Biology subjects

Dietze, M. C.

Publications and source records attributed to Dietze, M. C..

8 recordsLinked to original sources

bioRxiv2026

Harmonizing Repeated Multi-Platform Optical and LiDAR Observations for Continental-Scale Terrestrial Carbon Cycle Data Assimilation.

Accurate monitoring of terrestrial carbon has attracted considerable attention due to the impacts of rapid climate change, disturbance, and land use on carbon storage and the growth of new global carbon markets and reporting requirements. While uncertainties in C estimates remain nontrivial, crucial opportunities exist to reduce uncertainty by rigorously harmonizing models and monitoring through data assimilation (DA). LiDAR has received considerable attention for its potential to inform carbon monitoring, and while there have been a number of papers on how to initialize models using LiDAR data, it has yet to be integrated into DA systems. In this study, we integrate GEDI LiDAR into an established DA system that already assimilates optical observations of aboveground biomass (AGB) and leaf area index (LAI), microwave soil moisture (SM), and ground-based soil organic carbon (SOC) to constrain C cycling processes across the entire North American continent. Notably, this is the first time both optical and LiDAR AGB products have been iteratively harmonized within a DA system. Results show that by assimilating both optical and LiDAR observations, we reduced the uncertainty in woody biomass by 63.7% relative to unconstrained model simulations and 18% relative to our previous DA system, and achieved an average RMSE reduction of 38.3 kg/m2 across five AGB products. Importantly, LiDAR reduces DA biases in high AGB regions where optical AGB estimates were exhibiting saturation. Analysis further identified the leading factors driving spatial error patterns within the DA system to be data quality, cloud coverage, and elevation. Overall, by bridging the spatial and temporal limitations of individual satellite products, this approach offers a practical, generalized workflow for high-resolution terrestrial carbon MRV in complex, dynamic ecosystems.

ecology↗

bioRxiv2026

Mapping the North American Terrestrial Carbon Cycle: A Process-based Reanalysis Using State Data Assimilation (SDA)

AbstractThe ability to accurately assess ecosystem C budgets across scales from individual sites to continents is essential for C accounting, management, and ultimately mitigating climate change. State data assimilation (SDA) provides a framework for harmonizing observations with models, while robustly accounting for and reducing multiple sources of uncertainty. In this study, we employed a hybrid SDA framework that combines process-based terrestrial biosphere modeling, hierarchical Bayesian inference, and machine learning to harmonize bottom-up and remotely-sensed data streams for 8,000 pre-selected 1km2 locations across North America within a hybrid structure. Combining bottom-up soils data (SoilGrids) with spectral (MODIS and Landsat) and microwave (SMAP) remote sensing helps constrain the major C and water stocks through space and time. Machine learning is used both to identify and correct systematic errors in the process model (SIPNET) and to interpolate the pre-selected locations onto a 1km grid, making it computationally feasible to generate annual ensemble maps of the North American carbon budget. Furthermore, the uncertainties for each variable were reduced compared to those from observations or models alone. Spatiotemporal analysis showed a slight decrease in aboveground biomass (AGB) across the western US, a loss of leaf area across the boreal, and a slight greening of the Alaskan tundra. The uncertainty trends suggest a significant reduction in the uncertainty about soil organic carbon (SOC), the largest C reservoir. Validation results show that we accurately estimate C pools, compared to the assimilated data streams and held-out observations of AGB from GEDI, ICESat-2, and the US FIA, and SOC from the ISCN network. Our ML-debiasing algorithm further improved the accuracy of major C pools (AGB, SOC). In general, our continental SDA framework will facilitate global C MRV (monitoring, reporting, and verification) by providing accurate and precise C-cycle estimates, along with their corresponding spatiotemporal uncertainties.

ecology↗

bioRxiv2025

Incorporating weather and host abundance in an iterative subseasonal-to-interannual ecological forecast system for Ixodes scapularis, the vector of Lyme disease

Forecasting the population dynamics of disease vectors is critical for mitigating the risks of vector-borne diseases under a changing climate. We evaluate an iterative Bayesian forecast model of black-legged tick (Ixodes scapularis) phenology and population dynamics at near-term to interannual (e.g. 12 month) scales. The black-legged tick is the vector of Borrelia burgdorferi, the causative agent of human Lyme disease. Our forecasts consistently outperformed seasonal null models based on historical day-of-year averages, particularly during peak questing periods when disease risk is highest. Iterative data assimilation improved forecast performance over time, demonstrating the ability to adaptively learn about climate-driven shifts in demographic parameters, and reinforcing the value of long-term data to support management. Weather and climate variables emerged as the dominant predictors of nymph survival, with daily maximum temperature displacing humidity as the strongest predictor as the iterative forecast cycle evolved over time. Short-term forecasts driven by local weather observations were more accurate than those relying on seasonal climate forecasts, highlighting the importance of fine-scale weather dynamics and data for subannual predictions. At interannual scales, seasonal climate forecasts and vertebrate host (mouse) abundance were important for maintaining strong predictive skill in forecasting nymphal tick abundance, which is often used as a proxy for risk of human exposure to tick-borne disease, but forecasts were largely unaffected by larval abundance. Investment in monitoring efforts should prioritize observations of the nymphal stage to reduce forecast uncertainty. These results offer a path forward for operationalizing ecological forecasts of tick populations under environmental change and underscore the importance of adaptive, process-based models for managing tick-borne disease risk in a changing climate. Data availability statementThe data that support the findings of this study are openly available. The data from the Cary Institute of Ecosystem Studies are located on Figshare. Specifically, the tick data (Ostfeld and Oggenfuss 2023) is at doi.org/10.25390/caryinstitute.23611374.v1, the mouse data (Ostfeld et al. 2024) is at doi.org/10.25390/caryinstitute.25742778.v1, and the meteorological data (Kelly, 2020) is at doi.org/10.25390/caryinstitute.11553219.v6. The NMME system data are openly available at https://www.earthsystemgrid.org/search.html?Project=NMME. Code for this manuscript (Foster, 2025) is available from Zenodo at doi.org/10.5281/zenodo.17161317.

ecology↗

bioRxiv2025

The uncertainty partitioning of the terrestrial C cycling over CONUS NEON sites using model-data fusion

Accurate inventories of terrestrial carbon pools and fluxes are crucial for understanding ecosystem processes, tracking climate change impacts, and meeting the monitoring, reporting, and verification (MRV) requirements in international treaties and voluntary carbon markets. In meeting this need, the fusion of process-based modeling, field data, and remote sensing observations has the potential to provide more accurate and precise estimates than each alone. However, as the number of data constraints on a system increases, different sources of information can interact with each other in complex ways across space, time, and processes. In this study, we undertake a value-of-information analysis to assess the contribution of different observations to reducing carbon cycle uncertainties across pools, fluxes, and spatial domains within the PEcAn carbon cycle data assimilation system. We used a novel block-based Tobit Gamma Ensemble Filter to assimilate four synergistic data constraints, MODIS leaf area index, Landtrendr aboveground biomass, SMAP soil moisture, and SoilGrids soil organic C, into a process-based ecosystem model (SIPNET) at 39 National Ecological Observatory Network sites across the contiguous U.S. from 2012 to 2021. Results showed that not only did we greatly reduce uncertainty among the directly constrained pools but many observations were able to share information across variables and space. These indirect constraints helped identify synergies and conflicts among data streams and across space, which provides insights for further constraining carbon inventories. Overall, soil carbon remains the largest source of uncertainty in the overall carbon budget due to both its large size and limited observational constraints.

ecology↗

bioRxiv2023

A trigger may not be necessary to cause senescence in deciduous broadleaf forests

Plant phenological changes drive many ecosystem processes and are a key ecological indicator of climate change. Traditional models represent the onset of autumn leaf senescence, or color change, as a threshold response triggered by the accumulation of cold temperatures and declining day lengths, but the physiological mechanisms behind plant thermal memory and chilling thresholds remain elusive. Here we show that we can predict senescence in forest canopies by dynamically modeling daily greenness as the "memoryless" balance between chlorophyll synthesis, linearly-related to daily temperatures and day lengths, and breakdown. Indeed, summer-only data can be used to estimate breakdown and synthesis rates that in many cases successfully predict senescence at both calibration and validation sites. This mechanistic model shows that neither a trigger nor a physiological memory of coldness is necessary for senescence to start. These findings suggest that the start of senescence is not an irreversible transition, but a continuum of decreasing greenness where concurrent environmental conditions determine the rate of initial senescence. Furthermore, this emphasizes that in order to predict how senescence will shift in response to global change we likely need to focus on understanding the impacts on chlorophyll synthesis. Significance statementPlant phenology is a leading ecological indicator of climate change and has wide-ranging ecological and climatological impacts. Our findings here contradict the key assumption that senescence in deciduous broadleaf trees is actively triggered only when a threshold of cumulative cooling in combination of other stressors is reached. Instead we show that canopy greenness can be modeled as a passive process - balancing chlorophyll breakdown and temperature- and photoperiod-controlled synthesis - and still predict a rapid decline in greenness during senescence. This work is particularly important because it emphasizes that in order to understand climate change induced shifts in senescence, we need to focus on impacts on chlorophyll synthesis.

ecology↗

bioRxiv2021

Liana optical traits increase tropical forest albedo and reduce ecosystem productivity

Lianas are a key growth form in tropical forests. Their lack of self-supporting tissues and their vertical position on top of the canopy make them strong competitors of resources. A few pioneer studies have shown that liana optical traits differ on average from those of colocated tree. Those trait discrepancies were hypothesized to be responsible for the competitive advantage of lianas over trees. Yet, in the absence of reliable modelling tools, it is impossible to unravel their impact on the forest energy balance, light competition and on the liana success in Neotropical forests. To bridge this gap, we performed a meta-analysis of the literature to gather all published liana leaf optical spectra, as well as all canopy spectra measured over different levels of liana infestation. We then used a Bayesian data assimilation framework applied to two radiative transfer models (RTMs) covering the leaf and canopy scales to derive tropical tree and liana trait distributions, which finally informed a full dynamic vegetation model. According to the RTMs inversion, lianas grew thinner, more horizontal leaves with lower pigment concentrations. Those traits made the lianas particularly efficient at light interception and completely modified the forest energy balance and its carbon cycle. While forest albedo increased by 14% in the shortwave, light availability was dramatically reduced in the understory (-30% of the PAR radiation) and soil temperature decreased by 0.5{degrees}C. Those liana-specific traits were also responsible for a significant reduction of tree (-19%) and ecosystem (-7%) gross primary productivity (GPP) while lianas benefited from them (their GPP increased by +27%). This study provides a novel mechanistic explanation to the increase in liana abundance, new evidence of the impact of structural parasitism on forest functioning, and paves the way for the evaluation of the large-scale impacts of woody vines on forest biogeochemical cycles.

plant biology↗

bioRxiv2021

Capturing site-to-site variability through Hierarchical Bayesian calibration of a process-based dynamic vegetation model

Process-based ecosystem models help us understand and predict ecosystem processes, but using them has long involved a difficult choice between performing data- and labor-intensive site-level calibrations or relying on general parameters that may not reflect local conditions. Hierarchical Bayesian (HB) calibration provides a third option that frees modelers from assuming model parameters to be completely generic or completely site-specific and allows a formal distinction between prediction at known calibration sites and "out-of-sample" prediction to new sites. Here, we compare calibrations of a process-based dynamic vegetation model to eddy-covariance data across 12 temperate deciduous Ameriflux sites fit using either site-specific, joint cross-site, or HB approaches. To be able to apply HB to computationally demanding process-based models we introduce a novel emulator-based HB calibration tool, which we make available through the PEcAn community cyberinfrastructure. Using these calibrations to make predictions at held-out tower sites, we show that the joint cross-site calibration is falsely over-confident because it neglects parameter variability across sites and therefore underestimates variance in parameter distributions. By showing which parameters show high site-to-site variability, HB calibration also formally gives us a structure that can detect which process representations are missing from the models and prioritize errors based on the magnitude of the associated uncertainty. For example, in our case-study, we were able to identify large site-to-site variability in the parameters related to the temperature responses of respiration and photosynthesis, associated with a lack of thermal acclimation and adaptation in the model. Moving forward, HB approaches present important new opportunities for statistical modeling of the spatiotemporal variability in modeled parameters and processes that yields both new insights and improved predictions.

ecology↗

bioRxiv2018

Does the leaf economic spectrum hold within plant functional types? A Bayesian multivariate trait meta-analysis

We investigated whether global leaf economic relationships are also present within plant functional types (PFTs), and the extent to which this hierarchical structure can be used to constrain trait estimates. We developed a hierarchical multivariate Bayesian model that assumes separate means and covariance structures within and across PFTs and fit this model to seven leaf traits from the TRY database related to leaf morphology, biochemistry, and photosynthetic metabolism. Trait correlations were generally consistent in direction within and across PFTs, and consistent with predictions of the leaf economic spectrum. However, correlation strength varied substantially across PFTs indicating that leaf economic relationships within PFTs are often confounded by the unique physiology of certain plant types or environmental conditions in certain biomes. Leveraging covariance in multivariate models reduced uncertainties in mean trait estimates, particularly for undersampled trait-PFT combinations. However, additional constraint from the across-PFT hierarchy was limited.\n\nData accessibilityThe R code and ancillary data for running these analyses is publicly available online via the Open Science Framework at https://osf.io/w8y73/. The TRY data request used for this analysis has been archived at http://try-db.org, and can be retrieved by providing the TRY data request ID (#1584). Alternatively, the exact preformatted data used in this analysis are available on request to Alexey Shiklomanov (alexey.shiklomanov@pnnl.gov).

ecology↗