Search

Natural and anthropogenic disturbances are important drivers of tree mortality, shaping the structure, composition, and biomass distribution of forest ecosystems. Differences in disturbance regimes, characterized by the frequency, extent, and intensity of disturbance events, result in structurally different landscapes. In this study, we design a model‐based experiment to investigate the links between disturbance regimes and spatial biomass patterns. First, the effects of disturbance events on biomass patterns are simulated using a simple dynamic carbon cycle model based on different disturbance regime attributes, which are characterized via three parameters: μ (probability scale), α (clustering degree), and β (intensity slope). 856,800 dynamically stable biomass patterns were then simulated using combined disturbance regime, primary productivity, and background mortality. As independent variables, we use biomass synthesis statistics from simulated biomass patterns to retrieve three disturbance regime parameters. Results show confident inversion of all three "true" disturbance parameters, with Nash‐Sutcliffe efficiency of 94.8% for μ, 94.9% for α, and 97.1% for β. Biomass histogram statistics primarily dominate the prediction of μ and β, while texture features have a more substantial influence on α. Overall, these results demonstrate the association between biomass patterns and disturbance regimes. Given the increasing availability of Earth observation of biomass, our findings open a new avenue to understand better and parameterize disturbance regimes and their links with vegetation dynamics under climate change. Ultimately, at a large scale, this approach would improve our current understanding of controls and feedback at the biosphere‐atmosphere interface in the present Earth system models. Plain Language Summary: Forest dynamics are shaped by different disturbances, which are challenging to monitor and predict. Identifying individual disturbance occurrences and their impact on forest carbon stocks (biomass) is complex. However, our study deciphers the characteristics of disturbance occurrence, that is, disturbance regime, from biomass pattern. We characterized this regime across three dimensions: extent (μ), frequency (α), and intensity (β). Through a 200‐year landscape experiment, we explored the synthetic dynamically stable biomass under different disturbance regimes. Statistical features from biomass simulations revealed distinct spatial patterns, forming a connection between these patterns and the disturbance regime parameters via machine learning. Notably, specific biomass pattern statistics influence distinct disturbance regime parameters: μ and β are linked to histogram stats, while α is tied to texture statistics. This approach establishes a framework to diagnose disturbance regimes from biomass patterns, offering a way to incorporate these regimes into Earth system models. Key Points: We investigate the link between disturbance regimes and spatial patterns of aboveground biomass emerging from diverse primary productivityThe proposed framework allows for inferring disturbance probability, size and intensity from spatial features in aboveground biomassDisturbance regimes from high‐res Earth observations can enhance carbon cycle dynamics prediction from interannual to longer time scales [ABSTRACT FROM AUTHOR]

Wood density is a fundamental property related to tree biomechanics and hydraulic function while playing a crucial role in assessing vegetation carbon stocks by linking volumetric retrieval and a mass estimate. This study provides a high‐resolution map of the global distribution of tree wood density at the 0.01° (~1 km) spatial resolution, derived from four decision trees machine learning models using a global database of 28,822 tree‐level wood density measurements. An ensemble of four top‐performing models combined with eight cross‐validation strategies shows great consistency, providing wood density patterns with pronounced spatial heterogeneity. The global pattern shows lower wood density values in northern and northwestern Europe, Canadian forest regions and slightly higher values in Siberia forests, western United States, and southern China. In contrast, tropical regions, especially wet tropical areas, exhibit high wood density. Climatic predictors explain 49%–63% of spatial variations, followed by vegetation characteristics (25%–31%) and edaphic properties (11%–16%). Notably, leaf type (evergreen vs. deciduous) and leaf habit type (broadleaved vs. needleleaved) are the most dominant individual features among all selected predictive covariates. Wood density tends to be higher for angiosperm broadleaf trees compared to gymnosperm needleleaf trees, particularly for evergreen species. The distributions of wood density categorized by leaf types and leaf habit types have good agreement with the features observed in wood density measurements. This global map quantifying wood density distribution can help improve accurate predictions of forest carbon stocks, providing deeper insights into ecosystem functioning and carbon cycling such as forest vulnerability to hydraulic and thermal stresses in the context of future climate change. [ABSTRACT FROM AUTHOR]

Fire is a crucial factor in terrestrial ecosystems playing a role in disturbance for vegetation dynamics. Process‐based fire models quantify fire disturbance effects in stand‐alone dynamic global vegetation models (DGVMs) and their advances have incorporated both descriptions of natural processes and anthropogenic drivers. Nevertheless, these models show limited skill in modeling fire events at the global scale, due to stochastic characteristics of fire occurrence and behavior as well as the limits in empirical parameterizations in process‐based models. As an alternative, machine learning has shown the capability of providing robust diagnostics of fire regimes. Here, we develop a deep‐learning‐based fire model (DL‐fire) to estimate daily burnt area fraction at the global scale and couple it within JSBACH4, the land surface model used in the ICON‐ESM. The stand‐alone DL‐fire model forced with meteorological, terrestrial and socio‐economic variables is able to simulate global total burnt area, showing 0.8 of monthly correlation (rm) with GFED4 during the evaluation period (2011–2015). The performance remains similar with the hybrid modeling approach JSB4‐DL‐fire (rm = 0.79) outperforming the currently used uncalibrated standard fire model in JSBACH4 (rm = −0.07). We further quantify the importance of each predictor by applying layer‐wise relevance propagation (LRP). Overall, land properties, such as fuel amount and water content in soil layers, stand out as the major factors determining burnt fraction in DL‐fire, paralleled by meteorological conditions over tropical and high latitude regions. Our study demonstrates the potential of hybrid modeling in advancing fire prediction in ESMs by integrating deep learning approaches in physics‐based dynamical models. Plain Language Summary: We develop a fire‐vegetation model based on a hybrid approach integrating artificial intelligence (AI) techniques into physics‐based models. Given the weather conditions, vegetation states, and human factors, our model estimates daily burned area fraction. The spatiotemporal variations in burned area are closely reproduced, especially over fire‐prone regions, such as Africa, South America, and Australia. Our model is able to represent regional variations in the drivers of fire occurrence, showing different importance of input predictors for different regions. This approach shows the possibilities of using deep learning (DL) models to provide in‐depth fire predictions in Earth system models. Key Points: Deep neural networks (DNN) can accurately predict global burnt area fraction on a daily scaleIntegration of the DNN into a physics‐based land model significantly improves the estimation of biomass burnt damage in vegetation dynamicsThe DNN accounts for regional fire variations by assigning varying degrees of importance to each predictor [ABSTRACT FROM AUTHOR]

Copyright of Methods in Ecology & Evolution is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In a model simulating dynamics of a system, parameters can represent system sensitivities and unresolved processes, therefore affecting model accuracy and uncertainty. Taking a light use efficiency (LUE) model as an example, which is a typical approach for estimating gross primary productivity (GPP), we propose a Simultaneous Parameter Inversion and Extrapolation approach (SPIE) to overcome issues stemming from plant‐functional‐type (PFT)‐dependent parameterizations. SPIE refers to predicting model parameters using an artificial neural network based on collected variables, including PFT, climate types, bioclimatic variables, vegetation features, atmospheric nitrogen and phosphorus deposition, and soil properties. The neural network was optimized to minimize GPP errors and constrain LUE model sensitivity functions. We compared SPIE with 11 typical parameter extrapolating methods, including PFT‐ and climate‐specific parameterizations, global and PFT‐based parameter optimization, site‐similarity, and regression approaches. All methods were assessed using Nash‐Sutcliffe model efficiency (NSE), determination coefficient and normalized root mean squared error, and contrasted with site‐specific calibrations. Ten‐fold cross‐validated results showed that SPIE had the best performance across sites, various temporal scales and assessing metrics. Taking site‐level calibrations as a benchmark (NSE = 0.95), SPIE performed with an NSE of 0.68, while all the other investigated approaches showed negative NSE. The Shapley value, layer‐wise relevance and partial dependence showed that vegetation features, bioclimatic variables, soil properties and some PFTs determine parameters. SPIE overcomes strong limitations observed in many standard parameterization methods. We argue that expanding SPIE to other models overcomes current limits and serves as an entry point to investigate the robustness and generalization of different models. Plain Language Summary: Parameters can represent ecosystem properties and sensitivities of ecosystem processes to environmental changes, affecting model accuracy and output uncertainties. Therefore determining parameters is of great importance for applying models. Current ecosystem‐level models mostly determine parameters according to biomes. For example, light use efficiency (LUE) models, a typical tool to estimate gross primary productivity (GPP), use plant‐functional‐type (PFT)‐specific parameters. However, PFT‐specific parameters cannot represent the spatial variance of GPP sensitivities to environmental conditions within PFT and introduce significant estimation errors. To overcome these issues, we propose a Simultaneous Parameter Inversion and Extrapolation method (SPIE). Taking an LUE model as an example, we estimated model parameters using SPIE based on the input features representing vegetation, climate and soil properties at 196 observational sites. We compared SPIE with 11 other parameter extrapolating methods and all of these methods with site‐specific calibrations based on full‐time‐series observed GPP. The results were validated and showed that SPIE was the best method to extrapolate parameters across various temporal and spatial scales. According to the importance of input features, vegetation features, bioclimatic variables, soil properties and some PFTs are dominating spatial changes of parameters. Overall, SPIE is a robust method that overcomes strong limitations in many standard parameter extrapolating methods. Key Points: We propose a machine‐learning‐based parameterization method to model gross primary productivityThe method overcomes significant biases in extrapolations using assumptions of biome‐dependent and globally fixed parameterizationsVegetation, climate and soil properties explain most variability in model parameters, challenging current prescriptions of their patterns [ABSTRACT FROM AUTHOR]

Negative extreme anomalies in vegetation growth (NEGs) usually indicate severely impaired ecosystem services. These NEGs can result from diverse natural and anthropogenic causes, especially climate extremes (CEs). However, the relationship between NEGs and many types of CEs remains largely unknown at regional and global scales. Here, with satellite‐derived vegetation index data and supporting tree‐ring chronologies, we identify periods of NEGs from 1981 to 2015 across the global land surface. We find 70% of these NEGs are attributable to five types of CEs and their combinations, with compound CEs generally more detrimental than individual ones. More importantly, we find that dominant CEs for NEGs vary by biome and region. Specifically, cold and/or wet extremes dominate NEGs in temperate mountains and high latitudes, whereas soil drought and related compound extremes are primarily responsible for NEGs in wet tropical, arid and semi‐arid regions. Key characteristics (e.g., the frequency, intensity and duration of CEs, and the vulnerability of vegetation) that determine the dominance of CEs are also region‐ and biome‐dependent. For example, in the wet tropics, dominant individual CEs have both higher intensity and longer duration than non‐dominant ones. However, in the dry tropics and some temperate regions, a longer CE duration is more important than higher intensity. Our work provides the first global accounting of the attribution of NEGs to diverse climatic extremes. Our analysis has important implications for developing climate‐specific disaster prevention and mitigation plans among different regions of the globe in a changing climate. [ABSTRACT FROM AUTHOR]

Climate change exacerbates the occurrence of extreme droughts and heatwaves, increasing the frequency and intensity of large wildfires across the globe. Forecasting wildfire danger and uncovering the drivers behind fire events become central for understanding relevant climate‐land surface feedback and aiding wildfire management. In this work, we leverage Deep Learning (DL) to predict the next day's wildfire danger in a fire‐prone part of the Eastern Mediterranean and explainable Artificial Intelligence (xAI) to diagnose model attributions. We implement DL models that capture the temporal and spatio‐temporal context, generalize well for extreme wildfires, and demonstrate improved performance over the traditional Fire Weather Index. Leveraging xAI, we identify the substantial contribution of wetness‐related variables and unveil the temporal focus of the models. The variability of the contribution of the input variables across wildfire events hints into different wildfire mechanisms. The presented methodology paves the way to more robust, accurate, and trustworthy data‐driven anticipation of wildfires. Plain Language Summary: Climate change drives the aggravation of wildfires globally. In this context, it is crucial to better predict and understand wildfires. Our work proposes methods to accurately forecast wildfire danger and hints into the mechanisms that drive it. We use Machine Learning (ML) to predict next‐day wildfire danger, using meteorological, vegetation, and human‐related data, improving the results of the widely used Fire Weather Index, which is based only on meteorological forecasts. We then look to understand what the models have learned and how the different conditioning factors contribute to the final prediction. We find that ML models focus more on soil moisture and relative humidity for their predictions. They also take into account the cumulative temperature and precipitation, as well as the last day's wind speed and the relative humidity in the 2 days preceding fire ignition. These associations lead to new, data‐driven ways to anticipate wildfires. Key Points: When forecasting fires that lead to large burned areas, Deep Learning models indicate higher predictive skill than the Fire Weather IndexSoil moisture, Normalized Difference Vegetation Index, and weather are the most important predictors; changes in their importance across events reveal diverse wildfire typesModels' explainability uncovers physically‐consistent associations and temporal dynamics of the fire drivers [ABSTRACT FROM AUTHOR]

Hydrological interactions between vegetation, soil, and topography are complex, and heterogeneous in semi‐arid landscapes. This along with data scarcity poses challenges for large‐scale modeling of vegetation‐water interactions. Here, we exploit metrics derived from daily Meteosat data over Africa at ca. 5 km spatial resolution for ecohydrological analysis. Their spatial patterns are based on Fractional Vegetation Cover (FVC) time series and emphasize limiting conditions of the seasonal wet to dry transition: the minimum and maximum FVC of temporal record, the FVC decay rate and the FVC integral over the decay period. We investigate the relevance of these metrics for large scale ecohydrological studies by assessing their co‐variation with soil moisture, and with topographic, soil, and vegetation factors. Consistent with our initial hypothesis, FVC minimum and maximum increase with soil moisture, while the FVC integral and decay rate peak at intermediate soil moisture. We find evidence for the relevance of topographic moisture variations in arid regions, which, counter‐intuitively, is detectable in the maximum but not in the minimum FVC. We find no clear evidence for wide‐spread occurrence of the "inverse texture effect" on FVC. The FVC integral over the decay period correlates with independent data sets of plant water storage capacity or rooting depth while correlations increase with aridity. In arid regions, the FVC decay rate decreases with canopy height and tree cover fraction as expected for ecosystems with a more conservative water‐use strategy. Thus, our observation‐based products have large potential for better understanding complex vegetation‐water interactions from regional to continental scales. Plain Language Summary: Local‐scale processes controlling vegetation dynamics under water limitation are highly uncertain at large scales, despite their importance on global carbon and water cycles. This is particularly pronounced in Africa due to the scarcity of ground measurements despite the importance of African ecosystems due to their contribution to global cycles and their services to population. In order to overcome this problem, we developed a set of metrics based on the fractional vegetation cover observed from the European geostationary satellite. The metrics, derived from the daily fractional vegetation cover data help diagnose ecohydrological processes thanks to their high spatiotemporal resolution. Initial analyses show consistent continental gradients in the metrics together with strong local variations and corroboration with different data sets from independent sources, in agreement with the literature. Key Points: We derive ecohydrological metrics from seasonal decays in geostationary satellite data of vegetation cover over Africa at 5 km resolutionThe metrics covary with soil moisture, contain information on water sources apart from precipitation and vegetation access by deep rootingLand surface models can benefit from these metrics by tailored evaluation, calibration, and parameterization of ecohydrological processes [ABSTRACT FROM AUTHOR]

Much uncertainty remains in measuring the inter‐annual and longer‐term dynamics of vegetation gross and net primary productivity (GPP, NPP) and the connected land carbon sink. Potential for better GPP estimation lies in newer satellite products representing different processes or vegetation states, but how they capture interannual GPP dynamics remains to be explored. Here, we differentiate shorter‐ and longer‐term vegetation dynamics and their drivers for several Earth‐observation‐based vegetation proxies and a GPP estimate using time series decomposition. We find that relations between proxies can significantly differ between time scales, along land cover and climate gradients. For GPP estimated at FLUXNET sites, seasonal and multiannual slopes differ by at least 25% for half of the cases investigated, indicating considerable mismatch if multiannual relations were derived from seasonal slopes. Considering time‐scale‐specific sensitivities between proxies of vegetation productivity may improve estimates of interannual variability in vegetation productivity. Plain Language Summary: How ecosystems will develop in the future is still difficult to predict, in part because the factors that influence ecosystems over several years or decades may differ from those influencing them in the course of a day or a year. Several satellites monitor vegetation growth on Earth and report, for example, leaf greenness or fluorescence. Fluorescence is often superior in tracking seasonal vegetation dynamics, but when it comes to understanding the long‐term imprint of climate, it remains to be clarified how different data products relate to each other. In this study, we compare how much changes in interannual and longer vegetation activity are captured by different satellite products. We find that the relationship between satellite proxies for vegetation differs between monthly, yearly and long‐term scales. These findings may help to better understand and predict vegetation growth and carbon uptake from the atmosphere in the future. Key Points: Contrary to seasonal time scales, it is poorly understood how different vegetation productivity proxies relate at interannual time scalesInterannual relations between vegetation proxies need to be considered: Seasonal relations do not generally hold at multiannual scaleTime‐scale‐specific slopes between vegetation proxies vary along gradients of tree cover, vegetation type, climate, and across FLUXNET sites [ABSTRACT FROM AUTHOR]

The responses of forest carbon dynamics to fluctuations in environmental conditions at a global scale remain elusive. Despite the understanding that favourable environmental conditions promote forest growth, these responses have been challenging to observe across different ecosystems and climate gradients. Based on a global annual time series of aboveground biomass (AGB) estimated from radar satellites between 1992 and 2018, we present forest carbon changes and provide insights on their sensitivities to environmental conditions across scales. Our findings indicate differences in forest carbon changes across AGB classes, with regions with carbon stocks of 50-125 MgC ha-1 depict the highest forest carbon gains and losses, while regions with 125-150 MgC ha-1 have the lowest forest carbon gains and losses in absolute terms. Net forest carbon change estimates show that the arc-of-deforestation and the Congo Basin were the main hotspots of forest carbon loss, while a substantial part of European forest gained carbon during the last three decades. Furthermore, we observe that changes in forest carbon stocks were systematically positively correlated with changes in forest cover fraction. At the same time, it was not necessarily the case with other environmental variables, such as air temperature and water availability at the bivariate level. We also used a model attribution method to demonstrate that atmospheric conditions were the dominant control of forest carbon changes (56% of the total study area) followed by water-related (29% of the total study area) and vegetation (15% of the total study area) conditions. Regionally, we find evidence that carbon gains from long-term forest growth covary with long-term carbon sinks inferred from atmospheric inversions. Our results describe the contributions from the atmosphere, water-related and vegetation conditions to forest carbon changes and provide new insights into the underlying mechanisms of the coupling between forest growth and the global carbon cycle. [ABSTRACT FROM AUTHOR]

Plant transpiration (T), biologically controlled movement of water from soil to atmosphere, currently lacks sufficient estimates in space and time to characterize global ecohydrology. Here we describe the Transpiration Estimation Algorithm (TEA), which uses both the signals of gross primary productivity and evapotranspiration (ET) to estimate temporal patterns of water use efficiency (WUE, i.e., the ratio between gross primary productivity and T) from which T is calculated. The method first isolates periods when T is most likely to dominate ET. Then, a Random Forest Regressor is trained on WUE within the filtered periods and can thus estimate WUE and T at every time step. Performance of the method is validated using terrestrial biosphere model output as synthetic flux data sets, that is, flux data where WUE dynamics are encoded in the model structure and T is known. TEA reproduced temporal patterns of T with modeling efficiencies above 0.8 for all three models: JSBACH, MuSICA, and CASTANEA. Algorithm output is robust to data set noise but shows some sensitivity to sites and model structures with relatively constant evaporation levels, overestimating values of T while still capturing temporal patterns. The ability to capture between‐site variability in the fraction of T to total ET varied by model, with root‐mean‐square error values between algorithm predicted and modeled T/ET ranging from 3% to 15% depending on the model. TEA provides a widely applicable method for estimating WUE while requiring minimal data and/or knowledge on physiology which can complement and inform the current understanding of underlying processes. Plain Language Summary: While it is widely known that plants need water to survive, exactly how much water plants in an ecosystem use is hard to quantify. However, many places have been measuring how much total water leaves an ecosystem, both the water plants use directly and the water that simply evaporates from the soil or the surfaces of leaves, using eddy covariance towers. These eddy covariance towers also measure the coming and going of carbon, such as the total amount of carbon taken up by photosynthesis. Here we present the idea that by using the signals from both photosynthesis and total water losses together, we can capture the water signal related to plants, namely, transpiration, using an algorithm called Transpiration Estimation Algorithm (TEA). To verify that TEA is working the way we expect, we test it out using artificial ecosystem simulations where transpiration and photosynthesis come from mathematical models. By thoroughly testing TEA, we have a better idea of how it will work in a real world situation, hopefully opening the door for a better understanding on how much water ecosystems are using and how it might affect our changing planet. Key Points: TEA is a method for extracting water use efficiency (WUE) dynamics from flux data with minimal assumptionsValidation shows that TEA is able to derive patterns of WUE and transpiration from three different modelsMethod is applicable to eddy covariance data sets, opening the door to wide‐scale transpiration estimates [ABSTRACT FROM AUTHOR]

Turnover concepts in state-of-the-art global vegetation models ( GVMs) account for various processes, but are often highly simplified and may not include an adequate representation of the dominant processes that shape vegetation carbon turnover rates in real forest ecosystems at a large spatial scale. Here, we evaluate vegetation carbon turnover processes in GVMs participating in the Inter-Sectoral Impact Model Intercomparison Project ( ISI- MIP, including HYBRID4, JeDi, JULES, LPJml, ORCHIDEE, SDGVM, and VISIT) using estimates of vegetation carbon turnover rate ( k) derived from a combination of remote sensing based products of biomass and net primary production ( NPP). We find that current model limitations lead to considerable biases in the simulated biomass and in k (severe underestimations by all models except JeDi and VISIT compared to observation-based average k), likely contributing to underestimation of positive feedbacks of the northern forest carbon balance to climate change caused by changes in forest mortality. A need for improved turnover concepts related to frost damage, drought, and insect outbreaks to better reproduce observation-based spatial patterns in k is identified. As direct frost damage effects on mortality are usually not accounted for in these GVMs, simulated relationships between k and winter length in boreal forests are not consistent between different regions and strongly biased compared to the observation-based relationships. Some models show a response of k to drought in temperate forests as a result of impacts of water availability on NPP, growth efficiency or carbon balance dependent mortality as well as soil or litter moisture effects on leaf turnover or fire. However, further direct drought effects such as carbon starvation (only in HYBRID4) or hydraulic failure are usually not taken into account by the investigated GVMs. While they are considered dominant large-scale mortality agents, mortality mechanisms related to insects and pathogens are not explicitly treated in these models. [ABSTRACT FROM AUTHOR]

Aim While fire is recognized as an integral part of the Earth system, the ability of humans to shape fire regimes both spatially and temporally remains poorly understood. Our goals were to identify the extent of fire regimes exhibiting two annual fire seasons and to investigate the environmental correlates of such regimes at the global scale. Location All areas of the globe exhibiting relevant fire activity, at 0.5º spatial resolution. Time period 2002-2012. Major taxa studied (not applicable). Methods The modality of fire seasonality at the global scale was classified using a 10-year record of satellite-derived fire activity and model fitting of circular statistical distributions. The main environmental correlates controlling global fire regimes were then analysed over bimodal and unimodal areas using the Kolmogorov-Smirnov test. Results About 25% of the global land surface with relevant fire activity has two significantly distinct fire seasons per year, with at least one of these seasons occurring under sub-optimal fire weather conditions. In these bimodal areas, population density and the fraction of fires occurring in actively managed land, especially in croplands and pastures, are significantly higher than in neighbouring unimodal areas. Results reveal that through these land-use and management practices humans have a strong influence on global patterns of fire seasonality. Main conclusions We identified a bimodal seasonality pattern, previously unreported at the global scale, and show that it reveals an anthropogenic fingerprint on fire regimes. Insights into where and when fire is actively employed as a land management tool enhance our understanding of the role of fire in the Earth system, and highlight the need to better understand how fire practices may change in the future. [ABSTRACT FROM AUTHOR]

Soil carbon (C) is a critical component of Earth system models (ESMs), and its diverse representations are a major source of the large spread across models in the terrestrial C sink from the third to fifth assessment reports of the Intergovernmental Panel on Climate Change (IPCC). Improving soil C projections is of a high priority for Earth system modeling in the future IPCC and other assessments. To achieve this goal, we suggest that (1) model structures should reflect real-world processes, (2) parameters should be calibrated to match model outputs with observations, and (3) external forcing variables should accurately prescribe the environmental conditions that soils experience. First, most soil C cycle models simulate C input from litter production and C release through decomposition. The latter process has traditionally been represented by first-order decay functions, regulated primarily by temperature, moisture, litter quality, and soil texture. While this formulation well captures macroscopic soil organic C (SOC) dynamics, better understanding is needed of their underlying mechanisms as related to microbial processes, depth-dependent environmental controls, and other processes that strongly affect soil C dynamics. Second, incomplete use of observations in model parameterization is a major cause of bias in soil C projections from ESMs. Optimal parameter calibration with both pool- and flux-based data sets through data assimilation is among the highest priorities for near-term research to reduce biases among ESMs. Third, external variables are represented inconsistently among ESMs, leading to differences in modeled soil C dynamics. We recommend the implementation of traceability analyses to identify how external variables and model parameterizations influence SOC dynamics in different ESMs. Overall, projections of the terrestrial C sink can be substantially improved when reliable data sets are available to select the most representative model structure, constrain parameters, and prescribe forcing fields. [ABSTRACT FROM AUTHOR]

Identifying the relative importance of climatic and other environmental controls on the interannual variability and trends in global land surface phenology and greenness is challenging. Firstly, quantifications of land surface phenology and greenness dynamics are impaired by differences between satellite data sets and phenology detection methods. Secondly, dynamic global vegetation models (DGVMs) that can be used to diagnose controls still reveal structural limitations and contrasting sensitivities to environmental drivers. Thus, we assessed the performance of a new developed phenology module within the LPJmL (Lund-Potsdam-Jena managed Lands) DGVM with a comprehensive ensemble of three satellite data sets of vegetation greenness and ten phenology detection methods, thereby thoroughly accounting for observational uncertainties. The improved and tested model allows us quantifying the relative importance of environmental controls on interannual variability and trends of land surface phenology and greenness at regional and global scales. We found that start of growing season interannual variability and trends are in addition to cold temperature mainly controlled by incoming radiation and water availability in temperate and boreal forests. Warming-induced prolongations of the growing season in high latitudes are dampened by a limited availability of light. For peak greenness, interannual variability and trends are dominantly controlled by water availability and land-use and land-cover change (LULCC) in all regions. Stronger greening trends in boreal forests of Siberia than in North America are associated with a stronger increase in water availability from melting permafrost soils. Our findings emphasize that in addition to cold temperatures, water availability is a codominant control for start of growing season and peak greenness trends at the global scale. [ABSTRACT FROM AUTHOR]

Understanding the environmental and biotic drivers of respiration at the ecosystem level is a prerequisite to further improve scenarios of the global carbon cycle. In this study we investigated the relevance of physiological phenology, defined as seasonal changes in plant physiological properties, for explaining the temporal dynamics of ecosystem respiration ( RECO) in deciduous forests. Previous studies showed that empirical RECO models can be substantially improved by considering the biotic dependency of RECO on the short-term productivity (e.g., daily gross primary production, GPP) in addition to the well-known environmental controls of temperature and water availability. Here, we use a model-data integration approach to investigate the added value of physiological phenology, represented by the first temporal derivative of GPP, or alternatively of the fraction of absorbed photosynthetically active radiation, for modeling RECO at 19 deciduous broadleaved forests in the FLUXNET La Thuile database. The new data-oriented semiempirical model leads to an 8% decrease in root mean square error ( RMSE) and a 6% increase in the modeling efficiency ( EF) of modeled RECO when compared to a version of the model that does not consider the physiological phenology. The reduction of the model-observation bias occurred mainly at the monthly time scale, and in spring and summer, while a smaller reduction was observed at the annual time scale. The proposed approach did not improve the model performance at several sites, and we identified as potential causes the plant canopy heterogeneity and the use of air temperature as a driver of ecosystem respiration instead of soil temperature. However, in the majority of sites the model-error remained unchanged regardless of the driving temperature. Overall, our results point toward the potential for improving current approaches for modeling RECO in deciduous forests by including the phenological cycle of the canopy. [ABSTRACT FROM AUTHOR]

Aim To infer a forest carbon density map at 0.01° resolution from a radar remote sensing product for the estimation of carbon stocks in Northern Hemisphere boreal and temperate forests. Location The study area extends from 30° N to 80° N, covering three forest biomes - temperate broadleaf and mixed forests ( TBMF), temperate conifer forests ( TCF) and boreal forests ( BFT) - over three continents ( North America, Europe and Asia). Methods This study is based on a recently available growing stock volume ( GSV) product retrieved from synthetic aperture radar data. Forest biomass and spatially explicit uncertainty estimates were derived from the GSV using existing databases of wood density and allometric relationships between biomass compartments (stem, branches, roots, foliage). We tested the resultant map against inventory-based biomass data from Russia, Europe and the USA prior to making intercontinent and interbiome carbon stock comparisons. Results Our derived carbon density map agrees well with inventory data at regional scales ( r2 = 0.70-0.90). While 40.7 ± 15.7 petagram of carbon ( Pg C) are stored in BFT, TBMF and TCF contain 24.5 ± 9.4 Pg C and 14.5 ± 4.8 Pg C, respectively. In terms of carbon density, we found 6.21 ± 2.07 kg C m−2 retained in TCF and 5.80 ± 2.21 kg C m−2 in TBMF, whereas BFT have a mean carbon density of 4.00 ± 1.54 kg C m−2. Indications of a higher carbon density in Europe compared with the other continents across each of the three biomes could not be proved to be significant. Main conclusions The presented carbon density and corresponding uncertainty map give an insight into the spatial patterns of biomass and stand as a new benchmark to improve carbon cycle models and carbon monitoring systems. In total, we found 79.8 ± 29.9 Pg C stored in northern boreal and temperate forests, with Asian BFT accounting for 22.1 ± 8.3 Pg C. [ABSTRACT FROM AUTHOR]

In this study we examined ecosystem respiration ( R) data from 104 sites belonging to FLUXNET, the global network of eddy covariance flux measurements. The goal was to identify the main factors involved in the variability of R: temporally and between sites as affected by climate, vegetation structure and plant functional type (PFT) (evergreen needleleaf, grasslands, etc.). We demonstrated that a model using only climate drivers as predictors of R failed to describe part of the temporal variability in the data and that the dependency on gross primary production (GPP) needed to be included as an additional driver of R. The maximum seasonal leaf area index (LAI) had an additional effect that explained the spatial variability of reference respiration (the respiration at reference temperature T=15 °C, without stimulation introduced by photosynthetic activity and without water limitations), with a statistically significant linear relationship ( r=0.52, P<0.001, n=104) even within each PFT. Besides LAI, we found that reference respiration may be explained partially by total soil carbon content (SoilC). For undisturbed temperate and boreal forests a negative control of total nitrogen deposition (N) on reference respiration was also identified. We developed a new semiempirical model incorporating abiotic factors (climate), recent productivity (daily GPP), general site productivity and canopy structure (LAI) which performed well in predicting the spatio-temporal variability of R, explaining >70% of the variance for most vegetation types. Exceptions include tropical and Mediterranean broadleaf forests and deciduous broadleaf forests. Part of the variability in respiration that could not be described by our model may be attributed to a series of factors, including phenology in deciduous broadleaf forests and management practices in grasslands and croplands. [ABSTRACT FROM AUTHOR]

Assumptions of steady-state conditions in biogeochemical modelling are often invoked because knowledge on the development status of the modelling domain is generally unavailable. Here, we investigate the role of vegetation pool sizes on nonequilibrium conditions through model-data integration approaches for a set of sites using eddy covariance CO2 flux data. The study is based on the Carnegie–Ames–Stanford Approach (CASA) model, modified (CASAG) in order to evaluate the sensitivity of simulated net ecosystem production (NEP) fluxes to vegetation pool sizes. The experimental design is based on the inverse model optimization of different parameter vectors performed at the measurement site level. Each parameter vector prescribes different simulation dynamics that embody different model structural assumptions concerning (non)steady-state conditions in vegetation and soil carbon pools. We further explore the potential of assimilating biometric constraints through the cost function for sites where in situ information on aboveground biomass or wood pools is available. The integration of biometric data yields marked improvements in the simulation of vegetation C pools compared to single constraints with eddy flux data. Overall, it is necessary to relax both vegetation and soil carbon pools for consistency with the observed data streams. Multiple constraints approaches also leads to variable model performance among the different experimental setups and model structures. We identify and assess the limitations of various model structures and the role of multiple constraints approaches for tackling issues of equifinality. These studies emphasize the need for establishing consistent data sets of fluxes and biometric data for successful model-data fusion. [ABSTRACT FROM AUTHOR]

We analyze the impacts of the steady state assumption on inverse model parameter retrieval from biogeochemical models. An inverse model parameterization study using eddy covariance CO2 flux data was performed with the Carnegie Ames Stanford Approach (CASA) model under conditions of strict and relaxed carbon cycle steady state assumption (CCSSA) in order to evaluate both the robustness of the model's structure for the simulation of net ecosystem carbon fluxes and the assessment of the CCSSA effects on simulations and parameter estimation. Net ecosystem production (NEP) measurements from several eddy covariance sites were compared with NEP estimates from the CASA model driven by local weather station climate inputs as well as by remotely sensed fraction of photosynthetically active radiation absorbed by vegetation and leaf area index. The parameters considered for optimization are directly related to aboveground and belowground modeled responses to temperature and water availability, as well as a parameter (η) that relaxed the CCSSA in the model, allowing for site level simulations to be initialized either as net sinks or sources. A robust relationship was observed between NEP observations and predictions for most of the sites through the range of temporal scales considered (daily, weekly, biweekly, and monthly), supporting the conclusion that the model structure is able to capture the main processes explaining NEP variability. Overall, relaxing CCSSA increased model efficiency (21%) and decreased normalized average error (-92%). Intersite variability was a major source of variance in model performance differences between fixed (CCSSAf) and relaxed (CCSSAr) CCSSA conditions. These differences were correlated with mean annual NEP observations, where an average increase in modeling efficiency of 0.06 per 100 g C m-2 a-1 (where a is years) of NEP is observed (α < 0.003). The parameter η was found to be a key parameter in the optimization exercise, generating significant model efficiency losses when removed from the initial parameter set and parameter uncertainties were significantly lower under CCSSAr. Moreover, modeled soil carbon stocks were generally closer to observations once the steady state assumption was relaxed. Finally, we also show that estimates of individual parameters are affected by the steady state assumption. For example, estimates of radiation-use efficiency were strongly affected by the CCSSAf indicating compensation effects for the inadequate steady state assumption, leading to effective and thus biased parameters. Overall, the importance of model structural evaluation in data assimilation approaches is thus emphasized. [ABSTRACT FROM AUTHOR]