Cooperative Institute for Research in Environmental Sciences

A winning proposal for the Innovative Research Program, 2005:

Anthropogenic carbon forcing in a simple climate model with coupled hydrological and carbon cycles

Investigators: V. Gupta, T. Chase, K. Nordstrom
CIRES' Center for the Study of Earth from Space

Introduction The Earth’s climate is a highly non-linear system operating through complex feedback loops. This system has been generally stable over billions of years, despite a slowly warming sun and other perturbations. Thus it can be viewed as dominated by long periods of negative feedbacks, punctuated by periods in which positive feedbacks change the system dramatically. Since these positive feedback regimes have not run away, it is clear that negative feedbacks must once again come to dominate after some time. More recently, the postulated climate change due to increased anthropogenic carbon forcing in the last century relies heavily on assumptions of positive feedbacks from the hydrological cycle, which generate large surface warming. The hydrological cycle is a pervasive and highly non-linear portion of the climate system. It is compelling, therefore, to assume that it plays a key role in both positive and negative feedbacks, and hence in regulating the climate. This has been demonstrated in a simple climate model by Nordstrom et al. [3], described below. The carbon cycle, another key component of the global climate system, is also dominated by both positive and negative feedbacks, an issue featured prominently in a recent distinguished lecture at CIRES [5]. On earth, these two cycles are coupled in a complex manner. Our broad objective in this proposal is to study the global climate as a coupled system incorporating both positive and negative feedbacks from both the hydrological and carbon cycles. In particular, we want to investigate the impact of anthropogenic carbon forcing on extreme floods and droughts in such a climate system.

Background

Recently, Nordstrom et al. [3] published a simple, nonlinear climate model called a Dynamical Area Fraction Model (DAFM). This represented a generalization of their first paper [2], which laid the basic theoretical framework and was published in the prestigious volume Scientists on Gaia II. The generalization in [3] contained a full surfacial energy balance; stratiform rainfall; a dynamic ocean; dynamically adjusting clouds with height-dependent albedo; fully dynamic ice caps; and two types of competing biota, representing one species of tree and one of grass. It also contained a static carbon cycle that did not respond to external forcing. All parameterizations were constructed on dynamic boxes, which lend the model its name.

Nordstrom et al. [3] performed two experiments with the DAFM. In both cases, results were surprising. In the first experiment, [3] varied the solar input over the range of .7 to 1.3 times the present day values for the full model and compared to an identical run with all water-related variables fixed to their steady-state values at present-day earthlike conditions. Despite the presence of some very powerful positive hydrological feedbacks like the ice-albedo and hydrological greenhouse feedbacks, the active hydrological cycle reduced system response by up to 50% (fig. 1).

In the second experiment, [3] tested the response of the model to changes in the optimal growing temperature for biota. In this case, despite an overwhelmingly powerful contribution to energy balance from the hydrological cycle at large, the biota were capable of adjusting the mean global temperature by approximately 1.6K. This is on the order of the change expected by the IPCC [1] in response to a doubling of CO2.

Objective In the earth climate system, there exist some very important couplings between the hydrological cycle and the carbon cycle. Chief among these are the sequestering of carbon in the surface waters of the terrestrial ocean; the sequestering and re-release of carbon due to the birth-death cycle in the planet’s biomass; and the behavior of stomatal density in response to local changes in carbon, which directly affects evapotranspiration in vegetated regions. These couplings have frequently been overlooked in the literature, as people have tended to focus on either the carbon cycle or the hydrological cycle. We propose to include first order parameterizations of a dynamic carbon cycle, as discussed in eg. [4], coupled to the fully dynamic hydrological cycle presented in [3] in order to study the effect on the global means of temperature and rainfall under changes in the anthropogenic carbon forcing.

Methodology

Simple coupled biosphere and hydrosphere parameterizations have already been developed as part of the DAFM described in [3]. Similar to a box model, a DAFM’s boxes are sized at some fraction of the global area and contain a representation of their local surfacial mean climate state. However, a DAFM’s boxes, or "area fractions", are allowed to dynamically expand and contract against one another following physical parameterizations of their own. Thus dynamic boundaries between ecosystems are designed into the model at a fundamental level.

Figure 1: Comparison of global mean temperature T vs. luminosity L in a DAFM integration with fixed hydrological cycle to a DAFM with a dynamic hydrological cycle. The carbon cycle is static.

A carbon cycle model has been studied in some detail in the model of [4] and was featured prominently in a recent talk [5]. Fluxes of carbon include temperature-dependent oceanic sequestration and outgassing from the atmosphere; removal of soil carbon when plants are born; sequestration of atmospheric carbon during plant life cycles; return of carbon to both soil and atmosphere upon plant death; and sequestration of soil carbon in ocean water due to runoff. Both trees and grasses will be considered C3 species for purposes of estimating carbon uptake, with tree mass representing a much larger effect on the carbon cycle.

References

[1] IPCC, 1996. Second Assessment Report. Climate Change 1995.

[2] Nordstrom, K.; V. Gupta, and T. Chase, 2003. Salvaging the Daisyworld parable under the Dynamic Area Fraction Framework. in Scientists Debate Gaia: the next century, eds. J. Miller, P. Boston, S. Schneider, E. Crist. Cambridge: MIT Press.

[3] Nordstrom, K.; V. Gupta, and T. Chase, 2005. Role of the hydrological cycle in regulating the climate of a simple Dynamic Area Fraction model. Nonlinear Processes in Geophysics, in press.

[4] Svirezhev, Y. and W. von Bloh, 1998. A zero-dimensional climate-vegetation model containing global carbon and hydrological cycle. Ecological Modelling, v. 106, p. 119-27

[5] Zachos, J., 2005. A Rapid Rise in Greenhouse Gas Concentrations 55 Million Years Ago: Lessons for the Future. CIRES Distinguished Lecture Series, March 11, 2005, CIRES Auditorium, University of Colorado, Boulder, CO.