Publication Date

2018-11-26

Availability

Open access

Embargo Period

2018-11-26

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PHD)

Department

Meteorology and Physical Oceanography (Marine)

Date of Defense

2018-11-02

First Committee Member

Brian J. Soden

Second Committee Member

Amy C. Clement

Third Committee Member

Paquita Zuidema

Fourth Committee Member

Tristan S. L'Ecuyer

Fifth Committee Member

Angeline G. Pendergrass

Abstract

Due to considerable societal implications, accurately simulating the response of the hydrological cycle to climate change is an important focus of climate research. Currently, climate model simulations predict the hydrological cycle will strengthen globally with global warming, but the magnitude of this change remains highly uncertain. Here we identify sources of this uncertainty by diagnosing individual components of the radiative changes that constrain the hydrological cycle response. We show that the differing influences of CO2 increases on long-term anthropogenic climate change versus short-term internal climate variability explains differences in precipitation sensitivity on these time scales. We investigate the response of the hydrological cycle to CO2 increases further by using radiative kernels to quantify radiative forcing and feedbacks. We show instantaneous radiative forcing contributes substantially to the magnitude and inter-model spread of both hydrological cycle responses and climate projections more broadly. This indicates that diversity in the implementation of radiative transfer among climate models serves as an important source of uncertainty in climate change projections. This is confirmed in a comparison of radiative kernels, which are shown to differ more than previously documented, in part due to differences in radiative transfer modeling. We also show that model bias in the distribution of climatological clouds is a substantial source of radiative kernel differences, which contributes to inconsistencies in estimates of cloud feedback. We introduce a new set of observation-based radiative kernels free from model bias in the base state, which will serve as a neutral radiative kernel for diagnosing radiative changes in models and observations.

Keywords

Hydrological Cycle; Feedback; Forcing; Radiative Transfer; Climate Models; Satellite Observations

Share

COinS