Estimation of forest aboveground biomass and uncertainties by integration of field measurements, airborne LiDAR, and SAR and optical satellite data in Mexico

A comprehensive dataset of above-ground forest biomass from field observations, machine learning and topographically augmented allometric models over the Kashmir Himalaya

Accurate estimates of forest dynamics and above-ground forest biomass for the topographically challenging Himalaya are crucial for understanding carbon storage potential, assessing ecosystem services, and guiding conservation efforts in response to climate change. This dataset provides a manually delineated multi-temporal forest inventory and a comprehensive record of above-ground biomass (AGB) across the Kashmir Himalaya, generated from field observations, advanced remote sensing and machine learning. Data were collected and generated through remote sensing techniques and extensive in-situ measurements of 6220 trees (n=275 plots), including tree diameter at breast height, species composition, and tree density to map forest area and model AGB across varied terrain. The dataset captures major forest types and species-specific AGB variation influenced by elevation, slope, and aspect. Additionally, newly developed species-specific allometric models, improved through the integration of normalized difference vegetation index (NDVI) and topographical augmentation are provided to improve AGB estimation accuracy. This dataset serves as a crucial resource for forest management, carbon monitoring, and ecological modeling, with broad applications in regional conservation strategies, biodiversity planning, and climate policy development in mountainous ecosystems.

Improved aboveground biomass estimation and regional assessment with aerial lidar in California's subalpine forests

BACKGROUND

Understanding the impacts of climate change on forest aboveground biomass is a high priority for land managers. High elevation subalpine forests provide many important ecosystem services, including carbon sequestration, and are vulnerable to climate change, which has altered forest structure and disturbance regimes. Although large, regional studies have advanced aboveground biomass mapping with satellite data, typically using a general approach broadly calibrated or trained with available field data, it is unclear how well these models work in less prevalent and highly heterogeneous forest types such as the subalpine. Monitoring biomass using methods that model uncertainty at multiple scales is critical to ensure that local relationships between biomass and input variables are retained. Forest structure metrics from lidar are particularly valuable alongside field data for mapping aboveground biomass, due to their high correlation with biomass.

RESULTS

We estimated aboveground woody biomass of live and dead trees and uncertainty at 30 m resolution in subalpine forests of the Sierra Nevada, California, from aerial lidar data in combination with a collection of field inventory data, using a Bayesian geostatistical model. The ten-fold cross-validation resulted in excellent model calibration of our subalpine-specific model (94.7% of measured plot biomass within the predicted 95% credible interval). When evaluated against two commonly referenced regional estimates based on Landsat optical imagery, root mean square error, relative standard error, and bias of our estimations were substantially lower, demonstrating the benefits of local modeling for subalpine forests. We mapped AGB over four management units in the Sierra Nevada and found variable biomass density ranging from 92.4 to 199.2 Mg/ha across these management units, highlighting the importance of high quality, local field and remote sensing data.

CONCLUSIONS

By applying a relatively new Bayesian geostatistical modeling method to a novel forest type, our study produced the most accurate and precise aboveground biomass estimates to date for Sierra Nevada subalpine forests at 30 m pixel and management unit scales. Our estimates of total aboveground biomass within the management units had low uncertainty and can be used effectively in carbon accounting and carbon trading markets.

Biome evolution in subfamily Cercidoideae (Leguminosae): a tropical arborescent clade with a relictual depauperate temperate lineage

Some plant lineages remain within the same biome over time (biome conservatism), whereas others seem to adapt more easily to new biomes. The c. 398 species (14 genera) of subfamily Cercidoideae (Leguminosae or Fabaceae) are found in many biomes around the world, particularly in the tropical regions of South America, Asia and Africa, and display a variety of growth forms (small trees, shrubs, lianas and herbaceous perennials). Species distribution maps derived from cleaned occurrence records were compiled and compared with existing biome maps and with the literature to assign species to biomes. Rainforest (144 species), succulent (44 species), savanna (36 species), and temperate (10 species) biomes were found to be important in describing the global distribution of Cercidoideae, with many species occurring in more than one biome. Two phylogenetically isolated species-poor temperate (Cercis) and succulent (Adenolobus) biome lineages are sister to two broadly distributed species-rich tropical clades. Ancestral state reconstructions on a time-calibrated phylogeny suggest biome shifts occurred throughout the evolutionary history of the subfamily, with shifts between the succulent and rainforest biomes, from the rainforest to savanna, from the succulent to savanna biome, and one early occurring shift into (or from) the temperate biome. Of the 26 inferred shifts in biome, three are closely associated with a shift from the ancestral tree/shrub growth form to a liana or herbaceous perennial habit. Only three of the 13 inferred transcontinental dispersal events are associated with biome shifts. Overall, we find that biome shifts tend to occur within the same continent and that dispersals to new continents tend to occur within the same biome, but that nonetheless the biome-conserved and biogeographically structured Cercidoideae have been able to adapt to different environments through time.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40415-024-01058-z.

Total and component forest aboveground biomass inversion via LiDAR-derived features and machine learning algorithms

Forest aboveground biomass (AGB) and its biomass components are key indicators for assessing forest ecosystem health, productivity, and carbon stocks. Light Detection and Ranging (LiDAR) technology has great advantages in acquiring the vertical structure of forests and the spatial distribution characteristics of vegetation. In this study, the 56 features extracted from airborne LiDAR point cloud data were used to estimate forest total and component AGB. Variable importance-in-projection values calculated through a partial least squares regression algorithm were utilized for LiDAR-derived feature ranking and optimization. Both leave-one-out cross-validation (LOOCV) and cross-validation methods were applied for validation of the estimated results. The results showed that four cumulative height percentiles (AIH 30, AIH 40, AIH 20, and AIH 25), two height percentiles (H 8 and H 6), and four height-related variables (H mean, H sqrt, H mad, and H curt) are ranked more frequently in the top 10 sensitive features for total and component forest AGB retrievals. Best performance was acquired by random forest (RF) algorithm, with R 2 = 0.75, root mean square error (RMSE) = 22.93 Mg/ha, relative RMSE (rRMSE) = 25.30%, and mean absolute error (MAE) = 19.26 Mg/ha validated by the LOOCV method. For cross-validation results, R 2 is 0.67, RMSE is 24.56 Mg/ha, and rRMSE is 25.67%. The performance of support vector regression (SVR) for total AGB estimation is R 2 = 0.66, RMSE = 26.75 Mg/ha, rRMSE = 28.62%, and MAE = 22.00 Mg/ha using LOOCV validation and R 2 = 0.56, RMSE = 30.88 Mg/ha, and rRMSE = 31.41% by cross-validation. For the component AGB estimation, the accuracy from both RF and SVR algorithms was arranged as stem > bark > branch > leaf. The results confirmed the sensitivity of LiDAR-derived features to forest total and component AGBs. They also demonstrated the worse performance of these features for retrieval of leaf component AGB. RF outperformed SVR for both total and component AGB estimation, the validation difference from LOOCV and cross-validation is less than 5% for both total and component AGB estimated results.

Robust spatial estimates of biomass carbon on farms

The drive for farm businesses to move towards net zero greenhouse gas emissions means that there is a need to develop robust methods to quantify the amount of biomass carbon (C) on farms. Direct measurements can be destructive and time-consuming and some prediction methods provide no assessment of uncertainty. This study describes the development, validation, and use of an integrated spatial approach, including the use of lidar data, and Bayesian Belief Networks (BBNs) to quantify total biomass carbon stocks (C_total) of i) land cover and ii) landscape features such as hedges and lone trees for five case study sites in lowland England. The results demonstrated that it was possible to develop and use a remote integrated approach to estimate biomass carbon at a farm scale. The highest achievable prediction accuracy was attained from models using the variables AGBC, BGBC, DOMC, age, height, species and land cover, derived from measured information and from literature review. The two BBN models successfully predicted the test values of the total biomass carbon with propagated error rates of 6.7 % and 4.3 % for the land cover and landscape features respectively. These error rates were lower than in other studies indicating that the seven predictors are strong determinants of biomass carbon. The lidar data also enabled the spatial presentation and calculation of the variable C stocks along the length of hedges and within woodlands.

Spatial Scale Effect and Correction of Forest Aboveground Biomass Estimation Using Remote Sensing

Forest biomass is critically important for forest dynamics in the carbon cycle. However, large-scale AGB mapping applications from remote sensing data still carry large uncertainty. In this study, an AGB estimation model was first established with three different remote sensing datasets of GF-2, Sentinel-2 and Landsat-8. Next, the optimal scale estimation result was considered as a reference AGB to obtain the relative true AGB distribution at different scales based on the law of conservation of mass, and the error of the scale effect of AGB estimation at various spatial resolutions was analyzed. Then, the information entropy of land use type was calculated to identify the heterogeneity of pixels. Finally, a scale conversion method for the entropy-weighted index was developed to correct the scale error of the estimated AGB results from coarse-resolution remote sensing images. The results showed that the random forest model had better prediction accuracy for GF-2 (4 m), Sentinel-2 (10 m) and Landsat-8 (30 m) AGB mapping. The determination coefficient between predicted and measured AGB was 0.5711, 0.4819 and 0.4321, respectively. Compared to uncorrected AGB, R2 between scale-corrected results and relative true AGB increased from 0.6226 to 0.6725 for Sentinel-2, and increased from 0.5910 to 0.6704 for Landsat-8. The scale error was effectively corrected. This study can provide a reference for forest AGB estimation and scale error reduction for AGB production upscaling with consideration of the spatial heterogeneity of the forest surface.

Improved Object-Based Estimation of Forest Aboveground Biomass by Integrating LiDAR Data from GEDI and ICESat-2 with Multi-Sensor Images in a Heterogeneous Mountainous Region

Accurate and effective mapping of forest aboveground biomass (AGB) in heterogeneous mountainous regions is a huge challenge but an urgent demand for resource managements and carbon storage monitoring. Conventional studies have related the plot-measured or LiDAR-based biomass to remote sensing data using pixel-based approaches. The object-based relationship between AGB and multi-source data from LiDAR, multi-frequency radar, and optical sensors were insufficiently studied. It deserves the further exploration that maps forest AGB using the object-based approach and combines LiDAR data with multi-sensor images, which has the smaller uncertainty of positional discrepancy and local heterogeneity, in heterogeneous mountainous regions. To address the improvement of mapping accuracy, satellite LiDAR data from GEDI and ICEsat-2, and images of ALOS-2 yearly mosaic L band SAR (Synthetic Aperture Radar), Sentinel-1 C band SAR, Sentinel-2 MSI, and ALOS-1 DSM were combined for pixel- and object-based forest AGB mapping in a vital heterogeneous mountainous forest. For the object-based approach, optimized objects during a multiresolution segmentation were acquired by the ESP (Estimation of the Scale Parameter) tool, and suitable predictors were selected using an algorithm named VSURF (Variable Selection Using Random Forests). The LiDAR variables at the footprint-level were extracted to connect field plots to the multi-sensor objects as a linear bridge. It was shown that forests’ AGB values varied by elevations with a mean value of 142.58 Mg/ha, ranging from 12.61 to 514.28 Mg/ha. The north slope with the lowest elevation (<1100 m) had the largest mean AGB, while the smallest mean AGB was located in the south slope with the altitude above 2000 m. Using independent validation samples, it was indicated by the accuracy comparison that the object-based approach performed better on the precision with relative improvement based on root-mean-square errors (RIRMSE) of 4.46%. The object-based approach also selected more optimized predictors and markedly decreased the prediction time than the pixel-based analysis. Canopy cover and height explained forest AGB with their effects on biomass varying according to the elevation. The elevation from DSM and variables involved in red-edge bands from MSI were the most contributive predictors in heterogeneous temperate forests. This study is a pioneering exploration of object-based AGB mapping by combining satellite data from LiDAR, MSI, and SAR, which offers an improved methodology for regional carbon mapping in the heterogeneous mountainous forests.

Modelling Aboveground Biomass of Miombo Woodlands in Niassa Special Reserve, Northern Mozambique

Aboveground biomass (AGB) estimation plays a crucial role in forest management and carbon emission reporting, especially for developing countries wishing to address REDD+ projects. Both passive and active remote-sensing technologies can provide spatially explicit information of AGB by using a limited number of field samples, thus reducing the substantial budgetary cost of field inventories. The aim of the current study was to estimate AGB in the Niassa Special Reserve (NSR) using fusion of optical (Landsat 8/OLI and Sentinel 2A/MSI) and radar (Sentinel 1B and ALOS/PALSAR-2) data. The performance of multiple linear regression models to relate ground biomass with different combinations of sensor data was assessed using root-mean-square error (RMSE), and the Akaike and Bayesian information criteria (AIC and BIC). The mean AGB and carbon stock (CS) estimated from field data were estimated at 56 Mg ha−1 (ranging from 11 to 95 Mg ha−1) and 28 MgC ha−1, respectively. The best model estimated AGB at 63 ± 20.3 Mg ha−1 for NSR, ranging from 0.6 to 200 Mg ha−1 (r2 = 87.5%, AIC = 123, and BIC = 51.93). We obtained an RMSE % of 20.46 of the mean field estimate of 56 Mg ha−1. The estimation of AGB in this study was within the range that was reported in the existing literature for the miombo woodlands. The fusion of vegetation indices derived from Landsat/OLI and Sentinel 2A/MSI, and backscatter from ALOS/PALSAR-2 is a good predictor of AGB.

An Improved Generalized Hierarchical Estimation Framework with Geostatistics for Mapping Forest Parameters and Its Uncertainty: A Case Study of Forest Canopy Height

Forest canopy height is an essential parameter in estimating forest aboveground biomass (AGB), growing stock volume (GSV), and carbon storage, and it can provide necessary information in forest management activities. Light direction and ranging (LiDAR) is widely used for estimating canopy height. Considering the high cost of acquiring LiDAR data over large areas, we took a two-stage up-scaling approach in estimating forest canopy height and aimed to develop a method for quantifying the uncertainty of the estimation result. Based on the generalized hierarchical model-based (GHMB) estimation framework, a new estimation framework named RK-GHMB that makes use of a geostatistical method (regression kriging, RK) was developed. In this framework, the wall-to-wall forest canopy height and corresponding uncertainty in map unit scale are generated. This study was carried out by integrating plot data, sampled airborne LiDAR data, and wall-to-wall Ziyuan-3 satellite (ZY3) stereo images. The result shows that RK-GHMB can obtain a similar estimation accuracy (r = 0.92, MAE = 1.50 m) to GHMB (r = 0.92, MAE = 1.52 m) with plot-based reference data. For LiDAR-based reference data, the accuracy of RK-GHMB (r = 0.78, MAE = 1.75 m) is higher than that of GHMB (r = 0.75, MAE = 1.85 m). The uncertainties for all map units range from 1.54 to 3.60 m for the RK-GHMB results. The values change between 1.84 and 3.60 m for GHMB. This study demonstrates that this two-stage up-scaling approach can be used to monitor forest canopy height. The proposed RK-GHMB approach considers the spatial autocorrelation of neighboring data in the second modeling stage and can achieve a higher accuracy.

Estimation of Boreal Forest Growing Stock Volume in Russia from Sentinel-2 MSI and Land Cover Classification

Growing stock volume (GSV) is a fundamental parameter of forests, closely related to the above-ground biomass and hence to carbon storage. Estimation of GSV at regional to global scales depends on the use of satellite remote sensing data, although accuracies are generally lower over the sparse boreal forest. This is especially true of boreal forest in Russia, for which knowledge of GSV is currently poor despite its global importance. Here we develop a new empirical method in which the primary remote sensing data source is a single summer Sentinel-2 MSI image, augmented by land-cover classification based on the same MSI image trained using MODIS-derived data. In our work the method is calibrated and validated using an extensive set of field measurements from two contrasting regions of the Russian arctic. Results show that GSV can be estimated with an RMS uncertainty of approximately 35–55%, comparable to other spaceborne estimates of low-GSV forest areas, with 70% spatial correspondence between our GSV maps and existing products derived from MODIS data. Our empirical approach requires somewhat laborious data collection when used for upscaling from field data, but could also be used to downscale global data.

Machine Learning Techniques for Fine Dead Fuel Load Estimation Using Multi-Source Remote Sensing Data

Fine dead fuel load is one of the most significant components of wildfires without which ignition would fail. Several studies have previously investigated 1-h fuel load using standard fuel parameters or site-specific fuel parameters estimated ad hoc for the landscape. On the one hand, these methods have a large margin of error, while on the other their production times and costs are high. In response to this gap, a set of models was developed combining multi-source remote sensing data, field data and machine learning techniques to quantitatively estimate fine dead fuel load and understand its determining factors. Therefore, the objectives of the study were to: (1) estimate 1-h fuel loads using remote sensing predictors and machine learning techniques; (2) evaluate the performance of each machine learning technique compared to traditional linear regression models; (3) assess the importance of each remote sensing predictor; and (4) map the 1-h fuel load in a pilot area of the Apulia region (southern Italy). In pursuit of the above, fine dead fuel load estimation was performed by the integration of field inventory data (251 plots), Synthetic Aperture Radar (SAR, Sentinel-1), optical (Sentinel-2), and Light Detection and Ranging (LIDAR) data applying three different algorithms: Multiple Linear regression (MLR), Random Forest (RF), and Support Vector Machine (SVM). Model performances were evaluated using Root Mean Squared Error (RMSE), Mean Squared Error (MSE), the coefficient of determination (R2) and Pearson’s correlation coefficient (r). The results showed that RF (RMSE: 0.09; MSE: 0.01; r: 0.71; R2: 0.50) had more predictive power compared to the other models, while SVM (RMSE: 0.10; MSE: 0.01; r: 0.63; R2: 0.39) and MLR (RMSE: 0.11; MSE: 0.01; r: 0.63; R2: 0.40) showed similar performances. LIDAR variables (Canopy Height Model and Canopy cover) were more important in fuel estimation than optical and radar variables. In fact, the results highlighted a positive relationship between 1-h fuel load and the presence of the tree component. Conversely, the geomorphological variables appeared to have lower predictive power. Overall, the 1-h fuel load map developed by the RF model can be a valuable tool to support decision making and can be used in regional wildfire risk management.

Comparison of Modeling Algorithms for Forest Canopy Structures Based on UAV-LiDAR: A Case Study in Tropical China

Knowledge of forest structure is vital for sustainable forest management decisions. Terrestrial laser scanning cannot describe the canopy trees in a large area, and it is unclear whether unmanned aerial vehicle-light detection and ranging (UAV-LiDAR) data have the ability to capture the forest canopy structural parameters in tropical forests. In this study, we estimated five forest canopy structures (stand density (N), basic area (G), above-ground biomass (AGB), Lorey’s mean height (HL), and under-crown height (hT)) with four modeling algorithms (linear regression (LR), bagged tree (BT), support vector regression (SVR), and random forest (RF)) based on UAV-LiDAR data and 60 sample plot data from tropical forests in Hainan and determined the optimal algorithms for the five canopy structures by comparing the performance of the four algorithms. First, we defined the canopy tree as a tree with a height ≥70% HL. Then, UAV-LiDAR metrics were calculated, and the LiDAR metrics were screened by recursive feature elimination (RFE). Finally, a prediction model of the five forest canopy structural parameters was established by the four algorithms, and the results were compared. The metrics’ screening results show that the most important LiDAR indexes for estimating HL, AGB, and hT are the leaf area index and some height metrics, while the most important indexes for estimating N and G are the kurtosis of heights and the coefficient of variation of height. The relative root mean squared error (rRMSE) of five structure parameters showed the following: when modeling HL, the rRMSEs (10.60%–12.05%) obtained by the four algorithms showed little difference; when N was modeled, BT, RF, and SVR had lower rRMSEs (26.76%–27.44%); when G was modeled, the rRMSEs of RF and SVR (15.37%–15.87%) were lower; when hT was modeled, BT, RF, and SVR had lower rRMSEs (10.24%–11.07%); when AGB was modeled, RF had the lowest rRMSE (26.75%). Our results will help facilitate choosing LiDAR indexes and modeling algorithms for tropical forest resource inventories.

A Model-Based Volume Estimator that Accounts for Both Land Cover Misclassification and Model Prediction Uncertainty

Forest/non-forest and forest species maps are often used by forest inventory programs in the forest estimation process. For example, some inventory programs establish field plots only on lands corresponding to the forest portion of a forest/non-forest map and use species-specific area estimates obtained from those maps to support the estimation of species-specific volume (V) totals. Despite the general use of these maps, the effects of their uncertainties are commonly ignored with the result that estimates might be unreliable. The goal of this study is to estimate the effects of the uncertainty of forest species maps used in the sampling and estimation processes. Random forest (RF) per-pixel predictions were used with model-based inference to estimate V per unit area for the six main forest species of La Rioja, Spain. RF models for predicting V were constructed using field plot information from the Spanish National Forest Inventory and airborne laser scanning data. To limit the prediction of V to pixels classified as one of the main forest species assessed, a forest species map was constructed using Landsat and auxiliary information. Bootstrapping techniques were implemented to estimate the total uncertainty of the V estimates and accommodated both the effects of uncertainty in the Landsat forest species map and the effects of plot-to-plot sampling variability on training data used to construct the RF V models. Standard errors of species-specific total V estimates increased from 2–9% to 3–22% when the effects of map uncertainty were incorporated into the uncertainty assessment. The workflow achieved satisfactory results and revealed that the effects of map uncertainty are not negligible, especially for open-grown and less frequently occurring forest species for which greater variability was evident in the mapping and estimation process. The effects of forest map uncertainty are greater for species-specific area estimation than for the selection of field plots used to calibrate the RF model. Additional research to generalize the conclusions beyond Mediterranean to other forest environments is recommended.

Multispectral LiDAR-Based Estimation of Surface Fuel Load in a Dense Coniferous Forest

Surface fuel load (SFL) constitutes one of the most significant fuel components and is used as an input variable in most fire behavior prediction systems. The aim of the present study was to investigate the potential of discrete-return multispectral Light Detection and Ranging (LiDAR) data to reliably predict SFL in a coniferous forest characterized by dense overstory and complex terrain. In particular, a linear regression analysis workflow was employed with the separate and combined use of LiDAR-derived structural and pulse intensity information for the load estimation of the total surface fuels and individual surface fuel types. Following a leave-one-out cross-validation (LOOCV) approach, the models developed from the different sets of predictor variables were compared in terms of their estimation accuracy. LOOCV indicated that the predictive models produced by the combined use of structural and intensity metrics significantly outperformed the models constructed with the individual sets of metrics, exhibiting an explained variance (R2) between 0.59 and 0.71 (relative Root Mean Square Error (RMSE) 19.3–37.6%). Overall, the results of this research showcase that both structural and intensity variables provided by multispectral LiDAR data are significant for surface fuel load estimation and can successfully contribute to effective pre-fire management, including fire risk assessment and behavior prediction in case of a fire event.

Woody Aboveground Biomass Mapping of the Brazilian Savanna with a Multi-Sensor and Machine Learning Approach

The tropical savanna in Brazil known as the Cerrado covers circa 23% of the Brazilian territory, but only 3% of this area is protected. High rates of deforestation and degradation in the woodland and forest areas have made the Cerrado the second-largest source of carbon emissions in Brazil. However, data on these emissions are highly uncertain because of the spatial and temporal variability of the aboveground biomass (AGB) in this biome. Remote-sensing data combined with local vegetation inventories provide the means to quantify the AGB at large scales. Here, we quantify the spatial distribution of woody AGB in the Rio Vermelho watershed, located in the centre of the Cerrado, at a high spatial resolution of 30 metres, with a random forest (RF) machine-learning approach. We produced the first high-resolution map of the AGB for a region in the Brazilian Cerrado using a combination of vegetation inventory plots, airborne light detection and ranging (LiDAR) data, and multispectral and radar satellite images (Landsat 8 and ALOS-2/PALSAR-2). A combination of random forest (RF) models and jackknife analyses enabled us to select the best remote-sensing variables to quantify the AGB on a large scale. Overall, the relationship between the ground data from vegetation inventories and remote-sensing variables was strong (R2 = 0.89), with a root-mean-square error (RMSE) of 7.58 Mg ha−1 and a bias of 0.43 Mg ha−1.

Improving aboveground biomass maps of tropical dry forests by integrating LiDAR, ALOS PALSAR, climate and field data

BACKGROUND

Reliable information about the spatial distribution of aboveground biomass (AGB) in tropical forests is fundamental for climate change mitigation and for maintaining carbon stocks. Recent AGB maps at continental and national scales have shown large uncertainties, particularly in tropical areas with high AGB values. Errors in AGB maps are linked to the quality of plot data used to calibrate remote sensing products, and the ability of radar data to map high AGB forest. Here we suggest an approach to improve the accuracy of AGB maps and test this approach with a case study of the tropical forests of the Yucatan peninsula, where the accuracy of AGB mapping is lower than other forest types in Mexico. To reduce the errors in field data, National Forest Inventory (NFI) plots were corrected to consider small trees. Temporal differences between NFI plots and imagery acquisition were addressed by considering biomass changes over time. To overcome issues related to saturation of radar backscatter, we incorporate radar texture metrics and climate data to improve the accuracy of AGB maps. Finally, we increased the number of sampling plots using biomass estimates derived from LiDAR data to assess if increasing sample size could improve the accuracy of AGB estimates.

RESULTS

Correcting NFI plot data for both small trees and temporal differences between field and remotely sensed measurements reduced the relative error of biomass estimates by 12.2%. Using a machine learning algorithm, Random Forest, with corrected field plot data, backscatter and surface texture from the L-band synthetic aperture radar (PALSAR) installed on the on the Advanced Land Observing Satellite-1 (ALOS), and climatic water deficit data improved the accuracy of the maps obtained in this study as compared to previous studies (R² = 0.44 vs R² = 0.32). However, using sample plots derived from LiDAR data to increase sample size did not improve accuracy of AGB maps (R² = 0.26).

CONCLUSIONS

This study reveals that the suggested approach has the potential to improve AGB maps of tropical dry forests and shows predictors of AGB that should be considered in future studies. Our results highlight the importance of using ecological knowledge to correct errors associated with both the plot-level biomass estimates and the mismatch between field and remotely sensed data.

Mangrove forest classification and aboveground biomass estimation using an atom search algorithm and adaptive neuro-fuzzy inference system

Background

Advances in earth observation and machine learning techniques have created new options for forest monitoring, primarily because of the various possibilities that they provide for classifying forest cover and estimating aboveground biomass (AGB).

Methods

This study aimed to introduce a novel model that incorporates the atom search algorithm (ASO) and adaptive neuro-fuzzy inference system (ANFIS) into mangrove forest classification and AGB estimation. The Ca Mau coastal area was selected as a case study since it has been considered the most preserved mangrove forest area in Vietnam and is being investigated for the impacts of land-use change on forest quality. The model was trained and validated with a set of Sentinel-1A imagery with VH and VV polarizations, and multispectral information from the SPOT image. In addition, feature selection was also carried out to choose the optimal combination of predictor variables. The model performance was benchmarked against conventional methods, such as support vector regression, multilayer perceptron, random subspace, and random forest, by using statistical indicators, namely, root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R²).

Results

The results showed that all three indicators of the proposed model were statistically better than those from the benchmarked methods. Specifically, the hybrid model ended up at RMSE = 70.882, MAE = 55.458, R² = 0.577 for AGB estimation.

Conclusion

From the experiments, such hybrid integration can be recommended for use as an alternative solution for biomass estimation. In a broader context, the fast growth of metaheuristic search algorithms has created new scientifically sound solutions for better analysis of forest cover.

Mapping and Monitoring Fractional Woody Vegetation Cover in the Arid Savannas of Namibia Using LiDAR Training Data, Machine Learning, and ALOS PALSAR Data

Namibia is a very arid country, which has experienced significant bush encroachment and associated decreased livestock productivity. Therefore, it is essential to monitor bush encroachment and widespread debushing activities, including selective bush thinning and complete bush clearing. The aim of study was to develop a system to map and monitor fractional woody cover (FWC) at national scales (50 m and 75 m resolution) using Synthetic Aperture Radar (SAR) satellite data (Advanced Land Observing Satellite (ALOS) Phased Arrayed L-band Synthetic Aperture Radar (PALSAR) global mosaics, 2009, 2010, 2015, 2016) and ancillary variables (mean annual precipitation—MAP, elevation), with machine learning models that were trained with diverse airborne Light Detection and Ranging (LiDAR) data sets (244,032 ha, 2008–2014). When only the SAR variables were used, an average R2 of 0.65 (RSME = 0.16) was attained. Adding either elevation or MAP, or both ancillary variables, increased the mean R2 to 0.75 (RSME = 0.13), and 0.79 (RSME = 0.12). The inclusion of MAP addressed the overestimation of FWC in very arid areas, but resulted in anomalies in the form of sharp gradients in FWC along a MAP contour which were most likely caused by to the geographic distribution of the LiDAR training data. Additional targeted LiDAR acquisitions could address this issue. This was the first attempt to produce SAR-derived FWC maps for Namibia and the maps contain substantially more detailed spatial information on woody vegetation structure than existing national maps. During the seven-year study period the Shrubland–Woodland Mosaic was the only vegetation structural class that exhibited a regional net gain in FWC of more than 0.2 across 9% (11,906 km2) of its area that may potentially be attributed to bush encroachment. FWC change maps provided regional insights and detailed local patterns related to debushing and regrowth that can inform national rangeland policies and debushing programs.

Canopy Height and Above-Ground Biomass Retrieval in Tropical Forests Using Multi-Pass X- and C-Band Pol-InSAR Data

Globally available high-resolution information about canopy height and AGB is important for carbon accounting. The present study showed that Pol-InSAR data from TS-X and RS-2 could be used together with field inventories and high-resolution data such as drone or LiDAR data to support the carbon accounting in the context of REDD+ (Reducing Emissions from Deforestation and Forest Degradation) projects.

An End to End Process Development for UAV-SfM Based Forest Monitoring: Individual Tree Detection, Species Classification and Carbon Dynamics Simulation

To promote Bio-Energy with Carbon dioxide Capture and Storage (BECCS), which aims to replace fossil fuels with bio energy and store carbon underground, and Reducing Emissions from Deforestation and forest Degradation (REDD+), which aims to reduce the carbon emissions produced by forest degradation, it is important to build forest management plans based on the scientific prediction of forest dynamics. For Measurement, Reporting and Verification (MRV) at an individual tree level, it is expected that techniques will be developed to support forest management via the effective monitoring of changes to individual trees. In this study, an end-to-end process was developed: (1) detecting individual trees from Unmanned Aerial Vehicle (UAV) derived digital images; (2) estimating the stand structure from crown images; (3) visualizing future carbon dynamics using a forest ecosystem process model. This process could detect 93.4% of individual trees, successfully classified two species using Convolutional Neural Network (CNN) with 83.6% accuracy and evaluated future ecosystem carbon dynamics and the source-sink balance using individual based model FORMIND. Further ideas for improving the sub-process of the end to end process were discussed. This process is expected to contribute to activities concerned with carbon management such as designing smart utilization for biomass resources and projecting scenarios for the sustainable use of ecosystem services.

Supervised cross-fusion method: a new triplet approach to fuse thermal, radar, and optical satellite data for land use classification

This study presents a new fusion method namely supervised cross-fusion method to improve the capability of fused thermal, radar, and optical images for classification. The proposed cross-fusion method is a combination of pixel-based and supervised feature-based fusion of thermal, radar, and optical data. The pixel-based fusion was applied to fuse optical data of Sentinel-2 and Landsat 8. According to correlation coefficient (CR) and signal to noise ratio (SNR), among the used pixel-based fusion methods, wavelet obtained the best results for fusion. Considering spectral and spatial information preservation, CR of the wavelet method is 0.97 and 0.96, respectively. The supervised feature-based fusion method is a fusion of best output of pixel-based fusion level, land surface temperature (LST) data, and Sentinel-1 radar image using a supervised approach. The supervised approach is a supervised feature selection and learning of the inputs based on linear discriminant analysis and sparse regularization (LDASR) algorithm. In the present study, the non-negative matrix factorization (NMF) was utilized for feature extraction. A comparison of the obtained results with state of the art fusion method indicated a higher accuracy of our proposed method of classification. The rotation forest (RoF) classification results improvement was 25% and the support vector machine (SVM) results improvement was 31%. The results showed that the proposed method is well classified and separated four main classes of settlements, barren land, river, river bank, and even the bridges over the river. Also, a number of unclassified pixels by SVM are very low compared to other classification methods and can be neglected. The study results showed that LST calculated using thermal data has had positive effects on improving the classification results. By comparing the results of supervised cross-fusion without using LST data to the proposed method results, SVM and RoF classifiers showed 38% and 7% of classification improvement, respectively.

AGB Estimation in a Tropical Mountain Forest (TMF) by Means of RGB and Multispectral Images Using an Unmanned Aerial Vehicle (UAV)

The present investigation evaluates the accuracy of estimating above-ground biomass (AGB) by means of two different sensors installed onboard an unmanned aerial vehicle (UAV) platform (DJI Inspire I) because the high costs of very high-resolution imagery provided by satellites or light detection and ranging (LiDAR) sensors often impede AGB estimation and the determination of other vegetation parameters. The sensors utilized included an RGB camera (ZENMUSE X3) and a multispectral camera (Parrot Sequoia), whose images were used for AGB estimation in a natural tropical mountain forest (TMF) in Southern Ecuador. The total area covered by the sensors included 80 ha at lower elevations characterized by a fast-changing topography and different vegetation covers. From the total area, a core study site of 24 ha was selected for AGB calculation, applying two different methods. The first method used the RGB images and applied the structure for motion (SfM) process to generate point clouds for a subsequent individual tree classification. Per the classification at tree level, tree height (H) and diameter at breast height (DBH) could be determined, which are necessary input parameters to calculate AGB (Mg ha−1) by means of a specific allometric equation for wet forests. The second method used the multispectral images to calculate the normalized difference vegetation index (NDVI), which is the basis for AGB estimation applying an equation for tropical evergreen forests. The obtained results were validated against a previous AGB estimation for the same area using LiDAR data. The study found two major results: (i) The NDVI-based AGB estimates obtained by multispectral drone imagery were less accurate due to the saturation effect in dense tropical forests, (ii) the photogrammetric approach using RGB images provided reliable AGB estimates comparable to expensive LiDAR surveys (R2: 0.85). However, the latter is only possible if an auxiliary digital terrain model (DTM) in very high resolution is available because in dense natural forests the terrain surface (DTM) is hardly detectable by passive sensors due to the canopy layer, which impedes ground detection.

Improving Aboveground Forest Biomass Maps: From High-Resolution to National Scale

Forest aboveground biomass (AGB) estimation over large extents and high temporal resolution is crucial in managing Mediterranean forest ecosystems, which have been predicted to be very sensitive to climate change effects. Although many modeling procedures have been tested to assess forest AGB, most of them cover small areas and attain high accuracy in evaluations that are difficult to update and extrapolate without large uncertainties. In this study, focusing on the Region of Murcia in Spain (11,313 km2), we integrated forest AGB estimations, obtained from high-precision airborne laser scanning (ALS) data calibrated with plot-level ground-based measures and bio-geophysical spectral variables (eight different indices derived from MODIS computed at different temporal resolutions), as well as topographic factors as predictors. We used a quantile regression forest (QRF) to spatially predict biomass and the associated uncertainty. The fitted model produced a satisfactory performance (R2 0.71 and RMSE 9.99 t·ha−1) with the normalized difference vegetation index (NDVI) as the main vegetation index, in combination with topographic variables as environmental drivers. An independent validation carried out over the final predicted biomass map showed a satisfactory statistically-robust model (R2 0.70 and RMSE 10.25 t·ha−1), confirming its applicability at coarser resolutions.

Potential of Multi-Temporal ALOS-2 PALSAR-2 ScanSAR Data for Vegetation Height Estimation in Tropical Forests of Mexico

Information on the spatial distribution of forest structure parameters (e.g., aboveground biomass, vegetation height) are crucial for assessing terrestrial carbon stocks and emissions. In this study, we sought to assess the potential and merit of multi-temporal dual-polarised L-band observations for vegetation height estimation in tropical deciduous and evergreen forests of Mexico. We estimated vegetation height using dual-polarised L-band observations and a machine learning approach. We used airborne LiDAR-based vegetation height for model training and for result validation. We split LiDAR-based vegetation height into training and test data using two different approaches, i.e., considering and ignoring spatial autocorrelation between training and test data. Our results indicate that ignoring spatial autocorrelation leads to an overoptimistic model’s predictive performance. Accordingly, a spatial splitting of the reference data should be preferred in order to provide realistic retrieval accuracies. Moreover, the model’s predictive performance increases with an increasing number of spatial predictors and training samples, but saturates at a specific level (i.e., at 12 dual-polarised L-band backscatter measurements and at around 20% of all training samples). In consideration of spatial autocorrelation between training and test data, we determined an optimal number of L-band observations and training samples as a trade-off between retrieval accuracy and data collection effort. In summary, our study demonstrates the merit of multi-temporal ScanSAR L-band observations for estimation of vegetation height at a larger scale and provides a workflow for robust predictions of this parameter.

Determinants of Above-Ground Biomass and Its Spatial Variability in a Temperate Forest Managed for Timber Production

The proper estimation of above-ground biomass (AGB) stocks of managed forests is a prerequisite to quantifying their role in climate change mitigation. The aim of this study was to analyze the spatial variability of AGB and its uncertainty between actively managed pine and unmanaged pine-oak reference forests in central Mexico. To investigate the determinants of AGB, we analyzed variables related to forest management, stand structure, topography, and climate. We developed linear (LM), generalized additive (GAM), and Random Forest (RF) empirical models to derive spatially explicit estimates and their uncertainty, and compared them. AGB was strongly influenced by forest management, as LiDAR-derived stand structure and stand age explained 80.9% to 89.8% of its spatial variability. The spatial heterogeneity of AGB varied positively with stand structural complexity and age in the managed forests. The type of predictive model had an impact on estimates of total AGB in our study site, which varied by as much as 19%. AGB densities varied from 0 to 492 ± 17 Mg ha−1 and the highest values were predicted by GAM. Uncertainty was not spatially homogeneously distributed and was higher with higher AGB values. Spatially explicit AGB estimates and their association with management and other variables in the study site can assist forest managers in planning thinning and harvesting schedules that would maximize carbon stocks on the landscape while continuing to provide timber and other ecosystem services. Our study represents an advancement toward the development of efficient strategies to spatially estimate AGB stocks and their uncertainty, as the GAM approach was used for the first time with improved results for such a purpose.