Understanding the role of climate uncertainty in analyses of coupled human-environment systems

Lafferty, David Conway

Permalink

https://hdl.handle.net/2142/124312 Copy

Description

Title

Understanding the role of climate uncertainty in analyses of coupled human-environment systems

Author(s)

Lafferty, David Conway

Issue Date

2024-04-18

Director of Research (if dissertation) or Advisor (if thesis)

Sriver, Ryan L

Doctoral Committee Chair(s)

Sriver, Ryan L

Committee Member(s)

• Dominguez, Francina
• Proistosescu, Cristian
• Keller, Klaus

Department of Study

Climate Meteorology & Atm Sci

Discipline

Atmospheric Sciences

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• climate change
• uncertainty
• soil moisture
• agriculture
• climate impacts
• downscaling
• bias-correction

Abstract

Anthropogenic climate change is altering the frequency and intensity of physical hazards across the world and magnifying the risks to many environmental systems that provide critical services to humanity. This has motivated an increased focus on understanding and more accurately quantifying these risks---a challenging task given the profound uncertainties associated with future climate change and its impact on natural systems. Robust long-term decision-making requires a sound understanding of these uncertainties, especially at local scales where adaptation strategies are often implemented. This dissertation contributes three case studies exploring the role of climate uncertainty in risk analyses of coupled human-environment systems. Particular attention is paid to the uncertainties associated with downscaling and bias-correction, two commonly-used post-processing techniques that aim to improve the resolution and accuracy, respectively, of climate model projections. The first case study focuses on characterizing climate impacts on maize production in the United States (U.S.). We use a multi-model ensemble of statistically bias-corrected and downscaled climate models, as well as the corresponding parent models from the Coupled Model Intercomparison Project Phase 5 (CMIP5), to drive a statistical panel model of U.S. maize yields that incorporates season-wide measures of temperature and precipitation. We analyze uncertainty in annual yield hindcasts, finding that the CMIP5 models considerably overestimate historical yield variability while the bias-corrected and downscaled versions underestimate the largest weather-induced yield declines. We also find large differences in projected yields and other decision-relevant metrics throughout this century, leaving stakeholders with modeling choices that require navigating trade-offs in resolution, historical accuracy, and projection confidence. The second case study looks globally at the uncertainties introduced by downscaling and bias-correction. Here, we perform a variance decomposition to partition uncertainty in global climate projections and quantify the relative importance of downscaling and bias-correction. We analyze simple climate metrics such as annual temperature and precipitation averages, as well as several indices of climate extremes. We find that downscaling and bias-correction often contribute substantial uncertainty to local decision-relevant climate outcomes, though our results are strongly heterogeneous across space, time, and climate metrics. Our results can provide guidance to impact modelers and decision-makers regarding the uncertainties associated with downscaling and bias-correction when performing local-scale analyses, as neglecting to account for these uncertainties may risk overconfidence relative to the full range of possible climate futures. Finally, the third case study examines the combined role of climate and hydrologic uncertainties in shaping future projections of agriculture-relevant soil moisture extremes, focusing on the central and eastern U.S. given its global relevance in maize and soybean production. We encode a simple water-balance soil moisture model in a differentiable programming framework that facilitates an efficient calibration procedure, and explore uncertainty in the model parameters by calibrating against several different observational datasets as well as using several different error metrics. We then convolve this parameter ensemble with a set of downscaled and bias-corrected climate projections. We find that accounting for soil parameter uncertainties can induce meaningful increases in the severity of future soil moisture extremes, with the choice of calibration dataset playing a significant role. Our results highlight the importance of considering combined hydrologic and climate uncertainties when constructing projections of decision-relevant hydroclimatic outcomes. Collectively, these case studies underscore the importance of a holistic accounting of uncertainty when analyzing coupled human-environment systems, particularly during risk assessments for which characterizing tail outcomes is crucial. They also elevate the role of downscaling and bias-correction, and suggest that end-users and decision-makers should consider this a central source of uncertainty in local climate projections. For a variety of metrics of climate change, the second case study can provide heuristic guidance regarding modeling decisions related to downscaled and bias-corrected data sources. The first and third case studies provide more focused insights related to climate impacts on U.S. agriculture.

Graduation Semester

2024-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2024 David Conway Lafferty

Owning Collections