Impacts of substrate quality on soil organic carbon decomposition dynamics: Modeling representation and data-model integration

Thumbnail Image
Yi_iastate_0097E_21358.pdf (10.37 MB)

File Embargoed Until: (2026-06-05)
Yi, Bo
Major Professor
Lu, Chaoqun
Lu, Chaoqun
Hall, Steven J.
McDaniel, Marshall D.
Hu, Guiping
Vanloocke, Andrew
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Is Version Of
Ecology, Evolution, and Organismal Biology
As an important carbon pool, soil contains more carbon than the sum of plants and atmosphere and plays a significant role in affecting climate system. Even a small change in soil organic carbon (SOC) pool could potentially alter the land carbon cycling and carbon budget. The balance of organic carbon in soil results from the interplay between input gains (e.g., plant and animal residues) and losses (e.g., erosion and CO2 emissions). The composition of litterfall and soil carbon pools determines carbon decomposition rates, affecting carbon density and stability. Mechanistic models are critical in bridging theory and measurements to provide quantitative assessment and decision support to inform SOC management. However, current models either lack adequate representation or oversimplify the impact of substrate quality, such as lignin in litterfall, soil carbon components, and plant residue compositions, on the SOC decomposition dynamics. This study focuses on two main research questions: 1) What are the roles of lignin, different SOC components (specifically, POC: particulate organic carbon and MAOC: mineral-associated organic carbon in this case), and plant residues in conventional and diversified cropping systems in affecting the SOC decomposition rate? 2) How can we enhance modeling representations and reduce uncertainties in estimating SOC decomposition by integrating measurement data into the modeling framework? The first study addressed long-standing questions in ecology: Is lignin an important control of soil organic carbon decomposition? If so, by what mechanisms, how can we represent these in a carbon decomposition model? We evaluated the impacts of different hypotheses of lignin’s role on modeling estimates through a process-based model and 18-month lab-incubated SOC decomposition data with soil samples collected from 20 NEON sites across the continental US. We reconciled competing hypotheses to re-produce the different decomposition patterns (with and without lagged peak) for lignin. Our study indicated the need to precisely represent the role of lignin and soil geochemical and microbial characteristics for accurately estimating SOC decomposition trajectories and attributing sources of carbon fluxes in earth-system modeling. The second study explored the controlling factors for two key modeling parameters (i.e., decay rate and carbon use efficiency) for POC and MAOC decomposition. Our study found distinct controls between them, underscoring the need to differentiate SOC components and incorporate the impacts of local soil properties on their decomposability in mechanistic models. We also call for future modeling research to integrate soil geochemical and microbial characteristics to enhance the projections of SOC decomposition and CO2 emissions from soils. The third study examined the effects of a diversified cropping system on soil C sequestration. We quantified SOC decomposition rates in various cropping systems utilizing a 20-year agricultural experiment study, lab incubation with stable isotope measurements, and isotope-enabled mechanistic models. Our result revealed that diversifying cropping does not enhance soil carbon sequestration but significantly increases nitrogen supply capacity, due to increased SOC decomposition from fast-cycling soil C pools (e.g., POC). We elucidated a critical trade-off between soil organic carbon storage and nitrogen supply within diversified cropping contexts. Taken together, these findings offer quantitative insights on how to reduce SOC modeling uncertainties derived from mechanism understanding, parameterization, and modeling structure (i.e., isotope vs non-isotope). It suggests that a critical knowledge gap in carbon modeling could be filled by 1) representing the substrate quality (e.g., lignin in litterfall, POC and MAOC, plant residue from various cropping systems) and their impacts on SOC decomposition, and 2) integrating the controls of local soil geochemical and microbial properties that have been ignored in previous modeling work.