A Comprehensive Comparison of Artificial Neural Networks and Some Regression Modeling Techniques on Dominant Height Predictions of Oriental Spruce in a Mixed Forest, the Northeast Turkey

Background: Dominant height is needed for assessing silvicultural practices in sustainable wood production management. Also, dominant height is used as an important explanatory variable in forest growth and yield models. This paper introduces the evaluation for Artificial Neural Networks and Some Regression Modeling Techniques on Dominant Height Predictions of Oriental Spruce in a Mixed Forest, the Northeast Turkey. Methods: In this study, 873 height-age pairs were obtained from oriental spruce trees in a mixed forest stand. Nonlinear mixed-effects models (NLMEs), autoregressive models (ARM), dummy variable method (DVM), and artificial neural networks (ANNs) were compared to predict dominant height growth. Results: The best predictive model was NLME with single random parameter (root mean square error, RMSE: 0.68 m). The results showed that NLMEs outperformed ARM (RMSE: 1.09 m), DVM in conjunction with ARM (RMSE: 1.09 m), and ANNs (RMSE: from 1.11 to 2.40 m) in majority of the cases. Whereas considering variations among observations by random parameter(s) significantly improved predictions of dominant height, taking into account correlated error terms by autoregressive correlation parameter(s) enhanced slightly the predictions. ANNs generally underperformed compared to NLMEs, ARM, and DVM with ARM. Conclusion: All regression techniques fulfilled the desirable characteristics such as sigmoidal pattern, polymorphism, multiple asymptote, base-age invariance, and inflection point. However, ANNs could not most of these features excluding sigmoidal pattern. Accordingly, ANNs seem to insufficient to assure biological growth assumptions regarding dominant height growth.

Evaluation of the mixed effects model and quantile regression approaches for predicting tree height in larch (Larix olgensis) plantations in northeastern China

Tree height (<i>H</i>) is one of the most important tree variables and is widely used in growth and yield models, and its measurement is often time-consuming and costly. Hence, height-diameter (H-D) models have become a great alternative, providing easy-to-use and accurate tools for <i>H</i> prediction. In this study, H-D models were developed for <i>Larix olgensis</i> in Northeast China. The Chapman-Richards function with three predictors (diameter at breast height, dominant tree height, and relative size of individual trees) performed best. Nonlinear mixed effects (NLME) models and nonlinear quantile regressions (NQR9, 9 quantiles; NQR5, 5 quantiles; and NQR3, 3 quantiles) were further used and improved the generalized H-D model, successfully providing accurate <i>H</i> predictions. In addition, the <i>H</i> predictions were calibrated using several measurements from subsamples, which were obtained from different sampling designs and sizes. The results indicated that the predictive accuracy was higher when calibrated by using any number of height measurements for the NLME model and more than 3 height measurements for the NQR3, NQR5 and NQR9 models. The best sampling strategy for the NLME and NQR models involved sampling the medium-sized trees. Overall, the newly developed H-D models can provide highly accurate height predictions for <i>L. olgensis</i>.

Near Adult Heights and Predictions of Prospective Adult Heights using Automated and Conventional Greulich-Pyle Bone Age Determinations in children with chronic endocrine diseases

Calculation of prospective adult heights (PAH) is associated with considerable bone age interrater variability. Therefore, the new PAH method based on automated bone age (BA) determination (BoneXpert™) was compared to the conventional PAH method by Bayley- Pinneau (BP) based on BA determination according to Greulich and Pyle (GP) and to observed near adult heights. Heights and near adult heights were measured in 82 patients (48 females) with chronic endocrinopathies at age of 10.45 ± 2.12 years and at time of transition to adult care (17.98 ± 3.02 years). Further, BA were assessed according to conventional GP - by three experts- and by BoneXpert™. PAH were calculated using conventional BP tables and BoneXpert™. The conventional and the automated BA determinations revealed a mean difference of 0.25 ± 0.72 years (p = 0.0027). The automated PAH by BoneXpert™ were 156.96 ± 0.86 cm in females and 171.75 ± 1.6 cm in males compared to 153.95 ± 1.12 cm in females and 169.31 ± 1.6 cm in males by conventional BP, respectively, and in comparison to near adult heights 156.38 ± 5.84 cm in females and 168.94 ± 8.18 cm in males, respectively. Conclusion: BA ratings and adult height predictions by BoneXpert™ in children with chronic endocrinopathies abolish rater dependent variability and enhance reproducibility of estimates thereby refining care in growth disorders. Conventional methods may outperform automated analyses in specific cases.

Wildfire smoke-plume rise: a simple energy balance parameterization

The buoyant rise and the resultant vertical distribution of wildfire smoke in the atmosphere have a strong influence on downwind pollutant concentrations at the surface. The amount of smoke injected vs. height is a key input into chemical transport models and smoke modelling frameworks. Due to scarcity of model evaluation data as well as the inherent complexity of wildfire plume dynamics, smoke injection height predictions have large uncertainties. In this work we use the coupled fire–atmosphere model WRF-SFIRE configured in large-eddy simulation (LES) mode to develop a synthetic plume dataset. Using this numerical data, we demonstrate that crosswind integrated smoke injection height for a fire of arbitrary shape and intensity can be modelled with a simple energy balance. We introduce two forms of updraft velocity scales that exhibit a linear dimensionless relationship with the plume vertical penetration distance through daytime convective boundary layers. Lastly, we use LES and prescribed burn data to constrain and evaluate the model. Our results suggest that the proposed simple parameterization of mean plume rise as a function of vertical velocity scale offers reasonable accuracy (30 m errors) at little computational cost.

Sensitivity of Flood Hazard and Damage to Modelling Approaches

Combination of uncertainties in water level and wave height predictions for extreme storms can result in unacceptable levels of error, rendering flood hazard assessment frameworks less useful. A 2D inundation model, LISFLOOD-FP, was used to quantify sensitivity of flooding to uncertainty in coastal hazard conditions and method used to force the coastal boundary of the model. It is shown that flood inundation is more sensitive to small changes in coastal hazard conditions due to the setup of the regional model, than the approach used to apply these conditions as boundary forcing. Once the threshold for flooding is exceeded, a few centimetres increase in combined water level and wave height increases both the inundation and consequent damage costs. Improved quantification of uncertainty in inundation assessments can aid long-term coastal flood hazard mitigation and adaptation strategies, to increase confidence in knowledge of how coastlines will respond to future changes in sea-level.

Wildfire smoke-plume rise: a simple energy balance parameterization

The buoyant rise and the resultant vertical distribution of wildfire smoke in the atmosphere have a strong influence on downwind pollutant concentrations at the surface. The amount of smoke injected vs. height is a key input into chemical transport models and smoke modelling frameworks. Due to scarcity of model evaluation data as well as inherent complexity of wildfire plume dynamics, smoke injection height predictions have large uncertainties. In this work we use a coupled fire-atmosphere model WRF-SFIRE configured in large eddy simulation (LES) mode to develop a synthetic plume dataset. Using this numerical data, we demonstrate that crosswind integrated smoke injection height for a fire of arbitrary shape and intensity can be modelled with a simple energy balance. We introduce two forms of updraft velocity scales that exhibit a linear dimensionless relationship with the plume vertical penetration distance through daytime convective boundary layers. Lastly, we use LES and prescribed burn data to constrain and evaluate the model. Our results suggest that the proposed simple parameterization of mean plume rise as a function of vertical velocity scale offers reasonable accuracy (30 m errors) at little computational cost.