Prediction and model-assisted estimation of diameter distributions using Norwegian national forest inventory and airborne laser scanning data

Räty, Janne; Astrup, Rasmus; Breidenbach, Johannes

doi:10.1139/cjfr-2020-0440

Cited by 8 publications

(6 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There exist few comparable studies that have carried out MA estimations using field and ALS data. Räty et al (2021a) reported an RE value of 2.11 for N with the MA estimator that used a model constructed for the prediction of DBH distributions using Norwegian NFI and ALS data. Räty et al (2021a) also reported a much smaller systematic error that was 1.4% compared with 39.0% in our study.…”

Section: Discussionmentioning

confidence: 99%

“…Räty et al (2021a) reported an RE value of 2.11 for N with the MA estimator that used a model constructed for the prediction of DBH distributions using Norwegian NFI and ALS data. Räty et al (2021a) also reported a much smaller systematic error that was 1.4% compared with 39.0% in our study. We also re-analyzed a dataset of NFI and ALS data, and corresponding forest attribute models, originally used by Hauglin et al (2021) for large-scale mapping of forest attributes in Norway.…”

Section: Discussionmentioning

confidence: 99%

“…There exists no design-unbiased variance estimator available for systematic sampling designs. However, simulation studies and empirical analysis have shown that a variance estimator assuming simple random sampling (SRS) provides acceptable variance estimates (always conservative) also under systematic sampling designs in the case of low sampling proportions that are typical for NFIs (Magnussen et al 2020;Räty et al 2021a). A variance estimator for μ assuming SRS (Cochran 1977, sec.…”

Section: Direct Estimationmentioning

confidence: 99%

See 2 more Smart Citations

Assessing and mitigating systematic errors in forest attribute maps utilizing harvester and airborne laser scanning data

et al. 2023

Self Cite

View full text Add to dashboard Cite

Cut-to-length harvesters collect useful information for modeling relationships between forest attributes and airborne laser scanning (ALS) data. However, harvesters operate in mature forests, which may introduce selection biases that can result in systematic errors in harvester data-based forest attribute maps. We fitted regression models (harvester models) for volume (V), height (HL), stem frequency (N), above-ground biomass, basal area, and quadratic mean diameter (QMD) using harvester and ALS data. Performances of the harvester models were evaluated using national forest inventory plots in an 8.7 Mha study area. We estimated biases of large-area synthetic estimators and compared efficiencies of model-assisted (MA) estimators with field data-based direct estimators. The harvester models performed better in productive than unproductive forests, but systematic errors occurred in both. The use of MA estimators resulted in efficiency gains that were largest for HL (relative efficiency, RE=6.0) and smallest for QMD (RE=1.5). The bias of the synthetic estimator was largest for N (39%) and smallest for V (1%). The latter was due to an overestimation of deciduous and an underestimation of spruce forests that by chance balanced. We conclude that a probability sample of reference observations may be required to ensure the unbiasedness of estimators utilizing harvester data.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Direct Estimationmentioning

confidence: 99%

See 1 more Smart Citation

Assessing and mitigating systematic errors in forest attribute maps utilizing harvester and airborne laser scanning data

et al. 2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…These results indicate that adding spatially explicit information on forest attributes are beneficial for terrain evaluation. Other forest properties such as diameter distributions can be estimated and mapped from laser scanner data (Räty et al 2021) and should be considered as in put to avalanche hazard models. Furthermore, avalanche hazard models can be further developed to use forest properties are used as continuous values as input.…”

Section: Discussionmentioning

confidence: 99%

The utility of forest attribute maps for automated Avalanche Terrain Exposure Scale (ATES) modelling

Schumacher

Larsen

McLean

et al. 2022

Scandinavian Journal of Forest Research

Self Cite

View full text Add to dashboard Cite

Fatalities associated with recreational activities occur every year as a result of snow avalanches. Terrain classification systems, such as the Avalanche Terrain Exposure Scale (ATES) are designed to provide guidance for safe route finding and this system has been automated (AutoATES). ATES classifies terrain into the three classes simple, challenging, and complex. Forests can provide some protection from avalanches, and these can be incorporated into avalanche hazard models. The objective of this study was to map relevant forest attributes (stem density and canopy cover) based on National Forest Inventory and remote sensing data and, subsequently, use these forest attributes as input to the AutoATES model to improve avalanche hazard maps. We predicted stem density with species-specific mixed-effects models and directly calculated canopy cover using airborne laser scanning data in a 20 Mha study area ranging from the arctic circle to southern Norway. We mapped these forest attributes for 16 m x 16 m pixels, which were used as input for the AutoATES model. The uncertainty of the stem number and canopy cover maps were 30% and 32%, respectively. The overall classification accuracy of 52 ski touring routes in Western Norway with a total length of 282 km increased by up to 12% when utilizing the mapped forest attributes, compared to the model without forest information. The F1 score for the three predicted ATES classes improved by up to 31%, 9%, and 6% for the three classes, respectively, when including a forest attribute in the AutoATES model. We conclude that large-scale fine-resolution forest attribute maps are valuable data in the modelling of avalanche hazards.

show abstract

“…Knowing more about how non-normal tree size distributions influence machine productivity at a stand level could contribute to more accurate operations planning and assortment forecasting, but a priori estimates of diameter distributions would be required if they are to be used in prediction. Aerial laser scanning (ALS) has proven to be useful in generating such distributions in coniferous forests (Gobakken & Naesset 2004;Räty et al 2021). Productivity models can be developed on the basis of tree distribution data collected from the onboard computer and StanForD data library (Kemmerer & Labelle 2021), or further enhanced by combining it with the aforementioned ALS data, as demonstrated by Söderberg et al (2021).…”

Section: Productivity Of Machines In Target Tree Dimensionsmentioning

confidence: 99%

Understanding the Influence of Tree Size Distribution on the CTL Harvesting Productivity of Two Different Size Harvesting Machines

Ackerman

Bekker

Astrup

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Tree size determines harvester productivity. Stand management determines tree size distributions. Understanding the effect of tree size distribution on harvesting productivity is central for optimizing management of operations. To investigate tree size distribution on harvester productivity, productivity functions for a medium and larger-sized harvesters were applied to harvester derived tree size distributions from 35 clearfelled pine stands. These functions were applied to a normal distribution of trees covering the same tree size ranges. Productivity differences were analysed on a stand-by-stand basis. Results showed that for the large harvester, productivity rates remained constant (67.1 vs 67.6 m3 PMH− 1) indicating relatively little sensitivity to variations in tree size distributions. Although the standard deviation (SD) halved from 11.6 to 5.6 in the case of the uniform tree distribution. The smaller harvesters, productivity decreased by 15% from 47.3 to 40.1 m3 PMH− 1 and the coefficient of variation (CV) by 6% in the same transition to a uniform distribution. Further investigation was done on more skewed tree size distributions, a family of nine Weibull distributions was generated, representing combinations of three mean DBH classes (25cm, 30 cm, and 35cm) and three levels of CV (15%, 20%, 25%), for each DBH class. Results clearly indicate that different distribution shapes have different effects on different machine sizes, and that a low CV correlates to a higher productivity in larger tree sizes. A more uniform tree size distribution also provides more predictable results (lower CV), which would promote machine scheduling and result in fewer discrepancies on production rates.

show abstract

Prediction and model-assisted estimation of diameter distributions using Norwegian national forest inventory and airborne laser scanning data

Cited by 8 publications

References 50 publications

Assessing and mitigating systematic errors in forest attribute maps utilizing harvester and airborne laser scanning data

Assessing and mitigating systematic errors in forest attribute maps utilizing harvester and airborne laser scanning data

The utility of forest attribute maps for automated Avalanche Terrain Exposure Scale (ATES) modelling

Understanding the Influence of Tree Size Distribution on the CTL Harvesting Productivity of Two Different Size Harvesting Machines

Contact Info

Product

Resources

About