Understanding Extremes in Biogeochemical Cycles and Their Driving Factors
Date Issued
August 2026
Author(s)
Kandel, Pragya
The University of Tennessee
Advisor(s)
Forrest M. Hoffman
Additional Advisor(s)
Mona Papes
Joshua S. Fu
Simone Tilmes
Abstract
Terrestrial ecosystems absorb a substantial fraction of anthropogenic CO2 emissions. Still, the future strength of this land carbon sink depends on how vegetation productivity and carbon cycling respond to warming, drought, and changing climate forcing. This dissertation examines carbon cycle extremes across models, scenarios, and regions to assess ecosystem vulnerability under climate change.
The first study quantifies positive and negative extremes in gross primary productivity (GPP) and net biome productivity (NBP) across multiple Earth system models under moderate warming, high-emission, and stratospheric aerosol intervention (SAI) scenarios from the Geoengineering Model Intercomparison Project (GeoMIP) G6sulfur experiment. GPP and NBP generally increase under higher radiative forcing, and SAI reduces the intensity and frequency of negative carbon cycle extremes in many regions. However, parts of North America and Europe show productivity losses and more frequent negative extremes even under SAI, with temperature, precipitation, evaporation, and solar radiation interacting to shape spatial patterns.
The second study investigates consequences of abrupt SAI termination on terrestrial biogeochemistry using the Energy Exascale Earth System Model (E3SM). SAI is applied under SSP5-8.5 and SSP5-3.4-OS through 2150, with abrupt termination imposed in 2100. Termination produces rapid warming that elevates respiration, suppresses productivity, and intensifies drought, heatwaves, and wildfire. Under SSP5-8.5, the land shifts from a net sink to a source; under SSP5-3.4-OS it remains a sink but weakens. Tropical forests benefit from continued SAI, whereas non-arctic grasses, C4 grasses, and temperate deciduous trees are vulnerable to termination. Machine learning attribution identifies temperature, moisture, vapor pressure deficit, and radiation as dominant drivers of post-termination GPP and NBP.
The third study evaluates satellite constraints on vegetation water stress, using Evaporative Stress Index data from ECOSTRESS to assess drought susceptibility across grasslands, shrublands, and forests in California from 2018 to 2026. Evaporative stress varies with vegetation type and region, reflecting differences in rooting depth, canopy structure, phenology, and soil moisture access, and responds distinctly to temperature, soil moisture, and precipitation.
Together, these studies address gaps in understanding how terrestrial carbon climate feedbacks respond to future warming, stratospheric geoengineering, and severe drought.
Degree
Doctor of Philosophy
Major
Data Science and Engineering
Embargo Date
August 1, 2027