Received: 30 September 2021  |  Accepted: 16 October 2021 DOI: 10.1111/2041-210X.13756 A P P L I C A T I O N allodb: An R package for biomass estimation at globally distributed extratropical forest plots Erika Gonzalez- Akre1  | Camille Piponiot1,2,3  | Mauro Lepore4  | Valentine Herrmann1  | James A. Lutz5  | Jennifer L. Baltzer6  | Christopher W. Dick7  | Gregory S. Gilbert8  | Fangliang He9  | Michael Heym10  | Alejandra I. Huerta11 | Patrick A. Jansen2,12  | Daniel J. Johnson13  | Nikolai Knapp14,15  | Kamil Král16  | Dunmei Lin17  | Yadvinder Malhi18  | Sean M. McMahon19  | Jonathan A. Myers20  | David Orwig21  | Diego I. Rodríguez- Hernández22  | Sabrina E. Russo23,24  | Jessica Shue19 | Xugao Wang25  | Amy Wolf26 | Tonghui Yang27 | Stuart J. Davies2  | Kristina J. Anderson- Teixeira1,2 1Conservation Ecology Center, Smithsonian National Zoo & Conservation Biology Institute, Front Royal, VA, USA; 2Forest Global Earth Observatory, Smithsonian Tropical Research Institute, Panama, Panama; 3UR Forests and Societies, Cirad, Univ Montpellier, Montpellier, France; 4Forest Global Earth Observatory, Smithsonian Institution, Washington, DC, USA; 5Wildland Resources Department, Utah State University, Logan, UT, USA; 6Department of Biology, Wilfrid Laurier University, Waterloo, ON, Canada; 7Ecology and Evolutionary Biology, University of Michigan, Ann Harbor, MI, USA; 8Department of Environmental Studies, University of California, Santa Cruz, CA, USA; 9Biodiversity & Landscape Modeling Group, University of Alberta, Edmonton, AB, Canada; 10Faculty of Forest Science and Resource Management, Technical University of Munich, Freising, Germany; 11Deptartment of Entomology and Plant Pathology, North Carolina State University, Raleigh, NC, USA; 12Department of Environmental Sciences, Wageningen University, Wageningen, Netherlands; 13School of Forest, Fisheries, and Geomatics Sciences, University of Florida, Gainesville, FL, USA; 14Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany; 15Thünen Institute of Forest Ecosystems, Eberswalde, Germany; 16Department of Forest Ecology, Silva Tarouca Research Institute, Brno, Czech Republic; 17Key Laboratory of the Three Gorges Reservoir Region's Eco-E nvironment, Ministry of Education, Chongqing University, Chongqing, China; 18School of Geography and the Environment, University of Oxford, Oxford, UK; 19Smithsonian Environmental Research Center, Edgewater, MD, USA; 20Department of Biology, Washington University, St. Louis, MO, USA; 21The Harvard Forest, Petersham, MA, USA; 22Department of Ecology, Sun Yat- sen University, Guangzhou, China; 23School of Biological Sciences, University of Nebraska, Lincoln, NE, USA; 24University of Nebraska– Lincoln, Lincoln, NE, USA; 25Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, China; 26Natural & Applied Sciences, University of Wisconsin, Green Bay, WI, USA and 27Forestry Institute, Ningbo Academy of Agricultural Science, Ningbo, China Correspondence Kristina J. Anderson- Teixeira Abstract Email: teixeirak@si.edu 1. Allometric equations for calculation of tree above- ground biomass (AGB) form the Funding information basis for estimates of forest carbon storage and exchange with the atmosphere. Smithsonian Forest Global Earth While standard models exist to calculate forest biomass across the tropics, we Observatory (ForestGEO); Smithsonian Institution lack a standardized tool for computing AGB across boreal and temperate regions that comprise the global extratropics. Handling Editor: Laura Graham 2. Here we present an integrated R package, allodb, containing systematically se- lected published allometric equations and proposed functions to compute AGB. The data component of the package is based on 701 woody species identified at 24 large Forest Global Earth Observatory (ForestGEO) forest dynamics plots representing a wide diversity of extratropical forests. © 2021 British Ecological Society. This article has been contributed to by US Government employees and their work is in the public domain in the USA. Methods Ecol Evol. 2021;00:1–9. wileyonlinelibrary.com/journal/mee3  |  1 2  |   M ethods in Ecology and Evolu on GONZALEZ-A KRE Et AL. 3. A total of 570 parsed allometric equations to estimate individual tree biomass were retrieved, checked and combined using a weighting function designed to ensure optimal equation selection over the full tree size range with smooth transitions across equations. The equation dataset can be customized with built- in functions that subset the original dataset and add new equations. 4. Although equations were curated based on a limited set of forest communities and number of species, this resource is appropriate for large portions of the global extratropics and can easily be expanded to cover novel forest types. K E Y W O R D S above- ground biomass, extratropics, forest carbon storage, Forest Global Earth Observatory (ForestGEO), R, temperate forest, tree allometry, tree biomass 1  | INTRODUC TION biomass data have been pooled to form the basis of a standard- ized approach to biomass estimation across the tropics (Chave Forest trees account for 70%–9 0% of the land biomass of earth et al., 2005, 2014; Réjou- Méchain et al., 2017), no such standard- (Houghton, 2008). The quantification of forest above- ground bio- ized approach currently exists for extratropical regions (above 23.5° mass (AGB) is an essential step to understand the sources, sinks and latitude N or S). Rather, a wide variety of allometries developed for flow of carbon world-w ide and, more importantly, how carbon stor- various levels of taxonomic and geographic organization, and of age and fluxes are changing through time (Houghton, 2005). Changes variable quality, are scattered throughout the literature (Chojnacky in forest carbon storage will strongly influence the course of climate et al., 2014; Conti et al., 2019; Jenkins et al., 2004; Luo et al., 2018, change (Friedlingstein et al., 2006), and forest conservation, man- 2020; Muukkonen, 2007; Návar, 2009; Paul et al., 2016; Rojas- agement and restoration are among the most cost- effective tools for García et al., 2015). These equations differ in functional form, input climate change mitigation (Griscom et al., 2017). Indeed, changes in and output variables, units and size range across which they can be forest carbon are emphasized in the guidelines for national green- applied. This makes identification and application of appropriate al- house gas inventories by the Intergovernmental Panel on Climate lometries a time- consuming and error- prone process (van Breugel Change (IPCC, Buendia et al., 2019), and account for approximately et al., 2011) with low reproducibility and little standardization across one-q uarter of national emission reductions planned by countries studies (Somogyi et al., 2007). While challenging for studies at indi- under the Paris Climate Agreement (Grassi et al., 2017). Thus, accu- vidual sites, this becomes particularly problematic for studies aiming rate estimates of tree biomass are critical to understanding forest to apply an approach that is both locally optimized and standardized carbon dynamics and managing forests for climate change mitigation. across multiple forest types and regions (e.g. Lutz et al., 2017). Despite rapidly developing technology focusing on remote sens- Several key principles should guide the development of temper- ing to estimate forest biomass over large areas (Knapp et al., 2020; ate and boreal allometries. First, larger sample sizes of trees used to Zolkos et al., 2013), ground-b ased assessments that combine tree develop allometric equations greatly reduce biases and systematic er- census data and allometric equations remain the most widely applied rors (Duncanson, Rourke, et al., 2015), and are particularly important indirect method to estimate forest biomass and are still required in constraining the uncertainty in AGB estimates of large trees (Chave to calibrate remote sensing data (Chave et al., 2014, 2019). These et al., 2004; Stovall et al., 2018; Sullivan et al., 2018). For example, models are based on common biomass predictors including DBH pantropical models based on large datasets (Chave et al., 2005; and height (H) (e.g. Feldpausch et al., 2012), sometimes incorpo- Feldpausch et al., 2011) give reliable results with smaller errors com- rating wood density and crown structure (Chave et al., 2005, 2014; pared to regional models (Rutishauser et al., 2013). Second, the pre- Goodman et al., 2014). Although ground- based LiDAR is emerging cision of predictions can be improved by using equations calibrated as a promising technique for non- destructive allometry develop- with trees from a similar taxonomic group, and that grew in similar ment (Stovall et al., 2018), the vast majority of biomass allometries climatic conditions (Daba & Soromessa, 2019; Ngomanda et al., 2014; have been created through destructive tree harvest. Yet, the devel- Roxburgh et al., 2015). In practice, these two principles are in con- opment of reliable allometric equations requires large sample sizes flict, in that taxa- or location- specific allometries are usually con- (Duncanson et al., 2015), particularly for large trees that are the structed based on a much lower sample size than generic allometries. most problematic to sample (Stovall et al., 2018) and usually under- Furthermore, specific allometries are often limited in the size range represented (Burt et al., 2020). Moreover, allometric relationships over which they were calibrated and are largely driven by a very small vary across species (Poorter et al., 2015; but see Paul et al., 2016) number of large trees, leading to potentially large errors if extrap- and with environmental factors such as climate and nutrient avail- olated beyond their size range, or to discontinuous functions if an ability (Duncanson et al., 2015; Lines et al., 2012), stand age (Fatemi alternative equation is applied beyond the calibrated range. Lastly, et al., 2011) and stand density (Gower et al., 1992). Whereas tropical biomass allometries should be continuous functions of tree size. This GONZALEZ- AKRE Et AL. Methods in Ecology and Evolu  on    |  3 is especially critical for applications using records of tree diameter 2  | SOF T WARE DE VELOPMENT AND growth to estimate woody productivity (e.g. Anderson-T eixeira et al., WORKFLOW 2021; Helcoski et al., 2019) or to compare carbon stocks or fluxes across tree size classes (e.g. Lutz et al., 2018; Meakem et al., 2018; 2.1 | Focal sites and species Piponiot, C. unpubl. data). Ideally, continuous functions based on suf- ficient sample sizes would be derived from re-a nalysis of data col- We focus on multiple sites within the Forest Global Earth lected to produce existing sets of allometric equations, as has been Observatory (ForestGEO), the largest world-w ide network of long- done for the tropics (Chave et al., 2014), but unfortunately original term forest monitoring sites using standardized methods (Anderson- data are often difficult to access, lack proper documentation or are Teixeira et al., 2015; Davies et al., 2021). As such, it is a good model unavailable. Although there has been some progress in developing for assembling and applying allometric equations across a wide comprehensive databases to support the development of allometries range of species, forest environments and to understand associated (Falster et al., 2015; Henry et al., 2013; Schepaschenko et al., 2017), challenges in calculating biomass. ForestGEO currently includes 33 these are not yet comparable in coverage to the existing set of al- extratropical forests across North America (n = 17), Europe (n = 4) lometric models. Thus, for now, a standardized process for applying and Asia (n = 12), ranging in latitude from 23 to 61 degrees N. At biomass allometries across extratropical forests must draw upon ex- each site, all stems ≥1 cm DBH within 5– 50 ha plots are censused isting sets of allometric equations. following standardized protocols, including identification to spe- Here we present a framework aimed at facilitating tree biomass cies level (Condit, 1998). From the 24 participant sites included in estimation across globally distributed extratropical forests. To stan- allodb (Table S1), there are 1109 species- location combinations, 701 dardize and simplify the biomass estimation process, we developed woody species, 248 genera and 86 plant families represented (see allodb (Table 1, https://docs.ropens ci.org/allod b/) as an open- source site- species table in allodb). application aiming to: (a) compile relevant published and unpublished allometric equations, focusing on AGB but structured to handle other variables (e.g. height and biomass components); (b) objectively 2.2 | Systematic search for biomass allometries select and integrate appropriate available equations across the full range of tree sizes; and (c) serve as a platform for future updates and We compiled 570 allometric equations from the literature, focus- expansion to other research sites globally. ing on retrieving equations to estimate AGB based on DBH and TA B L E 1   Description of data and functions in allodb. A detailed explanation Name Description of functions and data can be found in the Data allodb R package documentation (https:// equations A dataframe with retrieved equations from literature and auxiliary data docs.ropens ci.org/allod b/refer ence/index. html) references A dataframe listing all references by reference ID used in equation table site- species A dataframe listing focal sites in this study and the identified family, genus and species per site Metadata equations_ A dataframe explaining fields in the equation table metadata missing_values A dataframe describing the use of codes for missing values used in the equation table reference_ A dataframe explaining fields in the reference table metadata site- species_ A dataframe explaining fields in the site- species table metadata Functions est_params Estimates the parameters (slope, intercept, sigma) of the recalibrated allometric equations get_biomass Executes the AGB calculation per stem (kg) illustrate_allodb Produces illustrative graphs of the recalibration process new_equations Customizes the original set of allometric equations by subsetting it and/ or by adding new equations resample_agb Resamples the original equations weight_allom Combines multiple variables (taxa, climate and sample size) to attribute a weight to each equation 4  |   M ethods in Ecology and Evolu on GONZALEZ- AKRE Et AL. developed primarily in extratropical regions (Chojnacky et al., 2014; use of such a function to homogenize and correct taxonomic in- Forrester et al., 2017; Jenkins et al., 2004; Luo et al., 2018), and drew formation prior to using allodb. upon these and local expertise to help identify original, species- c. Site coordinates: These are needed to account for climate zones. specific and location- specific allometries (Figure S1). Three of our The Köppen classification scheme (Köppen, 2011) provides an focal sites have local biomass allometries (SCBI: Stovall et al., 2018; efficient way to describe climatic conditions defined by multiple Wytham Woods: Fenn et al., 2015; and Yosemite: Lutz et al., 2014). variables with a single and ecologically relevant metric (Chen & For eighteen species found at the University of California Santa Cruz Chen, 2013) and allows the assignment to a particular climate ForestGEO site (UCSC, Table S1), we include new local allometric based on site coordinates. allodb obtains the Köppen climate zone equations to estimate H, which is an independent variable in some of a given site using the kgc R package (Bryant et al., 2017). The allometric models. In some cases, equations were only available for obtained climate is then compared to the allometric equations’ separate tree components (stem, bark, branches, foliage); these Köppen zone(s) and used in the weighting scheme. By including a were summed to obtain AGB. For each equation, we retrieved stand- climate input, we are able to represent bioclimatic variables oth- ard information including location, taxa, units, DBH ranges, sample erwise not included in original publications. size (see allodb equations table for other categories), which are used in the proposed weighting scheme. We assigned Köppen climate A user constructs a table with DBH, species and site coordinates, zones to each equation using the R package kgc (Bryant et al., 2017; as in the example provided in the allodb package: Köppen, 2011). When equations were calibrated for broad regions (e.g. North America, Northern Germany) or vaguely defined loca- install.packages("remotes") tions, we estimated their location from brief descriptions or regional remotes::install _ github("ropensci/allodb") maps in the original publication and included all possible Köppen library(allodb) zones. Details on all equations are available in the equations.csv file data(scbi _ stem1) within allodb. scbi _ stem1$agb = get _ biomass( dbh = scbi _ stem1$dbh, 2.3 | Inputs for calculating biomass genus = scbi _ stem1$genus, species = scbi _ stem1$species, Prior to calculating tree biomass using allodb, users need to provide: coords = c(-78.2, 38.9) (a) DBH (cm), (b) parsed species Latin names and (c) site coordinates ) (Figure 1). a. DBH: allodb makes consistent calculations of AGB (kg) based on DBH (cm) as the primary predictor. In some instances, avail- 2.4 | AGB estimation in allodb able allometric equations include H as an additional predictor (e.g. Jansen et al., 1996), for these cases, inputs of H (m) refine allodb estimates AGB (or any other dependent variable) by calibrat- predictions. We structured allodb expecting that the input DBH ing a new allometric equation for each taxon and location in the from plot inventories is checked in advance. For sites where user-p rovided census data. The new allometric equation is based trees are commonly measured at heights other than the standard on a set of allometric equations that can be customized using the 1.3 m (e.g. buttresses, trunk irregularities, differing census pro- new_equations() function. Each equation is then given a weight by tocols), we recommend users to apply a taper correction func- the function weight_allom() based on: (1) its original sample size tion to improve the estimates of biomass changes (see Cushman (numbers of trees used to develop a given allometry), (2) its cli- et al., 2014) before using allodb. As many forest census proto- matic similarity with the target location and (3) its taxonomic simi- cols recommend measuring stems at 1.3 m (including shrubs), we larity with the target taxon (see weighting scheme below). The final provided additional equations to convert DBH into diameter at weight attributed to each equation is the product of those three base (dba, i.e. diameter conversion models by Lutz, 2005; Paul weights. Equations are then resampled with the function resam- et al., 2016) for those allometries that use dba or diameter at ple_agb(): the number of samples per equation is proportional to stump height (20– 30 cm above the ground) to predict biomass. its weight, and the total number of samples is 104 by default. The b. Latin species names: Species identification is critical for selecting resampling is done by drawing DBH values from a uniform distri- appropriate allometric equations. To standardize spelling and bution on the DBH range of the equation, and estimating the AGB nomenclature, plant names for all sites were checked using the with the equation. The pairs of values (DBH, AGB) obtained are function correctTaxo from the BIOMASS package (Réjou- Méchain then used in the function est_params() to recalibrate a new allo- et al., 2017). Accepted family names (used in the weighting metric equation: this is done by applying a linear regression to the scheme) were retrieved using the function tax_name from the log- transformed data (see example in Figure 1). The parameters package taxize (Chamberlain et al., 2020). We recommend the of the new allometric equations are then used in the get_biomass() GONZALEZ- AKRE Et AL. Methods in Ecology and Evolu  on    |  5 F I G U R E 1   Illustration of allodb workflow and predictions. User provides a dataframe with DBH (cm), parsed species Latin names and site coordinates. allodb estimates AGB by calibrating a new allometric equation for each taxon in the user-p rovided data. The equations table in allodb can be customized using the new_equations() function. Each equation is given a weight by the weight_allom() function and then resampled with the function resample_agb(). The values obtained are used in the function est_params() to recalibrate a new allometric equation and then used in the get_biomass() function. illustrate_allodb() is used to visualize the top resampled equations (details for each equation can be found in the equations table within allodb) and the final fitted equation function by back- transforming the AGB predictions based on the three- letter system of Köppen climate scheme (Köppen, 2011). user- provided DBHs. By using the function illustrate_allodb(), the This weight is calculated in three steps: (1) if the main climate user can visualize in a plot the top 10 resampled equations and the group (first letter) is the same, the climate weight starts at 0.4; final fitted equation (e.g. Figure 1; Figure S3). if one of the groups is ‘C’ (temperate climate) and the other is ‘D’ (continental climate), the climate weight starts at 0.2 because the two groups are considered similar enough; otherwise, the weight 2.5 | Weighting scheme of allometric equations is 1e-6 ; (2) if the equation and site belong to the same group, the weight is incremented by an additional value between 1e- 6 and Each equation is given a weight by the function weight_allom(), calcu- 0.3 based on precipitation pattern similarity (second letter of the lated as the product of the following components: Köppen zone); and (3) if the equation and site belong to the same group, the weight is incremented by an additional value between 1. Sample- size weight: because larger sample sizes greatly reduce 1e- 6 and 0.3 based on temperature pattern similarity (third letter biases and systematic errors (Duncanson, Rourke, et al., 2015), of the Köppen zone). The resulting weight has a value between we attribute a larger weight to equations calibrated with a larger 1e- 6 (different climate groups) and 1 (exactly the same climate number of trees. This weight is calculated as an asymptotic classification). When an equation was calibrated with trees from ( ) log(20) function of the sample size n: −n⋅1 − e w95 . The sample-s ize several locations with different Köppen climates, the maximum weight increases sharply at low sample sizes and gets close value out of all pairwise equation- site climate weights is used. to 1 (its asymptotic value) for sample sizes >w95. w95 is 500 3. Taxonomic weight: equations calibrated with trees from a similar by default, and may be adjusted by the user. Equations with taxonomic group as the target taxon are given a higher weight no sample size information are given a sample-s ize weight of (Figure S2). The taxonomic weight is equal to 1 for same species 0.1 by default: this value can be adjusted by the user using equations, 0.8 for same genus equations and 0.5 for same family the argument wna. equations and for equations calibrated for the same broad func- 2. Climatic weight: equations calibrated in similar climatic condi- tional or taxonomic group (e.g. shrubs, conifers, angiosperms). All tions as the target location are given a higher weight, using the other equations are given a low taxonomic weight of 10−6: these 6  |   M ethods in Ecology and Evolu on GONZALEZ- AKRE Et AL. equations will have a significant relative weight in the final predic- occurred for the largest DBH trees in the plot, for which absolute tion only when no other more specific equation is available. differences could be large (>3,000 kg) for a couple of species (e.g. Quercus velutina), with the Chojnacky et al. (2014) allometries pre- The choices of weighting functions and parameter values are dicting higher AGB. Across smaller and intermediate tree sizes, al- selected based on our current understanding of the principles of al- lodb predictions could be higher or lower depending on the species, lometric equations and experimentation with various options, and with an overall tendency for allodb predictions to be higher. Both of weightings may be adjusted based on user discretion. However, ad- these differences align with the findings of a terrestrial LiDAR study justing these values can result in unsatisfactory predictions: alter- at this site (Stovall et al., 2018), which found that the Chojnacky et al. native weighting schemes should be checked before being used for (2014) equations underestimated biomass overall while overestimat- predictions. ing biomass of the largest individuals. Summing across all trees in the In particular, we use taxonomic similarity as an easily measur- SCBI plot, allodb predicted a total AGB of ss307.6 Mg/ha, which is able proxy of expected similarity among species’ allometries, but 19% higher than a published estimate of 258.9 Mg/ha that applies the assumption that related species have similar allometries does Chojnacky et al. (2014) equations to the same data (Lutz et al., 2018). not always hold. For example, the North American high- elevation Finally, we tested the accuracy of allodb predictions against a com- five-n eedle pines (Pinus longaeva, P. aristata, P. albicaulis and P. bal- prehensive compilation on destructive sampling by Schepaschenko fouriana) are morphologically similar to one another but extremely et al. (2017). A subset (n = 6266 trees) from the original dataset was different from the more common Pinus species (e.g. Pinus strobus). used providing DBH (>1 cm), H (m) and measured AGB (kg) at 176 Because generic genus- level equations are usually based on the sites distributed in Eurasia (Figure S5). The allodb predictions were more common species (e.g. Chojnacky et al., 2014), biased predic- reasonable across the tree size range, with root- mean- square error tions can result where the target species has vastly different mor- (RMSE) of 87.02 kg on a linear scale (and a mean relative error [MRE] phology or wood density from the genus-l evel mean, particularly if of 72%) and 0.71 kg on a logarithmic scale. they grow in similar climate zones. The resulting errors can be espe- cially important when dealing with large trees. Using species’ phy- logenetic or morphological similarity and wood density could help 3  | CONCLUSIONS AND FUTURE reduce such biases, but this information is not always available for IMPROVEMENTS all species and equations. We recommend that researchers working with species that do not conform to generalized allometric models The calculation of tree biomass has multiple challenges that we tried for their taxa and climate zone (i.e. ~8% of species in analysis of Paul to overcome when designing allodb. The allodb package makes it et al., 2016) carefully evaluate the weighting of allodb equations and possible to obtain consistent, reproducible AGB estimates for ex- apply alternative allometric models if needed. tratropical forests, noting that careful attention to versioning (i.e. citation of package version) will be necessary to ensure reproduc- ibility. We believe that these estimates are as accurate as possible 2.6 | Evaluation and validation of methods given the issues that currently plague the field (e.g. limited diam- eter ranges, allometries based on low sample sizes, lack of harvested To validate and evaluate allodb, we (a) screened for equation errors; data; Burt et al., 2020). In addition, the allodb platform and scope can (b) evaluated against widely used regional allometric models; and (c) be expanded to include more equations and thereby represent more compared allodb predictions against raw data. species and sites. It can also be expanded to cover more response As a preliminary test to detect preventable equation errors (e.g. variables (e.g. roots, foliage, heights and crown dimensions) so that unit conversion issues, typos when transcribing, errors within origi- we can better predict AGB (or below ground biomass) on an ecosys- nal publications), we manually evaluated each equation in R (R Core tem scale, characterize forest structure and potentially link it with Team, 2018) as it was entered into our dataset to ensure that predic- LiDAR applications and more general remote sensing methods. With tions were within reasonable range. We identified outliers through appropriate accounting for snags and down wood (Janik et al., 2017) plotting of each species per focal ForestGEO site to compare bio- and appropriate reduction factors (e.g. Harmon et al., 2011), allodb mass values predicted by the different equations on a hypothetical can also form the basis for calculating dead woody biomass. We DBH range between 1 and 200 cm (e.g. Figure S3). Through this pro- encourage the user community to contribute to building allodb into cess, equation errors were corrected when possible, and problem- an increasingly useful resource for estimating extratropical forest atic equations removed. biomass, thereby better meeting the challenge of characterizing and Next, we evaluated how AGB estimates using allodb compare to managing forest carbon stocks and fluxes in an era of climate change. those obtained from the widely used regional equations for North America of Chojnacky et al. (2014). Using the SCBI ForestGEO plot ACKNOWLEDG EMENTS as a test case, we found that allodb predictions aligned reasonably The authors acknowledge all authors of equations included in allodb with those of the Chojnacky et al. (2014) equations (Figure S4), but for their contributions, without which this synthetic effort would not with differences that can be meaningful. The most notable departure be possible. This study was funded by the Smithsonian Forest Global GONZALEZ-A KRE Et AL. Methods in Ecology and Evolu  on    |  7 Earth Observatory (ForestGEO) and by two Smithsonian Scholarly Zambrano, A. M., Alonso, A., Baltzer, J. L., Basset, Y., Bourg, N. A., Studies grants to K.J.A.- T. and collaborators. Broadbent, E. N., Brockelman, W. Y., Bunyavejchewin, S., Burslem, D. F. R. P., Butt, N., Cao, M., Cardenas, D., … Zimmerman, J. (2015). CTFS-F orestGEO: A worldwide network monitoring forests in an CONFLIC T OF INTERE S T era of global change. Global Change Biology, 21(2), 528– 549. https:// The authors have no conflict of interest to declare. doi.org/10.1111/gcb.12712 Anderson-T eixeira, K. J., Herrmann, V., Rollinson, C. R., Gonzalez, B., Gonzalez-A kre, E. B., Pederson, N., Alexander, M. R., Allen, C. D., AUTHORS' CONTRIBUTIONS Alfaro- Sánchez, R., Awada, T., Baltzer, J. L., Baker, P. J., Birch, J. D., E.G.- A. and K.J.A.- T. conceived the idea; C.P., M.L., E.G.-A . and Bunyavejchewin, S., Cherubini, P., Davies, S. J., Dow, C., Helcoski, K.J.A.-T . designed the software; V.H. contributed with workflow im- R., Kašpar, J., … Zuidema, P. A. (2021). Joint effects of climate, tree provements; K.J.A.-T ., E.G.-A . and C.P. led the writing of the manu- size, and year on annual tree growth derived from tree-r ing records of ten globally distributed forests. Global Change Biology, https:// script. And all other authors contributed critically and approved for doi.org/10.1111/gcb.15934 publication. Bryant, C., Wheeler, N., Rubel, F., & French, R. (2017). Kgc: Koeppen- Geiger climatic zones. R package version 1.0.0.2. https://CRAN.R- PEER RE VIE W projec t.org/packag e=kgc Buendia, C., Eduardo, S. G., Limmeechokchai, B., Pipatti, R., Rojas, Y., The peer review history for this article is available at https://publo Sturgiss, R., Tanabe, K., & Wirth, T., (2019).2019 Refinement to ns.com/publo n/10.1111/2041-2 10X.13756. the 2006 IPCC Guidelines for National Greenhouse Gas Inventories. https://www.ipcc.ch/site/asset s/uploa ds/2019/12/19R_V0_01_ DATA AVAIL ABILIT Y S TATEMENT Overvi ew.pdf The allodb source code and data are published under the GNU Burt, A., Calders, K., Cuni- Sanchez, A., Gómez- Dans, J., Lewis, P., Lewis, S. L., Malhi, Y., & Phillips, O. L., Disney, M. (2020). Assessment of General Public License 3. The version described in this paper (version Bias in Pan- Tropical Biomass Predictions. Frontiers in Forests and 0.0.0.9000) can be accessed at https://docs.ropen sci.org/allodb /. Global Change, 3, https://doi.org/10.3389/ffgc.2020.00012 Chamberlain, S., Szoecs, E., Foster, Z., Arendsee, Z., Boettiger, C., Ram, ORCID K., Baumgartner, J., O'Donnell, J., Oksanen, J., Greshake Tzovaras, B., Marchand, P., Tran, V., Salmon, M., Li, G., & Grenié, M. (2020). Erika Gonzalez-A kre https://orcid.org/0000-0001-8305-6672 Taxize: Taxonomic information from around the web. R package v0.9.9. Camille Piponiot https://orcid.org/0000-0002-3473-1982 Retrieved from https://taxize.dev/ Mauro Lepore https://orcid.org/0000-0002-1986-7988 Chave, J., Andalo, C., Brown, S., Cairns, M. A., Chambers, J. Q., Eamus, D., Valentine Herrmann https://orcid.org/0000-0002-4519-481X Fölster, H., Fromard, F., Higuchi, N., Kira, T., Lescure, J.- P., Nelson, B. W., Ogawa, H., Puig, H., Riéra, B., & Yamakura, T. (2005). Tree James A. Lutz https://orcid.org/0000-0002-2560-0710 allometry and improved estimation of carbon stocks and balance in Jennifer L. Baltzer https://orcid.org/0000-0001-7476-5928 tropical forests. Oecologia, 145(1), 87–9 9. https://doi.org/10.1007/ Christopher W. Dick https://orcid.org/0000-0001-8745-9137 s0044 2- 005- 0100- x Gregory S. Gilbert https://orcid.org/0000-0002-5195-9903 Chave, J., Condit, R., Aguilar, S., Hernandez, A., Lao, S., & Perez, R. (2004). Error propagation and scaling for tropical forest bio- Fangliang He https://orcid.org/0000-0003-0774-4849 mass estimates. Philosophical Transactions of the Royal Society B: Michael Heym https://orcid.org/0000-0002-1314-2257 Biological Sciences, 359(1443), 409–4 20. https://doi.org/10.1098/ Patrick A. Jansen https://orcid.org/0000-0002-4660-0314 rstb.2003.1425 Daniel J. Johnson https://orcid.org/0000-0002-8585-2143 Chave, J., Davies, S. J., Phillips, O. L., Lewis, S. L., Sist, P., Schepaschenko, D., Armston, J., Baker, T. R., Coomes, D., Disney, M., Duncanson, L., Nikolai Knapp https://orcid.org/0000-0001-5065-9979 Hérault, B., Labrière, N., Meyer, V., Réjou- Méchain, M., Scipal, K., Kamil Král https://orcid.org/0000-0002-3848-2119 & Saatchi, S. (2019). Ground data are essential for biomass remote Dunmei Lin https://orcid.org/0000-0001-6112-7783 sensing missions. Surveys in Geophysics, 40(4), 863–8 80. https://doi. Yadvinder Malhi https://orcid.org/0000-0002-3503-4783 org/10.1007/s1071 2-0 19-0 9528- w Sean M. McMahon https://orcid.org/0000-0001-8302-6908 Chave, J., Réjou- Méchain, M., Búrquez, A., Chidumayo, E., Colgan, M. S., Delitti, W. B. C., Duque, A., Eid, T., Fearnside, P. M., Goodman, R. C., Jonathan A. Myers https://orcid.org/0000-0002-2058-8468 Henry, M., Martínez-Y rízar, A., Mugasha, W. A., Muller-L andau, H. C., David Orwig https://orcid.org/0000-0001-7822-3560 Mencuccini, M., Nelson, B. W., Ngomanda, A., Nogueira, E. M., Ortiz- Diego I. Rodríguez-H ernández https://orcid. Malavassi, E., … Vieilledent, G. (2014). Improved allometric models to org/0000-0002-1125-2399 estimate the aboveground biomass of tropical trees. Global Change Biology, 20(10), 3177– 3190. https://doi.org/10.1111/gcb.12629 Sabrina E. Russo https://orcid.org/0000-0002-6788-2410 Chen, D., & Chen, H. W. (2013). Using the Köppen classification to quan- Xugao Wang https://orcid.org/0000-0003-1207-8852 tify climate variation and change: An example for 1901–2 010. Stuart J. Davies https://orcid.org/0000-0002-8596-7522 Environmental Development, 6, 69– 79. https://doi.org/10.1016/j. Kristina J. Anderson- Teixeira https://orcid. envdev.2013.03.007 Chojnacky, D. C., Heath, L. S., & Jenkins, J. C. (2014). Updated general- org/0000-0001-8461-9713 ized biomass equations for North American tree species. Forestry, 87(1), 129– 151. https://doi.org/10.1093/forest ry/cpt053 R E FE R E N C E S Condit, R. (1998). Tropical Forest Census Plots. Methods and Results from Anderson- Teixeira, K. J., Davies, S. J., Bennett, A. C., Gonzalez-A kre, E. Barro Colorado Island, Panama and a Comparison with Other Plots. B., Muller- Landau, H. C., Joseph Wright, S., Abu Salim, K., Almeyda Springer. https://doi.org/10.1007/978- 3-6 62- 03664 -8 8  |   M ethods in Ecology and Evolu on GONZALEZ- AKRE Et AL. Conti, G., Gorné, L. D., Zeballos, S. R., Lipoma, M. L., Gatica, G., Kowaljow, Friedlingstein, P., Cox, P., Betts, R., Bopp, L., von Bloh, W., Brovkin, V., E., Whitworth-H ulse, J. I., Cuchietti, A., Poca, M., Pestoni, S., & Cadule, P., Doney, S., Eby, M., Fung, I., Bala, G., John, J., Jones, C., Fernandes, P. M. (2019). Developing allometric models to predict the Joos, F., Kato, T., Kawamiya, M., Knorr, W., Lindsay, K., Matthews, individual aboveground biomass of shrubs worldwide. Global Ecology H. D., … Zeng, N. (2006). Climate– carbon cycle feedback analysis: and Biogeography, 28(7), 961–9 75. https://doi.org/10.1111/geb.12907 Results from the C4MIP model intercomparison. Journal of Climate, Cushman, K. C., Muller- Landau, H. C., Condit, R. S., & Hubbell, S. P. 19(14), 3337– 3353. https://doi.org/10.1175/jcli38 00.1 (2014). Improving estimates of biomass change in buttressed trees Goodman, R. C., Phillips, O. L., & Baker, T. R. (2014). The importance of crown using tree taper models. Methods in Ecology and Evolution, 5(6), 573– dimensions to improve tropical tree biomass estimates. Ecological 582. https://doi.org/10.1111/2041-2 10X.12187 Applications, 24(4), 680– 698. https://www.jstor.org/stabl e/24432182 Daba, D. E., & Soromessa, T. (2019). The accuracy of species- specific Gower, S. T., Vogt, K. A., & Grier, C. C. (1992). Carbon dynamics of allometric equations for estimating aboveground biomass in trop- rocky mountain Douglas- Fir: Influence of water and nutrient ical moist montane forests: Case Study of albizia grandibracteata availability. Ecological Monographs, 62(1), 43– 65. https://doi. and trichilia dregeana. Carbon Balance and Management, 14(1), 18. org/10.2307/2937170 https://doi.org/10.1186/s1302 1-0 19- 0134- 8 Grassi, G., House, J., Dentener, F., Federici, S., den Elzen, M., & Penman, Davies, S. J., Abiem, I., Abu Salim, K., Aguilar, S., Allen, D., Alonso, A., J. (2017). The key role of forests in meeting climate targets requires Anderson- Teixeira, K., Andrade, A., Arellano, G., Ashton, P. S., science for credible mitigation. Nature Climate Change, 7(3), 220– Baker, P. J., Baker, M. E., Baltzer, J. L., Basset, Y., Bissiengou, P., 226. https://doi.org/10.1038/nclim ate3227 Bohlman, S., Bourg, N. A., Brockelman, W. Y., Bunyavejchewin, Griscom, B. W., Adams, J., Ellis, P. W., Houghton, R. A., Lomax, G., Miteva, S., … Zuleta, D. (2021). ForestGEO: Understanding forest di- D. A., Schlesinger, W. H., Shoch, D., Siikamäki, J. V., Smith, P., versity and dynamics through a global observatory network. Woodbury, P., Zganjar, C., Blackman, A., Campari, J., Conant, R. T., Biological Conservation, 253, 108907. https://doi.org/10.1016/j. Delgado, C., Elias, P., Gopalakrishna, T., Hamsik, M. R., … Fargione, biocon.2020.108907 J. (2017). Natural climate solutions. Proceedings of the National Duncanson, L. I., Dubayah, R. O., & Enquist, B. J. (2015). Assessing the Academy of Sciences of the United States of America, 114(44), 11645– general patterns of forest structure: Quantifying tree and forest al- 11650. https://doi.org/10.1073/pnas.171046 5114 lometric scaling relationships in the United States. Global Ecology and Harmon, M. E., Woodall, C. W., Fasth, B., Sexton, J., & Yatkov, M. (2011). Biogeography, 24(12), 1465–1 475. https://doi.org/10.1111/geb.12371 Differences between standing and downed dead tree wood density re- Duncanson, L. I., Rourke, O., & Dubayah, R. (2015). Small sample sizes duction factors: A comparison across decay classes and tree species. yield biased allometric equations in temperate forests. Scientific U.S. Department of Agriculture, Forest Service, Northern Research Reports, 5(1), 17153. https://doi.org/10.1038/srep17 153 Station. 40 p. 15: 1–4 0. https://doi.org/10.2737/NRS- RP- 15 Falster, D. S., Duursma, R. A., Ishihara, M. I., Barneche, D. R., FitzJohn, R. Helcoski, R., Tepley, A. J., Pederson, N., McGarvey, J. C., Meakem, V., G., Vårhammar, A., Aiba, M., Ando, M., Anten, N., Aspinwall, M. J., Herrmann, V., Thompson, J. R., & Anderson- Teixeira, K. J. (2019). Baltzer, J. L., Baraloto, C., Battaglia, M., Battles, J. J., Bond- Lamberty, Growing season moisture drives interannual variation in woody B., van Breugel, M., Camac, J., Claveau, Y., Coll, L., … York, R. A. productivity of a temperate deciduous forest. New Phytologist, (2015). BAAD: A Biomass And Allometry Database for woody plants. 223(3), 1204– 1216. https://doi.org/10.1111/nph.15906 Ecology, 96(5), 1445–1 445. https://doi.org/10.1890/14-1 889.1 Henry, M., Bombelli, A., Trotta, C., Alessandrini, A., Birigazzi, L., Sola, G., Fatemi, F. R., Yanai, R. D., Hamburg, S. P., Vadeboncoeur, M. A., Arthur, Vieilledent, G., Santenoise, P., Longuetaud, F., Valentini, R., Picard, M. A., Briggs, R. D., & Levine, C. R. (2011). Allometric equations N., & Saint-A ndré, L. (2013). GlobAllomeTree: international plat- for young northern hardwoods: The importance of age-s pecific form for tree allometric equations to support volume, biomass and equations for estimating aboveground biomass. Canadian Journal of carbon assessment. iForest - Biogeosciences and Forestry, 6(6), 326– Forest Research, 41(4), 881–8 91. https://doi.org/10.1139/x10- 248 330. https://doi.org/10.3832/ifor09 01-0 06 Feldpausch, T. R., Banin, L., Phillips, O. L., Baker, T. R., Lewis, S. L., Houghton, R. A. (2005). Aboveground forest biomass and the global Quesada, C. A., Affum-B affoe, K., Arets, E. J. M. M., Berry, N. J., carbon balance. Global Change Biology, 11(6), 945–9 58. https://doi. Bird, M., Brondizio, E. S., de Camargo, P., Chave, J., Djagbletey, org/10.1111/j.1365- 2486.2005.00955.x G., Domingues, T. F., Drescher, M., Fearnside, P. M., França, M. Houghton, R. A. (2008). Biomass. In S. E. Jørgensen & B. D. Fath (Eds.), B., Fyllas, N. M., … Lloyd, J. (2011). Height-d iameter allometry of Encyclopedia of ecology (pp. 448– 453). Academic Press. https://doi. tropical forest trees. Biogeosciences, 8(5), 1081–1 106. https://doi. org/10.1016/B978- 008045 405- 4.00462 - 6 org/10.5194/bg- 8-1 081- 2011 Janik, D., Kral, K., Adam, D., Vrska, T., & Lutz, J. A. (2017). ForestGEO Feldpausch, T. R., Lloyd, J., Lewis, S. L., Brienen, R. J. W., Gloor, M., dead wood census protocol. Digital Commons at Utah State Monteagudo Mendoza, A., Lopez- Gonzalez, G., Banin, L., Abu University. https://doi.org/10.26078/ vcdr- y089 Salim, K., Affum- Baffoe, K., Alexiades, M., Almeida, S., Amaral, I., Jansen, J., Sevenster, J., & Faber, P. J. (1996). Opbrengst Tabellen Voor Andrade, A., Aragão, L. E. O. C., Araujo Murakami, A., Arets, E. J. M. Belangrijke Boomsoorten in Nederland. IBN rapport nr. 221 tevens M., Arroyo, L., Aymard C., G. A., … Phillips, O. L. (2012). Tree height verschenen als: Hinkeloord Reports No. 17. ISSN: 0928– 6888. integrated into pantropical forest biomass estimates. Biogeosciences, Jenkins, J. C., Chojnacky, D. C., Heath, L. S., & Birdsey, R. A. (2004). 9(8), 3381– 3403. https://doi.org/10.5194/bg- 9-3 381- 2012 Comprehensive database of diameter- based biomass regressions for Fenn, K., Malhi, Y., Morecroft, M., Lloyd, C., & Thomas, M. (2015). The North American tree species (p. 45). Gen. Tech. Rep. NE- 319. U.S. carbon cycle of a maritime ancient temperate broadleaved wood- Department of Agriculture, Forest Service, Northeastern Research land at seasonal and annual scales. Ecosystems, 18(1), 1– 15. https:// Station. https://doi.org/10.2737/NE-G TR- 319 doi.org/10.1007/s10021 - 014- 9793- 1 Knapp, N., Fischer, R., Cazcarra- Bes, V., & Huth, A. (2020). Structure Forrester, D. I., Tachauer, I. H. H., Annighoefer, P., Barbeito, I., metrics to generalize biomass estimation from lidar across forest Pretzsch, H., Ruiz-P einado, R., Stark, H., Vacchiano, G., Zlatanov, types from different continents. Remote Sensing of Environment, T., Chakraborty, T., Saha, S., & Sileshi, G. W. (2017). Generalized 237, 111597. https://doi.org/10.1016/j.rse.2019.111597 biomass and leaf area allometric equations for European tree spe- Köppen, W. (2011). The thermal zones of the earth according to the du- cies incorporating stand structure, tree age and climate. Forest ration of hot, moderate and cold periods and to the impact of heat Ecology and Management, 396, 160– 175. https://doi.org/10.1016/j. on the organic world. Meteorologische Zeitschrift, 20(3), 351– 360. foreco.2017.04.011 https://doi.org/10.1127/0941- 2948/2011/105 GONZALEZ- AKRE Et AL. Methods in Ecology and Evolu  on    |  9 Lines, E. R., Zavala, M. A., Purves, D. W., & Coomes, D. A. (2012). Predictable its uncertainty in tropical forests. Methods in Ecology and Evolution, changes in aboveground allometry of trees along gradients of tempera- 8(9), 1163– 1167. https://doi.org/10.1111/2041- 210x.12753 ture, aridity and competition. Global Ecology and Biogeography, 21(10), Rojas- García, F., De Jong, B. H. J., Martínez-Z urimendí, P., & Paz-P ellat, F. 1017– 1028. https://doi.org/10.1111/j.1466- 8238.2011.00746.x (2015). Database of 478 allometric equations to estimate biomass Luo, Y., Wang, X., & Ouyang, Z. (2018). A China’s normalized tree biomass equa- for Mexican trees and forests. Annals of Forest Science, 72(6), 835– tion dataset. PANGAEA. https://doi.org/10.1594/PANGA EA.895244 864. https://doi.org/10.1007/s13595 - 015- 0456-y Luo, Y., Wang, X., Ouyang, Z., Lu, F., Feng, L., & Tao, J. (2020). A review Roxburgh, S. H., Paul, K. I., Clifford, D., Englana, J. R., & Raison, R. J. of biomass equations for China's tree species. Earth System Science (2015). Guidelines for constructing allometric models for the Data, 12(1), 21– 40. https://doi.org/10.5194/essd- 12- 21-2 020 prediction of woody biomass: How many individuals to harvest? Lutz, J. A. (2005). The contribution of mortality to early coniferous forest Ecosphere, 6(3), 38. https://doi.org/10.1890/ES14-0 0251.1 development. {PhD Thesis}, University of Washington. Thesis. 95 p. Rutishauser, E., Noor’an, F., Laumonier, Y., Halperin, J., Rufi’ie, Lutz, J. A., Furniss, T. J., Johnson, D. J., Davies, S. J., Allen, D., Alonso, Hergoualc’h, K., & Verchot, L. (2013). Generic allometric models A., Anderson- Teixeira, K. J., Andrade, A., Baltzer, J., Becker, K. M. including height best estimate forest biomass and carbon stocks in L., Blomdahl, E. M., Bourg, N. A., Bunyavejchewin, S., Burslem, D. Indonesia. Forest Ecology and Management, 307, 219– 225. https:// F. R. P., Cansler, C. A., Cao, K. E., Cao, M., Cárdenas, D., Chang, doi.org/10.1016/j.foreco.2013.07.013 L.- W., … Kerkhoff, A. (2018). Global importance of large-d iameter Schepaschenko, D., Shvidenko, A., Usoltsev, V., Lakyda, P., Luo, Y., trees. Global Ecology and Biogeography, 27(7), 849–8 64. https://doi. Vasylyshyn, R., Lakyda, I., Myklush, Y., See, L., McCallum, I., Fritz, org/10.1111/geb.12747 S., Kraxner, F., & Obersteiner, M. (2017). A dataset of forest bio- Lutz, J. A., Matchett, J. R., Tarnay, L. W., Smith, D. F., Becker, K. M. L., mass structure for Eurasia. Scientific Data, 4(1), 170070. https://doi. Furniss, T. J., & Brooks, M. L. (2017). Fire and the distribution and org/10.1038/sdata.2017.70 uncertainty of carbon sequestered as aboveground tree biomass in Somogyi, Z., Cienciala, E., Mäkipää, R., Muukkonen, P., Lehtonen, A., & Yosemite and Sequoia & Kings Canyon National Parks. Land, 6(1), Weiss, P. (2007). Indirect methods of large- scale forest biomass 10. https://doi.org/10.3390/land6 010010 estimation. European Journal of Forest Research, 126(2), 197– 207. Lutz, J. A., Schwindt, K. A., Furniss, T. J., Freund, J. A., Swanson, M. E., https://doi.org/10.1007/s1034 2- 006-0 125- 7 Hogan, K. I., Kenagy, G. E., & Larson, A. J. (2014). Community com- Stovall, A. E. L., Anderson- Teixeira, K. J., & Shugart, H. H. (2018). position and allometry of leucothoe Davisiae, Cornus Sericea, and Assessing terrestrial laser scanning for developing non- destructive Chrysolepis sempervirens. Canadian Journal of Forest Research, biomass allometry. Forest Ecology and Management, 427, 217–2 29. 44(6), 677–6 83. https://doi.org/10.1139/cjfr- 2013-0 524 https://doi.org/10.1016/j.foreco.2018.06.004 Meakem, V., Tepley, A. J., Gonzalez-A kre, E. B., Herrmann, V., Muller- Sullivan, M. J. P., Lewis, S. L., Hubau, W., Qie, L., Baker, T. R., Banin, L. Landau, H. C., Joseph Wright, S., Hubbell, S. P., Condit, R., & F., Chave, J., Cuni- Sanchez, A., Feldpausch, T. R., Lopez- Gonzalez, Anderson- Teixeira, K. J. (2018). Role of tree size in moist tropical G., Arets, E., Ashton, P., Bastin, J.-F ., Berry, N. J., Bogaert, J., Boot, forest carbon cycling and water deficit responses. New Phytologist, R., Brearley, F. Q., Brienen, R., Burslem, D. F. R. P., … Phillips, O. L. 219(3), 947–9 58. https://doi.org/10.1111/nph.14633 (2018). Field methods for sampling tree height for tropical forest Muukkonen, P. (2007). Generalized allometric volume and biomass equations biomass estimation. Methods in Ecology and Evolution, 9(5), 1179– for some tree species in Europe. European Journal of Forest Research, 1189. https://doi.org/10.1111/2041- 210X.12962 126(2), 157– 166. https://doi.org/10.1007/s1034 2- 007- 0168- 4 van Breugel, M., Ransijn, J., Craven, D., Bongers, F., & Hall, J. S. (2011). Návar, J. (2009). Biomass component equations for Latin American spe- Estimating carbon stock in secondary forests: Decisions and uncer- cies and groups of species. Annals of Forest Science, 66(2), 208–2 08. tainties associated with allometric biomass models. Forest Ecology https://doi.org/10.1051/fores t/2009001 and Management, 262(8), 1648– 1657. https://doi.org/10.1016/j. Ngomanda, A., Engone Obiang, N. L., Lebamba, J., Moundounga foreco.2011.07.018 Mavouroulou, Q., Gomat, H., Mankou, G., Loumeto, J., Midoko Zolkos, S. G., Goetz, S. J., & Dubayah, R. (2013). A meta- analysis of ter- Iponga, D., Kossi Ditsouga, F., Zinga Koumba, R., Botsika Bobé, K. restrial aboveground biomass estimation using lidar remote sens- H., Mikala Okouyi, C., Nyangadouma, R., Lépengué, N., Mbatchi, ing. Remote Sensing of Environment, 128, 289–2 98. https://doi. B., & Picard, N. (2014). Site- specific versus pantropical allometric org/10.1016/j.rse.2012.10.017 equations: Which option to estimate the biomass of a moist central African forest? Forest Ecology and Management, 312, 1– 9. https:// doi.org/10.1016/j.foreco.2013.10.029 SUPPORTING INFORMATION Paul, K. I., Roxburgh, S. H., Chave, J., England, J. R., Zerihun, A., Specht, Additional supporting information may be found in the online ver- A., Lewis, T., Bennett, L. T., Baker, T. G., Adams, M. A., Huxtable, sion of the article at the publisher’s website. D., Montagu, K. D., Falster, D. S., Feller, M., Sochacki, S., Ritson, P., Bastin, G., Bartle, J., Wildy, D., … Sinclair, J. (2016). Testing the gen- erality of above-g round biomass allometry across plant functional types at the continent scale. Global Change Biology, 22(6), 2106– How to cite this article: Gonzalez-A kre, E., Piponiot, C., 2124. https://doi.org/10.1111/gcb.13201 Lepore, M., Herrmann, V., Lutz, J. A., Baltzer, J. L., Dick, C. Poorter, H., Jagodzinski, A. M., Ruiz- Peinado, R., Kuyah, S., Luo, Y., Oleksyn, W., Gilbert, G. S., He, F., Heym, M., Huerta, A. I., Jansen, P. A., J., Usoltsev, V. A., Buckley, T. N., Reich, P. B., & Sack, L. (2015). How does biomass distribution change with size and differ among spe- Johnson, D. J., Knapp, N., Král, K., Lin, D., Malhi, Y., cies? An analysis for 1200 plant species from five continents. New McMahon, S. M., Myers, J. A., … Anderson- Teixeira, K. J. Phytologist, 208(3), 736–7 49. https://doi.org/10.1111/nph.13571 (2021). allodb: An R package for biomass estimation at R Core Team. (2018). R: A language and environment for statistical comput- globally distributed extratropical forest plots. Methods in ing (version 3.6.3). R Foundation for Statistical Computing. http:// www.R- proje ct.org/ Ecology and Evolution, 00, 1–9 . https://doi.org/10.1111/2041- Réjou- Méchain, M., Tanguy, A., Piponiot, C., Chave, J., & Hérault, B. (2017). 210X.13756 biomass : An R package for estimating above- ground biomass and