elements can provide valuable information in order to assess future forest planning and management goals. Forest structure analysis is mainly focused on gathering detailed, precise and updated information on vertical and horizontal structure of forest layers (Latifi et al, 2017). Among the different structural variables, four are investigated in this thesis to describe the vertical and horizontal structure of canopy at different layers, namely 1) regeneration coverage, 2) tree stem count per ha, 3) tree segmentation and 4) tree species classification.…”

Section: Importance Of Forest Structurementioning

confidence: 99%

Sensing forest structure from LIDAR

Amiri¹

View full text Add to dashboard Cite

Prozess beinhaltete die Untersuchung einer umfangreichen Merkmalsraums mithilfe einer vorwärtsgerichteten Merkmalsselektion, die den Merkmalsraum optimiert und die relevantesten Merkmale selektiert. Die Untersuchungen zeigen, dass die Kombination von verschiedenen strukturellen und spektralen Merkmalen, die aus multispektralem lidar berechnet werden, genauere Ergebnisse erbringt als die Fusion von multispektralen Luftbildern und lidar Daten einer einzelnen Wellenlänge. Die Intensität von multispektralem lidar (1064nm) erweist sich als das einflussreichste Merkmal, was zu einer 10%igen Genauigkeitssteigerung führt. Die experimentellen Ergebnisse zeigen, dass die lidar-basierten Merkmale die wichtigsten Informationsträger zur Waldstrukturanalyse sind. Die in dieser Arbeit entwickelten Methoden besitzen das Potential, auf andere Gebieteübertragen zu werden.

show abstract