Forest Volume Research Articles

Wood-living beetle diversity is enriched by surrounding broadleaf forest in a conifer forestry landscape

ContextHumans have drastically changed many of the world’s forests since the industrial revolution, with northern European forests among the most affected. Research is needed on consequent effects on forest biodiversity. Here I focus on the conservation-relevant wood-living beetles. Whilst previous studies have examined the effect of local stand characteristics on beetle diversity, the influence of the surrounding landscape on alpha diversity may potentially be more important.ObjectivesThe aim of this study is to judge the influence and scale of effect of four landscape variables (broadleaf forest volume, area of mature forest, area of conservation forest, area of clear-cut) on alpha diversity of wood-living beetles as a whole, and ecological groups reflecting potentially differing dispersal ecologies (fungivores, wood consumers, predators, red-listed species).MethodsThe landscape variables were analyzed at three scales (within 250, 1250 and 2500 m radius) and related to wood-living beetle diversity sampled using flight interception traps in 73 managed and unmanaged forest stands in the conifer forestry dominated hemiboreal landscape of south-central Sweden.ResultsWood-living beetle diversity as a whole was positively associated with broadleaf volume at the 2500 m scale, with a weaker association for fungivore diversity at a varying scale. Surprisingly, both predator diversity (varying scale) and red-listed diversity (2500 m scale) was weakly negatively associated with mature forest.ConclusionsThe results highlight the importance of broadleaf trees for the diversity of saproxylic beetles in conifer-dominated forestry landscapes. Further, research and management should more closely consider contrasting responses between species with potentially differing dispersal ecologies, and at different scales.

Improving airborne laser scanning-based species-specific forest volume estimation using sentinel-2 time series

ABSTRACT Species-specific timber volume estimates are required to support forest planning and conservation. We evaluated whether additional predictors from a Sentinel-2 time series can improve airborne laser scanning (ALS)-based estimation of species-specific timber volumes. Furthermore, we determined the satellite image dates that provided the greatest improvement in accuracy. The Sentinel-2 time series was constructed to cover 1 March–30 November time-period, with a focus on late spring and early summer. The estimation was done using the k Most Similar Neighbor method and predictors extracted from the ALS data and Sentinel-2 images. Our best model included both ALS and Sentinel-2 time-series predictors, and the relative root mean square error (RMSE) values for pine, spruce and deciduous timber volumes were 40.8%, 57.0% and 51.3%, respectively (mean 49.7%). All deciduous trees were treated as one species. When bands from an individual image were used instead of the time series, the best result was obtained with an image from September where the respective relative RMSE values were 42.2% (deciduous), 58.4% (pine) and 60.6% (spruce) with a mean value of 53.7%. A fusion of a Sentinel-2 time-series and ALS data can improve species-specific estimation results compared to the use of individual Sentinel-2 images or ALS only.

Species Substitution and Changes in the Structure, Volume, and Biomass of Forest in a Savanna

Research related to Cerradão vegetation focuses more on the floristic-structural aspect, with rare studies on the quantification of volume and biomass stocks, and even fewer investigating the increments of these attributes. Using a systematic sampling method with subdivided strips and 400 m2 plots, the density found was 1135, 1165, and 1229 trees/ha in 2012, 2020, and 2023, respectively, in Lajeado State Park, Tocantins State, Brazil. Volume was estimated using the equation v=0.000085D2.122270H0.666217, and biomass was estimated using the equation AGB=0.0673ρD2H0.976. Vegetation dynamics were assessed using growth increment, recruitment, mortality, turnover rate, and time. The results indicated that dynamics have increased since the start of monitoring. Typical Cerrado species, in the strict sense, were replaced by those from forest environments. The total production in volume and biomass was 160.91 m3/ha and 118.10 Mg/ha, respectively, in 2023. The species of Emmotum nitens, Mezilaurus itauba, Ocotea canaliculata, and Sacoglottis guianensis showed the highest increment values in volume and biomass. For the community, the average values were 4.04 m3/ha/year and 3.54 Mg/ha/year. The community has not yet reached its carrying capacity and stores a significant amount of biomass. This is influenced by the transition of the study area from an exploited environment to a conservation unit (park) and by its location in a transitional area with the Amazon biome.

Economic Valuation of Timber in Phratamnak Muang Nakhon Area, Phrom Khiri District, Nakhon Si Thammarat Province

This research aimed to explore the diversity of tree species and evaluate the economic value of wood in the area of Phratamnak Mueang Nakhon. A sample plot of 20 x 50 meters with a total of 5 plots was designed for surveying plant species and number of plants. The value of wood, logging costs including net worth were evaluated. The results showed that Pratumnak Muang Nakhon had a total area of 52 rai with 3,777 trees of 16 species and 12 families. Champathongs was found to be the most abundance followed by Mangosteen, Niang, Longkong, Sato, durian, Ma Fai Krathon, Mahogany, Rambutan, Langsat, Yang Na, Artificial, and Mafaika, respectively. The total forest volume was 3,399.28 cubic meters, which is an economic value. The total forest timber cost was 22,232,800.65 baht, with a logging cost of 2,726,322.93 baht and a net worth of 19,506,477.72 baht.

Quantification of tree growth change under climate change using National Forest Inventory of Korea

To achieve carbon neutrality in the face of climate change, accurately estimating forest carbon uptake is crucial. Environmental changes, such as temperature increases or precipitation fluctuations caused by climate change, can alter the growth responses of each species and act as a driver for changes in forest productivity. In particular, using a single-species growth model increases uncertainty in predicting stand growth, given the high species diversity in temperate forests in East Asia. Therefore, in this study, we identified the growth curve of 17 species-specific distributed in Korean temperate forest as well as the growth trend over the last 30 years (1976–2005). For this purpose, the species distribution and growth characteristics of 17 major Korean species of trees were analyzed using 13,808 tree-ring series from the 5th National Forest Inventory, which was conducted between 2006 and 2010. The growth characteristics of each species were analyzed using the average growth curve specific to each species, and the yearly growth trend was examined through residual analysis of the average growth curve. The distribution range of the 17 major tree species varied by species, as did the growth rate, ranging from 7 cm² · year⁻¹ to 17 cm² · year⁻¹ for the 30-year-old average basal area increment (BAI). Over time, fast-growing tree species exhibited initially high BAI, followed by a sharp decline from the 2000s. In contrast, broadleaf trees showed a continuous increase. Furthermore, a comparison of the total volume of individual trees and the volume of the yield table revealed that accurately estimating the actual forest volume was challenging. This study contributes to improving the accuracy of productivity predictions for temperate forests, underscoring the necessity to investigate the interaction between climate and forest succession in the future.

Spatio-Temporal Assessment of Heterogeneity by Logging Intensity in a Federal Concession Area in the Brazilian Amazon

The logging intensity often does not take into account the spatial heterogeneity of the forest volume of commercial native species in the Brazilian Amazon. This study aims to evaluate the spatio-temporal heterogeneity distribution by assessing logging intensity and its effects on the volumetric stock and abundance of commercial species, with a focus on sustainable management practices. This study was conducted in the Saracá-Taquera National Forest in the Brazilian Amazon. Forest inventory data, elevation, and PlanetScope satellite images were integrated into a geographic information system. The information was aggregated into regular 1-hectare cells for the times before, during, and after logging (t0, t1, and t2). The unsupervised classification algorithm k-means with four clusters was used to analyze heterogeneity. Before logging, areas with higher commercial volumes were distant from water bodies, while areas with lower elevation had lower wood stocks. Logging intensity was generally low, concentrating on a few trees per hectare. Logging in the study area revealed a heterogeneous spatial distribution by intensifying in areas with the highest wood stocks. These results suggest that, in addition to the recommended logging intensity according to legislation, forest heterogeneity should be considered by the manager, promoting adaptive strategies to ensure the conservation of forest resources.

What is the potential of bioCCS to deliver negative emissions in Norway? From biomass mapping to a window of negative emissions potential

Negative emissions have been highlighted as a key component of achieving the net-zero ambition. However, ground-up approaches are necessary to better understand the realistic potential of negative emissions technologies at the national or continental level. Such an approach was applied in the present study to understand the potential of bioenergy with carbon capture and storage to deliver negative emissions in Norway, starting from mapping and quantification of biomass until the derivation of a window of negative emission potential.The results indicate that bioenergy with carbon capture and storage could enable between 1 and 13 MtCO2/y of negative emissions, with a more probable range between 2 and 8 MtCO2/y, at least in the coming decades. These values are drastically higher than the potential identified in previous studies, thus highlighting the importance of bottom-up approaches, like the one adopted here, to better estimate the negative emissions potential that could be delivered by bioenergy with carbon capture and storage.In terms of biomass, the strongest potential for negative emissions comes from the integration of forestry resources and activities with bioenergy with carbon capture and storage. However, it is important to ensure that this integration takes place in a sustainable way and does not result in a decrease in the standing volume of the Norwegian forest for multiple reasons. Integrating waste with bioenergy with carbon capture and storage also represents a significant potential to enable negative emissions, especially as a substantial fraction of waste is already integrated with energy production. Finally, biomasses from agriculture and seaweed farming are expected to have a limited potential to enable negative emissions, although seaweed farming could take a more significant role towards the second half of the century, depending on the development of this sector.

Mapping and Estimating Forest Stand Volume using Machine Learning Methods and Multi-Spectral Sentinel 2 Data

Sustainable forest management necessitates the mapping and estimation of forest stand attributes such as density, volume, basal area, and aboveground biomass. This study was conducted to explore the potential of geographic information systems (GIS), remote sensing, machine learning, and field inventories to estimate the forest stand volume of natural and plantation forests within watersheds in the Abra River Basin. The common machine learning regression techniques, which are random forest (RF), k-nearest neighbors (KNN), and support vector machines (SVM), were used to model and predict forest stand volume. The validation of the three machine learning methods showed that the best model to estimate and map forest stand volume is the RF algorithm (R2 = 0.42, RMSE = 0.40 m3/plot, MAE = 0.31 m3/plot). Topographic variables such as the Digital Elevation Model (DEM) and the spectral band Near Infrared (NIR) were the most important variables in predicting forest stand volume. The estimated forest stand volume using the RF model ranged from 33 to 115 m3/ha, with a mean of 59 m3/ha. The results of this study revealed that forest volume can be measured using freely available satellite data and machine learning techniques.

Effects of Climate Change on Scots Pine (Pinus sylvestris L.) Growth across Europe: Decrease of Tree-Ring Fluctuation and Amplification of Climate Stress

From an economic perspective, Scots pine (Pinus sylvestris L.) is one of Europe’s most important tree species. It is characterized by its wide ecological adaptability across its natural range. This research aimed to evaluate the forest structure, productivity and especially radial growth of heterogenous pine stands in 16 research plots in the Czech Republic, Poland, Spain and Great Britain. The study assessed the tree-ring formation and its relationship to climate change for each country, using 163 dendrochronological samples. The stand volume of mature pine forest ranged between 91 and 510 m3 ha−1, and the carbon sequestration in the tree biomass was 40–210 t ha−1. The stands had a prevailing random distribution of trees, with a high vertical structure close to selection forests (forest stands with typical very diverse height, diameter and age structure). Spectral analyses showed a substantial decrease in fluctuations in the tree-ring index and a loss in natural growth cyclicity in the last thirty years. The results also evinced that mean air temperature was the most important factor influencing the radial growth compared to precipitation totals. Pine thrives in precipitation-stable locations, as shown by the results from Great Britain. The conclusions of this study confirm the fundamental effect of ongoing global climate change on the dynamics and growth of pine forests in Europe.

Patterns of deadwood volume and dynamics in Slovenian forests

Patterns of deadwood volume and dynamics in Slovenian forests

A Fourier–Legendre Polynomial Forest Height Inversion Model Based on a Single-Baseline Configuration

In this article, we propose a Fourier–Legendre (FL) polynomial forest height estimation algorithm based on low-frequency single-baseline polarimetric interferometric synthetic aperture radar (PolInSAR) data. The algorithm can obtain forest height with a single-baseline PolInSAR configuration while capturing a high-resolution vertical profile for the forest volume. This is based on the consideration that the forest height remains constant within neighboring pixels. Meanwhile, we also assume that the coefficients of the FL polynomials remain unchanged within neighboring pixels, except for the last polynomial coefficient. The idea of using neighboring pixels to increase the observations provides us with the possibility to obtain high-order FL polynomials. With this approach, it is possible to obtain a high-resolution vertical profile that is suitable for forest height estimation without losing too much spatial resolution. P-band PolInSAR data acquired in Mabounie in Gabon and Krycklan in Sweden were selected for testing the proposed algorithm. The results show that the algorithm outperforms the random volume over ground (RVoG) model by 18% and 16.7% in forest height estimation for the Mabounie and Krycklan study sites, respectively.

Agroforestry Harvesting Residue: A Case Study in Private Forests in Probolinggo, East Java, Indonesia

One of the wood sources in Indonesia is derived from private forests managed with agroforestry systems. The harvesting operation in these forests will produce harvesting residues. The study aimed to quantify the harvesting residue volume and to construct a prediction model for harvesting residue volume in a private forest managed with an agroforestry system. The study was conducted in a private forest managed with an agroforestry system in Probolinggo, East Java. The method employed for quantifying harvesting residue was a whole-tree method. The harvesting residue was classified into stumps, butt ends, top logs, and branches. Harvesting residues with dbh >= 4 cm were measured after harvesting. The best model was selected based on the Root Mean Squared Error and the adjusted coefficient of determination values. The number of sample trees was 31 trees, which were chosen by farmers to be felled. The study found that the harvesting residues were 6% of the total harvested volume. The harvesting residue was predominantly the stumps. The best model for estimating harvesting residue volume in agroforestry private forests was V = 0.042VolTot1.248. The study concluded that harvesting operations in private forests managed with agroforestry systems produced a small number of harvesting residues. The harvested volume was the most significant indicator for estimating the harvesting residue volume. Keywords: agroforestry, harvesting residue, private forests, predicting models

Estimation of Forest Stock Volume Combining Airborne LiDAR Sampling Approaches with Multi-Sensor Imagery

Timely and reliable estimation of forest stock volume is essential for sustainable forest management and conservation. Light detection and ranging (LiDAR) data can provide an effective depiction of the three-dimensional structure information of forests, but its large-scale application is hampered by spatial continuity. This study aims to construct a LiDAR sampling framework, combined with multi-sensor imagery, to estimate the regional forest stock volume of natural secondary forests in Northeast China. Two sampling approaches were compared, including systematic sampling and classification-based sampling. First, the forest stock volume was mapped using a combination of field measurement data and full-coverage LiDAR data. Then, the forest stock volume obtained in the first step of estimation was used as a reference value, and optical images and topographic features were combined for secondary modeling to compare the effectiveness and accuracy of different sampling methods, including 12 systematic sampling and classification-based sampling methods. Our results show that the root mean square error (RMSE) of the 12 systematic sampling approaches ranged from 55.81 to 57.42 m3/ha, and the BIAS ranged from 21.55 to 24.89 m3/ha. The classification-based LiDAR sampling approach outperformed systematic sampling, with an RMSE of 55.56 (<55.81 m3/ha) and a BIAS of 20.68 (<21.55 m3/ha). This study compares different LiDAR sampling approaches and explores an effective LiDAR sample collection scheme for estimating forest stock, while balancing cost and accuracy. The classification-based LiDAR sampling approach described in this study is easy to apply and portable and can provide a reference for future LiDAR sample collection.

Predicting the Stand Growth and Yield of Mixed Chinese Fir Forests Based on Their Site Quality, Stand Density, and Species Composition

The Chinese fir (Cunninghamia lanceolata) is the largest tree species used for afforestation in China. The purpose of this study was to explore the effects of site quality, stand density, and tree species composition on the growth and yield of mixed Chinese fir forests and to build prediction models for their stand average DBH (diameter at breast height), average height, and volume. Using 430 plots of mixed Chinese fir forests in the Fujian Province of China, the optimal base models for predicting stand average DBH, average height, and volume were selected from the Schumacher, Korf, Logistic, Mitscherlich, and Richards equations. On this basis, the site class index (SCI), stand density index (SDI), and tree species composition coefficient (TSCC) were introduced to improve the model’s performance, and the applicability of the different models was evaluated. The optimal base models for the average DBH, average height, and stand volume of mixed Chinese fir forests all used the Richards equation. The best fitting effect was obtained when the SCI was introduced into parameter a in the average height model, while the inclusion of the TSCC did not improve the model significantly. The fitting effects of the average DBH and stand volume models were both best in the form of y=a1SCIa2[1−exp⁡(−b1SDIb2)t]c when the SCI and SDI were introduced. When the TSCC was further included, the fitting effects of the stand average DBH and volume models were significantly improved, with their R2 increased by 47.47% and 58.45%, respectively, compared to the base models. The optimal models developed in this study showed good applicability; the residuals were small and distributed uniformly. We found that the SCI had an impact on the maximum values of the stand average DBH, average height, and volume; the SDI was closely related to the growth rate of the diameter and volume, while the TSCC influenced the maximum values of the stand average DBH and volume. The model system established in this study can provide a reference for the harvest prediction and mixing ratio optimization of mixed Chinese fir forests.

Forest management plan based on TOPSIS analysis method

Climate change poses a huge threat to life, especially climate change caused by greenhouse gas emissions. Positive actions should be taken to reduce greenhouse gas emissions. But simply reducing emissions is not enough. It is necessary to increase carbon dioxide reserves through biosphere or mechanical means. Forests in the biosphere can store carbon dioxide. Moderate deforestation can increase the amount of fixed carbon dioxide and have little impact on society. For this issue, this article selects Saihanba Forest Farm as the research object. By establishing a mathematical model, the impact of different forest management schemes on the carbon dioxide absorption capacity of Saihanba Forest Farm was analyzed. This paper selected four main variables to reflect their impact: forest coverage, forest volume, forest volume at different ages, and water conservation. The entropy method and TOPSIS comprehensive evaluation method are used for the data in the variables. The results showed that near mature and mature forests contributed the most to carbon dioxide absorption, while over mature trees contributed the least to carbon dioxide absorption. Based on the results, this paper presents the optimal forest management plan, and suggests that managers should appropriately cut down over mature forests to increase the stock of near mature and mature forests.

3D visualization technology for rubber tree forests based on a terrestrial photogrammetry system

IntroductionRubber trees are an important cash crop in Hainan Province; thus, monitoring sample plots of these trees provides important data for determining growth conditions. However, existing monitoring technology and rubber forest sample plot analysis methods are relatively simple and present widespread issues, such as limited monitoring equipment, transportation difficulties, and relatively poor three-dimensional visualization effects in complex environments. These limitations have complicated the development of rubber forest sample plot monitoring.MethodThis study developed a terrestrial photogrammetry system combined with 3D point-cloud reconstruction technology based on the structure from motion with multi-view stereo method and sample plot survey data. Deviation analyses and accuracy evaluations of sample plot information were performed in the study area for trees to explore the practical significance of this method for monitoring rubber forest sample plots. Furthermore, the relationship between the height of the first branch, diameter at breast height (DBH), and rubber tree volume was explored, and a rubber tree standard volume model was established.ResultsThe Bias, relative Bias, RMSE, and RRMSE of the height of the first branch measured by this method were −0.018 m, −0.371%, 0.562 m, and 11.573%, respectively. The Bias, relative Bias, RMSE, and RRMSE of DBH were −0.484 cm, −1.943%, −2.454 cm, and 9.859%, respectively, which proved that the method had high monitoring accuracy and met the monitoring requirements of rubber forest sample plots. The fitting results of rubber tree standard volume model had an R2 value of 0.541, and the estimated values of each parameter were 1.745, 0.115, and 0.714. The standard volume model accurately estimated the volume of rubber trees and forests using the first branch height and DBH.DiscussionThis study proposed an innovative planning scheme for a terrestrial photogrammetry system for 3D visual monitoring of rubber tree forests, thus providing a novel solution to issues observed in current sample plot monitoring practices. In the future, the application of terrestrial photogrammetry systems to monitor other types of forests will be explored.

A non-destructive approach to estimate buttress volume using 3D point cloud data

Buttressed trees provide mechanical support for themselves and offer essential ecological functions, such as nutrient acquisition, while being one of the largest sources of volume or biomass estimation variation in tropical forests. In this study, we collected 51 buttressed trees from (33) Democratic Republic of Congo, (12) Indonesia, and (6) Costa Rica, including (48) point clouds, and (3) destructive measurement. Specifically, we compared the performance of the Alpha Shape Algorithm (ASA) and the Slice Triangulation (ST) method on buttress volume estimation based on 30 point clouds with two species. Six point clouds from Costa Rica were used to validate the 3D surface reconstruction method. Meanwhile, we developed three allometric models based on 36 point clouds: a diameter above the buttress-based (DAB, 39 to 203 cm) model, a diameter computed from the non-convex area (Darea130) model, and the convex hull perimeter (Dconvex130) of the breast height model. The developed models were validated with independent data, including (6) point clouds and (3) destructive measurements, to highlight the broader contextualization and application of these methods. Volume estimated by the ASA and ST showed a high agreement with the reference volume acquired using the Smalian formula (relative RMSE of 0.07 and 0.11, respectively, regardless of species effect). ASA was also robust when modeling trees with more and shallower horizontal buttresses. Darea130 was the most accurate predictor to estimate buttress volume, with a lower Akaike information criterion (−66.25) than DAB (−59.55) and Dconvex130 (30.56); however, DAB and Darea130 (relative RMSE of 0.21 and 0.23, respectively) showed similar performance when validated with independent datasets. Our results indicate that the ASA approach performs better than both the ST and allometric models used in this study. Furthermore, the ASA method can help correct the bias in the present and past estimates of volume and biomass of large trees, which are foundational components for understanding biomass allocation and dynamics in tropical forests contemporary fields.

Sentinel-1 Time Series for Predicting Growing Stock Volume of Boreal Forest: Multitemporal Analysis and Feature Selection

Copernicus Sentinel-1 images are widely used for forest mapping and predicting forest growing stock volume (GSV) due to their accessibility. However, certain important aspects related to the use of Sentinel-1 time series have not been thoroughly explored in the literature. These include the impact of image time series length on prediction accuracy, the optimal feature selection approaches, and the best prediction methods. In this study, we conduct an in-depth exploration of the potential of long time series of Sentinel-1 SAR data to predict forest GSV and evaluate the temporal dynamics of the predictions using extensive reference data. Our boreal coniferous forests study site is located near the Hyytiälä forest station in central Finland and covers an area of 2500 km2 with nearly 17,000 stands. We considered several prediction approaches and fine-tuned them to predict GSV in various evaluation scenarios. Our analyses used 96 Sentinel-1 images acquired over three years. Different approaches for aggregating SAR images and choosing feature (predictor) variables were evaluated. Our results demonstrate a considerable decrease in the root mean squared errors (RMSEs) of GSV predictions as the number of images increases. While prediction accuracy using individual Sentinel-1 images varied from 85 to 91 m3/ha RMSE, prediction accuracy with combined images decreased to 75.6 m3/ha. Feature extraction and dimension reduction techniques facilitated the achievement of near-optimal prediction accuracy using only 8–10 images. Examined methods included radiometric contrast, mutual information, improved k-Nearest Neighbors, random forests selection, Lasso, and Wrapper approaches. Lasso was the most optimal, with RMSE reaching 77.1 m3/ha. Finally, we found that using assemblages of eight consecutive images resulted in the greatest accuracy in predicting GSV when initial acquisitions started between September and January.

Estimation of Forest Stock Volume Using Sentinel-2 MSI, Landsat 8 OLI Imagery and Forest Inventory Data

Forest stock volume (FSV) is a key indicator for measuring forest quality, evaluating forest management capabilities, and the main factor for evaluating forest carbon sequestration levels. In this study, to achieve an accurate estimation of FSV, we used Ninth Beijing Forest Inventory data (FID), and Landsat 8 OLI and Sentinel-2 MSI imagery to establish FSV models. The performance of Landsat 8 and Sentinel-2 imagery data in estimating forest volume in Huairou District, Beijing, China was compared. The combination of Landsat 8 and Sentinel-2 satellite data was employed to create a new data source. Two variable selection methods, linear stepwise regression (LSR) and recursive feature elimination (RFE), were used to extract feature variables. The multiple linear regression(MLR) models, Back Propagation (BP) neural network models, and Random Forest (RF) models were employed to estimate forest volume in the study area based on the feature variables obtained from both data sources. The research results indicate (1) the Sentinel-2-based model achieved higher accuracy compared to the same model based on the Landsat 8 factor set. The correlation between the red-edge band of Sentinel-2 imagery and FSV is more significant than that of other characteristic variables used. Variables derived from the red-edge band have the potential to reduce model errors; (2) the estimation accuracy of the model can be significantly improved by using the RFE (Recursive Feature Elimination) method to select remote sensing feature variables. RFE is based on the importance ranking of all feature variables and selects the feature variables that contribute the most to the model. In the variable group selected by RFE, the texture features and the derived features from the red-edge band, such as SenB5, SenRVI, SenmNDVIre, and SenB5Mean, contribute the most to the improvement of model accuracy. Furthermore, in the optimal Landsat 8–Sentinel-2 RFE-RF model, where texture features are involved, the rRMSE is greatly reduced by 3.7% compared to the joint remote sensing RFE-RF model without texture features; (3) the MLR, BP, and RF models based on the modeling factor set established on Sentinel-2 have accuracy superior to the model accuracy established based on the modeling factor set of Landsat 8. Among them, the Random Forest (RF) method inverted by the recursive feature elimination (RFE) method using Sentinel-2A image has the best inversion accuracy effect (R2 = 0.831, RMSE = 12.604 m3 ha−1, rRMSE = 36.411%, MAE = 9.366 m3 ha−1). Comparing the performance of the models on the test set, the ranking is as follows, Random Forest (RF) model > Back Propagation (BP) neural network model > multiple linear regression (MLR) model. The feature variable screening based on the Random Forest’s recursive feature elimination (RFE) method is better than the linear stepwise regression (LSR). Therefore, the RFE-RF method based on the joint variables from Landsat 8 and Sentinel-2 satellite data to establish a new remote sensing data source provides the possibility to improve the estimation accuracy of FSV and provides reference for forest dynamic monitoring.

Assessing Forest Quality through Forest Growth Potential, an Index Based on Improved CatBoost Machine Learning

Human activities have always depended on nature, and forests are an important part of this; the determination and improvement of forest quality is therefore highly significant. Currently, domestic and foreign research on forest quality focuses on the current states of forests. We propose a new research direction based on the future states. By referencing and analyzing the forest quality standards of domestic and foreign experts and institutions, the concept and model for calculating forest growth potential were constructed. Forest growth potential is a new forest quality indicator. Based on the data of 110,000 subcompartments of forest resources from the Lin’an and Landsat8 satellites’ remote sensing data, the unit volume was predicted using three machine-learning algorithms: random gradient descent SGD, the integrated machine learning algorithm CatBoost, and deep learning CNN. The CatBoost algorithm model was improved based on Optuna; then the improved CatBoost algorithm was selected through evaluation indicators for the prediction of forest volume and finally incorporated into the calculation model for forest growth-potential value. The forest growth-potential value was calculated, and an accurate forest quality improvement scheme based on the subcompartments is preliminarily discussed. The successful calculation of forest growth potential values has a certain reference significance, providing guidance for accurately improving forest quality and forest management. The improved CatBoost calculation model is effective in the prediction of forest growth potential, and the determination coefficient R2 reaches 0.89, a value that compares favorably with those in other studies.