Improved Estimation of Aboveground Biomass in Rubber Plantations Using Deep Learning on UAV Multispectral Imagery

Drones, Journal Year: 2025, Volume and Issue: 9(1), P. 32 - 32

Published: Jan. 6, 2025

The accurate estimation of aboveground biomass (AGB) in rubber plantations is essential for predicting production and assessing carbon storage. Multispectral sensors mounted on unmanned aerial vehicles (UAVs) can obtain high spatiotemporal resolution imagery plantations, offering significant advantages capturing fine structural details heterogeneity. However, most previous studies primarily focused developing models using machine learning (ML) algorithms conjunction with feature selection methods based UAV-acquired multispectral imagery. reliance limits the model’s generalizability, robustness, predictive accuracy. In contrast, deep (DL) exhibits considerable promise extracting features from high-resolution UAV-based without need manual selection. Nonetheless, it remains unclear whether DL surpass traditional ML improving AGB accuracy plantations. To address this, our study evaluated performance three (random forest regression, RFR; XGBoost XGBR; categorical boosting CatBoost) combined techniques a convolutional neural network (DCNN) obtained UAV results indicate that RFR principal component analysis (PCA) yielded best (R2 = 0.81, RMSE 11.63 t/ha, MAE 9.27 t/ha) between algorithms. Meanwhile, DCNN model derived G, R, NIR spectral bands achieved highest 0.89, 6.44 5.72 t/ha), where outperformed other methods. Our highlights great potential combining to improve new perspective estimating physiological biochemical growth parameters forests.

Language: Английский

---

From Air to Space: A Comprehensive Approach to Optimizing Aboveground Biomass Estimation on UAV-Based Datasets

Forests, Journal Year: 2025, Volume and Issue: 16(2), P. 214 - 214

Published: Jan. 23, 2025

Estimating aboveground biomass (AGB) is vital for sustainable forest management and helps to understand the contributions of forests carbon storage emission goals. In this study, effectiveness plot-level AGB estimation using height crown diameter derived from UAV-LiDAR, calibration GEDI-L4A GEDI-L2A rh98 heights, spectral variables UAV-multispectral RGB data were assessed. These calibrated values UAV-derived used fit estimations a random (RF) regression model in Fuling District, China. Using Pearson correlation analysis, we identified 10 most important predictor prediction model, including GEDI height, Visible Atmospherically Resistant Index green (VARIg), Red Blue Ratio (RBRI), Difference Vegetation (DVI), canopy cover (CC), (ARVI), Red-Edge Normalized (NDVIre), Color (CIVI), elevation, slope. The results showed that, general, second based on Sentinel-2 indices, slope datasets with evaluation metric (for training: R2 = 0.941 Mg/ha, RMSE 13.514 MAE 8.136 Mg/ha) performed better than first prediction. result was between 23.45 Mg/ha 301.81 standard error 0.14 10.18 Mg/ha. This hybrid approach significantly improves accuracy addresses uncertainties modeling. findings provide robust framework enhancing stock assessment contribute global-scale monitoring, advancing methodologies ecological research.

Language: Английский

---

Integration of UAS and Backpack-LiDAR to Estimate Aboveground Biomass of Picea crassifolia Forest in Eastern Qinghai, China

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(4), P. 681 - 681

Published: Feb. 17, 2025

Precise aboveground biomass (AGB) estimation of forests is crucial for sustainable carbon management and ecological monitoring. Traditional methods, such as destructive sampling, field measurements Diameter at Breast Height with height (DBH H), optical remote sensing imagery, often fall short in capturing detailed spatial heterogeneity AGB are labor-intensive. Recent advancements technologies, predominantly Light Detection Ranging (LiDAR), offer potential improvements accurate Nonetheless, there limited research on the combined use UAS (Uncrewed Aerial System) Backpack-LiDAR technologies forest biomass. Thus, our study aimed to estimate plot level Picea crassifolia eastern Qinghai, China, by integrating UAS-LiDAR data. The Comparative Shortest Path (CSP) algorithm was employed segment point clouds from Backpack-LiDAR, detect seed points calculate DBH individual trees. After that, using these initial files, we segmented trees data employing Point Cloud Segmentation (PCS) method measured tree heights, which enabled calculation observed/measured across three specific areas. Furthermore, advanced regression models, Random Forest (RF), Multiple Linear Regression (MLR), Support Vector (SVR), used integrated both sources (UAS Backpack-LiDAR). Our results show that: (1) extracted compared shows about (R2 = 0.88, RMSE 0.04 m) whereas achieved accuracy 0.91, 1.68 m), verifies reliability abstracted obtained LiDAR (2) Individual Tree (ITS) a file X Y coordinates Backpack UAS-LiDAR, attaining total F-score 0.96. (3) Using allometric equation, ranges 9.95–409 (Mg/ha). (4) RF model demonstrated superior coefficient determination (R2) 89%, relative Root Mean Square Error (rRMSE) 29.34%, (RMSE) 33.92 Mg/ha MLR SVR models prediction. (5) combination enhanced ITS forests. This work highlights advance monitoring, can be very important climate change mitigation environmental monitoring practices.

Language: Английский

---

Remote sensing and integration of machine learning algorithms for above-ground biomass estimation in Larix principis-rupprechtii Mayr plantations: a case study using Sentinel-2 and Landsat-9 data in northern China

Frontiers in Environmental Science, Journal Year: 2025, Volume and Issue: 13

Published: April 2, 2025

Estimating above-ground biomass (AGB) is important for ecological assessment, carbon stock evaluation, and forest management. This research assesses the performance of machine learning algorithms XGBoost, SVM, RF using data from Sentinel-2 Landsat-9 satellites. The study influence significant spectral bands vegetation indices on accuracy AGB estimate. results presented in paper indicate that were more effective than data. mainly because it had higher spatial resolution, which enabled model gradients structural attributes accurately. XGBoost performed best with an R 2 0.82 RMSE 0.73 Mg/ha 0.80 0.71 Landsat-9. In current study, SVM also showed a substantial 0.79 0.76 For Sentinel-2, random achieved 0.74 0.93 Mg/ha, Landsat 9 yielded 0.72 0.88 Mg/ha. Thus, variable importance analysis, have predicting AGB. As expected their application research, these predictors consistently emerged as highly across models datasets. demonstrates potential integrating remote sensing to achieve accurate efficient assessment.

Language: Английский

---

Digital twin comprehensive models: a study of ancient tree ecological environment quality assessment based on a cyber-physical system

Environmental Monitoring and Assessment, Journal Year: 2025, Volume and Issue: 197(4)

Published: April 1, 2025

Language: Английский

---

Estimating Spatiotemporal Dynamics of Carbon Storage in Roinia pseudoacacia Plantations in the Caijiachuan Watershed Using Sample Plots and Uncrewed Aerial Vehicle-Borne Laser Scanning Data

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(8), P. 1365 - 1365

Published: April 11, 2025

Forest ecosystems play a pivotal role in the global carbon cycle and climate change mitigation. aboveground biomass (AGB), critical indicator of storage sequestration capacity, has garnered significant attention ecological research. Recently, uncrewed aerial vehicle-borne laser scanning (ULS) technology emerged as promising tool for rapidly acquiring three-dimensional spatial information on AGB vegetation storage. This study evaluates applicability accuracy UAV-LiDAR estimating spatiotemporal dynamics Robinia pseudoacacia (R. pseudoacacia) plantations gully regions Loess Plateau, China. At sample plot scale, optimal parameters individual tree segmentation (ITS) based canopy height model (CHM) were determined, was validated. The results showed root mean square error (RMSE) values 13.17 trees (25.16%) count, 0.40 m (3.57%) average (AH), 320.88 kg (16.94%) AGB. regression model, which links with AH generated estimates that closely matched observed values. watershed ULS data used to estimate R. Caijiachuan watershed. analysis revealed total 68,992 trees, 2890.34 Mg density 62.46 ha−1. Low-density forest areas (<1500 ha−1) dominated landscape, accounting 94.38% 82.62% area, 92.46% Analysis tree-ring variation onset growth decline across different classes aged 0–30 years, higher-density stands exhibiting delayed compared lower-density stands. Compared traditional methods diameter at breast (DBH), assessments demonstrated superior scientific validity. underscores feasibility potential estimation complex terrain, such Plateau. It highlights importance topographic factors enhance accuracy. findings provide valuable support management high-quality development present an efficient approach precise sink accounting.

Language: Английский

---

Forest aboveground biomass estimation using deep learning data fusion of ALS, multispectral, and topographic data

International Journal of Remote Sensing, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 39

Published: April 22, 2025

Language: Английский

---

Developing mixed-effects aboveground biomass model using biotic and abiotic variables for moso bamboo in China

Journal of Environmental Management, Journal Year: 2025, Volume and Issue: 384, P. 125544 - 125544

Published: April 28, 2025

Language: Английский

---

Energy Transformation in the Construction Industry: Integrating Renewable Energy Sources

Energies, Journal Year: 2025, Volume and Issue: 18(9), P. 2363 - 2363

Published: May 6, 2025

The development of the building sector to use renewable energy, more so in photovoltaic (PV) systems, is a great step toward enhanced environmental sustainability and improved energy efficiency. This study seeks determine economic, environmental, operational effects integrating PV system into Polish production plant for buildings. Case methodology was followed with help actual operating histories simulation modeling present estimates carbon emission savings, cost power Key findings illustrate that 31.8% business’s full-year supply electricity through utilization solar it saves as much 10,366 kg CO2 emissions every year. economic rationale provided form 3.6-year payback period against long-term savings over EUR 128,000 26 years. work also addresses broader implications storage management systems on basis scalability reproducibility intervention at construction scale. provides evidence towards requirement informing decision-making by business managers policy decisionmakers solution issues interest industrial levels world agenda harmonization practice.

Language: Английский

---

Modeling forest structural variables of Eucalyptus dunnii Maiden stands under short-rotation management using SAR, multispectral, soil-derived, and field-based data

Forest Ecology and Management, Journal Year: 2025, Volume and Issue: 588, P. 122759 - 122759

Published: May 9, 2025

Language: Английский

---