Global sensitivity analysis for terrestrial carbon cycle simulations under present and future climate conditions
| dc.contributor.author | Suruli Nagarajan, Raj Deepak | |
| dc.contributor.supervisor | Seiler, Christian | |
| dc.contributor.supervisor | Monahan, Adam Hugh | |
| dc.date.accessioned | 2025-06-16T19:14:44Z | |
| dc.date.available | 2025-06-16T19:14:44Z | |
| dc.date.issued | 2025 | |
| dc.degree.department | School of Earth and Ocean Sciences | |
| dc.degree.level | Doctor of Philosophy PhD | |
| dc.description.abstract | In this dissertation, I assessed the sensitivity of the land surface carbon, water, and energy fluxes to variations in model input parameters when simulated by a land surface model (LSM). The terrestrial biosphere currently uptakes approximately 30% of anthropogenic CO2 emissions. LSMs project that the biosphere will continue to take up carbon till early to mid 22nd century, making it a net carbon sink. These carbon sink projections are important for improving the future carbon predictions and informing mitigation strategies. But, there are substantial uncertainties in the strength of the simulated sink. For instance, the spread in the inter-model carbon sink is 1 to 3.2 PgC yr-1 during 2014-2023 (Global Carbon Budget), and 2 to 7 PgC yr-1 for the end of the 21st century (Intergovernmental Panel on Climate Change's Sixth Assessment Report). Some of the mentioned uncertainties in the simulated carbon sink arises from parameter uncertainties. While parameter tuning can help reduce these uncertainties, optimizing all input parameters in a complex, non-linear LSM is computationally prohibitive. Identifying influential parameters and understanding their influence on the model output(s) is an essential step before tuning the parameters. The influence of parameter uncertainties on the terrestrial carbon cycle output variables can be assessed using global sensitivity analysis (GSA). In this dissertation, I apply a two-step GSA to the output variables simulated by an LSM, the Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC). This research is divided into three parts, each applying GSA to CLASSIC output variables under different conditions. The questions asked are: (1) Is there a common set of parameters that substantially influence the majority of ecosystem output variables simulated at an eddy covariance site?, (2) Which parameters substantially affect the uncertainty of the historical carbon sink for different biomes?, and (3) Which parameters substantially affect the uncertainty of future carbon sink projections for different biomes? Through GSA's first step, a coarse sampled screening test, I found that only 15–17% of input parameters show appreciable influence on any of the simulated output variables. Through the second fine sampled quantitative analysis, I further narrowed this subset, and identified between two and 15 parameters as the most influential for different output variables and statistical measures. The influential parameters varied depending on the meteorological forcing used. The maximum rate at which CO2 is used during photosynthesis (vmax) and the loss of light along the canopy depth (kn) are the most recurring influential parameters across all forcing scenarios, and statistical measures. Additionally, other photosynthetic parameters, as well as those related to rooting and phenology, play an important role when CLASSIC is forced using reanalysis and Earth system model data. The sensitivity of the terrestrial carbon sink to the uncertainty in $vmax$ reduces by the end of the 21st century. In many cases the analysis is unable to rank the most influential parameters because of large sampling variations in the sensitivity indices. GSA is a stepping stone before performing model optimization. However, the computational demands of GSA are substantial. In this study, performing GSA for just seven grid cells required approximately 25 CPU years. Scaling such analyses to a global level using the full model would be computationally prohibitive. However, advancements in machine learning and emulator-based approaches present a promising alternative for GSA and optimization efforts, drastically reducing computational costs by requiring fewer input-output simulations than the full model. These innovations could enable large-scale assessments of parameter uncertainty, ultimately leading to more robust predictions of the terrestrial carbon sink, which will help in the shaping of better mitigation efforts. | |
| dc.description.scholarlevel | Graduate | |
| dc.identifier.bibliographicCitation | SN, Raj Deepak, Christian Seiler, and Adam H. Monahan. "A global sensitivity analysis of parameter uncertainty in the CLASSIC model." Atmosphere-Ocean (2024): 1-13, https://doi.org/10.1080/07055900.2024.2396426 | |
| dc.identifier.uri | https://hdl.handle.net/1828/22386 | |
| dc.language | English | eng |
| dc.language.iso | en | |
| dc.rights | Available to the World Wide Web | |
| dc.subject | Global sensitivity analysis | |
| dc.subject | Land surface model | |
| dc.subject | Terrestrial carbon cycle | |
| dc.subject | Net biome productivity | |
| dc.subject | Climate modeling | |
| dc.subject | Future climate conditions | |
| dc.subject | Terrestrial carbon sink | |
| dc.title | Global sensitivity analysis for terrestrial carbon cycle simulations under present and future climate conditions | |
| dc.type | Thesis |