The impact of feedbacks involving water vapor on shifts in the ITCZ

written by Spencer K. Clark on 2018-05-01

It is well-known that shifts in the intertropical convergence zone (ITCZ) can be induced by hemispherically asymmetric forcings, i.e. when the Northern or Southern Hemisphere is warmed or cooled relative to the other. The extent of a shift of the ITCZ in response to a given forcing magnitude and/or location has been shown to be sensitive to climate model physics, in particular the parameterization of clouds and their subsequent radiative effects. Less attention has been given, however, to the role of feedbacks involving water vapor in setting the sensitivity of ITCZ shifts to hemispherically asymmetric forcings. In this study, we make use of the Geophysical Fluid Dynamics Laboratory's (GFDL's) idealized moist aquaplanet general circulation model coupled to a full radiative transfer code to conduct a large suite of simulations to address this problem.

Our approach is to induce a shift in the ITCZ by introducing a reduction in incoming solar radiation in either the tropical or extratropical Northern Hemisphere in two different model configurations. In the first configuration, the radiation code sees the active water vapor tracer in the model (we call this the "interactive water vapor" configuration); in the second configuration the radiation code instead sees a zonally and hemispherically symmetric pattern of water vapor taken from a control simulation with no perturbation to the solar insolation (we call this the "prescribed water vapor" configuration). We find that for a given forcing magnitude and location, the ITCZ shifts about twice as far in the interactive water vapor configuration as it does in the prescribed water vapor configuration. Based on analysis using energy flux equator theory for the latitude of the ITCZ, we conclude that the difference in sensitivity is due mainly to the outgoing-longwave-radiation-inhibiting impact of the water-vapor-rich ITCZ shifting into the already warmer hemisphere. This is illustrated in the schematic above. There Q, the green curve, represents the net column heating; the shaded gradient above the falling rain represents the water vapor field seen by the radiation code; and the red arrows represent outgoing longwave radiation, with longer arrows representing greater values. More details can be found in our manuscript published in the Journal of Climate.

Uncertainty in future lightning flash density

written by Spencer K. Clark on 2017-05-02

Lightning is a source of tropospheric NOx, which is important for ozone chemistry. In addition lightning serves as the primary natural ignition source for forest and grass fires. A number of parameterizations for lightning flash density have been proposed in the last thirty years. Here we implemented 8 lightning parameterizations in NCAR's CAM5, and tested their performance in the present day in simulations with meteorology nudged to match that in the ERA-Interim reanalysis, and also tested what they projected for future lightning flash density in fully-coupled simulations using the RCP4.5 and RCP8.5 emissions scenarios.

We found that the two parameterizations that perform best in the present-day (one based on cloud top height, and one based on cold cloud depth) have substantially different projections for future lightning, with the cloud top height scheme projecting an increase of over 12% per Kelvin increase in global mean temperature and the cold cloud depth scheme projecting an increase of just under 4% per Kelvin. See more details in our published manuscript.

Episodicity of fire aerosol emissions

written by Spencer Clark on 2015-12-15

Somewhat inspired by then-recent progress in the prognostic fire modeling realm (Kloster et al., 2010; Li et al., 2012) this study was formulated to help guide the priorities in the next generation of fire models in ESMs. A particular aspect of fires that might be difficult to simulate prognostically or in prescribed emissions in an ESM is their episidocity (from aerosol modeling perspective, how frequently and intensely emissions are released). Commonly spatial variation in fire days in presecribed emissions schemes looks like what is shown in panel (b) in the figure below.

In this project we investigate how important the frequency of emissions (i.e. number of fire days per month) is in simulating the climate impacts of fire aerosols (in terms of radiative forcing and impacts on the mean meridional circulation in the tropics). We find that at a coarse gridscale (1.9 x 2.5 degree) fire frequency is most important to resolve in the high latitudes in the Northern Hemisphere, where fire is most infrequent in observations.

The bottom line, however, is that we find that in most locations at this coarse a grid-scale, the uncertainty in climate impacts due to approximations in emission episodicity is smaller than the uncertainty due to other factors such as quantity, composition, or optical properties; the quantitative details can be found within our published manuscript.

© Copyright (2021) by Spencer Clark. Adapted from the hyde theme; built using Lektor.