Search

Managing forests for climate change mitigation requires action by diverse stakeholders undertaking different activities with overlapping objectives and spatial impacts. To date, several forest carbon monitoring systems have been developed for different regions using various data, methods and assumptions, making it difficult to evaluate mitigation performance consistently across scales. Here, we integrate ground and Earth observation data to map annual forest-related greenhouse gas emissions and removals globally at a spatial resolution of 30 m over the years 2001–2019. We estimate that global forests were a net carbon sink of −7.6 ± 49 GtCO2e yr−1, reflecting a balance between gross carbon removals (−15.6 ± 49 GtCO2e yr−1) and gross emissions from deforestation and other disturbances (8.1 ± 2.5 GtCO2e yr−1). The geospatial monitoring framework introduced here supports climate policy development by promoting alignment and transparency in setting priorities and tracking collective progress towards forest-specific climate mitigation goals with both local detail and global consistency.

Management of forest carbon stocks on public lands is critical to maintaining or enhancing carbon dioxide removal from the atmosphere. Acknowledging this, an array of federal regulations and policies have emerged that requires US National Forests to report baseline carbon stocks and changes due to disturbance and management and assess how management activities and forest plans affect carbon stocks. To address these requirements with the best-available science, we compiled empirical and remotely sensed data covering the National Forests (one fifth of the area of US forest land) and analyzed this information using a carbon modeling framework. We demonstrate how integration of various data and models provides a comprehensive evaluation of key drivers of observed carbon trends, for individual National Forests. The models in this framework complement each other with different strengths: the Carbon Calculation Tool uses inventory data to report baseline carbon stocks; the Forest Carbon Management Framework integrates inventory data, disturbance histories, and growth and yield trajectories to report relative effects of disturbances on carbon stocks; and the Integrated Terrestrial Ecosystem Carbon Model incorporates disturbance, climate, and atmospheric data to determine their relative impacts on forest carbon accumulation and loss. We report results for several National Forests across the USA and compare their carbon dynamics. Results show that recent disturbances are causing some forests to transition from carbon sinks to sources, particularly in the West. Meanwhile, elevated atmospheric carbon dioxide and nitrogen deposition are consistently increasing carbon stocks, partially offsetting declines due to disturbances and aging. Climate variability introduces concomitant interannual variability in net carbon uptake or release. Targeting forest disturbance and post-disturbance regrowth is critical to management objectives that involve maintaining or enhancing future carbon sequestration.

There is an increasing need for approaches to determine reference emission levels and implement policies to address the objectives of Reducing Emissions from Deforestation and Forest Degradation, plus improving forest management, carbon stock enhancement and conservation (REDD+). Important aspects of approaching emissions reductions include coordination and sharing of technology, data, protocols and experiences within and among countries to maximize resources and apply knowledge to build robust monitoring, reporting and verification (MRV) systems. We propose that enhancing the multiple facets of interoperability could facilitate implementation of REDD+ programs and actions. For this case, interoperability is a collective effort with the ultimate goal of sharing and using information to produce knowledge and apply knowledge gained, by removing conceptual, technological, organizational and cultural barriers. These efforts must come from various actors and institutions, including government ministrie...

Assessing carbon (C) sequestration in forest ecosystems is fundamental to supply information to monitoring, reporting and verification (MRV) for reducing deforestation and forest degradation (REDD). The spatially-explicit version of Forest-DNDC (FDNDC) was evaluated using plot-based observations from Nez Perce-Clearwater National Forest (NPCNF) in Idaho of United States and used to assess C stocks in about 16,000 km2. The model evaluation indicated that the FDNDC can be used to assess C stocks with disturbances in this temperate forest with a proper model performance efficiency and small error between observations and simulations. Aboveground biomass in this forest was 85.1 Mg C ha-1 in 2010. The mean aboveground biomass in the forest increased by about 0.6 Mg C ha-1 yr-1 in the last 20 years from 1990 to 2010 with spatial mean stand age about 98 years old in 2010. Spatial differences in distributions of biomass, net primary production and net ecosystem product are substantial. The spatial divergence in C sequestration is mainly associated with the spatial disparities in stand age due to disturbances, secondly with ecological drivers and species. Climate variability and change can substantially impact C stocks in the forest based on the climatic variability of spatial climate data for a 33-year period from 1981 to 2013. Temperature rise can produce more biomass in NPCNF, but biomass cannot increase with an increase in precipitation in this forest. The simulation with disturbances using observations and estimates for the time period from 1991 to 2011 showed the effects of disturbances on C stocks in forests. The impacts of fires and insects on C stocks in this forest are highly dependent on the severity, the higher, the more C loss to atmosphere due to fires, and the more dead woods produced by fires and insects. The rates of biomass increase with an increase in stand age are different among the species. The changes in forest C stocks in the forest are almost species specific, non-linear and complex. The increase in aboveground biomass with an increase in stand age can be described by a high-order polynomial.

This report assesses how carbon stocks at regional scales and in individual national forests are affected by factors such as timber harvesting, natural disturbances, climate variability, increasing atmospheric carbon dioxide concentrations, and nitrogen deposition. Previous baseline assessments of carbon stocks (https://www.fs.fed.us/managing-land/sc/carbon) evaluated observed trends based on forest inventory data but were limited in ability to reveal detailed causes of these trends. The expanded assessments reported here are based on an extensive disturbance and climate history for each national forest, and two forest carbon models, to estimate the relative impacts of disturbance (e.g., fires, harvests, insect outbreaks, disease) and nondisturbance factors (climate, carbon dioxide concentration, nitrogen deposition). Results are summarized for each region of the National Forest System in the main document. A set of regional appendices to this report provides more detailed information about individual national forests within each region. Results are highly variable across the United States. Generally, carbon stocks are increasing in forests of the eastern United States as these forests continue to recover and grow older after higher historical harvesting rates and periods of nonforest land use. In contrast, carbon stocks in forests of the western United States may be either increasing or decreasing, depending on recent effects of natural disturbances and climate change. The information supports national forest units in assessing carbon stocks, quantifying carbon outcomes of broad forest management strategies and planning, and meeting carbon assessment requirements of the 2012 Planning Rule and directives. Results of these expanded assessments will provide context for project-level decisions, separated from the effects of factors that are beyond land managers’ control.

Currently, US forests constitute a large carbon sink, comprising about 9 % of the global terrestrial carbon sink. Wildfire is the most significant disturbance influencing carbon dynamics in US forests. Our objective is to estimate impacts of climate change, CO2 concentration, and nitrogen deposition on the future net biome productivity (NBP) of US forests until the end of twenty-first century under a range of disturbance conditions. We designate three forest disturbance scenarios under one future climate scenario to evaluate factor impacts for the future period (2011–2100): (1) no wildfires occur but forests continue to age (Saging), (2) no wildfires occur and forest ages are fixed in 2010 (Sfixed_nodis), and (3) wildfires occur according to a historical pattern, consequently changing forest age (Sdis_age_change). Results indicate that US forests remain a large carbon sink in the late twenty-first century under the Sfixed_nodis scenario; however, they become a carbon source under the Saging and Sdis_age_change scenarios. During the period of 2011 to 2100, climate is projected to have a small direct effect on NBP, while atmospheric CO2 concentration and nitrogen deposition have large positive effects on NBP regardless of the future climate and disturbance scenarios. Meanwhile, responses to past disturbances under the Sfixed_nodis scenario increase NBP regardless of the future climate scenarios. Although disturbance effects on NBP under the Saging and Sdis_age_change scenarios decrease with time, both scenarios experience an increase in NBP prior to the 2050s and then a decrease in NBP until the end of the twenty-first century. This study indicates that there is potential to increase or at least maintain the carbon sink of conterminous US forests at the current level if future wildfires are reduced and age structures are maintained at a productive mix. The effects of CO2 on the future carbon sink may overwhelm effects of other factors at the end of the twenty-first century. Although our model in conjunction with multiple disturbance scenarios may not reflect the true conditions of future forests, it provides a range of potential conditions as well as a useful guide to both current and future forest carbon management.

Quantifying forest biomass carbon (C) stock change is important for understanding forest dynamics and their feedbacks with climate change. Forests in the northeastern U.S. have been a net carbon sink in recent decades, but C accumulation in some northern hardwood forests has been halted due to the impact of emerging stresses such as invasive pests, land use change and climate change. The Delaware River Basin (DRB), sited in the southern edge of the northern hardwood forest, features diverse forest types and land-use histories. In 2001–2003, the DRB Monitoring and Research Initiative established 61 forest plots in three research sites, using Forest Service inventory protocols and enhanced measurements. These plots were revisited and re-measured in 2012–2014. By comparing forest biomass C stocks in the two measurements, our results suggest that the biomass C stock of the DRB forest increased, and was thus a carbon sink over the past decade. The net biomass C stock change in the Neversink area in the north of the DRB was 1.94 Mg C ha−1 yr−1, smaller than the biomass C change in the French Creek area (2.52 Mg C ha−1 yr−1, southern DRB), and Delaware Water Gap Area (2.68 Mg C ha−1 yr−1, central DRB). An increase of dead biomass C accounted for 20% of the total biomass C change. The change of biomass C stocks did not correlate with any climatic or topographic factors, but decreased with increasing stand age, and with tree mortality rate. Mortality rates were highest in the smallest size class. In most of the major tree species, stem density decreased, but the loss of biomass from mortality was offset by recruitment and growth. The demographic changes differ dramatically among species. The living biomass of chestnut oak, white oak and black oak decreased because of the large mortality rate, while white pine, American beech and sweet birch increased in both biomass and stem density. Our results suggest that forests in the DRB could continue to be a carbon sink in the coming decades, because they are likely at a middle successional stage. The linkage between demography of individual trees species and biomass C change underscores the effects of species-specific disturbances such as non-native insects and pathogens on forest dynamics, and highlights the need for forest managers to anticipate these effects in their management plans.

Assessing the effect of climate change on carbon sequestration in tropical forest ecosystems is important to inform monitoring, reporting, and verification (MRV) for reducing deforestation and forest degradation (REDD), and to effectively assess forest management options under climate change. Two process-based models, Forest-DNDC and Biome-BGC, with different spatial modeling scales were evaluated to estimate the potential effect of climate change on carbon sequestration in a tropical dry semi-deciduous forest in the Yucatan Peninsula of Mexico. The results from the simulations using the two models show that carbon sequestration in this dry forest is highly sensitive to warming. Carbon uptake in this forest may increase or decrease slightly with a corresponding increase or decrease in precipitation; however, with an increase in temperature, carbon uptake may decrease significantly, showing that warming may be the main climate factor that impacts carbon storage in this tropical dry forest. Model performance evaluation indicates that both models may be used to estimate C stocks, but DNDC may be better than BGC for assessing the effect of climate change on C dynamics.

Large trees are important and unique organisms in forests, providing ecosystem services including carbon dioxide removal from the atmosphere and long-term storage. Some reports have raised concerns about the global decline of large trees. Based on observations from two regions in Finland and three regions in the United States we report that trends of large trees during recent decades have been surprisingly variable among regions. In southern Finland, the growing stock volume of trees larger than 30 cm at breast height increased nearly five-fold during the second half of the 20th century, yet more recently ceased to expand. In the United States, large hardwood trees have become increasingly common in the Northeast since the 1950s, while large softwood trees declined until the mid 1990s as a consequence of harvests in the Pacific region, and then rebounded when harvesting there was reduced. We conclude that in the regions studied, the history of land use and forest management governs changes of the diameter-class distributions of tree populations. Large trees have significant benefits; for example, they can constitute a large proportion of the carbon stock and affect greatly the carbon density of forests. Large trees usually have deeper roots and long lifetimes. They affect forest structure and function and provide habitats for other species. An accumulating stock of large trees in existing forests may have negligible direct biophysical effects on climate through transpiration or forest albedo. Understanding changes in the demography of tree populations makes a contribution to estimating the past impact and future potential of forests in the global carbon budget and to assessing other ecosystem services of forests.

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 &plusmn, 17 Mg ha&minus, 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.

Many studies and production inventory systems have shown the utility of coupling covariates derived from Light Detection and Ranging (LiDAR) data with forest variables measured on georeferenced inventory plots through regression models. The objective of this study was to propose and assess the use of a Bayesian hierarchical modeling framework that accommodates both residual spatial dependence and non-stationarity of model covariates through the introduction of spatial random effects. We explored this objective using four forest inventory datasets that are part of the North American Carbon Program, each comprising point-referenced measures of above-ground forest biomass and discrete LiDAR. For each dataset, we considered at least five regression model specifications of varying complexity. Models were assessed based on goodness of fit criteria and predictive performance using a 10-fold cross-validation procedure. Results showed that the addition of spatial random effects to the regression model intercept improved fit and predictive performance in the presence of substantial residual spatial dependence. Additionally, in some cases, allowing either some or all regression slope parameters to vary spatially, via the addition of spatial random effects, further improved model fit and predictive performance. In other instances, models showed improved fit but decreased predictive performance—indicating over-fitting and underscoring the need for cross-validation to assess predictive ability. The proposed Bayesian modeling framework provided access to pixel-level posterior predictive distributions that were useful for uncertainty mapping, diagnosing spatial extrapolation issues, revealing missing model covariates, and discovering locally significant parameters.

Global forests are increasingly affected by land-use change, fragmentation, changing management objectives, and degradation. In this paper we broadly characterize trends in global forest area by intensity of management, and provide an overview of changes in global carbon stocks associated with managed forests. We discuss different interpretations of “management” and highlight some important accounting and analysis issues. The area of global forests has declined by 3% since 1990 but the area of planted forest has increased in all regions of the world and now accounts for almost 7% of global forest land. The area of primary forest, which is typically defined as lacking direct human influence, is about 34% of all forest land according to country reports, but the area is declining especially in South America and Africa because of human-caused habitat fragmentation and degradation. Concurrently, the area of naturally regenerated forest that is not classified as primary forest has declined. As a result of increasing management intensity, the area of unmanaged forest, typically defined as land lacking protected status or a management plan, dropped significantly since 1990 and now comprises only 21% of global forests. There have been significant increases in areas of forest used for non-wood forest products such as protection of soil and water, conservation of biodiversity, and provision of social services. Globally, timber production has been relatively stable since 1990, but increasing areas of forest used for non-wood forest products indicates that harvesting is taking place on a smaller proportion of the total forest area. Based on trends in the area of managed forest and regional studies, it is clear that historical and current forest management has been a very significant determining factor of current carbon stocks. Established forests currently offset about 30% of global emissions of CO 2 from fossil fuel use, and there are mitigation opportunities involving forests that could increase the gross terrestrial C uptake from roughly 4.0 to 6.2 Pg C annually. However, our results suggest that a diversifying use of forest land may have significant consequences for maintaining or increasing the current rate of terrestrial carbon sequestration. In the future, indirect human influences such as increasing atmospheric CO 2 and climate change, along with the direct effects of land management and projected increasing demand for wood biofuel, are likely to become increasingly important elements that influence land management strategies and the role of forests in the global carbon cycle.

Questions Is the accuracy of predictions of above-ground biomass (AGB) and plant species richness of tropical dry forests from LIDAR data compromised during leaf-off canopy period, when most of the vegetation is leafless, compared to the leaf-on period? How does topographic position affect prediction accuracy of AGB for leaf-off and leaf-on canopy conditions? Location Tropical dry forest, Yucatan Peninsula, Mexico. Methods We evaluated the accuracy of predictions using both leaf-on and leaf-off LIDAR estimates of biomass and species richness, and assessed the adequacy of both LIDAR data sets for characterizing these vegetation attributes in tropical dry forests using multiple regression analysis and ANOVA. The performance of the models was assessed by leave-one-out cross-validation. We also investigated differences in vegetation structure between two topographic conditions using PCA and ANOSIM. Finally, we evaluated the influence of topography on the accuracy of biomass estimates from LIDAR using multiple regression analysis and ANOVA. Results A higher overall accuracy was obtained with leaf-on vs leaf-off conditions for AGB (root mean square error (RMSE) = 21.6 vs 25.7 ton·ha−1), as well as for species richness (RMSE = 5.5 vs 5.8 species, respectively). However, no significant differences in mean dissimilarities between biomass estimates from LIDAR and in situ biomass estimates comparing the two canopy conditions were found (F1,39 = 0.03, P = 0.87). In addition, no significant differences in dissimilarities of AGB estimation were found between flat and hilly areas (F1,39 = 1.36, P = 0.25). Conclusions Our results suggest that estimates of species richness and AGB from LIDAR are not significantly influenced by canopy conditions or slope, indicating that both leaf-on and leaf-off models are appropriate for these variables regardless of topographic position in these tropical dry forests.

Stand age and disturbance data have become more available in recent years and can facilitate modeling studies that integrate and quantify effects of disturbance and nondisturbance factors on carbon dynamics. Since high-quality disturbance and forest age data to support forest dynamic modeling are lacking before 1950, we assumed dynamic equilibrium (carbon neutrality) for the starting conditions of forests with unknown historical disturbance and forest age information. The impacts of this assumption on forest carbon cycle estimation for recent decades have not been systematically examined. In this study, we tested an assumption of disequilibrium conditions for forests with unknown disturbance and age data by randomly assigning ages to them in the model initial year (1900) and analyzed uncertainties for 1951–2010 carbon dynamic simulations compared with the equilibrium assumption. Results show that with the dynamic equilibrium assumption, the total net biome productivity (NBP) of conterminous U.S. forests was 188 ± 60 Tg C yr−1 with 185 ± 56 Tg C yr−1 in living biomass and 3 ± 23 Tg C yr−1 in soil. The C release due to disturbance on average was about 68 ± 55 Tg C yr−1. The disequilibrium assumption causes annual NBP from 1951 to 2010 in conterminous U.S. forests to vary by an average of 13% with the largest impact on the soil carbon component. Uncertainties related to nondisturbance factors have relatively small impacts on NBP estimation (within 10%), while uncertainties related to disturbances cause biases in NBP of 4 to 28%. We conclude that the dynamic equilibrium assumption for the model initialization in 1900 is acceptable for simulating 1951–2010 forest carbon dynamics as long as disturbance and age data are reliable for this later period, although caution should be taken regarding the prior-1950 simulation results because of their greater uncertainties.

Trees outside forests (TOF) play important roles in national economies, ecosystem services, and international efforts for mitigating climate warming. Detailed assessment of the dynamics of carbon (C) stocks in China's TOF is necessary for fully evaluating the role of the country's trees in the national C cycle. This study is the first to explore the changes in biomass C stocks of China's TOF over the last three decades, using the national forest inventory data in six periods from 1977 to 2008. According to the definition of the forest inventory, China's TOF could be categorized into three groups: woodlands, shrubberies, and trees on non-forest land (including four-side greening trees, defined in the article, and scattered trees). We estimated biomass C stocks of woodlands and trees on non-forest land by using the provincial biomass-volume conversion equations derived from the data of low-canopy forests, and estimated the biomass C stocks of shrubberies using the provincial mean biomass density. Total TOF biomass C stock increased by 62.7% from 823 Tg C (1 Tg = 1012 g) in the initial period of 1977–1981 to 1339 Tg C in the last period of 2004–2008. As a result, China's TOF have accumulated biomass C of 516 Tg during the study period, with 12, 270, and 234 Tg in woodlands, shrubberies, and trees on non-forest land, respectively. The annual biomass C sink of China's TOF averaged 19.1 Tg C yr−1, offsetting 2.1% of the contemporary fossil-fuel CO2 emissions in the country. These estimates are equal to 16.5–20.7% of the contemporary total forest biomass C stock and 27.2% of the total forest biomass C sink in the country, suggesting that TOF are substantial components in China's tree C budget.

We estimated aboveground tree biomass and change in aboveground tree biomass using repeated airborne laser scanner (ALS) acquisitions and temporally coincident ground observations of forest biomass, for a relatively undisturbed period (2004-2007; ?07-04), a contrasting period of disturbance (2007-2009; ?09-07), and an integrated period (2004-2009; ?09-04). A simple randomsampling (SRS) estimator was used to estimate means and variances of biomass and biomass change for each measurement occasion and interval. For each year, linear regression models were used to predict mean total aboveground tree biomass for live, dead, and total biomass components from ALS-derived variables. These models predicted biomass with R2 = 0.68, 0.59, and 0.70 and RMSEs of 32.7, 30.5, and 31.7 Mg ha-1 for 2004, 2007 and 2009, respectively. A model assisted indirect estimatorwas then used to estimate biomass and biomass change for comparison to the field-based SRS estimator. This model assisted indirect approach decreased standard errors of biomass estimation relative to the SRS estimator, but had larger variances for biomass change estimation. Linear regression models were also used to directly predict field-estimated biomass change using ALS ?-variables, calculated as the difference between multi-temporal ALS variables, for the study area. Integrated over the 6 year period, these change models had R2 = 0.81, 0.72, and 0.68 with RMSEs of 2.0, 9.3, and 1.0 Mg ha-1 yr-1 for live, dead, and total aboveground tree biomass, respectively. A model assisted direct estimator reduced standard errors of change estimates by 100-200% compared to the field-based estimates. We discuss several potential advantages and limitations of the direct and indirect approaches. Our primary finding is that model assisted direct estimation of biomass change decreased estimation uncertainty relative to both field and model assisted indirect estimation.

Efforts to test and improve terrestrial biosphere models (TBMs) using a variety of data sources have become increasingly common. Yet, geographically extensive forest inventories have been under-exploited in previous model-data fusion efforts. Inventory observations of forest growth, mortality, and biomass integrate processes across a range of timescales, including slow timescale processes such as species turnover, that are likely to have important effects on ecosystem responses to environmental variation. However, the large number (thousands) of inventory plots precludes detailed measurements at each location, so that uncertainty in climate, soil properties, and other environmental drivers may be large. Errors in driver variables, if ignored, introduce bias into model-data fusion. We estimated errors in climate and soil drivers at U.S. Forest Inventory and Analysis (FIA) plots, and we explored the effects of these errors on model-data fusion with the Geophysical Fluid Dynamics Laboratory LM3V dynamic global vegetation model. When driver errors were ignored or assumed small at FIA plots, responses of biomass production in LM3V to precipitation and soil available water capacity appeared steeper than the corresponding responses estimated from FIA data. These differences became nonsignificant if driver errors at FIA plots were assumed to be large. Ignoring driver errors when optimizing LM3V parameter values yielded estimates for fine-root allocation that were larger than biometric estimates, which is consistent with the expected direction of bias. To explore whether complications posed by driver errors could be circumvented by relying on intensive study sites where driver errors are small, we performed a power analysis. To accurately quantify the response of biomass production to spatial variation in mean annual precipitation within the eastern United States would require at least 40 intensive study sites, which is larger than the number of sites typically available for individual biomes in existing plot networks. Driver errors may be accommodated by several existing model-data fusion approaches, including hierarchical Bayesian methods and ensemble filtering methods; however, these methods are computationally expensive. We propose a new approach, in which the TBM functional response is fit directly to the driver-error-corrected functional response estimated from data, rather than to the raw observations.

a b s t r a c t Spatial and temporal environmental heterogeneity is known to play an important role in the dynam- ics of populations and communities. However, the implications of this heterogeneity for developing and testing regional- to global-scale forest dynamics models are largely unexplored. Predictions from forest dynamics models are typically compared to chronosequences assembled from forest inventory data using the space-for-time substitution approach, which assumes that different-aged stands across space have followed (and will follow) the same dynamics. Often, this assumption is invalid in the pres- ence of spatial and/or temporal heterogeneity. We used the perfect plasticity approximation (PPA) forest dynamics model, parameterized with forest inventory data for > 10 forest types in the eastern U.S., to diagnose spatial and temporal heterogeneity in forest dynamics, and to explore how this heterogeneity can affect comparisons between predicted dynamics and chronosequence observations. Our results pro- vided evidence that spatial and temporal heterogeneity are widespread in eastern U.S. forests. Temporal heterogeneity was apparent because species whose observed abundances have increased over recent decades tended to be the same species for which predicted abundance (derived from individual-level growth and mortality rates estimated from the recent decades of inventory data) were greater than observed abundance. Spatial heterogeneity was apparent because species had more competitive param- eter estimates (higher growth and/or lower mortality) on inventory plots where they are most abundant, relative to other plots in the same forest type. Spatial and temporal heterogeneity both contributed to mismatches between predicted dynamics and chronosequence observations. Predictions of canopy struc- ture (proportion of individuals in the upper canopy vs. the understory) were well-matched to inventory data, suggesting that the PPA's simple space-filling algorithm was not an important source of error in predicting forest dynamics.

NASA Carbon Monitoring System projects NNX12AN07G, NNX14AP12G, and 80NSSC17K0710; National Science Foundation Grant No. DGE1322106

Forests are the dominant terrestrial ecosystem on Earth. We review the environmental factors controlling their structure and global distribution and evaluate their current and future trajectory. Adaptations of trees to climate and resource gradients, coupled with disturbances and forest dynamics, create complex geographical patterns in forest assemblages and structures. These patterns are increasingly discernible through new satellite and airborne observation systems, improved forest inventories, and global ecosystem models. Forest biomass is a complex property affected by forest distribution, structure, and ecological processes. Since at least 1990, biomass density has consistently increased in global established forests, despite increasing mortality in some regions, suggesting that a global driver such as elevated CO2 may be enhancing biomass gains. Global forests have also apparently become more dynamic. Advanced information about the structure, distribution, and biomass of the world's forests provides critical ecological insights and opportunities for sustainable forest management and enhancing forest conservation and ecosystem services.

Reducing emissions from deforestation and forest degradation plus improving forest-management (REDD+) is a mechanism to facilitate tropical countries’ participation in climate change mitigation. In this review we focus on the current state of monitoring systems to support implementing REDD+. The main elements of current monitoring systems – Landsat satellites and traditional forest inventories – will continue to be the backbone of many forest-monitoring systems around the world, but new remote-sensing and analytical approaches are addressing monitoring problems specific to the tropics and implementing REDD+. There is increasing recognition of the utility of combining remote sensing with field data using models that integrate information from many sources, which will continue to evolve as new sensors are deployed and as the availability of field data increases.

Atmospheric inversions have become an important tool in quantifying carbon dioxide (CO2) sinks and sources at a variety of spatiotemporal scales, but associated large uncertainties restrain the inversion research community from reaching agreement on many important subjects. We enhanced an atmospheric inversion of the CO2 flux for North America by introducing spatially explicit information on forest stand age for US and Canada as an additional constraint, since forest carbon dynamics are closely related to time since disturbance. To use stand age information in the inversion, we converted stand age into an age factor, and included the covariances between subcontinental regions in the inversion based on the similarity of the age factors. Our inversion results show that, considering age factors, regions with recently disturbed or old forests are often nudged towards carbon sources, while regions with middle-aged productive forests are shifted towards sinks. This conforms to stand age effects observed in flux networks. At the subcontinental level, our inverted carbon fluxes agree well with continuous estimates of net ecosystem carbon exchange (NEE) upscaled from eddy covariance flux data based on MODIS data. Inverted fluxes with the age constraint exhibit stronger correlation to these upscaled NEE estimates than those inverted without the age constraint. While the carbon flux at the continental and subcontinental scales is predominantly determined by atmospheric CO2 observations, the age constraint is shown to have potential to improve the inversion of the carbon flux distribution among subcontinental regions, especially for regions lacking atmospheric CO2 observations.

In forests, total belowground carbon (C) flux (TBCF) is a large component of the C budget and represents a critical pathway for delivery of plant C to soil. Reducing uncertainty around regional estimates of forest C cycling may be aided by incorporating knowledge of controls over soil respiration and TBCF. Photosynthesis, and presumably TBCF, declines with advancing tree size and age, and photosynthesis increases yet C partitioning to TBCF decreases in response to high soil fertility. We hypothesized that these causal relationships would result in predictable patterns of TBCF, and partitioning of C to TBCF, with natural variability in leaf area index (LAI), soil nitrogen (N), and tree height in subalpine forests in the Rocky Mountains, USA. Using three consecutive years of soil respiration data collected from 22 0.38-ha locations across three 1-km2 subalpine forested landscapes, we tested three hypotheses: (1) annual soil respiration and TBCF will show a hump-shaped relationship with LAI; (2) variability in TBCF unexplained by LAI will be related to soil nitrogen (N); and (3) partitioning of C to TBCF (relative to woody growth) will decline with increasing soil N and tree height. We found partial support for Hypothesis 1 and full support for Hypotheses 2 and 3. TBCF, but not soil respiration, was explained by LAI and soil N patterns (r2 = 0.49), and the ratio of annual TBCF to TBCF plus aboveground net primary productivity (ANPP) was related to soil N and tree height (r2 = 0.72). Thus, forest C partitioning to TBCF can vary even within the same forest type and region, and approaches that assume a constant fraction of TBCF relative to ANPP may be missing some of this variability. These relationships can aid with estimates of forest soil respiration and TBCF across landscapes, using spatially explicit forest data such as national inventories or remotely sensed data products.

We develop an approach for estimating net ecosystem exchange (NEE) using inventory-based information over North America (NA) for a recent 7-year period (ca. 2000–2006). The approach notably retains information on the spatial distribution of NEE, or the vertical exchange between land and atmosphere of all non-fossil fuel sources and sinks of CO2, while accounting for lateral transfers of forest and crop products as well as their eventual emissions. The total NEE estimate of a � 327 ± 252 TgC yr � 1 sink for NA was driven primarily by CO2 uptake in the Forest Lands sector (� 248 TgC yr � 1 ), largely in the Northwest and Southeast regions of the US, and in the Crop Lands sector (� 297 TgC yr � 1 ), predominantly in the Midwest US states. These sinks are counteracted by the carbon source estimated for the Other Lands sector (+218 TgC yr � 1 ), where much of the forest and crop products are assumed to be returned to the atmosphere (through livestock and human consumption). The ecosystems of Mexico are estimated to be a small net source (+18 TgC yr � 1 ) due to land use change between 1993 and 2002. We compare these inventorybased estimates with results from a suite of terrestrial biosphere and atmospheric inversion models, where the mean continental-scale NEE estimate for each ensemble is � 511 TgC yr � 1 and � 931 TgC yr � 1 , respectively. In the modeling approaches, all sectors, including Other Lands, were generally estimated to be a carbon sink, driven in part by assumed CO2 fertilization and/or lack of consideration of carbon sources from disturbances and product emissions. Additional fluxes not measured by the inventories, although highly uncertain, could add an additional � 239 TgC yr � 1 to the inventory-based NA sink estimate, thus suggesting some convergence with the modeling approaches.

Using forests to mitigate climate change has gained much interest in science and policy discussions. We examine the evidence for carbon benefits, environmental and monetary costs, risks and trade-offs for a variety of activities in three general strategies: (1) land use change to increase forest area (afforestation) and avoid deforestation; (2) carbon management in existing forests; and (3) the use of wood as biomass energy, in place of other building materials, or in wood products for carbon storage. We found that many strategies can increase forest sector carbon mitigation above the current 162-256 Tg C/yr, and that many strategies have co-benefits such as biodiversity, water, and economic opportunities. Each strategy also has trade-offs, risks, and uncertainties including possible leakage, permanence, disturbances, and climate change effects. Because approximately 60% of the carbon lost through deforestation and harvesting from 1700 to 1935 has not yet been recovered and because some strategies store carbon in forest products or use biomass energy, the biological potential for forest sector carbon mitigation is large. Several studies suggest that using these strategies could offset as much as 10-20% of current U.S. fossil fuel emissions. To obtain such large offsets in the United States would require a combination of afforesting up to one-third of cropland or pastureland, using the equivalent of about one-half of the gross annual forest growth for biomass energy, or implementing more intensive management to increase forest growth on one-third of forestland. Such large offsets would require substantial trade-offs, such as lower agricultural production and non-carbon ecosystem services from forests. The effectiveness of activities could be diluted by negative leakage effects and increasing disturbance regimes. Because forest carbon loss contributes to increasing climate risk and because climate change may impede regeneration following disturbance, avoiding deforestation and promoting regeneration after disturbance should receive high priority as policy considerations. Policies to encourage programs or projects that influence forest carbon sequestration and offset fossil fuel emissions should also consider major items such as leakage, the cyclical nature of forest growth and regrowth, and the extensive demand for and movement of forest products globally, and other greenhouse gas effects, such as methane and nitrous oxide emissions, and recognize other environmental benefits of forests, such as biodiversity, nutrient management, and watershed protection. Activities that contribute to helping forests adapt to the effects of climate change, and which also complement forest carbon storage strategies, would be prudent.

Forest disturbances such as harvesting, wildfire and insect infestation are critical ecosystem processes affecting the carbon cycle. Because carbon dynamics are related to time since disturbance, forest stand age that can be used as a surrogate for major clear-cut/fire disturbance information has recently been recognized as an important input to forest carbon cycle models for improving prediction accuracy. In this study, forest disturbances in the USA for the period of ~1990-2000 were mapped using 400+ pairs of re-sampled Landsat TM/ETM scenes in 500m resolution, which were provided by the Landsat Ecosystem Disturbance Adaptive Processing System project. The detected disturbances were then separated into two five-year age groups, facilitated by Forest Inventory and Analysis (FIA) data, which was used to calculate the area of forest regeneration for each county in the USA. In this study, a disturbance index (DI) was defined as the ratio of the short wave infrared (SWIR, band 5) to near-infrared (NIR, band 4) reflectance. Forest disturbances were identified through the Normalized Difference of Disturbance Index (NDDI) between circa 2000 and 1990, where a positive NDDI means disturbance and a negative NDDI means regrowth. Axis rotation was performed on the plot between DIs of the two matched Landsat scenes in order to reduce any difference of DIs caused by non-disturbance factors. The threshold of NDDI for each TM/ETM pair was determined by analysis of FIA data. Minor disturbances affecting small areas may be omitted due to the coarse resolution of the aggregated Landsat data, but the major stand-clearing disturbances (clear-cut harvest, fire) are captured. The spatial distribution of the detected disturbed areas was validated by Monitoring Trends in Burn Severity fire data in four States of the western USA (Washington, Oregon, Idaho, and California). Results indicate omission errors of 66.9%. An important application of this remote sensing-based disturbance map is to associate with FIA forest age data for developing a US forest age map. The US forest age map was also combined with the Canadian forest age map to produce a continent-wide forest map, which becomes a remarkable data layer for North America carbon cycle modeling.

Total forest carbon (C) storage is determined by succession, disturbances, climate, and the edaphic properties of a site or region. Forest harvesting substantially affects C dynamics; these effects may be amplified if forest harvesting is intensified to provide biofuel feedstock. We tested the effects of harvest intensity on landscape C using a simulation modeling approach that included C dynamics, multiple disturbances, and successional changes in composition. We developed a new extension for the LANDIS-II forest landscape disturbance and succession model that incorporates belowground soil C dynamics derived from the CENTURY soil model. The extension was parameterized and calibrated using data from an experimental forest in northeastern Wisconsin, USA. We simulated a 9800 ha forested landscape over 400 years with wind disturbance combined with no harvesting, harvesting with residual slash left on site (‘standard harvest’), and whole-tree harvesting. We also simulated landscapes without wind disturbance and without eastern hemlock (Tsuga canadensis) to examine the effects of detrital quantity and quality on C dynamics. We estimated changes in live C, detrital C, soil organic C, total C, and forest composition. Overall, the simulations without harvesting had substantially greater total C and continued to sequester C. Standard harvest simulations had more C than the whole tree harvest simulations. Under both harvest regimes, C accrual was not evident after 150 years. Without hemlock, SOC was reduced due to a decline in detritus and a shift in detrital chemistry. In conclusion, if the intensity of harvesting increases we can expect a corresponding reduction in potential C storage. Compositional changes due to historic circumstances (loss of hemlock) may also affect forest C although to a lesser degree than harvesting. The modeling approach presented enabled us to consider multiple, interacting drivers of landscape change and the subsequent changes in forest C.

Several studies have reported different estimates for forest biomass carbon (C) stocks in China. The discrepancy among these estimates may be largely attributed to the methods used. In this study, we used three methods [mean biomass density method (MBM), mean ratio method (MRM), and continuous biomass expansion factor (BEF) method (abbreviated as CBM)] applied to forest inventory data to estimate China's forest biomass C stocks and their changes from 1984 to 2003. The three methods generated various estimates of the biomass C stocks: the lowest (4.0–5.9 Pg C) from CBM and the highest (5.7–7.7 Pg C) from MBM, with an intermediate estimate (4.2–6.2 Pg C) from MRM. Forest age class is a major factor responsible for these method-induced differences. MBM overestimates biomass for young-aged forests, but underestimates biomass for old-aged forests; while the reverse is true for MRM. Further, the three methods resulted in different estimates of biomass C stocks for different forest types. For temperate/subtropical mixed forests, MBM generated a 92% higher estimate than CBM and MRM generated a 14% lower than CBM. The degree of the overestimates is closely related with the proportion of young-aged forest within total area of each forest type.

Assessing forest carbon storage and cycling over large areas is a growing challenge that is complicated by the inherent heterogeneity of forest systems. Field measurements must be conducted and analyzed appropriately to generate precise estimates at scales large enough for mapping or comparison with remote sensing data. In this study we examined spatial variability in three small temperate forest landscapes. Our objectives were (1) to quantify the magnitude and scale of variability in stand structure, carbon pools and carbon fluxes and (2) to assess how this variability influences both optimal sampling strategy and required sampling intensity. Stand structure was consistently less variable than carbon pools or fluxes, suggesting that measuring carbon dynamics may require more intense sampling than traditional forestry inventories. Likewise, the magnitude of variability differed substantially among response variables, implying that sampling efficiency can be enhanced by adopting a flexible sampling strategy that is optimized for each carbon pool. Our results indicate that plots dispersed across the study area are generally more effective than clustered plots for characterizing carbon dynamics. Published by Elsevier B.V.

Despite the key role that understory vegetation plays in ecosystems and the terrestrial carbon cycle, it is often overlooked and has few quantitative measurements, especially at national scales. To understand the contribution of understory carbon to the United States (US) carbon budget, we developed an approach that relies on field measurements of understory vegetation cover and height on US Department of Agriculture Forest Service, Forest Inventory and Analysis (FIA) subplots. Allometric models were developed to estimate aboveground understory carbon. A spatial model based on stand characteristics and remotely sensed data was also applied to estimate understory carbon on all FIA plots. We found that most understory carbon was comprised of woody shrub species (64%), followed by nonwoody forbs and graminoid species (35%) and seedlings (1%). The largest estimates were found in temperate or warm humid locations such as the Pacific Northwest and southeastern US, thus following the same broad trend as aboveground tree biomass. The average understory aboveground carbon density was estimated to be 0.977 Mg ha-1, for a total estimate of 272 Tg carbon across all managed forest land in the US (approximately 2% of the total aboveground live tree carbon pool). This estimate is more than twice as low as previous FIA modeled estimates that did not rely on understory measurements, suggesting that this pool may currently be overestimated in US National Greenhouse Gas reporting.

Terrestrial carbon dynamics have been vastly modified because of changes in atmospheric composition, climate, and land-use. However, few studies provide a complete analysis of the factors and interactions that affect carbon dynamics over a large landscape. This study examines how changes in atmospheric composition (CO2 ,O 3 and N deposition), climate and land-use affected carbon dynamics and sequestration in Mid-Atlantic temperate forests during the 20th century. We modified and applied the PnET-CN model, a well established process-based ecosystem model with a strong foundation of ecosystem knowledge from experimental studies. We validated the model results using the U.S. Forest Inventory and Analysis (FIA) data. Our results suggest that chronic changes in atmospheric chemistry over the past century markedly affected carbon dynamics and sequestration in Mid-Atlantic temperate forests, while climate change only had a minor impact although inter-annual climatic variability had a far more substantial effect. The NPP response to a century of chronic change in atmospheric composition at the regional scale was an increase of 29%, of which, 14% was from elevated CO2, 17% from N deposition, 6% from the interaction between CO2 and N deposition, and minus 8% from tropospheric ozone. Climate change increased NPP by only 4%. Disturbed forests had 6% lower NPP than undisturbed forests after seven decades. Regrowing forests after harvesting and natural disturbances had much greater capacity for sequestering carbon than undisturbed old-growth forests even though the newer forests had slightly lower net primary production (NPP). The modeling results indicated that N deposition was a stronger force than elevated CO2 for increasing NPP and fast turnover tissues, while elevated CO2 favored more sustainable carbon storage and sequestration. The model results are consistent with various experiments and observations and demonstrate a powerful approach to integrate and expand our knowledge of complex interactive effects of multiple environmental changes on forest carbon dynamics.

Reliably estimating carbon storage and cycling in detrital biomass is an obstacle to carbon accounting. We examined carbon pools and fluxes in three small temperate forest landscapes to assess the magnitude of carbon stored in detrital biomass and determine whether detrital carbon storage is related to stand structural properties (leaf area, aboveground biomass, primary production) that can be estimated by remote sensing. We characterized these relationships with and without forest age as an additional predictive variable. Results depended on forest type. Carbon in dead woody debris was substantial at all sites, accounting for ∼17% of aboveground carbon, whereas carbon in forest floor was substantial in the subalpine Rocky Mountains (36% of aboveground carbon) and less important in northern hardwoods of New England and mixed forests of the upper Midwest (∼7%). Relationships to aboveground characteristics accounted for between 38% and 59% of the variability in carbon stored in forest floor and between 21% and 71% of the variability in carbon stored in dead woody material, indicating substantial differences among sites. Relating dead woody debris or forest floor carbon to other aboveground characteristics and (or) stand age may, in some forest types, provide a partial solution to the challenge of assessing fine-scale variability.

Forest carbon stocks and fluxes vary with forest age, and relationships with forest age are often used to estimate fluxes for regional or national carbon inventories. Two methods are commonly used to estimate forest age: observed tree age or time since a known disturbance. To clarify the relationships between tree age, time since disturbance and forest carbon storage and cycling, we examined stands of known disturbance history in three landscapes of the southern Rocky Mountains. Our objectives were to assess the similarity between carbon stocks and fluxes for these three landscapes that differed in climate and disturbance history, characterize the relationship between observed tree age and time since disturbance and quantify the predictive capability of tree age or time since disturbance on carbon stocks and fluxes. Carbon pools and fluxes were remarkably similar across the three landscapes, despite differences in elevation, climate, species composition, disturbance history, and forest age. Observed tree age was a poor predictor of time since disturbance. Maximum tree age overestimated time since disturbance for young forests and underestimated it for older forests. Carbon pools and fluxes were related to both tree age and disturbance history, but the relationships differed between these two predictors and were generally less variable for pools than for fluxes. Using tree age in a relationship developed with time since disturbance or vice versa increases errors in estimates of carbon stocks or fluxes. Little change in most carbon stocks and fluxes occurs after the first 100 years following stand-replacing disturbance, simplifying landscape scale estimates. We conclude that subalpine forests in the Central Rocky Mountains can be treated as a single forest type for the purpose of assessment and modeling of carbon, and that the critical period for change in carbon is < 100 years.

Forest inventories are commonly used to estimate total tree biomass of forest land even though they are not traditionally designed to measure biomass of trees outside forests (TOF). The consequence may be an inaccurate representation of all of the aboveground biomass, which propagates error to the outputs of spatial and process models that rely on the inventory data. An ideal approach to fill this data gap would be to integrate TOF measurements within a traditional forest inventory for a parsimonious estimate of total tree biomass. In this study, Light Detection and Ranging (LIDAR) data were used to predict biomass of TOF in all "nonforest" Forest Inventory and Analysis (FIA) plots in the state of Maryland. To validate the LIDAR-based biomass predictions, a field crew was sent to measure TOF on nonforest plots in three Maryland counties, revealing close agreement at both the plot and county scales between the two estimates. Total tree biomass in Maryland increased by 25.5 Tg, or 15.6%, when biomass of TOF were included. In two counties (Carroll and Howard), there was a 47% increase. In contrast, counties located further away from the interstate highway corridor showed only a modest increase in biomass when TOF were added because nonforest conditions were less common in those areas. The advantage of this approach for estimating biomass of TOF is that it is compatible with, and explicitly separates TOF biomass from, forest biomass already measured by FIA crews. By predicting biomass of TOF at actual FIA plots, this approach is directly compatible with traditionally reported FIA forest biomass, providing a framework for other states to follow, and should improve carbon reporting and modeling activities in Maryland.

Knowing the uncertainty for biomass equations is critical for their use and error propagation of biomass estimates. Presented here is a method to estimate uncertainty for equations where only n and R 2 values from the original equations are available. Tree allometric equations form the basis of research and assessments of forest biomass. Frequently, uncertainty estimations do not propagate errors from these equations since the necessary information about sampling and tree measurements is not included in the original publication. Many biomass studies were conducted decades ago and the original, raw data is unavailable. Because of this information deficiency, and to improve error estimates in applications, a system to estimate the error structures of such equations is presented. A pseudo-data approach involving the creation of possible (pseudo) data using only R 2 and n with a simple Monte-Carlo process generates probable error structures that can be used to propagate errors. In a test of five different species with varying n input data and population variability, the original error structures were successfully recreated. This method of creating pseudo-data is simple and extensible and requires commonly published information about the original dataset. The method can be employed to create new ecosystem-level equations from species-specific equations, implemented in systems to select allometric equations to reduce uncertainty, and aid in the design of large-scale campaigns to generate new allometric equations for improving local to national scale estimates of forest biomass. The R code will be made freely available to anyone upon request to the authors.

The United States Climate Change Initiative includes improvements to the U.S. Department of Energy's Voluntary Greenhouse Gas Reporting Program. The program includes specific accounting rules and guidelines for reporting and registering forestry activities that reduce atmospheric CO2 by increasing carbon sequestration or reducing emissions. In the forestry sector, there is potential for the economic value of emissions credits to provide increased income for landowners, to support rural development, to facilitate the practice of sustainable forest management, and to support restoration of ecosystems. Forestry activities with potential for achieving substantial reductions include, but are not limited to: afforestation, mine land reclamation, forest restoration, agroforestry, forest management, short-rotation biomass energy plantations, forest protection, wood production, and urban forestry. To be eligible for registration, the reported reductions must use methods and meet standards contained in the guidelines. Forestry presents some unique challenges and opportunities because of the diversity of activities, the variety of practices that can affect greenhouse gases, year-to-year variability in emissions and sequestration, the effects of activities on different forest carbon pools, and accounting for the effects of natural disturbance.

This paper reviews the effects of past forest management on carbon stocks in the United States, and the challenges for managing forest carbon resources in the 21st century. Forests in the United States were in approximate carbon balance with the atmosphere from 1600-1800. Utilization and land clearing caused a large pulse of forest carbon emissions during the 19th century, followed by regrowth and net forest carbon sequestration in the 20th century. Recent data and knowledge of the general behavior of forests after disturbance suggest that the rate of forest carbon sequestration is declining. A goal of an additional 100 to 200 Tg C/yr of forest carbon sequestration is achievable, but would require investment in inventory and monitoring, development of technology and practices, and assistance for land managers.