The three major axes of terrestrial ecosystem function.

The leaf economics spectrum1,2 and the global spectrum of plant forms and functions3 revealed fundamental axes of variation in plant traits, which represent different ecological strategies that are shaped by the evolutionary development of plant species2. Ecosystem functions depend on environmental conditions and the traits of species that comprise the ecological communities4. However, the axes of variation of ecosystem functions are largely unknown, which limits our understanding of how ecosystems respond as a whole to anthropogenic drivers, climate and environmental variability4,5. Here we derive a set of ecosystem functions6 from a dataset of surface gas exchange measurements across major terrestrial biomes. We find that most of the variability within ecosystem functions (71.8%) is captured by three key axes. The first axis reflects maximum ecosystem productivity and is mostly explained by vegetation structure. The second axis reflects ecosystem water-use strategies and is jointly explained by variation in vegetation height and climate. The third axis, which represents ecosystem carbon-use efficiency, features a gradient related to aridity, and is explained primarily by variation in vegetation structure. We show that two state-of-the-art land surface models reproduce the first and most important axis of ecosystem functions. However, the models tend to simulate more strongly correlated functions than those observed, which limits their ability to accurately predict the full range of responses to environmental changes in carbon, water and energy cycling in terrestrial ecosystems7,8.

Globally rising soil heterotrophic respiration over recent decades.

Global soils store at least twice as much carbon as Earth's atmosphere1,2. The global soil-to-atmosphere (or total soil respiration, RS) carbon dioxide (CO2) flux is increasing3,4, but the degree to which climate change will stimulate carbon losses from soils as a result of heterotrophic respiration (RH) remains highly uncertain5-8. Here we use an updated global soil respiration database9 to show that the observed soil surface RH:RS ratio increased significantly, from 0.54 to 0.63, between 1990 and 2014 (P = 0.009). Three additional lines of evidence provide support for this finding. By analysing two separate global gross primary production datasets10,11, we find that the ratios of both RH and RS to gross primary production have increased over time. Similarly, significant increases in RH are observed against the longest available solar-induced chlorophyll fluorescence global dataset, as well as gross primary production computed by an ensemble of global land models. We also show that the ratio of night-time net ecosystem exchange to gross primary production is rising across the FLUXNET201512 dataset. All trends are robust to sampling variability in ecosystem type, disturbance, methodology, CO2 fertilization effects and mean climate. Taken together, our findings provide observational evidence that global RH is rising, probably in response to environmental changes, consistent with meta-analyses13-16 and long-term experiments17. This suggests that climate-driven losses of soil carbon are currently occurring across many ecosystems, with a detectable and sustained trend emerging at the global scale.

Disturbance-accelerated succession increases the production of a temperate forest.

Many secondary deciduous forests of eastern North America are approaching a transition in which mature early-successional trees are declining, resulting in an uncertain future for this century-long carbon (C) sink. We initiated the Forest Accelerated Succession Experiment (FASET) at the University of Michigan Biological Station to examine the patterns and mechanisms underlying forest C cycling following the stem girdling-induced mortality of >6,700 early-successional Populus spp. (aspen) and Betula papyrifera (paper birch). Meteorological flux tower-based C cycling observations from the 33-ha treatment forest have been paired with those from a nearby unmanipulated forest since 2008. Following over a decade of observations, we revisit our core hypothesis: that net ecosystem production (NEP) would increase following the transition to mid-late-successional species dominance due to increased canopy structural complexity. Supporting our hypothesis, NEP was stable, briefly declined, and then increased relative to the control in the decade following disturbance; however, increasing NEP was not associated with rising structural complexity but rather with a rapid 1-yr recovery of total leaf area index as mid-late-successional Acer, Quercus, and Pinus assumed canopy dominance. The transition to mid-late-successional species dominance improved carbon-use efficiency (CUE = NEP/gross primary production) as ecosystem respiration declined. Similar soil respiration rates in control and treatment forests, along with species differences in leaf physiology and the rising relative growth rates of mid-late-successional species in the treatment forest, suggest changes in aboveground plant respiration and growth were primarily responsible for increases in NEP. We conclude that deciduous forests transitioning from early to middle succession are capable of sustained or increased NEP, even when experiencing extensive tree mortality. This adds to mounting evidence that aging deciduous forests in the region will function as C sinks for decades to come.

Structure and parameter uncertainty in centennial projections of forest community structure and carbon cycling.

Secondary forest regrowth shapes community succession and biogeochemistry for decades, including in the Upper Great Lakes region. Vegetation models encapsulate our understanding of forest function, and whether models can reproduce multi-decadal succession patterns is an indication of our ability to predict forest responses to future change. We test the ability of a vegetation model to simulate C cycling and community composition during 100 years of forest regrowth following stand-replacing disturbance, asking (a) Which processes and parameters are most important to accurately model Upper Midwest forest succession? (b) What is the relative importance of model structure versus parameter values to these predictions? We ran ensembles of the Ecosystem Demography model v2.2 with different representations of processes important to competition for light. We compared the magnitude of structural and parameter uncertainty and assessed which sub-model-parameter combinations best reproduced observed C fluxes and community composition. On average, our simulations underestimated observed net primary productivity (NPP) and leaf area index (LAI) after 100 years and predicted complete dominance by a single plant functional type (PFT). Out of 4,000 simulations, only nine fell within the observed range of both NPP and LAI, but these predicted unrealistically complete dominance by either early hardwood or pine PFTs. A different set of seven simulations were ecologically plausible but under-predicted observed NPP and LAI. Parameter uncertainty was large; NPP and LAI ranged from ~0% to >200% of their mean value, and any PFT could become dominant. The two parameters that contributed most to uncertainty in predicted NPP were plant-soil water conductance and growth respiration, both unobservable empirical coefficients. We conclude that (a) parameter uncertainty is more important than structural uncertainty, at least for ED-2.2 in Upper Midwest forests and (b) simulating both productivity and plant community composition accurately without physically unrealistic parameters remains challenging for demographic vegetation models.

COSORE: A community database for continuous soil respiration and other soil-atmosphere greenhouse gas flux data.

Globally, soils store two to three times as much carbon as currently resides in the atmosphere, and it is critical to understand how soil greenhouse gas (GHG) emissions and uptake will respond to ongoing climate change. In particular, the soil-to-atmosphere CO2 flux, commonly though imprecisely termed soil respiration (RS ), is one of the largest carbon fluxes in the Earth system. An increasing number of high-frequency RS measurements (typically, from an automated system with hourly sampling) have been made over the last two decades; an increasing number of methane measurements are being made with such systems as well. Such high frequency data are an invaluable resource for understanding GHG fluxes, but lack a central database or repository. Here we describe the lightweight, open-source COSORE (COntinuous SOil REspiration) database and software, that focuses on automated, continuous and long-term GHG flux datasets, and is intended to serve as a community resource for earth sciences, climate change syntheses and model evaluation. Contributed datasets are mapped to a single, consistent standard, with metadata on contributors, geographic location, measurement conditions and ancillary data. The design emphasizes the importance of reproducibility, scientific transparency and open access to data. While being oriented towards continuously measured RS , the database design accommodates other soil-atmosphere measurements (e.g. ecosystem respiration, chamber-measured net ecosystem exchange, methane fluxes) as well as experimental treatments (heterotrophic only, etc.). We give brief examples of the types of analyses possible using this new community resource and describe its accompanying R software package.

Defining a spectrum of integrative trait-based vegetation canopy structural types.

Vegetation canopy structure is a fundamental characteristic of terrestrial ecosystems that defines vegetation types and drives ecosystem functioning. We use the multivariate structural trait composition of vegetation canopies to classify ecosystems within a global canopy structure spectrum. Across the temperate forest sub-set of this spectrum, we assess gradients in canopy structural traits, characterise canopy structural types (CST) and evaluate drivers and functional consequences of canopy structural variation. We derive CSTs from multivariate canopy structure data, illustrating variation along three primary structural axes and resolution into six largely distinct and functionally relevant CSTs. Our results illustrate that within-ecosystem successional processes and disturbance legacies can produce variation in canopy structure similar to that associated with sub-continental variation in forest types and eco-climatic zones. The potential to classify ecosystems into CSTs based on suites of structural traits represents an important advance in understanding and modelling structure-function relationships in vegetated ecosystems.

Forest aging, disturbance and the carbon cycle.

Contents Summary 1188 I. Introduction 1188 II. Forest aging and carbon storage 1189 III. Successional trends of NEP in northern deciduous forests 1190 IV. Mechanisms sustaining NEP in aging deciduous forests 1191 Acknowledgements 1192 References 1192 SUMMARY: Large areas of forestland in temperate North America, as well as in other parts of the world, are growing older and will soon transition into middle and then late successional stages exceeding 100 yr in age. These ecosystems have been important regional carbon sinks as they recovered from prior anthropogenic and natural disturbance, but their future sink strength, or annual rate of carbon storage, is in question. Ecosystem development theory predicts a steady decline in annual carbon storage as forests age, but newly available, direct measurements of forest net CO2 exchange challenge that prediction. In temperate deciduous forests, where moderate severity disturbance regimes now often prevail, there is little evidence for any marked decline in carbon storage rate during mid-succession. Rather, an increase in physical and biological complexity under these disturbance regimes may drive increases in resource-use efficiency and resource availability that help to maintain significant carbon storage in these forests well past the century mark. Conservation of aging deciduous forests may therefore sustain the terrestrial carbon sink, whilst providing other goods and services afforded by these biologically and structurally complex ecosystems.

Effects of canopy structure and species diversity on primary production in upper Great Lakes forests.

Canopy structure and tree species diversity, shaped by succession, disturbance, and community composition, are linked to numerous ecosystem functions, including net primary production (NPP). Understanding of how ecosystem structural metrics are interrelated and mechanistically link to NPP, however, is incomplete. We characterized leaf area index (LAI), Simpson's index of Diversity (D', a measure of species diversity), and canopy rugosity (Rc, a measure of canopy physical complexity) in 11 forest stands comprising two chronosequences varying in establishing disturbance, and in three late successional communities. We related LAI, D', and Rc to wood NPP (NPPw), and examined whether absorption of photosynthetically active radiation and light use-efficiency (LUE) link NPPw with ecosystem structure. We found that recovery of LAI and D' was delayed following more severe establishing disturbances, but that the development of Rc was strikingly conserved regardless of disturbance, converging on a common mean value in late-successional stands irrespective of differences in leaf area index and species diversity. LAI was significantly correlated with NPPw in each stage of ecosystem development, but NPPw was only correlated with Rc in early successional stages and with D' in late successional stages. Across all stands, NPPw was coupled with LAI and Rc, (but not D') through positive relationships with light absorption and LUE. We conclude by advocating for better integration of ecological disciplines investigating structure-function interactions, suggesting that improved understanding of such relationships will require ecologists to traverse disciplinary boundaries.

Joint control of terrestrial gross primary productivity by plant phenology and physiology.

Terrestrial gross primary productivity (GPP) varies greatly over time and space. A better understanding of this variability is necessary for more accurate predictions of the future climate-carbon cycle feedback. Recent studies have suggested that variability in GPP is driven by a broad range of biotic and abiotic factors operating mainly through changes in vegetation phenology and physiological processes. However, it is still unclear how plant phenology and physiology can be integrated to explain the spatiotemporal variability of terrestrial GPP. Based on analyses of eddy-covariance and satellite-derived data, we decomposed annual terrestrial GPP into the length of the CO2 uptake period (CUP) and the seasonal maximal capacity of CO2 uptake (GPPmax). The product of CUP and GPPmax explained >90% of the temporal GPP variability in most areas of North America during 2000-2010 and the spatial GPP variation among globally distributed eddy flux tower sites. It also explained GPP response to the European heatwave in 2003 (r(2) = 0.90) and GPP recovery after a fire disturbance in South Dakota (r(2) = 0.88). Additional analysis of the eddy-covariance flux data shows that the interbiome variation in annual GPP is better explained by that in GPPmax than CUP. These findings indicate that terrestrial GPP is jointly controlled by ecosystem-level plant phenology and photosynthetic capacity, and greater understanding of GPPmax and CUP responses to environmental and biological variations will, thus, improve predictions of GPP over time and space.

Net primary production of a temperate deciduous forest exhibits a threshold response to increasing disturbance severity.

The global carbon (C) balance is vulnerable to disturbances that alter terrestrial C storage. Disturbances to forests occur along a continuum of severity, from low-intensity disturbance causing the mortality or defoliation of only a subset of trees to severe stand- replacing disturbance that kills all trees; yet considerable uncertainty remains in how forest production changes across gradients of disturbance intensity. We used a gradient of tree mortality in an upper Great Lakes forest ecosystem to: (1) quantify how aboveground wood net primary production (ANPP,) responds to a range of disturbance severities; and (2) identify mechanisms supporting ANPPw resistance or resilience following moderate disturbance. We found that ANPPw declined nonlinearly with rising disturbance severity, remaining stable until >60% of the total tree basal area senesced. As upper canopy openness increased from disturbance, greater light availability to the subcanopy enhanced the leaf-level photosynthesis and growth of this formerly light-limited canopy stratum, compensating for upper canopy production losses and a reduction in total leaf area index (LAI). As a result, whole-ecosystem production efficiency (ANPPw/LAI) increased with rising disturbance severity, except in plots beyond the disturbance threshold. These findings provide a mechanistic explanation for a nonlinear relationship between ANPPw, and disturbance severity, in which the physiological and growth enhancement of undisturbed vegetation is proportional to the level of disturbance until a threshold is exceeded. Our results have important ecological and management implications, demonstrating that in some ecosystems moderate levels of disturbance minimally alter forest production.

Disturbance theory for ecosystem ecologists: A primer.

Understanding what regulates ecosystem functional responses to disturbance is essential in this era of global change. However, many pioneering and still influential disturbance-related theorie proposed by ecosystem ecologists were developed prior to rapid global change, and before tools and metrics were available to test them. In light of new knowledge and conceptual advances across biological disciplines, we present four disturbance ecology concepts that are particularly relevant to ecosystem ecologists new to the field: (a) the directionality of ecosystem functional response to disturbance; (b) functional thresholds; (c) disturbance-succession interactions; and (d) diversity-functional stability relationships. We discuss how knowledge, theory, and terminology developed by several biological disciplines, when integrated, can enhance how ecosystem ecologists analyze and interpret functional responses to disturbance. For example, when interpreting thresholds and disturbance-succession interactions, ecosystem ecologists should consider concurrent biotic regime change, non-linearity, and multiple response pathways, typically the theoretical and analytical domain of population and community ecologists. Similarly, the interpretation of ecosystem functional responses to disturbance requires analytical approaches that recognize disturbance can promote, inhibit, or fundamentally change ecosystem functions. We suggest that truly integrative approaches and knowledge are essential to advancing ecosystem functional responses to disturbance.

Sustained carbon uptake and storage following moderate disturbance in a Great Lakes forest.

Carbon (C) uptake rates in many forests are sustained, or decline only briefly, following disturbances that partially defoliate the canopy. The mechanisms supporting such functional resistance to moderate forest disturbance are largely unknown. We used a large-scale experiment, in which > 6700 Populus (aspen) and Betula (birch) trees were stem-girdled within a 39-ha area, to identify mechanisms sustaining C uptake through partial canopy defoliation. The Forest Accelerated Succession Experiment in northern Michigan, USA, employs a suite of C-cycling measurements within paired treatment and control meteorological flux tower footprints. We found that enhancement of canopy light-use efficiency and maintenance of light absorption maintained net ecosystem production (NEP) and aboveground wood net primary production (NPP) when leaf-area index (LAI) of the treatment forest temporarily declined by nearly half its maximum value. In the year following peak defoliation, redistribution of nitrogen (N) in the treatment forest from senescent early successional aspen and birch to non-girdled later successional species facilitated the recovery of total LAI to pre-disturbance levels. Sustained canopy physiological competency following disturbance coincided with a downward shift in maximum canopy height, indicating that compensatory photosynthetic C uptake by undisturbed, later successional subdominant and subcanopy vegetation supported C-uptake resistance to disturbance. These findings have implications for ecosystem management and modeling, demonstrating that forests may tolerate considerable leaf-area losses without diminishing rates of C uptake. We conclude that the resistance of C uptake to moderate disturbance depends not only on replacement of lost leaf area, but also on rapid compensatory photosynthetic C uptake during defoliation by emerging later successional species.

Lawn soil carbon storage in abandoned residential properties: an examination of ecosystem structure and function following partial human-natural decoupling.

Residential abandonment is on the rise in many urban areas, with unknown implications for ecosystem structure and function on land slated for partial or full restoration to native habitat. Partial decoupling of human and natural systems could reduce disturbance (e.g., trampling, recreational traffic) and modify vegetation structure in a way that alters soil carbon storage, an ecosystem function that many municipalities consider a management objective of growing importance. We quantified soil carbon percent and mass to 10 cm depth and examined vegetation structure in 50 vacant and 10 occupied residential lawns located in Richmond, VA, with the principal objective of determining whether occupancy status alters trajectories of soil carbon storage or its correspondence with household economic/demographic indicators and vegetation cover. Abandoned residential lawns supported significantly less grass cover, but these declines were largely offset by increases in emergent overstory (>1 m height) vegetation cover. Soil carbon percent and mass did not differ between lawns of occupied and abandoned residences, even though significant, but highly uncertain, increases in soil carbon mass occurred in the first decade following vacancy. Instead, all residential lawns exhibited similar significant increases in soil carbon percent and mass with increasing residence age and neighborhood affluence, the former indicating annual carbon accretion rates of 20 g m(-2). We conclude that in this early stage of vacancy, soil carbon storage is already subtly responding to declines in human intervention, with reduced soil disturbance and sustained vegetation cover in abandoned lawns playing likely roles in emerging soil carbon storage trajectories.

Climate Drives Modeled Forest Carbon Cycling Resistance and Resilience in the Upper Great Lakes Region, USA.

Forests dominate the global terrestrial carbon budget, but their ability to continue doing so in the face of a changing climate is uncertain. A key uncertainty is how forests will respond to (resistance) and recover from (resilience) rising levels of disturbance of varying intensities. This knowledge gap can optimally be addressed by integrating manipulative field experiments with ecophysiological modeling. We used the Ecosystem Demography-2.2 (ED-2.2) model to project carbon fluxes for a northern temperate deciduous forest subjected to a real-world disturbance severity manipulation experiment. ED-2.2 was run for 150 years, starting from near bare ground in 1900 (approximating the clear-cut conditions at the time), and subjected to three disturbance treatments under an ensemble of climate conditions. Both disturbance severity and climate strongly affected carbon fluxes such as gross primary production (GPP), and interacted with one another. We then calculated resistance and resilience, two dimensions of ecosystem stability. Modeled GPP exhibited a two-fold decrease in mean resistance across disturbance severities of 45%, 65%, and 85% mortality; conversely, resilience increased by a factor of two with increasing disturbance severity. This pattern held for net primary production and net ecosystem production, indicating a trade-off in which greater initial declines were followed by faster recovery. Notably, however, heterotrophic respiration responded more slowly to disturbance, and it's highly variable response was affected by different drivers. This work provides insight into how future conditions might affect the functional stability of mature forests in this region under ongoing climate change and changing disturbance regimes.

Mechanistically-grounded pathways connect remotely sensed canopy structure to soil respiration.

Variation in the soil-to-atmosphere C flux, or soil respiration (Rs), is influenced by a suite of biotic and abiotic factors, including soil temperature, soil moisture, and root biomass. However, whether light detection and ranging (lidar)-derived canopy structure is tied to soil respiration through its simultaneous influence over these drivers is not known. We assessed relationships between measures of above- and belowground vegetation density and complexity, and evaluated whether Rs is linked to remotely sensed canopy structure through pathways mediated by established biotic and abiotic mechanisms. Our results revealed that, at the stand-scale, canopy rugosity-a measure of complexity-and vegetation area index were coupled to soil respiration through their effects on light interception, soil microclimate, and fine root mass density, but this connection was stronger for complexity. Canopy and root complexity were not spatially coupled at the stand-scale, with canopy but not root complexity increasing through stand development. Our findings suggest that remotely sensed canopy complexity could be used to infer spatial variation in Rs, and that this relationship is grounded in known mechanistic pathways. The broad spatial inference of soil respiration via remotely sensed canopy complexity requires multi-site observations of canopy structure and Rs, which is possible given burgeoning open data from ecological networks and satellite remote sensing platforms.

The role of canopy structural complexity in wood net primary production of a maturing northern deciduous forest.

The even-aged northern hardwood forests of the Upper Great Lakes Region are undergoing an ecological transition during which structural and biotic complexity is increasing. Early-successional aspen (Populus spp.) and birch (Betula papyrifera) are senescing at an accelerating rate and are being replaced by middle-successional species including northern red oak (Quercus rubra), red maple (Acer rubrum), and white pine (Pinus strobus). Canopy structural complexity may increase due to forest age, canopy disturbances, and changing species diversity. More structurally complex canopies may enhance carbon (C) sequestration in old forests. We hypothesize that these biotic and structural alterations will result in increased structural complexity of the maturing canopy with implications for forest C uptake. At the University of Michigan Biological Station (UMBS), we combined a decade of observations of net primary productivity (NPP), leaf area index (LAI), site index, canopy tree-species diversity, and stand age with canopy structure measurements made with portable canopy lidar (PCL) in 30 forested plots. We then evaluated the relative impact of stand characteristics on productivity through succession using data collected over a nine-year period. We found that effects of canopy structural complexity on wood NPP (NPPw) were similar in magnitude to the effects of total leaf area and site quality. Furthermore, our results suggest that the effect of stand age on NPPw is mediated primarily through its effect on canopy structural complexity. Stand-level diversity of canopy-tree species was not significantly related to either canopy structure or NPPw. We conclude that increasing canopy structural complexity provides a mechanism for the potential maintenance of productivity in aging forests.

Warming homogenizes apparent temperature sensitivity of ecosystem respiration.

Warming-induced carbon loss through terrestrial ecosystem respiration (Re) is likely getting stronger in high latitudes and cold regions because of the more rapid warming and higher temperature sensitivity of Re (Q 10). However, it is not known whether the spatial relationship between Q 10 and temperature also holds temporally under a future warmer climate. Here, we analyzed apparent Q 10 values derived from multiyear observations at 74 FLUXNET sites spanning diverse climates and biomes. We found warming-induced decline in Q 10 is stronger at colder regions than other locations, which is consistent with a meta-analysis of 54 field warming experiments across the globe. We predict future warming will shrink the global variability of Q 10 values to an average of 1.44 across the globe under a high emission trajectory (RCP 8.5) by the end of the century. Therefore, warming-induced carbon loss may be less than previously assumed because of Q 10 homogenization in a warming world.

Temperature thresholds of ecosystem respiration at a global scale.

Ecosystem respiration is a major component of the global terrestrial carbon cycle and is strongly influenced by temperature. The global extent of the temperature-ecosystem respiration relationship, however, has not been fully explored. Here, we test linear and threshold models of ecosystem respiration across 210 globally distributed eddy covariance sites over an extensive temperature range. We find thresholds to the global temperature-ecosystem respiration relationship at high and low air temperatures and mid soil temperatures, which represent transitions in the temperature dependence and sensitivity of ecosystem respiration. Annual ecosystem respiration rates show a markedly reduced temperature dependence and sensitivity compared to half-hourly rates, and a single mid-temperature threshold for both air and soil temperature. Our study indicates a distinction in the influence of environmental factors, including temperature, on ecosystem respiration between latitudinal and climate gradients at short (half-hourly) and long (annual) timescales. Such climatological differences in the temperature sensitivity of ecosystem respiration have important consequences for the terrestrial net carbon sink under ongoing climate change.

Forest structure, diversity, and primary production in relation to disturbance severity.

Differential disturbance severity effects on forest vegetation structure, species diversity, and net primary production (NPP) have been long theorized and observed. Here, we examined these factors concurrently to explore the potential for a mechanistic pathway linking disturbance severity, changes in light environment, leaf functional response, and wood NPP in a temperate hardwood forest.Using a suite of measurements spanning an experimental gradient of tree mortality, we evaluated the direction and magnitude of change in vegetation structural and diversity indexes in relation to wood NPP. Informed by prior observations, we hypothesized that forest structural and species diversity changes and wood NPP would exhibit either a linear, unimodal, or threshold response in relation to disturbance severity. We expected increasing disturbance severity would progressively shift subcanopy light availability and leaf traits, thereby coupling structural and species diversity changes with primary production.Linear or unimodal changes in three of four vegetation structural indexes were observed across the gradient in disturbance severity. However, disturbance-related changes in vegetation structure were not consistently correlated with shifts in light environment, leaf traits, and wood NPP. Species diversity indexes did not change in response to rising disturbance severity.We conclude that, in our study system, the sensitivity of wood NPP to rising disturbance severity is generally tied to changing vegetation structure but not species diversity. Changes in vegetation structure are inconsistently coupled with light environment and leaf traits, resulting in mixed support for our hypothesized cascade linking disturbance severity to wood NPP.

High rates of primary production in structurally complex forests.

Structure-function relationships are central to many ecological paradigms. Chief among these is the linkage of net primary production (NPP) with species diversity and canopy structure. Using the National Ecological Observatory Network (NEON) as a subcontinental-scale research platform, we examined how temperate-forest NPP relates to several measures of site-level canopy structure and tree species diversity. Novel multidimensional canopy traits describing structural complexity, most notably canopy rugosity, were more strongly related to site NPP than were species diversity measures and other commonly characterized canopy structural features. The amount of variation in site-level NPP explained by canopy rugosity alone was 83%, which was substantially greater than that explained individually by vegetation area index (31%) or Shannon's index of species diversity (30%). Forests that were more structurally complex, had higher vegetation-area indices, or were more diverse absorbed more light and used light more efficiently to power biomass production, but these relationships were most strongly tied to structural complexity. Implications for ecosystem modeling and management are wide ranging, suggesting structural complexity traits are broad, mechanistically robust indicators of NPP that, in application, could improve the prediction and management of temperate forest carbon sequestration.