Detection and Attribution of Regional Precipitation Change with Granger Causality: Approaches to Short- and Long-Lived Climate Forcers
Abstract
We present methods for detecting and attributing the effects of short- and long-lived climate forcers (SLCFs and LLCFs) in a Granger causality framework. For mean and extreme precipitation over the United States from 1900 to present, we show that the most important SLCF and LLCFs are sulfur dioxide (SO2) and the well-mixed greenhouse gases (WMGHGs), respectively. SO2 is challenging given the dominance of fast (microphysics-mediated) effects of SO2 on rainfall, the significant spatial variability in its anthropogenic emissions and atmospheric burdens, and the short characteristic lifetimes of both SO2 and cloud systems. Moreover, neither detailed long-term, spatially resolved records nor internally consistent multi-model simulations of atmospheric SO2 concentrations are currently available. We demonstrate how to turn this apparent limitation into an opportunity by using the diversity of responses to SLCFs across the CMIP6 multi-model ensemble (MME) to ensure our D&A framework is insensitive to structural uncertainty associated with SLCFs. Since the response of precipitation to LLCFs is dominated by the lagged response of SSTs to these forcings, and since these time lags range between 1 and 3 decades due to the thermal inertia of the upper ocean, D&A for LLCFs should incorporate this lagged response. We conclude by presenting a simple physical model for the lagged response and estimates of the lag times obtained from the CMIP6 MME.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.A15H1742C