Détail de l'auteur
Auteur Diogo N. Cosenza |
Documents disponibles écrits par cet auteur (2)
![](./images/expand_all.gif)
![](./images/collapse_all.gif)
![Tris disponibles](./images/orderby_az.gif)
Silvicultural experiment assessment using lidar data collected from an unmanned aerial vehicle / Diogo N. Cosenza in Forest ecology and management, vol 522 (October-15 2022)
![]()
[article]
Titre : Silvicultural experiment assessment using lidar data collected from an unmanned aerial vehicle Type de document : Article/Communication Auteurs : Diogo N. Cosenza, Auteur ; Jason Vogel, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 120489 Langues : Anglais (eng) Descripteur : [Termes IGN] croissance végétale
[Termes IGN] données allométriques
[Termes IGN] données lidar
[Termes IGN] données localisées 3D
[Termes IGN] modélisation de la forêt
[Termes IGN] Pinus taeda
[Termes IGN] plantation forestière
[Termes IGN] sylviculture
[Vedettes matières IGN] ForesterieRésumé : (auteur) Collecting field data in silvicultural experiments can be challenging and time-consuming. Alternatively, unmanned aerial vehicles using laser scanners (UAV-lidar) can be used for cost-effective data collection in forest stands. This work aims to assess the capability of UAV-lidar to estimate biophysical forest attributes in silvicultural experiments. The showcase experiment refers to the IMPAC II (Intensive Management Practices Assessment Center II), a long-term project of 24 plots aiming to assess the effects of fertilization and weed control on forest growth and nutrient cycling in past and ongoing silvicultural treatments in a second rotation of loblolly pine (Pinus taeda L.) plantation at age 12 years. Treatment performances were assessed based on four biometric attributes related to forest productivity: Growing stock biomass (Mg ha−1), stem volume (m3 ha−1), dominant height (m), and leaf area index (LAI, m2 m−2). We used the area-based approach (ABA) and multiple linear models to characterize these forest attributes in the different silvicultural treatments and use their predictions to run the experiment analysis. Two groups of ALS-derived metrics were tested in the modeling, traditional metrics and a novel group of metrics based on plant area density (PAD) distribution. Models using PAD-based metrics increased the correlation between observed and predicted values (R2) from 0.27–0.40 to 0.50–0.85 when compared to the same models using traditional metrics, while the relative root mean square errors (RMSE%) of the predictions were reduced from 6–18% to 4–12%. Experiment analysis using UAV-lidar data and PAD-based model predictors led to the same results as those using field observations: i) fertilization was the most effective treatment for enhancing stand attributes, especially in terms of biomass, stem volume, and LAI; ii) weed control alone provided marginal improvements in the stands; iii) actively retreating stands in both first and second rotation led to increased growth when compared to the carryover effects. UAV-lidar using PAD-based metrics was effective in characterizing enhanced silvicultural treatments and might benefit studies involving understory assessment. Numéro de notice : A2022-314 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.foreco.2022.120489 En ligne : https://doi.org/10.1016/j.foreco.2022.120489 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102250
in Forest ecology and management > vol 522 (October-15 2022) . - n° 120489[article]Effects of numbers of observations and predictors for various model types on the performance of forest inventory with airborne laser scanning / Diogo N. Cosenza in Canadian Journal of Forest Research, Vol 52 n° 3 (March 2022)
![]()
[article]
Titre : Effects of numbers of observations and predictors for various model types on the performance of forest inventory with airborne laser scanning Type de document : Article/Communication Auteurs : Diogo N. Cosenza, Auteur ; Petteri Packalen, Auteur ; Matti Maltamo, Auteur ; Petri Varvia, Auteur ; Janne Raty, Auteur ; Paola Soares, Auteur ; Margarida Tomé, Auteur ; Jacob L. Strunk, Auteur ; Lauri Korhonen, Auteur Année de publication : 2022 Article en page(s) : pp 385 - 395 Note générale : bibliographie Langues : Français (fre) Anglais (eng) Descripteur : [Termes IGN] forêt boréale
[Termes IGN] lasergrammétrie
[Vedettes matières IGN] Inventaire forestierRésumé : (auteur) Semi- and nonparametric models are popular in the area-based approach (ABA) using airborne laser scanning. It is unclear, however, how many predictors and training plots are needed to provide accurate predictions without overfitting. This work aims to explore these limits for various approaches: ordinary least squares regression (OLS), generalized additive models (GAM), least absolute shrinkage and selection operator (LASSO), random forest (RF), support vector machine (SVM), and Gaussian process regression (GPR). We modeled timber volume (m3·ha–1) for four boreal sites using ABA with 2–39 predictors and 20–500 training plots. OLS, GAM, LASSO, and SVM overfitted as the number of predictors approached the number of training plots. They required ≥15 plots per predictor to provide accurate predictions (RMSE ≤30%). GAM required ≥250 plots regardless of the number of predictors. The number of predictors only mildly affected RF and GPR, but they required ≥200 and ≥250 training plots, respectively. RF did not overfit in any circumstances, whereas GPR overfit even with 500 training plots. Overall, using up to 39 predictors did not generally result in overfit, and for most model types, it resulted in better accuracy for sufficiently large datasets (≥250 plots). Numéro de notice : A2022-948 Affiliation des auteurs : non IGN Thématique : FORET Nature : Article DOI : 10.1139/cjfr-2021-0192 En ligne : https://doi.org/10.1139/cjfr-2021-0192 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100413
in Canadian Journal of Forest Research > Vol 52 n° 3 (March 2022) . - pp 385 - 395[article]