Exploring the use of learning techniques for relating the site index of radiata pine stands with climate, soil and physiography

González-Rodríguez, Miguel A.; Diéguez-Aranda, Ulises

doi:10.1016/j.foreco.2019.117803

Cited by 16 publications

(7 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The geostatistical models of ordinary kriging or regression kriging provided the most precise predictions of Site Index among the compared methods. Previous studies have focussed primarily on comparisons of precision between models of Site Index, that do not include a geostatistical component, demonstrating that non-parametric generally outperform parametric models [28,40,41]. Our results extend this research through demonstrating that addition of a spatial component to both of these model types outperform models without this component.…”

Section: Discussionsupporting

confidence: 76%

“…A large number of modelling methods with varying levels of complexity have been used to predict Site Index for a wide range of forest species growing in Europe, North America and New Zealand. These methods range from relatively simple approaches such as multiple linear regression [3,[17][18][19][20][21][28][29][30][31][32][33][34][35][36][37][38][39] to more complex parametric methods such as Partial Least Squares, Lasso, Elastic Net, Least Angle Regression and Infinitesimal Forward Stagewise Regression [40]. A wide range of non-parametric methodologies have also been used to model Site Index including Random Forests [41,42], Boosted Trees [28,29], Classification and Regression Trees [28,29], Neural Networks [29], Generalised Additive Models [28,29,43], and Multivariate Adaptive Regression Splines [40].…”

Section: Introductionmentioning

confidence: 99%

“…A comparison of seven modelling methods using data collected from northwest Spain found the non-parametric Multivariate Adaptive Regression Splines (MARS) most precisely predicted Site Index, which was closely followed by the parametric methods of stepwise regression and PLS. The best MARS model accounted for 50% of the variation in the data using 13 predictors [40].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Reducing the Uncertainty of Radiata Pine Site Index Maps Using an Spatial Ensemble of Machine Learning Models

Gavilán-Acuña¹,

Olmedo²,

Mena-Quijada³

et al. 2020

Preprint

View full text Add to dashboard Cite

Site Index has been widely used as an age normalised metric to account for variation in forest height at a range of spatial scales. Although previous research has used a range of modelling methods to describe regional variation in Site Index little research has examined gains that can be achieved through use of regression kriging or spatial ensemble methods. In this study an extensive set of environmental surfaces were used as covariates to predict Site Index measurements covering the environmental range of \textit{Pinus radiata} D. Don plantations in Chile. Using this dataset, the objectives of this research were to (i) compare predictive precision of a range of geostatistical, parametric and non-parametric models, (ii) determine if significant gains in precision can be attained through use of regression kriging, (iii) evaluate the precision of a spatial ensemble model that utilises predictions from the five most precise models, through using the model prediction with lowest error for a given pixel and (iv) produce a map of Site Index across the study area. The five most precise models were all geostatistical and included ordinary kriging and four regression kriging models that were based on partial least squares or random forests. A spatial ensemble model constructed from these five models was the most precise of those developed (RMSE = 1.851 m, RMSE% = 6.38%) and had relatively little bias. Climatic and edaphic variables were the strongest determinants of Site Index and in particular, variables related to soil water balance were well represented within the most precise predictive models. These results highlight the utility of predicting Site Index using a range of approaches, as these can be used to construct a spatial ensemble that may be more precise than predictions from the constituent models.

show abstract

Section: Discussionsupporting

confidence: 76%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reducing the Uncertainty of Radiata Pine Site Index Maps Using an Spatial Ensemble of Machine Learning Models

Gavilán-Acuña¹,

Olmedo²,

Mena-Quijada³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The current research trend aims at developing growth-environment relationships through predictive modelling, mainly focusing on the site index (SI), the most frequent empirical indicator of forest productivity [9]. A variety of supervised learning techniques have been used for this purpose [10][11][12], yielding, overall, successful results (R 2 ∼ 0.3-0.7).…”

Section: Introductionmentioning

confidence: 99%

Forecasting Variations in Profitability and Silviculture under Climate Change of Radiata Pine Plantations through Differentiable Optimization

González-Rodríguez

Méndez

Diéguez-Aranda

2021

Forests

View full text Add to dashboard Cite

Climate change might entail significant alterations in future forest productivity, profitability and management. In this work, we estimated the financial profitability (Soil Expectation Value, SEV) of a set of radiata pine plantations in the northwest of Spain under climate change. We optimized silvicultural interventions using a differentiable approach and projected future productivity using a machine learning model basing on the climatic predictions of 11 Global Climate Models (GCMs) and two Representative Concentration Pathways (RCPs). The forecasted mean SEV for future climate was lower than current SEV (∼22% lower for RCP 4.5 and ∼29% for RCP 6.0, with interest rate = 3%). The dispersion of the future SEV distribution was very high, alternatively forecasting increases and decreases in profitability under climate change depending on the chosen GCM. Silvicultural optimization considering future productivity projections effectively mitigated the potential economic losses due to climate change; however, its ability to perform this mitigation was strongly dependent on interest rates. We conclude that the financial profitability of radiata pine plantations in this region might be significantly reduced under climate change, though further research is necessary for clearing the uncertainties regarding the high dispersion of profitability projections.

show abstract

“…The site index, which is the dominant tree height at a reference stand age, is used as an index of site productivity [3,4]. The relationship between environmental factors and site index in even-aged forests, such as plantations [5][6][7] and natural forests [8,9], is well studied.…”

Section: Introductionmentioning

confidence: 99%

Topographic Factors and Tree Heights of Aged Cryptomeria japonica Plantations in the Boso Peninsula, Japan

Tange

2020

Forests

View full text Add to dashboard Cite

This study aimed to clarify the environmental factors limiting the height of aged Cryptomeria japonica trees. The study was conducted on C. japonica plantations of about 100 years old at the Boso Peninsula, Japan, where the climatic conditions are almost uniform. Twenty-eight plots measuring 10 × 10 m were established on ridges, and 40 plots were established on the middle or lower sections of slopes. The stand ages ranged from 93 to 115 years old. The height of the tallest tree and soil depth (SD) were measured in each plot, and the wetness index (WI) and openness (OP) of each plot were calculated using a digital elevation model. The tree height at the 100-year age (H100) was estimated. The H100 ranged from 16.2 to 44.9 m and was significantly correlated with the logWI (r = 0.78) and OP (r = −0.70). SD and H100 were significantly correlated in the plots on the ridges but not in the plots on the middle or lower sections of slopes. It indicated that soil water retention capacity might limit tree height in the relatively dry soil conditions. The coefficient of determination adjusted by the number of parameters for H100 predicted using multiple regression analysis with environmental factors of logWI, logWI and OP, or logWI, OP and SD were 0.60, 0.69, and 0.73, respectively. The inclusion of OP and SD in the model improved the prediction of H100, suggesting that the wind and rooting depth could be the influencing factors in determining the height of aged trees. The findings of this study could be used in the planning and management of forestry plantations of long rotation system.

show abstract

Exploring the use of learning techniques for relating the site index of radiata pine stands with climate, soil and physiography

Cited by 16 publications

References 45 publications

Reducing the Uncertainty of Radiata Pine Site Index Maps Using an Spatial Ensemble of Machine Learning Models

Reducing the Uncertainty of Radiata Pine Site Index Maps Using an Spatial Ensemble of Machine Learning Models

Forecasting Variations in Profitability and Silviculture under Climate Change of Radiata Pine Plantations through Differentiable Optimization

Topographic Factors and Tree Heights of Aged Cryptomeria japonica Plantations in the Boso Peninsula, Japan

Contact Info

Product

Resources

About