A random forest model for basal area increment predictions from national forest inventory data

Jevšenak, Jernej; Skudnik, Mitja

doi:10.1016/j.foreco.2020.118601

Cited by 51 publications

(27 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Greater MAPE in terms of below-ground biomass are to be expected since the entire calibration of the model is based on the inventory database which only encompasses above-ground biomass. National reports in the framework of the UNFCCC are based on the National Forest Inventory (NFI) [42][43][44], which is a different data source than that used in our study. Our simulations are based on data of the SFS which is collected in the framework of forest inventories for the purpose of forest management planning.…”

Section: The Impact Of Harvesting On Carbon Stock Dynamics In Sloveni...mentioning

confidence: 99%

The Effect of Harvesting on National Forest Carbon Sinks up to 2050 Simulated by the CBM-CFS3 Model: A Case Study from Slovenia

Jevšenak

Klopčić

Mali

2020

Forests

Self Cite

View full text Add to dashboard Cite

With the advent of global warming, forests are becoming an increasingly important carbon sink that can mitigate the negative effects of climate change. An understanding of the carbon dynamics of forests is, therefore, crucial to implement appropriate forest management strategies and to meet the expectations of the Paris Agreement with respect to international reporting schemes. One of the most frequently used models for simulating the dynamics of carbon stocks in forests is the Carbon Budget Model of the Canadian Forest Sector (CBM-CFS3). We applied this model in our study to evaluate the effects of harvesting on the carbon sink dynamics in Slovenian forests. Five harvesting scenarios were defined: (1) business as usual (BAU), (2) harvesting in line with current forest management plans (PLAN), (3) more frequent natural hazards (HAZ), (4) high harvest (HH) and (5) low harvest (LH). The simulated forest carbon dynamics revealed important differences between the harvesting scenarios. Relative to the base year of 2014, by 2050 the carbon stock in above-ground biomass is projected to increase by 28.4% (LH), 19% (BAU), 10% (PLAN), 6.5% (HAZ) and 1.2% (HH). Slovenian forests can be expected to be a carbon sink until harvesting exceeds approximately 9 million m3 annually, which is close to the calculated total annual volume increase. Our results are also important in terms of Forest Reference Levels (FRL), which will take place in European Union (EU) member states in the period 2021–2025. For Slovenia, the FRL was set to −3270.2 Gg CO2 eq/year, meaning that the total timber harvested should not exceed 6 million m3 annually.

show abstract

Section: The Impact Of Harvesting On Carbon Stock Dynamics In Sloveni...mentioning

confidence: 99%

The Effect of Harvesting on National Forest Carbon Sinks up to 2050 Simulated by the CBM-CFS3 Model: A Case Study from Slovenia

Jevšenak

Klopčić

Mali

2020

Forests

Self Cite

View full text Add to dashboard Cite

show abstract

“…The explained variance is usually lower in mixed-species unevenly aged forests that in even-aged pure stands [2]. This result is attributed to the significant differenc the growth of various tree species in the mixed forest, which causes difficulties in pre tion.…”

Section: Evaluation Of Random-forest Modelmentioning

confidence: 99%

“…Forests with mixed species have higher productivity, higher temporal stability, a lower risk of biotic and abiotic disturbances, and a more diverse ecosystem. However, it is challenging to model their growth because of complex communities [2]. Basal area growth increment (BAI) is particularly suitable for modelling tree growth among the other measurements because it is directly related to the diameter at breast height (DBH), thus making it more reliable [3,4].…”

Section: Introductionmentioning

confidence: 99%

The Effect of Stand Density, Biodiversity, and Spatial Structure on Stand Basal Area Increment in Natural Spruce-Fir-Broadleaf Mixed Forests

Liu

Zhou

et al. 2022

Forests

View full text Add to dashboard Cite

Forest trees exhibit a large variation in the basal area increment (BAI), and the variation is attributed to the stand density, biodiversity, and stand spatial structure. Studying and quantifying the effect of these above variables on tree growth is vital for future forest management. However, the stand spatial structure based on neighboring trees has rarely been considered, especially in the mixed forests. This study adopted the random-forest (RF) algorithm to model and interpret BAI based on stand density, biodiversity, and spatial structure. Fourteen independent variables, including two stand density predictors, four biodiversity predictors, and eight spatial structure predictors, were evaluated. The RF model was trained for the whole stand, three tree species groups (gap, neutral, and shade_tolerant), and two tree species (spruce and fir). A 10-fold blocked cross-validation was then used to optimize the hyper-parameters and evaluate the models. The squared correlation coefficients (R2) for the six groups were 0.233 for the whole stand, 0.575 for fir, 0.609 for shade_tolerant, 0.622 for neutral, 0.722 for gap, and 0.730 for spruce. The Stand Density Index (SDI) was the most-important predictor, suggesting that BAI is primarily restricted by competition. BAI and species biodiversity were positively correlated for the whole stand. The stands were expected to be randomly distributed based on the relationship between the uniform angle index (W) and growth. The relationship between dominance (U) and BAI indicated that small trees should be planted around the light-demanding tree species and vice versa. Of note, these findings emphasize the need to consider the three types of variables in mixed forests, especially the spatial structure factors. This study may help make significant advances in species composition, spatial arrangement, and the sustainable development of mixed forests.

show abstract

“…For the former, although various machine learning models are employed for natural disasters [7], they are almost all related to BPNN and SVR in the field of economic loss forecasting. Ensemble learning is a widely-used algorithm [13][14][15][16] that combines several machine learning techniques into an ensemble model to reduce deviation and improve prediction accuracy [17]. Zhao et al [18] used an ensemble learning model Adaboost-BPNN for forecasting direct economic losses of marine disasters.…”

Section: Introductionmentioning

confidence: 99%

A Data Augmentation-Based Evaluation System for Regional Direct Economic Losses of Storm Surge Disasters

Sun

Wang

2021

IJERPH

View full text Add to dashboard Cite

The accurate prediction of storm surge disasters’ direct economic losses plays a positive role in providing critical support for disaster prevention decision-making and management. Previous researches on storm surge disaster loss assessment did not pay much attention to the overfitting phenomenon caused by the data scarcity and the excessive model complexity. To solve these problems, this paper puts forward a new evaluation system for forecasting the regional direct economic loss of storm surge disasters, consisting of three parts. First of all, a comprehensive assessment index system was established by considering the storm surge disasters’ formation mechanism and the corresponding risk management theory. Secondly, a novel data augmentation technique, k-nearest neighbor-Gaussian noise (KNN-GN), was presented to overcome data scarcity. Thirdly, an ensemble learning algorithm XGBoost as a regression model was utilized to optimize the results and produce the final forecasting results. To verify the best-combined model, KNN-GN-based XGBoost, we conducted cross-contrast experiments with several data augmentation techniques and some widely-used ensemble learning models. Meanwhile, the traditional prediction models are used as baselines to the optimized forecasting system. The experimental results show that the KNN-GN-based XGBoost model provides more precise predictions than the traditional models, with a 64.1% average improvement in the mean absolute percentage error (MAPE) measurement. It could be noted that the proposed evaluation system can be extended and applied to the geography-related field as well.

show abstract

A random forest model for basal area increment predictions from national forest inventory data

Cited by 51 publications

References 57 publications

The Effect of Harvesting on National Forest Carbon Sinks up to 2050 Simulated by the CBM-CFS3 Model: A Case Study from Slovenia

The Effect of Harvesting on National Forest Carbon Sinks up to 2050 Simulated by the CBM-CFS3 Model: A Case Study from Slovenia

The Effect of Stand Density, Biodiversity, and Spatial Structure on Stand Basal Area Increment in Natural Spruce-Fir-Broadleaf Mixed Forests

A Data Augmentation-Based Evaluation System for Regional Direct Economic Losses of Storm Surge Disasters

Contact Info

Product

Resources

About