Model-data Fusion Research Articles

Leveraging 35 years of <i>Pinus taeda</i> research in the southeastern US to constrain forest carbon cycle predictions: regional data assimilation using ecosystem experiments

Abstract. Predicting how forest carbon cycling will change in response to climate change and management depends on the collective knowledge from measurements across environmental gradients, ecosystem manipulations of global change factors, and mathematical models. Formally integrating these sources of knowledge through data assimilation, or model–data fusion, allows the use of past observations to constrain model parameters and estimate prediction uncertainty. Data assimilation (DA) focused on the regional scale has the opportunity to integrate data from both environmental gradients and experimental studies to constrain model parameters. Here, we introduce a hierarchical Bayesian DA approach (Data Assimilation to Predict Productivity for Ecosystems and Regions, DAPPER) that uses observations of carbon stocks, carbon fluxes, water fluxes, and vegetation dynamics from loblolly pine plantation ecosystems across the southeastern US to constrain parameters in a modified version of the Physiological Principles Predicting Growth (3-PG) forest growth model. The observations included major experiments that manipulated atmospheric carbon dioxide (CO2) concentration, water, and nutrients, along with nonexperimental surveys that spanned environmental gradients across an 8.6 × 105 km2 region. We optimized regionally representative posterior distributions for model parameters, which dependably predicted data from plots withheld from the data assimilation. While the mean bias in predictions of nutrient fertilization experiments, irrigation experiments, and CO2 enrichment experiments was low, future work needs to focus modifications to model structures that decrease the bias in predictions of drought experiments. Predictions of how growth responded to elevated CO2 strongly depended on whether ecosystem experiments were assimilated and whether the assimilated field plots in the CO2 study were allowed to have different mortality parameters than the other field plots in the region. We present predictions of stem biomass productivity under elevated CO2, decreased precipitation, and increased nutrient availability that include estimates of uncertainty for the southeastern US. Overall, we (1) demonstrated how three decades of research in southeastern US planted pine forests can be used to develop DA techniques that use multiple locations, multiple data streams, and multiple ecosystem experiment types to optimize parameters and (2) developed a tool for the development of future predictions of forest productivity for natural resource managers that leverage a rich dataset of integrated ecosystem observations across a region.

Constraining a complex biogeochemical model for CO<sub>2</sub> and N<sub>2</sub>O emission simulations from various land uses by model–data fusion

Abstract. This study presents the results of a combined measurement and modelling strategy to analyse N2O and CO2 emissions from adjacent arable land, forest and grassland sites in Hesse, Germany. The measured emissions reveal seasonal patterns and management effects, including fertilizer application, tillage, harvest and grazing. The measured annual N2O fluxes are 4.5, 0.4 and 0.1 kg N ha−1 a−1, and the CO2 fluxes are 20.0, 12.2 and 3.0 t C ha−1 a−1 for the arable land, grassland and forest sites, respectively. An innovative model–data fusion concept based on a multicriteria evaluation (soil moisture at different depths, yield, CO2 and N2O emissions) is used to rigorously test the LandscapeDNDC biogeochemical model. The model is run in a Latin-hypercube-based uncertainty analysis framework to constrain model parameter uncertainty and derive behavioural model runs. The results indicate that the model is generally capable of predicting trace gas emissions, as evaluated with RMSE as the objective function. The model shows a reasonable performance in simulating the ecosystem C and N balances. The model–data fusion concept helps to detect remaining model errors, such as missing (e.g. freeze–thaw cycling) or incomplete model processes (e.g. respiration rates after harvest). This concept further elucidates the identification of missing model input sources (e.g. the uptake of N through shallow groundwater on grassland during the vegetation period) and uncertainty in the measured validation data (e.g. forest N2O emissions in winter months). Guidance is provided to improve the model structure and field measurements to further advance landscape-scale model predictions.

Data assimilation strategies for volcano geodesy

Ground deformation observed using near-real time geodetic methods, such as InSAR and GPS, can provide critical information about the evolution of a magma chamber prior to volcanic eruption. Rapid advancement in numerical modeling capabilities has resulted in a number of finite element models targeted at better understanding the connection between surface uplift associated with magma chamber pressurization and the potential for volcanic eruption. Robust model-data fusion techniques are necessary to take full advantage of the numerical models and the volcano monitoring observations currently available. In this study, we develop a 3D data assimilation framework using the Ensemble Kalman Filter (EnKF) approach in order to combine geodetic observations of surface deformation with geodynamic models to investigate volcanic unrest. The EnKF sequential assimilation method utilizes disparate data sets as they become available to update geodynamic models of magma reservoir evolution. While the EnKF has been widely applied in hydrologic and climate modeling, the adaptation for volcano monitoring is in its initial stages. As such, our investigation focuses on conducting a series of sensitivity tests to optimize the EnKF for volcano applications and on developing specific strategies for assimilation of geodetic data. Our numerical experiments illustrate that the EnKF is able to adapt well to the spatial limitations posed by GPS data and the temporal limitations of InSAR, and that specific strategies can be adopted to enhance EnKF performance to improve model forecasts. Specifically, our numerical experiments indicate that: (1) incorporating additional iterations of the EnKF analysis step is more efficient than increasing the number of ensemble members; (2) the accuracy of the EnKF results are not affected by initial parameter assumptions; (3) GPS observations near the center of uplift improve the quality of model forecasts; (4) occasionally shifting continuous GPS stations to provide variability in the locations of observations results in better model predictions than utilizing fixed locations when the number of available instruments is limited; (5) spotty InSAR data coverage on the flanks of a volcano due to topographic shadows and/or atmospheric interference does not adversely impact the effectiveness of EnKF if the available coverage is >50%; and (6) snow or glacial obstruction in the center of uplift can adversely impact EnKF forecasts. By utilizing these strategies, we conclude that the EnKF is an effective sequential model-data fusion technique for assimilating multiple geodetic observations to forecast volcanic activity at restless volcanoes.

A 3D Scan Model and Thermal Image Data Fusion Algorithms for 3D Thermography in Medicine.

Objectives At present, medical thermal imaging is still considered a mere qualitative tool enabling us to distinguish between but lacking the ability to quantify the physiological and nonphysiological states of the body. Such a capability would, however, facilitate solving the problem of medical quantification, whose presence currently manifests itself within the entire healthcare system. Methods A generally applicable method to enhance captured 3D spatial data carrying temperature-related information is presented; in this context, all equations required for other data fusions are derived. The method can be utilized for high-density point clouds or detailed meshes at a high resolution but is conveniently usable in large objects with sparse points. Results The benefits of the approach are experimentally demonstrated on 3D thermal scans of injured subjects. We obtained diagnostic information inaccessible via traditional methods. Conclusion Using a 3D model and thermal image data fusion allows the quantification of inflammation, facilitating more precise injury and illness diagnostics or monitoring. The technique offers a wide application potential in medicine and multiple technological domains, including electrical and mechanical engineering.

基于模型数据融合的中国温带和亚热带典型森林生态系统碳通量模拟

PDF HTML阅读 XML下载 导出引用 引用提醒 基于模型数据融合的中国温带和亚热带典型森林生态系统碳通量模拟 DOI: 10.5846/stxb201510132066 作者: 作者单位: 中国科学院地理科学与资源研究所,中国科学院地理科学与资源研究所 作者简介: 通讯作者: 中图分类号: 基金项目: 中国科学院战略性先导科技专项（XDA05050600）；国家科技支撑计划（2013BAC03B00）；中国科学院科研信息化“科技领域云”项目（XXH12503-05-03） Carbon flux simulation of typical temperate and subtropical forest ecosystems in China based on model-data fusion approach Author: Affiliation: Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Science,Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Science Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:生态系统碳循环过程对水分响应的研究已成为全球变化关注的焦点问题之一。基于长白山温带针阔混交林与千烟洲亚热带人工针叶林观测站2003-2009年生长季的碳通量（NEE）和气象观测数据，综合考虑水分对光合、呼吸作用的影响，构建不同的NEE模型，并应用模型数据融合方法优化模型参数、遴选最适模型，系统分析了水分因子对不同森林生态系统碳循环的影响。结果表明：（1）优化后的模型参数均能被NEE实测数据较好约束。长白山生长季的光合、呼吸参数值均高于千烟洲，未考虑空气饱和水汽压差（VPD）的模型高估了千烟洲温度敏感性参数（Q10）值、低估了千烟洲基础呼吸速率参数（BR）值；（2）仅考虑VPD对光合作用影响的模型是长白山生长季碳通量模拟的最优模型，但模拟精度提高不显著。不同模型间碳通量组分模拟结果差异较小；（3）考虑VPD和土壤含水量对光合、呼吸作用共同影响的模型是千烟洲生长季碳通量模拟的最优模型，并且显著提高了模拟精度。未考虑水分的模型在生长季高估了总生态系统生产力（GEP）总量2.0%（21.85 g C/m2），同时更大幅度地高估了生态系统呼吸（RE）总量4.4%（38.02 g C/m2），从而导致NEE总量低估于实测值7.8%（18.55 g C/m2）。 Abstract:Moisture effect on the carbon balance of terrestrial ecosystems is a key issue in global change research. It is crucial to accurately analyze the response of terrestrial ecosystem carbon cycle to moisture. However, the carbon flux models responding to environmental factors rarely consider the moisture effects on photosynthesis and respiration simultaneously; meanwhile there are still large uncertainties in model structures and parameters. Thus, this study was designed to (1) choose the optimal carbon flux model with accurate parameters for different ecosystems through model-data fusion approach, reducing the uncertainties of modeled results; (2) systematically analyze the influence of water factors on carbon flux simulation, including gross ecosystem productivity (GEP), ecosystem respiration (RE) and net ecosystem exchange (NEE). To consider the effects of moisture on both GEP and RE, we developed four different NEE models. Then, based on carbon flux and meteorological data during growing season from 2003 to 2009 in Changbaishan temperate mixed forest (CBS) and Qianyanzhou subtropical coniferous plantation (QYZ), Markov Chain Monte Carlo was employed to estimate model parameters, and Bayesian Information Criterion was applied to choose the optimal model for two forest ecosystems. The results showed that (1) the posterior values of model parameters were normally distributed, indicating that the parameters were well constrained by NEE. Photosynthetic and respiratory parameter values of CBS were higher than those of QYZ during the growing season. The model without vapor pressure deficit (VPD) overestimated the value of temperature sensitivity (Q10) and underestimated the value of basal respiration rate (BR) in QYZ; (2) the model considering VPD only was the optimal model for CBS,but its performance was not improved much. The modeled flux components were similar among the four models; (3) the model considering both VPD and soil water content (Sw) was the optimal model for QYZ, and its performance was improved significantly. The model ignored water factors overestimated 2% (21.85 g C/m2) of the total GEP, and 4.4% (38.02 g C/m2) of the total RE, and therefore, underestimated 7.8% (18.55 g C/m2) of the total measured NEE during the growing season. 参考文献 相似文献 引证文献

The potential benefit of using forest biomass data in addition to carbon and water flux measurements to constrain ecosystem model parameters: Case studies at two temperate forest sites

Biomass as a resource, and as a vulnerable carbon pool, is a key variable to diagnose the impacts of global changes on the terrestrial biosphere, and therefore its proper description in models is crucial. Model-Data Fusion (MDF) or data assimilation methods are useful tools in improving ecosystem models that describe interactions between vegetation and atmosphere. We use a MDF method based on a Bayesian approach, in which data are combined with a process model in order to provide optimized estimates of model parameters and to better quantify model uncertainties, whilst taking into account prior information on the parameters. With this method we are able to use multiple data streams, which allows us to simultaneously constrain modeled variables at site level across different temporal scales. In this study both high frequency eddy covariance flux measurements of net CO2 and evapotranspiration (ET), and low frequency biometric measurements of total aboveground biomass and the annual increment (which includes all compartments), are assimilated with the ORCHIDEE model version “AR5” at a beech (Hesse) and a maritime pine (Le Bray) forest site using four to five years of flux data and nine years of biomass data. When assimilating the observed aboveground annual biomass increment (AGB_inc) together with net CO2 and ET flux, the RMSE of modelled AGB_inc was reduced from the a priori estimates by 37% at Hesse and 69% at Le Bray, without reducing the fit to the net CO2 and ET that can be achieved when assimilating flux data alone. Assimilating biomass increment data also provides insight in the performance of the allocation scheme of the model. Comparison with detailed site-based measurements at Hesse showed that the optimization reduced positive biases in the model, for example in fine root and leaf production. We also investigated how to use stand-scale total aboveground biomass in optimization (AGB_tot). However, this study demonstrated that assimilating AGB_tot measurements in the ORCHIDEE-AR5 model lead to some inconsistencies, particularly for the annual dynamics of the AGB_inc, partly because this version of the model lacked a realistic representation of forest stand processes including management and disturbances.

Environmental control of carbon allocation matters for modelling forest growth.

We aimed to evaluate the importance of modulations of within-tree carbon (C) allocation by water and low-temperature stress for the prediction of annual forest growth with a process-based model. A new C allocation scheme was implemented in the CASTANEA model that accounts for lagged and direct environmental controls of C allocation. Different approaches (static vs dynamic) to modelling C allocation were then compared in a model-data fusion procedure, using satellite-derived leaf production estimates and biometric measurements at c. 104 sites. The modelling of the environmental control of C allocation significantly improved the ability of CASTANEA to predict the spatial and year-to-year variability of aboveground forest growth along regional gradients. A significant effect of the previous year's water stress on the C allocation to leaves and wood was reported. Our results also are consistent with a prominent role of the environmental modulation of sink demand in the wood growth of the studied species. Data available at large scales can inform forest models about the processes driving annual and seasonal C allocation. Our results call for a greater consideration of C allocation drivers, especially sink-demand fluctuations, for the simulations of current and future forest productivity with process-based models.

A Model–Data Fusion Approach for Predicting Cover Crop Nitrogen Supply to Corn

One potential benefit of cover crops (CCs) is that N mineralization from decomposing CC residues may reduce the N fertilizer requirement of a subsequent crop, but predicting this credit remains a significant challenge. This study used a model–data fusion approach to calibrate a model of CC residue N mineralization and pre‐emptive competition for soil NO3− that occurs during CC growth to predict the yield response of an unfertilized corn (Zea mays L.) crop. The model was calibrated with a data set of 199 observations from four CC experiments in central Pennsylvania. The most parsimonious model explained 82% of the variation in corn yield response. Parameters representing the C humification coefficients for decomposed residues from winterkilled (εwk = 0.00) and winter‐hardy (εwh = 0.40) CCs suggest that all winterkilled CCs resulted in net N mineralization, probably due to the longer period of time for decomposition of winterkilled residues. However, the yield response per unit of potentially mineralized N was greater for winter‐hardy CCs (αwh = 0.034 with tillage, αwh = 0.020 with no‐till) than for winterkilled CCs (αwk = 0.0084), probably due to the improved synchrony between corn N demand and the decomposition of winter‐hardy CC residues relative to winterkilled residues. Pre‐emptive competition for soil NO3− led to a reduction in the corn yield response. Because the model is based on ecological processes and can be calibrated with data sets from simple field experiments, the model–data fusion approach could be widely used to guide adaptive management of CCs and N fertilizer applications in a subsequent corn crop.Core Ideas A simple model predicts how cover crops affect N availability to the next corn crop. The model highlights the ecological controls on N supply from cover crops. Site‐specific cover crop measurements can guide adaptive N management using the model. Regional model calibration could be achieved with easily collected data.

Review: Current and emerging methods for catchment-scale modelling of recharge and evapotranspiration from shallow groundwater

A review is provided of the current and emerging methods for modelling catchment-scale recharge and evapotranspiration (ET) in shallow groundwater systems. With increasing availability of data, such as remotely sensed reflectance and land-surface temperature data, it is now possible to model groundwater recharge and ET with more physically realistic complexity and greater levels of confidence. The conceptual representation of recharge and ET in groundwater models is critical in areas with shallow groundwater. The depth dependence of recharge and vegetation water-use feedback requires additional calibration to fluxes as well as heads. Explicit definition of gross recharge vs. net recharge, and groundwater ET vs. unsaturated zone ET, in preparing model inputs and reporting model results is necessary to avoid double accounting in the water balance. Methods for modelling recharge and ET include (1) use of simple surface boundary conditions for groundwater flow models, (2) coupling saturated groundwater models with one-dimensional unsaturated-zone models, and (3) more complex fully-coupled surface-unsaturated-saturated conceptualisations. Model emulation provides a means for including complex model behaviours with lower computational effort. A precise ET surface input is essential for accurate model outputs, and the model conceptualisation depends on the spatial and temporal scales under investigation. Using remote sensing information for recharge and ET inputs in model calibration or in model–data fusion is an area for future research development. Improved use of uncertainty analysis to provide probability bounds for groundwater model outputs, understanding model sensitivity and parameter dependence, and guidance for further field-data acquisition are also areas for future research.

Spatial representativeness and uncertainty of eddy covariance carbon flux measurements for upscaling net ecosystem productivity to the grid scale

Eddy covariance (EC) measurements are often used to validate net ecosystem productivity (NEP) estimated from satellite remote sensing data and biogeochemical models. However, EC measurements represent an integrated flux over their footprint area, which usually differs from respective model grids or remote sensing pixels. Quantifying the uncertainties of scale mismatch associated with gridded flux estimates by upscaling single EC tower NEP measurements to the grid scale is an important but not yet fully investigated issue due to limited data availability as well as knowledge of flux variability at the grid scale. The Heihe Watershed Allied Telemetry Experimental Research (HiWATER) Multi-Scale Observation Experiment on Evapotranspiration (MUSOEXE) built a flux observation matrix that includes 17 EC towers within a 5km×5km area in a heterogeneous agricultural landscape in northwestern China, providing an unprecedented opportunity to evaluate the uncertainty of upscaling due to spatial representative differences at the grid scale. Based on the HiWATER-MUSOEXE data, this study evaluated the spatial representativeness and uncertainty of EC CO2 flux measurements for upscaling to the grid scale using a scheme that combines a footprint model and a model-data fusion method. The results revealed the large spatial variability of gross primary productivity (GPP), ecosystem respiration (Re), and NEP within the study site during the growing season from 10 June to 14 September 2012. The variability of fluxes led to high variability in the representativeness of single EC towers for grid-scale NEP. The systematic underestimations of a single EC tower may reach 92(±11)%, 30(±11)%, and 165(±150)% and the overestimations may reach 25(±14)%, 20(±13)%, and 40(±33)% for GPP, Re, and NEP, respectively. This finding suggests that remotely sensed NEP at the global scale (e.g., MODIS products) should not be validated against single EC tower data in the case of heterogeneous surfaces. Any systematic bias should be addressed before upscaling EC data to grid scale. Otherwise, most of the systematic bias may be propagated to grid scale due to the scale dependence of model parameters. A systematic bias greater than 20% of the EC measurements can be corrected effectively using four indicators proposed in this study. These results will contribute to the understanding of spatial representativeness of EC towers within a heterogeneous landscape, to upscaling carbon fluxes from the footprint to the grid scale, to the selection of the location of EC towers, and to the reduction in the bias of NEP products by using an improved parameterization scheme of remote-sensing driven models, such as VPRM.

Bayesian Model–Data Fusion for Mechanistic Postearthquake Damage Assessment of Building Structures

AbstractThis paper presents a probabilistic framework for estimating seismic-induced damage in partially instrumented buildings. The proposed framework uses acceleration measurements at a limited number of stories and Bayesian filtering to estimate the response at all stories. The paper compares four Bayesian filters: the extended, unscented, and ensemble Kalman filters, and the particle filter. The estimated response throughout the building serves as input to a damage model that yields an estimate of structural damage and its uncertainty at all stories. The methodology is numerically verified in an elastoplastic 5-story shear building and in a 10-story inelastic moment-resisting frame under various types of model errors and minimal instrumentation. It was found that under ideal and mild modeling error conditions, the proposed methodology provides consistent estimates of damage and its uncertainty.

Short-term wind power forecasting using model structure selection and data fusion techniques

This paper proposes a novel strategy to forecast the short-term wind power using model structure selection in combined with data fusion technique. The available inputs are usually treated as an integrated one for the predictive modeling, which ignores the fusion of the different types of inputs. This results the relationship between the multi-inputs and desired outputs are not effectively reflected in forecasting procedure. Moreover, the performances of the various types of data fusion methods are susceptible to the wind power distribution and critical factors such as smoothness, overlapping, the intrinsic amount of information evaluation and the number of neighbor points. These outlined factors can result the lower generality ability of the proposed methods due to the insufficient optimization of the model structure. So this paper presents short-term wind power forecasting method using model structure selection technique in combined with four representative fusions of dimensionality reduction methods to optimize the model structure, promote the computational efficiency and improve the forecasting accuracy. Experimental evaluation based on the real data from the wind farm in Jiangsu province is given to verify the effectiveness of the proposed method by comparing the traditional techniques.

A multi-data stream assimilation framework for the assessment of volcanic unrest

Active volcanoes pose a constant risk to populations living in their vicinity. Significant effort has been spent to increase monitoring and data collection campaigns to mitigate potential volcano disasters. To utilize these datasets to their fullest extent, a new generation of model-data fusion techniques is required that combine multiple, disparate observations of volcanic activity with cutting-edge modeling techniques to provide efficient assessment of volcanic unrest. The purpose of this paper is to develop a data assimilation framework for volcano applications. Specifically, the Ensemble Kalman Filter (EnKF) is adapted to assimilate GPS and InSAR data into viscoelastic, time-forward, finite element models of an evolving magma system to provide model forecasts and error estimations. Since the goal of this investigation is to provide a methodological framework, our efforts are focused on theoretical development and synthetic tests to illustrate the effectiveness of the EnKF and its applicability in physical volcanology. The synthetic tests provide two critical results: (1) a proof of concept for using the EnKF for multi dataset assimilation in investigations of volcanic activity; and (2) the comparison of spatially limited, but temporally dense, GPS data with temporally limited InSAR observations for evaluating magma chamber dynamics during periods of volcanic unrest. Results indicate that the temporally dense information provided by GPS observations results in faster convergence and more accurate model predictions. However, most importantly, the synthetic tests illustrate that the EnKF is able to swiftly respond to data updates by changing the model forecast trajectory to match incoming observations. The synthetic results demonstrate a great potential for utilizing the EnKF model-data fusion method to assess volcanic unrest and provide model forecasts. The development of these new techniques provides: (1) a framework for future applications of rapid data assimilation and model development during volcanic crises; (2) a method for hind-casting to investigate previous volcanic eruptions, including potential eruption triggering mechanisms and precursors; and (3) an approach for optimizing survey designs for future data collection campaigns at active volcanic systems.

Применение пространственно-разнесённых радиолокационных станций для решения задачи селекции

Рассматриваются решения вопросов селекции с использованием информации от многопозиционных радиолокационных систем в автономном режиме для обеспечения рубежа принятия классификационного решения на участке с сильно разрежённой атмосферой. Описываются основные и дополнительные признаки селекции, математический аппарат принятия решения и модель системы объединения данных. Проводится анализ целесообразности использования информации от многопозиционных радиолокационных систем в автономном режиме.

Statistical uncertainty of eddy covariance CO2 fluxes inferred using a residual bootstrap approach

High-frequency eddy-covariance measurements of net ecosystem CO2 exchange (NEE) with the atmosphere are valuable resources for model parameterization, calibration, and validation. However, uncertainties in measured data, i.e., data gaps and inherent random errors, create problems for researchers attempting to quantify uncertainties in model projections of terrestrial ecosystem carbon cycling. Here, we demonstrate that a model-data fusion method (residual bootstrap) produces defensible annual NEE sums, through mimicking the behavior of random errors, filling missing values, and simulating gap-filling biases. This study estimated annual NEE sums for 53 site-years based on nine eddy-covariance tower sites in the USA, and found that our annual estimates were, in most cases, comparable in magnitude with those obtained from AmeriFlux gap-filled data. Additionally, compared to the AmeriFlux standardized gap-filling, our approach provides better NEE estimates for moderate to longer, and more frequent, data gaps. Annual accumulated uncertainties in NEE at the 95% confidence level were ±30gCm−2year−1 for evergreen needleleaf forests; ±60gCm−2year−1 for deciduous broadleaf forests; and ±80gCm−2year−1 for croplands. The residual bootstrap performed worst when gap length was greater than one month or data exclusion greater than 90% during the growing season, common to other gap-filling techniques. However, this study produced robust results for most site years when monthly data coverage during the growing season is not extremely low. We therefore suggest that the inclusion of NEE uncertainty estimates and better estimation for moderate to longer, and more frequent, data gaps as provided by the residual bootstrap approach can be beneficial for ecosystem model evaluation.

Constraining ecosystem carbon dynamics in a data-limited world: integrating ecological "common sense" in a model–data fusion framework

Abstract. Many of the key processes represented in global terrestrial carbon models remain largely unconstrained. For instance, plant allocation patterns and residence times of carbon pools are poorly known globally, except perhaps at a few intensively studied sites. As a consequence of data scarcity, carbon models tend to be underdetermined, and so can produce similar net fluxes with very different parameters and internal dynamics. To address these problems, we propose a series of ecological and dynamic constraints (EDCs) on model parameters and initial conditions, as a means to constrain ecosystem variable inter-dependencies in the absence of local data. The EDCs consist of a range of conditions on (a) carbon pool turnover and allocation ratios, (b) steady-state proximity, and (c) growth and decay of model carbon pools. We use a simple ecosystem carbon model in a model–data fusion framework to determine the added value of these constraints in a data-poor context. Based only on leaf area index (LAI) time series and soil carbon data, we estimate net ecosystem exchange (NEE) for (a) 40 synthetic experiments and (b) three AmeriFlux tower sites. For the synthetic experiments, we show that EDCs lead to an overall 34% relative error reduction in model parameters, and a 65% reduction in the 3 yr NEE 90% confidence range. In the application at AmeriFlux sites all NEE estimates were made independently of NEE measurements. Compared to these observations, EDCs resulted in a 69–93% reduction in 3 yr cumulative NEE median biases (–0.26 to +0.08 kg C m−2), in comparison to standard 3 yr median NEE biases (–1.17 to −0.84 kg C m−2). In light of these findings, we advocate the use of EDCs in future model–data fusion analyses of the terrestrial carbon cycle.

Differences in scale-dependent, climatological variation of mean areal precipitation based on satellite and radar-gauge observations

Summary This study compares the scale-dependent variation in hourly Mean Areal Precipitation (MAP) derived from a satellite (S) and a radar-gauge (R) Quantitative Precipitation Estimate (QPE), and seeks to explain the S–R differences on the basis of errors in the satellite QPE. This study employs an analytical framework to estimate the coefficient of variation (CV) of MAP for window sizes ranging from 4 km to 512 km, using the rainfall fields of the CPC MORPHing (CMORPH) satellite QPE and a radar-gauge Multisensor QPE (MQPE) over five domains centered in Texas, Oklahoma and New Mexico. CV values based on the analytical framework are first corroborated using empirical estimates. Then, S–R differences in CV are analyzed to determine the contributions of the S–R differences from empirical fractional coverage (FC) and spatial correlograms. Subsequently, sensitivity analyses are performed to isolate the impacts of false detections and long-term, magnitude-dependent bias in CMORPH on the inaccuracies in FC and correlograms. The results are stratified by domain and season (winter and summer) to highlight the impacts of differential accuracy of CMORPH under diverse rainfall regimes. Our analyses reveal that CMORPH-based CV tends to plateau at larger window sizes (referred to as critical window size, or CWS), and is broadly higher in magnitude. The mechanisms underlying the CV differences, however, differ between winter and summer. Over the winter, CMORPH suffers from severe underdetection, which yields suppressed FC across window sizes. This underestimation of FC, together with the lack of resolution of internal rainfall structure by CMORPH, leads to an magnification of both CWS and the magnitude of CV. By contrast, over the summer, widespread false detections in CMORPH lead to inflated FC, which tends to suppress CWS but this effect is outweighed by the opposing impacts of inflated outer and inner scales (i.e., distance parameters of indicator and conditional correlograms). Moreover, it is found that introducing false detection to MQPE via a simple expansion scheme is effective in increasing the FC and inner scale in tandem, and that histogram differences are a rather minor contributor to the S–R difference in inner scale. The implications of the findings for disaggregating climate model projection and data fusion are discussed.

Model–data fusion across ecosystems: from multisite optimizations to global simulations

Abstract. This study uses a variational data assimilation framework to simultaneously constrain a global ecosystem model with eddy covariance measurements of daily net ecosystem exchange (NEE) and latent heat (LE) fluxes from a large number of sites grouped in seven plant functional types (PFTs). It is an attempt to bridge the gap between the numerous site-specific parameter optimization works found in the literature and the generic parameterization used by most land surface models within each PFT. The present multisite approach allows deriving PFT-generic sets of optimized parameters enhancing the agreement between measured and simulated fluxes at most of the sites considered, with performances often comparable to those of the corresponding site-specific optimizations. Besides reducing the PFT-averaged model–data root-mean-square difference (RMSD) and the associated daily output uncertainty, the optimization improves the simulated CO2 balance at tropical and temperate forests sites. The major site-level NEE adjustments at the seasonal scale are reduced amplitude in C3 grasslands and boreal forests, increased seasonality in temperate evergreen forests, and better model–data phasing in temperate deciduous broadleaf forests. Conversely, the poorer performances in tropical evergreen broadleaf forests points to deficiencies regarding the modelling of phenology and soil water stress for this PFT. An evaluation with data-oriented estimates of photosynthesis (GPP – gross primary productivity) and ecosystem respiration (Reco) rates indicates distinctively improved simulations of both gross fluxes. The multisite parameter sets are then tested against CO2 concentrations measured at 53 locations around the globe, showing significant adjustments of the modelled seasonality of atmospheric CO2 concentration, whose relevance seems PFT-dependent, along with an improved interannual variability. Lastly, a global-scale evaluation with remote sensing NDVI (normalized difference vegetation index) measurements indicates an improvement of the simulated seasonal variations of the foliar cover for all considered PFTs.

Process models and model-data fusion in dendroecology

Dendrochronology (i.e. the study of annually dated tree-ring time series) has proved to be a powerful technique to understand tree-growth. This paper intends to show the interest of using ecophysiological modeling not only to understand and predict tree-growth (dendroecology) but also to reconstruct past climates (dendroclimatology). Process models have been used for several decades in dendroclimatology, but it is only recently that methods of model-data fusion have led to significant progress in modeling tree-growth as a function of climate and in reconstructing past climates. These model-data fusion (MDF) methods, mainly based on the Bayesian paradigm, have been shown to be powerful for both model calibration and model inversion. After a rapid survey of tree-growth modeling, we illustrate MDF with examples based on series of Southern France Aleppo pines and Central France oaks. These examples show that if plants experienced CO2 fertilization, this would have a significant effect on tree-growth which in turn would bias the climate reconstructions. This bias could be extended to other environmental non-climatic factors directly or indirectly affecting annual ring formation and not taken into account in classical empirical models, which supports the use of more complex process-based models. Finally, we conclude by showing the interest of the data assimilation methods applied in climatology to produce climate re-analyses.

Structural damage identification based on self-fitting ARMAX model and multi-sensor data fusion

Statistical time series methods have proven to be a promising technique in structural health monitoring, since it provides a direct form of data analysis and eliminates the requirement for domain transformation. Latest research in structural health monitoring presents a number of statistical models that have been successfully used to construct quantified models of vibration response signals. Although a majority of these studies present viable results, the aspects of practical implementation, statistical model construction and decision-making procedures are often vaguely defined or omitted from presented work. In this article, a comprehensive methodology is developed, which essentially utilizes an auto-regressive moving average with exogenous input model to create quantified model estimates of experimentally acquired response signals. An iterative self-fitting algorithm is proposed to construct and fit the auto-regressive moving average with exogenous input model, which is capable of integrally finding an optimum set of auto-regressive moving average with exogenous input model parameters. After creating a dataset of quantified response signals, an unlabelled response signal can be identified according to a ‘closest-fit’ available in the dataset. A unique averaging method is proposed and implemented for multi-sensor data fusion to decrease the margin of error with sensors, thus increasing the reliability of global damage identification. To demonstrate the effectiveness of the developed methodology, a steel frame structure subjected to various bolt-connection damage scenarios is tested. Damage identification results from the experimental study suggest that the proposed methodology can be employed as an efficient and functional damage identification tool.