Aboveground Biomass Models Research Articles

Satellite-Based Carbon Estimation in Scotland: AGB and SOC

The majority of state-of-the-art research employs remote sensing on AGB (Above Ground Biomass) and SOC (Soil Organic Carbon) separately, although some studies indicate a positive correlation between the two. We intend to combine the two domains in our research to improve state-of-the-art total carbon estimation. We begin by establishing a baseline model in our study area in Scotland, using state-of-the-art methodologies in the SOC and AGB domains. The effects of feature engineering techniques such as variance inflation factor and feature selection on machine learning models are then investigated. This is extended by combining predictor variables from the two domains. Finally, we leverage the possible correlation between AGB and SOC to establish a relationship between the two and propose novel models in an attempt to outperform the state-of-the-art results. We compared three machine learning techniques, boosted regression tree, random forest, and xgboost. These techniques have been demonstrated to be the most effective in both domains. This research makes three contributions: (i) Including Digital Elevation Map (DEM) as a predictor variable in the AGB model improves the model result by 13.5 % on average across the three machine learning techniques experimented, implying that DEM should be considered for AGB estimation as well, despite the fact that it has previously been used exclusively for SOC estimation. (ii) Using SOC and SOC Density improves the prediction of the AGB model by a significant 14.2% on average compared to the state-of-the-art baseline (When comparing the R2 value across all three modeling techniques in Model B and Model H, there is an increase from 0.5016 to 0.5604 for BRT, 0.4958 to 0.5925 for RF and 0.5161 to 0.5750 for XGB), which strengthens our experiment results and suggests a future research direction of combining AGB and SOC as a joint study domain. (iii) Including AGB as a predictor variable for SOC improves model performance for Random Forest, but reduced performance for Boosted Regression tree and XG Boost, indicating that the results are specific to ML models and more research is required on the feature space and modeling techniques. Additionally, we propose a method for estimating total carbon using data from Sentinel 1, Sentinel 2, Landsat 8, Digital Elevation, and the Forest Inventory.

Modelling Above-ground Biomass Using Machine Learning Algorithm: Case Study Miombo Woodlands of Malawi

Malawi’s Miombo Woodlands provide critical wood products and ecosystem services, including carbon storage through above ground-biomass (AGB). However, accurate and efficient estimation of AGB and C remains a huge challenge due to the complexity of heterogeneous natural and anthropogenic environmental factors exacerbated by choice of data collection and analysis techniques. A study was conducted to explore the potential of Machine Learning (ML) algorithmic on modelling AGB in Malawi’s Miombo Woodlands using remotely sensed data. A combination of AGB field measurements and Sentinel 2 Multi-Spectral Instrument (S2 MSI) imagery, and environmental data were employed to train and evaluate the performance of Random Forest (RF) regression ML model, thus against the backdrop of traditional simple linear and multiple linear regression models used in the past. The randomForest package in Rstudio R-3.6.1 software was used to model the S2 median seasonal composite band imagery datasets acquired in 2019 over Ntchisi Forest. Results demonstrate an outstanding performance of the RF with Mean Decrease in Accuracy (MDA) importance of >80 acquired from the Green, Blue and Near-Infra-Red (NIR) bands of the post-rainy season imagery while the dry season scene registered a MDA importance of >52 obtained from the Red, Blue and NIR bands. These findings underscore the superiority of the RF, revealing its ability to model AGB using different spectral bands, and more importantly, in different seasons. The outcomes have also revealed the better accuracy that the RF render in generating AGB maps (providing a minimum range of 25-33 tCha-1, which is in line with the ground reference estimates). Overall, the study has shown that RF Regression is a better technique to estimate AGB in Miombo Woodlands using the S2 MSI imagery under the given conditions. In conclusion, the research has further shown that the optical band combination of NIR and the Red regions of the Electro-Magnetic Spectrum are indispensable variables for successful modelling of AGB as well as delineating forest cover from non-forest attributes in the Miombo Woodlands. However, the study recommends exploring modelling the AGB using averaged vegetation indices i.e., NDVI, SAVI, and MSAVI, for robust results.

Evaluation of Hyperspectral Monitoring Model for Aboveground Dry Biomass of Winter Wheat by Using Multiple Factors

The aboveground dry biomass (AGDB) of winter wheat can reflect the growth and development of winter wheat. The rapid monitoring of AGDB by using hyperspectral technology is of great significance for obtaining the growth and development status of winter wheat in real time and promoting yield increase. This study analyzed the changes of AGDB based on a winter wheat irrigation experiment. At the same time, the AGDB and canopy hyperspectral reflectance of winter wheat were obtained. The effect of spectral preprocessing algorithms such as reciprocal logarithm (Lg), multiple scattering correction (MSC), standardized normal variate (SNV), first derivative (FD), and second derivative (SD); sample division methods such as the concentration gradient method (CG), the Kennard–Stone method (KS), and the sample subset partition based on the joint X–Y distances method (SPXY); sample division ratios such as 1:1 (Ratio1), 3:2 (Ratio2), 2:1 (Ratio3), 5:2 (Ratio4), and 3:1 (Ratio5); dimension reduction algorithms such as uninformative variable elimination (UVE); and modeling algorithms such as partial least-squares regression (PLSR), stepwise multiple linear regression (SMLR), artificial neural network (ANN), and support vector machine (SVM) on the hyperspectral monitoring model of winter wheat AGDB was studied. The results showed that irrigation can improve the AGDB and canopy spectral reflectance of winter wheat. The spectral preprocessing algorithm can change the original spectral curve and improve the correlation between the original spectrum and the AGDB of winter wheat and screen out the bands of 1400 nm, 1479 nm, 1083 nm, 741 nm, 797 nm, and 486 nm, which have a high correlation with AGDB. The calibration sets and validation sets divided by different sample division methods and sample division ratios have different data-distribution characteristics. The UVE method can obviously eliminate some bands in the full-spectrum band. SVM is the best modeling algorithm. According to the universality of data, the better sample division method, sample division ratio, and modeling algorithm are SPXY, Ratio4, and SVM, respectively. Combined with the original spectrum and by using UVE to screen bands, a model with stable performance and high accuracy can be obtained. According to the particularity of data, the best model in this study is FD-CG-Ratio4-Full-SVM, for which the R2c, RMSEc, R2v, RMSEv, and RPD are 0.9487, 0.1663 kg·m−2, 0.7335, 0.3600 kg·m−2, and 1.9226, respectively, which can realize hyperspectral monitoring of winter wheat AGDB. This study can provide a reference for the rational irrigation of winter wheat in the field and provide a theoretical basis for monitoring the AGDB of winter wheat by using hyperspectral remote sensing technology.

Inversion models of aboveground grassland biomass in Xinjiang based on multisource data.

Grassland biomass monitoring is essential for assessing grassland health and carbon cycling. However, monitoring grassland biomass in drylands based on satellite remote sensing is challenging.Statistical regression models and machine learning have been used for the construction of grassland biomass models, but the predictive power for different grassland types is unclear. Additionally, the selection of the most appropriate variables to construct a biomass inversion model for different grassland types must be explored. Therefore,1201 ground-truthed data points collected from 2014-2021,including 15 Moderate Resolution Imaging Spectroradiometer (MODIS) vegetation indices,geographic location and topographic data,and meteorological factors and vegetation biophysical indicators were screened for key variables using principal component analysis (PCA). The accuracy of multiple linear regression models, exponential regression models, power function models, support vector machine (SVM) models, random forest (RF) models, and neural network models was evaluated for the inversion of three types of grassland biomass. The results were as follows: (1) The biomass inversion accuracy of single vegetation indices was low, and the optimal vegetation indices were the soil-adjusted vegetation index (SAVI) (R2 = 0.255), normalized difference vegetation index (NDVI) (R2 = 0.372) and optimized soil-adjusted vegetation index (OSAVI) (R2 = 0.285). (2)Grassland above-ground biomass (AGB) was affected by various factors such as geographic location,topography, and meteorological factors, and the inverse models using a single environmental variable had large errors. (3) The main variables used to model biomass in the three types of grasslands were different. SAVI, aspect, slope, and precipitation (Prec.) were selected for desert grasslands; NDVI,shortwave infrared 2 (SWI2), longitude, mean temperature, and annual precipitation were selected for steppe;and OSAVI, phytochrome ratio (PPR), longitude, precipitation, and temperature were selected for meadows. (4) The non-parametric meadow biomass model was superior to the statistical regression model. (5) The RF model was the best model for the inversion of grassland biomass in Xinjiang, and this model had the highest accuracy for grassland biomass inversion (R2 = 0.656, root mean square error (RMSE) = 815.6 kg/ha),followed by meadow (R2 = 0.610, RMSE = 547.9 kg/ha) and desert grassland (R2 = 0.441, RMSE = 353.6 kg/ha).

Mapping individual tree and plot-level biomass using airborne and mobile lidar in piñon-juniper woodlands

Piñon-juniper (PJ) woodlands are a widespread dryland ecosystem in the US containing an immense but poorly-constrained amount of aboveground biomass (AGB). Found at the dry end of the climatic range within which trees can persist, PJ faces an uncertain future in a changing climate, giving unique importance to mapping its AGB, past, present, and future. Lidar remote sensing offers great potential towards that end with research in a range of tree-dominant ecosystems demonstrating a strong capacity for mapping AGB. However, studies applying lidar to the task of mapping AGB in PJ are few. Given the unique structural characteristics of PJ trees, which tend to be short in stature (<10 m), possess multiple stems often obscured by low crown base heights, and irregularly-shaped canopies, there remains a high degree of uncertainty surrounding tree- and stand-level structural mapping in these woodlands. To resolve our limited understanding of how to best quantify AGB in PJ using lidar, we conducted a field-validated analysis comparing lidar platform (airborne laser scanner, ALS vs. mobile laser scanner, MLS) and analytical framework (area-based modeling vs. individual tree-based modeling). We used a random forest machine learning approach to predicting tree- and stand-level structure with leave-one-out cross-validation to assess model performance. In an area-based context, MLS only marginally outperformed ALS, with both able to explain approximately 80 % of variance in AGB. However, in an individual tree-based context, MLS significantly outperforms ALS at several tasks, including: (1) delineating tree crowns (MLS delineated 100 % of trees; ALS delineated 69 %); (2) directly characterizing tree structural parameters including height (R2ALS = 0.52 vs. R2MLS = 0.74) and crown area (R2ALS = 0.26 vs. R2MLS = 0.68); and (3) modeling AGB (R2ALS = 0.38 vs. R2MLS = 0.62). We attribute this difference in performance to the vastly greater structural detail contained within the MLS point cloud data as compared to ALS, with the former possessing an average of approximately 30,000 pts/m2 and the latter 14 pts/m2. Based on our results, we make the following recommendation: if individual tree structure is needed, MLS is the far superior option; however, given the wider availability, ALS data are a suitable substitute when mapping PJ AGB, particularly under an area-based modeling analytical framework. This study represents the first time MLS data have been applied to tree structure mapping in PJ, and provides strong quantitative evidence towards best structural and AGB mapping practices in this unique, widespread, and important ecosystem. It serves as a solid foundation upon which future studies aiming to map AGB in PJ and other similar dryland woodland ecosystems can be built.

Aboveground Biomass Models for Indigenous Tree Species in the Dry Afromontane Forest, Central Ethiopia

The application of biomass models for quantifying forests’ above-ground biomass is essential for sustainable forest management. However, lack of knowledge in modelig biomass of individual tree growth hinders the sustainable management of Dry Afromontane forests. In this study, models to estimate above-ground biomass were developed for Rhus ruspolii, Ekebergia capensis, and Nuxia congesta. To develop the models, a total of 45 trees from different diameter classes were selected, felled, and divided into different biomass compartments. For the model’s development, diameter at breast height (DBH), total height (TH), diameter at stump height (DSH), and wood density (WD) were used as independent variables. Models’ performances were evaluated using RSE, adjusted coefficient of determination, and AIC. Also, model validations were done by using rRMSE, mean absolute deviation, bias, and coefficient of variation. Models 5 (Adj-R2 = 0.92), 6 (Adj-R2 = 0.97), and 8 (Adj-R22 = 0.82) were the best fitted models for Nuxia congesta, Ekebergia capensis, and Rhus ruspolii, respectively. The average wood densities of Ekebergia capensis, Nuxia congesta, and Rhus ruspolii were 0.59, 0.50, and 0.69, respectively. The variation between observed biomass and estimated biomass using new models was statistically not significant (p&gt;0.05). Thus, the biomass models developed here can be important tools to accurately estimate above-ground biomass in the Menagesha Suba forest and can be integrated into decision support tools.

Volume and aboveground biomass models for a dry evergreen montane forest in Tanzania

Abstract. Mwaluseke ML, Mwakalukwa EE, Maliondo SMS. 2023. Volume and aboveground biomass models for a dry evergreen montane forest in Tanzania. Asian J For 7: 45-53. Models available for accurately estimating trees and shrubs' volume and aboveground biomass from dry evergreen montane forests in Tanzania are largely lacking. Therefore, this study was conducted to develop volume, and aboveground biomass models for a dry evergreen montane forest of Lendikinya Forest Reserve found in Northern Tanzania. A total of thirty sample trees and shrubs with a diameter range of 5-58.5 cm were destructively harvested and used in this study. Specifically, the study developed (i) the height-diameter model and (ii) the total volume and aboveground biomass models. The following height, volume, and biomass models appear to be suitable for estimating tree height, volume, and biomass of tree and shrub species found in the study site: Height (m) =2.3249 + 6.6101/DBH + 0.2847DBH (R2 = 0.78, RMSE = 1.79, AIC = 164.37), ln(Volume, m3) = -9.845 + 1.915 ln(DBH) + 1.089ln(Ht) (R2 = 0.97, RMSE = 0.296, AIC = -144.18) and ln(Biomass, kg) = -1.666 + 0.853ln(WD x DBH2 x Ht) (R2 = 0.95, RMSE = 0.324, AIC = 224.13). Both models yielded low bias, hence indicating an excellent fit. These models will be useful in understanding the condition of the forest and the potential of this forest in storing carbon hence, the possibility of benefiting from the ongoing negotiations of REDD+ schemes for payment for avoided deforestation and degradation through sustainable management of the reserve.

ABOVEGROUND BIOMASS AND CARBON STOCK ESTIMATION OF FALCATA THROUGH THE SYNERGISTIC USE OF SENTINEL-1 AND SENTINEL-2 IMAGES

Abstract. Estimates of aboveground biomass (AGB) in forests have been made in the context of climate change mitigation. There were limited studies about Falcata aboveground biomass and carbon stock estimation using the traditional method; however, that is time-intensive and expensive. Hence, this study was executed to utilize remote sensing and assess the potential of the synergistic use of Sentinel-1 and Sentinel-2 images in estimating the AGB and carbon stock of Falcata. The methodology consists of boundary demarcation of Falcata plantations in Butuan City, development of aboveground biomass models, and estimation and mapping of aboveground biomass and carbon stock. Among the developed models, the model which is the combination of Sentinel-1 and Sentinel-2 has the highest coefficient of determination (R2) of 0.632 and lowest Root Mean Square Error (RMSE) of 1.94 ton/pixel and was found to perform best in predicting the AGB and carbon stock of 4-year-old Falcata and performed poorly in 1-year-old Falcata. Nevertheless, its overall R2 and RMSE have proven that the model is moderately good and acceptable in predicting AGB of all stand ages of Falcata and, indirectly, the carbon stock. This study demonstrates that combining satellites generates a robust and more accurate AGB and Carbon Stock model than the models derived from the individual satellites.

Allometric models for aboveground biomass estimation of small trees and shrubs in African savanna ecosystems

Quantification of plant biomass and carbon in ecosystems is critical for establishing climate change mitigation potential. For large trees in various ecosystems, allometric models for estimating biomass have been developed but few biomass equations exist for small trees and shrubby vegetation. Allometric above-ground biomass (AGB) models are needed for small trees and shrubs in order to improve the quantification of biomass, particularly for savanna ecosystems, where small trees and shrubs comprise a significant portion of the biomass. In this study we have developed species-specific and multi-species allometric models for biomass estimation of small tree species and shrubs in the savanna ecosystem of Lake Mburo National Park in South Western Uganda. For our models we selected 27 small tree species (N = 403 individuals) and 12 shrub species (N = 177) common in savanna ecosystems for destructive sampling. We developed species-specific and multi-species allometric AGB models to provide estimates of AGB using specific biometric variables recorded for the small trees (i.e. species, DBH, height and crown area), and shrubs (species, height and crown area). We found that crown area was the best single predictor of species-specific AGB for small trees and for species-specific and multi-species models for shrubs. Species-specific models had the best predictive capacity of AGB compared to multi-species biomass models for small trees and shrubs. Multiple-variable models had the best predictive capacity of AGB in both species-specific and multi-species modeling compared to single-variable models. Based on these findings we conclude that the evaluation of carbon stocks of tropical savanna ecosystems should use multi-variable species-specific models for AGB estimation at the individual level, and multi-species models for AGB at the ecosystem level.

Spatiotemporal fusion of multi-source remote sensing data for estimating aboveground biomass of grassland

Accurate estimation of aboveground biomass of grasslands is key to sustainable grassland utilization. However, most satellites cannot provide high temporal and spatial resolution data. Patterns of grassland dynamics associated with variability in climate conditions across spatiotemporal scales are yet to be adequately quantified. A spatiotemporal fusion model offers the opportunity to combine the resolution advantages of different remote sensing data to achieve a high frequency and high precision monitoring of vegetation. We test a flexible spatiotemporal data fusion (FSDAF) methodology to generate synthetic normalized difference vegetation index (NDVI) data from Moderate-Resolution Imaging Spectroradiometer (MODIS) and Landsat data sets. The methodology is tested for the semi-arid grassland of the Xilin River Basin, China. Based on NDVI data fusion and field measured aboveground biomass an aboveground biomass estimation model is established for the watershed. Exploring the temporal and spatial changes of biomass and its relationship with environmental factors. The results show that: (1) The FSDAF model performs well (R2 = 0.75) and has clear textural features. (2) The established Support Vector Machine Aboveground Biomass model not only ensured the accuracy of estimation (R2 = 0.78, RMSE = 15.43 g/m2), but also generated spatiotemporal maps of biomass with higher spatial (30 m) and temporal resolution (8 days). (3) The grassland aboveground biomass in this area decreases from southeast to northwest, and the grassland biomass reaches its peak at the end of July. The average biomass of different grasslands decreases in the order of meadow grassland > typical grassland > desert grassland. (4) Aboveground biomass increased linearly with increasing water content, organic carbon and total nitrogen, and was most sensitive to soil water content. During the early growing and rapid growing period, aboveground biomass is mainly affected by both air temperature and precipitation, while the effects of temperature and human activities gradually dominate in the middle and late growing periods. This study helps to improve the spatial and temporal resolution of dynamic monitoring of grassland biomass, and provides a scientific basis for grassland protection and management in arid and semi-arid regions.

Assessment of the grassland carrying capacity for winter-spring period in Mongolia

The grassland ecosystems of Mongolia are among the most sensitive to global climate change because of the arid and semiarid climate. As a key source of primary productivity for livestock, the quantification of the carrying capacity in grassland for the winter and early spring period is crucial for sustainable livestock management and livelihoods of herders in Mongolia. In this study, we used remote sensing data and ancillary data to propose a framework to estimate the aboveground biomass(AGB) and the carrying capacity of grassland (GCC) using the Google Earth Engine (GEE) environment. We analysed the spatial and temporal changes in the GCC for the winter-spring period in Mongolia during 2000–2020, and the grassland carrying status index for winter-spring period (GCSIW) was proposed to reflect grassland utilization and livestock carrying status over the past 21 years. Our study demonstrated the effectiveness of AGB and GCC estimation using the Carnegie-Ames-Stanford Approach (CASA) model with the root-to-crown ratio method within the GEE environment. The AGB model validation showed good performance with an R2 of 0.67–0.71 and RMSE of 22.91–28.94 g/m2. Significant increases in AGB and GCC over the 21 years were found in Mongolian grasslands and most provinces. The average GCSIW increased significantly during 2000–2020 in the whole country and all provinces, indicating the increasing stocking density and the overexploited status of grassland in recent years. The multiregression analysis further showed that the dramatic increase in livestock populations contributed 87.5% and 55%-99% to the variations in the GSCIW for the grassland and seventeen provinces, respectively. These results will be useful and helpful in supporting sustainable grassland management and the sustainable livelihoods of herders in Mongolia.

Estimating vertically growing crop above-ground biomass based on UAV remote sensing

The accurate estimation of crop above-ground biomass (AGB) can assist in crop growth monitoring and grain yield prediction. Remote sensing has been widely used for AGB estimation at regional and local scales in recent years. However, optical remote sensing spectral indices (SIs) become saturated at medium-to-high crop covers. The combined use of remote sensing techniques and statistical regression models is not based on an understanding of how crop leaves and vertical organs contribute to the crop AGB. This causes difficulties in measuring the biomass stored in vertical organs (e.g., plant stem, wheat-spike, maize-tassel; abbreviated as AGBv) using optical remote sensing. This study aims to develop an unmanned aerial vehicle (UAV)-based vertically growing crop AGB (VGC-AGB) model. We defined Csm (g/m) to describe the crop stem and reproductive organs’ average dry mass content. This was done to improve the estimation of AGBv. The crop leaf area index (LAI, m2/m2), leaf dry matter content (Cm, g/m2), height (Ch, m), and density (Cd, m−2) were used in the VGC-AGB. The VGC-AGB calculated crop leaf AGB (AGBl) using LAI × Cm (g/m2) and AGBv using Cd × Ch × Csm (g/m2). The proposed VGC-AGB (AGB = LAI × Cm + Cd × Ch × Csm) was verified using field and UAV-based hyperspectral datasets of winter-wheat and summer-maize at three growth stages. Our results indicate that UAV-based VGC-AGB (R2 = 0.92–0.93, RMSE = 68.82–75.15 g/m2) is superior to the statistical regression model that is based on remote sensing SIs and CSMs (R2 = 0.77, RMSE = 134.94 g/m2). The results indicate that the UAV-based VGC-AGB supports the analysis of crop photosynthetic product transfers and high-performance UAV-based high-performance non-destructive AGB monitoring.

Modeling forest above-ground biomass using freely available satellite and multisource datasets

Achieving the 2 °C global warming target will require all stakeholders, from local to global, to accelerate carbon reduction plans. To properly estimate carbon amounts sequestered and stored by forests, these stakeholders need accurate maps of forest above-ground biomass (AGB). However, AGB maps are developed solely by researchers using methodologies that are typically difficult for non-researchers to understand. To ensure that non-researchers can prepare such maps, freely available open-source datasets are needed along with a simple AGB estimation method. Remote sensing data have become widely used, and some satellite images are provided as open-source data. However, few AGB estimation studies have relied exclusively on freely available datasets. This study, therefore, examined the capability of using open-source satellite and multisource datasets to estimate AGB for two contrasting test sites in Hokkaido, Japan, and for the two sites combined (the unified sites). Test site 1 contained three forest types on a hilly terrain, while test site 2 contained one dominant forest type on a flat terrain. The datasets used were L-band backscatter, a canopy height model, slope, aspect, annual mean temperature, annual precipitation, and forest type. Three machine-learning algorithms and one parametric algorithm were compared. The results showed that the machine-learning models outperformed the parametric model in all instances, reproducing low and high AGB values properly with reduced saturation issues. A variable importance analysis revealed that in the best-performing machine-learning models, forest type and annual precipitation were the two most important variables for test site 1 and the unified sites, and another climatic factor, annual mean temperature, rose in rank for the unified sites. These variables had relatively large intra-variability and inter-variability in common, both of which appear to have offered advantages in modeling AGB for the sites where AGB variability was large. In contrast, the synthetic aperture radar (SAR)-based variables (the backscatter and the canopy height model) ranked lower in relative importance. This may reflect the negative influence of hilly terrain of test site 1 (and the unified sites) on the precision of SAR-based datasets. For test site 2, with its flat terrain, these SAR-based variables were the two dominant variables. Our results show the promise of freely available satellite and multisource datasets for modeling AGB and the important roles that non-SAR-based variables may play in large-scale AGB mapping that includes varied terrain.

A new strategy for improving the accuracy of forest aboveground biomass estimates in an alpine region based on multi-source remote sensing

ABSTRACT Spatially explicit information on the distribution of dominant tree species groups and aboveground biomass (AGB) in forested areas is essential for developing targeted forest management and biodiversity conservation measures, as well as assessing forest carbon sequestration capacity. There is a shortage of continuously updated 30-m spatial resolution products for mapping dominant tree species groups. The vast majority of remote sensing-based AGB estimation approaches have relatively low accuracy for dominant tree species groups or forest types and are unsuitable for AGB modeling. Therefore, this study aims to develop an integrated framework that considers the phenological characteristics of different tree species to improve the mapping accuracies of forest dominant tree groups and corresponding AGB estimates. Thirty-meter resolution maps of dominant tree species groups were created using machine learning algorithms and phenological parameters. Features extracted from optical and radar images and phenological characteristics were used to construct AGB estimation models in a temporally consistent manner to improve the AGB estimation accuracy and perform dynamic AGB monitoring. The proposed method accurately characterized the dynamic distribution of the dominant tree species groups in the study area. The traditional AGB model that does not consider different forest types or species had an R2 value of 0.52, whereas the proposed model that considers phenology and forest types had an R2 value of 0.67. This result indicates that incorporating information on phenology and dominant species improves the accuracy of AGB estimations. The AGB in most regions was 30–55 t/ha, showing that the majority of the forests were young or middle-aged stands, and the areal percentage of AGB greater than 30 t/ha increased during the study period, suggesting an improvement in forest quality. Furthermore, the oak AGB was the highest, indicating that oak afforestation should be encouraged to enhance the carbon sequestration capacity of future forest ecosystems. The results provide new insights for researchers and managers to understand the trends of forest development and forest health, as well as technical information and a database for formulating more rational forest management strategies.

Development of tools to estimate the contribution of young sweet chestnut plantations to climate-change mitigation

Sweet chestnut plantations, irrespective of their main productive orientation (nut or timber production), are key elements of the landscape as well as the cultural heritage of the areas where they are found and provide important functions and services. Hence, recent initiatives have been aimed at extending the area occupied by chestnut trees through forest plantations. Nevertheless, the role of these young chestnut plantations as carbon sinks has been often ignored. The National Inventory of Greenhouse Gas Emissions (GHG) must include estimates of the so-called ‘transition forests’ during the 20 years following their plantation.In this study, new tools for estimating the total amount of above and belowground biomass stored in young plantations of chestnut were developed to quantify the carbon storage capacity of these plantations. A new set of aboveground biomass and root-shoot ratio models were fitted for individual-tree level based on four different independent variables – root collar diameter, total height, diameter at breast height and crown projection area – and their combinations. The expansion to stand level was based on age, plantation density, productive orientation of the plantation (nut or timber), site index and climate covariates as possible independent variables. At tree level, the best aboveground biomass models were those that include the product of root-collar or breast height diameters and tree height, whereas for root-shoot ratio the best results were obtained when only diameter at breast height is included. At stand level, the most accurate models included age, plantation density and site index for aboveground biomass and only age for root-shoot ratio. The fitted models provided accurate and unbiased predictions of aboveground biomass in the first years of reforestations.The different fitted equations can be used to estimate carbon stocks in young plantations depending on the available data and the objective of the prediction. Individual tree-level equations are recommended when accurate estimates are needed and detailed inventory measurements are available. Stand level equations, only using plantation age, can be an appropriate alternative for use with forest statistics at national scale, although the inclusion of additional covariates can greatly improve the accuracy of the age-based stand level equations. Our results indicated that even low-density nut-oriented chestnut plantations can play a relevant role as C sinks.

Assessing and mitigating systematic errors in forest attribute maps utilizing harvester and airborne laser scanning data

Cut-to-length harvesters collect useful information for modeling relationships between forest attributes and airborne laser scanning (ALS) data. However, harvesters operate in mature forests, which may introduce selection biases that can result in systematic errors in harvester data-based forest attribute maps. We fitted regression models (harvester models) for volume (V), height (HL), stem frequency (N), above-ground biomass, basal area, and quadratic mean diameter (QMD) using harvester and ALS data. Performances of the harvester models were evaluated using national forest inventory plots in an 8.7 Mha study area. We estimated biases of large-area synthetic estimators and compared efficiencies of model-assisted (MA) estimators with field data-based direct estimators. The harvester models performed better in productive than unproductive forests, but systematic errors occurred in both. The use of MA estimators resulted in efficiency gains that were largest for HL (relative efficiency, RE = 6.0) and smallest for QMD (RE = 1.5). The bias of the synthetic estimator was largest for N (39%) and smallest for V (1%). The latter was due to an overestimation of deciduous and an underestimation of spruce forests that by chance balanced. We conclude that a probability sample of reference observations may be required to ensure the unbiasedness of estimators utilizing harvester data.

Machine learning prediction of above-ground biomass in pure Calabrian pine (Pinus brutia Ten.) stands of the Mediterranean region, Türkiye

Above-ground biomass (AGB) is an important component for identifying carbon stocks, monitoring the impacts of climate change, and evaluating merchantable timber. Accurate prediction of forest AGB is central to the correct interpretation of these components and to produce usable data for planners and researchers. In this study, remotely sensed time-series data derived from Landsat 8 (reflectance (R) and vegetation indices (VI)), topographic (T) and climate (C) data were used as independent variables to predict AGB of pure Calabrian pine (Pinus brutia Ten.) stands using multiple regression analysis (MLR) and support vector machines (SVM) methods. The AGB modeling was done by using independent variables individually and by combining variables, and the AGB maps of the most successful models obtained from MLR and SVM methods were produced. It was determined that the most successful variable group was the VI when the independent variables were used one by one (MLR Training R2 = 0.50, SVM Training R2 = 0.67). The most successful predictions in AGB modeling were obtained with combining all independent variables and using the SVM method (Training R2 = 0.85, Validation R2 = 0.69). In the combination of independent variables, VI and C data made the greatest contribution to the success of the AGB prediction. The ‘green leaf index’ vegetation indices had the most significant effect on the modeling AGB. In this study, T and C in addition to spectral data has increased the AGB estimation performance. It has been found that the SVM method yielded higher model accuracy than MLR method in predicting AGB. Overall, the spectral data and the SVM method can contribute to improving the accuracy of AGB estimates and provide an effective approach towards the capability for forest ecosystem monitoring.

Improved forest biomass estimation based on P-band repeat-pass PolInSAR data across different forest sites

• Only canopy volume backscatter is almost free from the topographic influence. • One exponential equation can link volume backscatter with AGB across different sites. • Multi-baseline PolInSAR can retrieve forest height without overall bias across sites. • PolInSAR forest height agreed more with the top forest height for open forests. • The proposed AGB model alleviated the influence of forest heterogeneity. The upcoming BIOMASS mission will provide P-band repeat-pass PolInSAR data from space for the improved mapping of global biomass. PolInSAR technique has been widely validated with the potential to invert forest height and estimate forest aboveground biomass (AGB). However, the robustness of PolInSAR-based AGB estimation across different sites still lacks full evaluation, especially for those with a varied forest type, heterogeneity (varied growth ratio between cover and height), and topographic relief. In this study, we concentrated on backscatter decomposition and forest height inversion, and developed a robust AGB estimation method that can be applied to different sites. Two dense and closed tropical forest sites (Paracou and Nouragues) and one open and heterogeneous boreal forest site (Krycklan) were selected as the study areas, and the corresponding airborne PolInSAR, LiDAR, and ground measured AGB data were used for validation and analysis. Results show that ground backscatter has the strongest correlation with AGB in boreal forests, but this correlation cannot be transferred to the tropical forests. Only canopy volume backscatter is almost free from topographic influence, and its relationship with AGB across three sites can be formulated using one exponential equation, producing the best estimation accuracy, with R 2 of 0.79 and RMSE of 61.5 tons/ha (relative RMSE of 20.0 %). Multi-baseline PolInSAR retrieved forest height with little bias in spite of the presence of temporal decorrelation. One power equation can be used to correlate PolInSAR forest height with AGB across three different sites, and LOO (leave-one-out) validation shows the R 2 of 0.85 and RMSE of 51.8 tons/ha (relative RMSE of 16.9 %). However, the RVoG-inverted PolInSAR FH was found to mainly represent the top forest height for open and heterogeneous forests, which means PolInSAR FH (forest height) lacks consideration for forest horizontal structure (e.g. forest density). In contrast, volume backscatter better captured forest density, and the proposed AGB model that combines PolInSAR FH and volume backscatter further improved the AGB estimation accuracy, especially for open forests: the plot-scale validation from all three sites shows R 2 was improved from 0.79 (volume backscatter) and 0.85 (PolInSAR FH) to 0.89, and RMSE decreased from 61.5 and 51.8 to 45.2 (relative RMSE of 14.7 %) tons/ha; for region-scale validation, R 2 was improved from 0.77 and 0.83 to 0.89, and RMSE decreased from 64.2 (relative RMSE of 39.0 %) and 54.5 (34.5 %) to 48.1 (29.4 %) tons/ha.

Distribution and Carbon Sequestration Potential of Cola Laurifolia Mast.: A Dominant Native Riparian Species Along Permanent Rivers in Sub-Saharan Africa

Species-specific models for estimating aboveground biomass (AGB) are the accurate means of quantifying species’ carbon pools. Cola laurifolia Mast., a dominant and multi-purpose riparian species along the Mouhoun River in Burkina Faso have a regressive population. Few scientific studies exist concerning this riparian species population and carbon stock capacity. This study aims to allow this gap by formulating a species-specific allometric model for assessing with direct method for Cola laurifolia leave, branches, stem and whole AGB. Parameters used to perform models are tree diameter at breast height (DBH), basal diameter at 20 cm (D20), height (H), and mean crown diameter (CD) using data from 30 trees. Population structure shows a low regeneration potential at all of the studied river zones (i.e. upstream, intermediate and downstream zones). The carbon stock was found to be 54.14 kg C tree-1 and 9.24 Mg C. ha-1. The density of C. laurifolia was higher in downstream zone, and consequently the carbon stock was higher in these areas. The log-log linear model is the best-fitted form incorporated DBH and H as predictors. This form is best fitted for the three tree components (i.e. leaves, branches, stem) and the AGB. The AGB model is more accurate with high coefficient of determination and low RSE (R²=0.92; RSE=0.28) contrasted with leaves models. The global model has the best goodness of fit because of a low relative error (-0.213 %) compared to the use of three component models. The accuracy of our species-specific model confirms the need to develop such models for greater accuracy in AGB estimations.

Predictive Modeling of Above-Ground Biomass in Brachiaria Pastures from Satellite and UAV Imagery Using Machine Learning Approaches

Grassland pastures are crucial for the global food supply through their milk and meat production; hence, forage species monitoring is essential for cattle feed. Therefore, knowledge of pasture above-ground canopy features help understand the crop status. This paper finds how to construct machine learning models to predict above-ground canopy features in Brachiaria pasture from ground truth data (GTD) and remote sensing at larger (satellite data on the cloud) and smaller (unmanned aerial vehicles (UAV)) scales. First, we used above-ground biomass (AGB) data obtained from Brachiaria to evaluate the relationship between vegetation indices (VIs) with the dry matter (DM). Next, the performance of machine learning algorithms was used for predicting AGB based on VIs obtained from ground truth and satellite and UAV imagery. When comparing more than twenty-five machine learning models using an Auto Machine Learning Python API, the results show that the best algorithms were the Huber with R2 = 0.60, Linear with R2 = 0.54, and Extra Trees with R2 = 0.45 to large scales using satellite. On the other hand, short-scale best regressions are K Neighbors with an R2 of 0.76, Extra Trees with an R2 of 0.75, and Bayesian Ridge with an R2 of 0.70, demonstrating a high potential to predict AGB and DM. This study is the first prediction model approach that assesses the rotational grazing system and pasture above-ground canopy features to predict the quality and quantity of cattle feed to support pasture management in Colombia.