The density of wood is a key indicator of the carbon investment strategies of trees, impacting productivity and carbon storage. Despite its importance, the global variation in wood density and its environmental controls remain poorly understood, preventing accurate predictions of global forest carbon stocks. Here we analyse information from 1.1 million forest inventory plots alongside wood density data from 10,703 tree species to create a spatially explicit understanding of the global wood density distribution and its drivers. Our findings reveal a pronounced latitudinal gradient, with wood in tropical forests being up to 30% denser than that in boreal forests. In both angiosperms and gymnosperms, hydrothermal conditions represented by annual mean temperature and soil moisture emerged as the primary factors influencing the variation in wood density globally. This indicates similar environmental filters and evolutionary adaptations among distinct plant groups, underscoring the essential role of abiotic factors in determining wood density in forest ecosystems. Additionally, our study highlights the prominent role of disturbance, such as human modification and fire risk, in influencing wood density at more local scales. Factoring in the spatial variation of wood density notably changes the estimates of forest carbon stocks, leading to differences of up to 21% within biomes. Therefore, our research contributes to a deeper understanding of terrestrial biomass distribution and how environmental changes and disturbances impact forest ecosystems.

Solar radiation is affected by absorption and emission phenomena during its downward trajectory from the Sun to the Earth’s surface and during the upward trajectory detected by satellite sensors. This leads to distortion of the ground radiometric properties (reflectance) recorded by satellite images, used in this study to estimate aboveground forest biomass (AGB). Atmospherically-corrected remote sensing data can be used to estimate AGB on a global scale and with moderate effort. The objective of this study was to evaluate four atmospheric correction algorithms (for surface reflectance), ATCOR2 (Atmospheric Correction for Flat Terrain), COST (Cosine of the Sun Zenith Angle), FLAASH (Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes) and 6S (Second Simulation of Satellite Signal in the Solar), and one radiometric correction algorithm (for reflectance at the sensor) ToA (Apparent Reflectance at the Top of Atmosphere) to estimate AGB in temperate forest in the northeast of the state of Durango, Mexico. The AGB was estimated from Landsat 5 TM imagery and ancillary information from a digital elevation model (DEM) using the non-parametric multivariate adaptive regression splines (MARS) technique. Field reference data for the model training were collected by systematic sampling of 99 permanent forest growth and soil research sites (SPIFyS) established during the winter of 2011. The following predictor variables were identified in the MARS model: Band 7, Band 5, slope (β), Wetness Index (WI), NDVI and MSAVI2. After cross-validation, 6S was found to be the optimal model for estimating AGB (R² = 0.71 and RMSE = 33.5 Mg·ha⁻¹; 37.61% of the average stand biomass). We conclude that atmospheric and radiometric correction of satellite images can be used along with non-parametric techniques to estimate AGB with acceptable accuracy.

An accurate estimation of forests’ aboveground biomass (AGB) is required because of its relevance to the carbon cycle, and because of its economic and ecological importance. The selection of appropriate variables from satellite information and physical variables is important for precise AGB prediction mapping. Because of the complex relationships for AGB prediction, non-parametric machine-learning techniques represent potentially useful techniques for AGB estimation, but their use and comparison in forest remote-sensing applications is still relatively limited. The objective of the present study was to evaluate the performance of automatic learning techniques, support vector regression (SVR) and random forest (RF), to predict the observed AGB (from 318 permanent sampling plots) from the Landsat 8 Landsat 8 Operational Land Imager (OLI) sensor, spectral indexes, texture indexes and physical variables the Sierra Madre Occidental in Mexico. The result showed that the best SVR model explained 80% of the total variance (root mean square error (RMSE) = 8.20 Mg ha⁻¹). The variables that best predicted AGB, in order of importance, were the bands that belong to the region of red and near and middle infrared, and the average temperature. The results show that the SVR technique has a good potential for the estimation of the AGB and that the selection of the model hyperparameters has important implications for optimizing the goodness of fit.

This paper presents new equations for estimating above-ground biomass (AGB) and biomass components of seventeen forest species in the temperate forests of northwestern Mexico. A data set corresponding to 1336 destructively sampled oak and pine trees was used to fit the models. The generalized method of moments was used to simultaneously fit systems of equations for biomass components and AGB, to ensure additivity. In addition, the carbon content of each tree component was calculated by the dry combustion method, in a TOC analyser. The results of cross-validation indicated that the fitted equations accounted for on average 91%, 82%, 83% and 76% of the observed variance in stem wood and stem bark, branch and foliage biomass, respectively, whereas the total AGB equations explained on average 93% of the total observed variance in AGB. The inclusion of total height (h) or diameter at breast height² × total height (d²h) as a predictor in the d-only based equations systems slightly improved estimates for stem wood, stem bark and total above-ground biomass, and greatly improved the estimates produced by the branch and foliage biomass equations. The predictive power of the proposed equations is higher than that of existing models for the study area. The fitted equations were used to estimate stand level AGB stocks from data on growing stock in 429 permanent sampling plots. Three machine-learning techniques were used to model the estimated stand level AGB and carbon contents; the selected models were used to map the AGB and carbon distributions in the study area, for which mean values of respectively 129.84 Mg ha⁻¹ and 63.80 Mg ha⁻¹ were obtained.