Roger Pielke Sr. (Colorado State) has a blog (Climate Science) that gives his personal perspective on climate change issues. In it, he has made clear that he feels that apart from greenhouse gases, other climate forcings (the changes that affect the energy balance of the planet) are being neglected in the scientific discussion. Specifically, he feels that many of these other forcings have sufficient ‘first-order’ effects to prevent a clear attribution of recent climate change to greenhouse gases.
In general, I heartily agree – other forcings are important, even essential, for understanding observed climate variability and, as a community, we are only just starting to get to grips with some of the more complicated effects. Obviously, though, not all forcings are of the same magnitude (either globally or regionally) and so it is useful to separate the ‘first-order’ forcings from those that are relatively minor. But what exactly is ‘first-order’ and what is not?
It is helpful to distinguish forcings that are important in the global mean, from those which might be important locally but not have much impact for ‘global warming’. A good metric for the importance of a global forcing is the radiative forcing (i.e. the global mean radiation impact at the tropopause for an instantaneous change – the so-called “instantaneous forcing”). There are other definitions (e.g. the “adjusted forcing” where stratospheric temperatures are allowed to adjust), but for the purpose of this article these differences are not that important (for those who are interested there is a good discussion of the different forcing definitions in Hansen et al (2005, JGR)).
While the definition of a forcing may appear a little arbitrary, the reason why radiative forcing is used is because it (conveniently) gives quite good predictions of what happens in models to the global mean temperature once the climate system has fully responded to the change. Thus the forcing can be used as a shorthand for the climate response without having to do the experiment.
Of course, the global mean temperature isn’t the only useful metric of climate change – regional temperatures and precipitation are arguably much more societally relevant – but it does have a good signal-to-noise ratio. That is, changes to the system are more clearly discerned in the global mean temperature than at a regional level, mainly because the noisy ‘weather’ component increases as you go to smaller scales.
Figure 1: Two breakdowns of the global mean forcings since the pre-industrial. Factors causing warming are red, cooling factors, blue. The top panel shows the direct effects of the individual components, while the second panel attributes various indirect factors (associated with atmospheric chemistry, aerosol cloud interactions and albedo effects) and includes a model estimate of the ‘efficacy’ of the forcing that depends on its spatial distribution. (From Hansen et al (2005, JGR)). Subjective uncertainties associated with the well-mixed GHGs are around ~0.15 W/m2, for ozone ~0.07 W/m2, tropospheric aerosols around ~1 W/m2 and solar ~0.3 W/m2.
The figure here gives one estimate for how many of those forcings have changed over the industrial period (1750-2000). This assessment is from my own lab and so I may be a little biased, but although there are significant uncertainties (particular for the aerosol indirect effects), it probably gives a reasonable idea of the current thinking. The forcings illustrated here are from the well mixed greenhouse gases (GHGs) (CO2, CH4, N2O, CFCs), tropospheric and stratospheric O3, direct aerosol effects (from sulphates, nitrates, organic and black carbon), land use change, solar irradiance, volcanic aerosols, and various indirect effects (on clouds, stratospheric water vapour, snow albedo etc.). Reasonable estimates of these 16 different effects (and counting…) were included in the GISS simulations for the upcoming IPCC assessment.
The land use change used in the figure is related to the deforestation dataset of Ramankutty and Foley (1999) and includes the effects of albedo and vegetation type change, but not the impacts of increased irrigation or the ‘greening’ of the high latitudes (due to climate changes and possible CO2 fertilisation effects) . These latter two effects are expected to lead to slight warming, but the overall impact of land use changes is expected to be negative (i.e. a cooling) (Myhre and Myhre, 2003), although the uncertainty is still significant (maybe 0.5 W/m2 either way).
To my mind, the ‘first-order’ forcings would be the ones without which you can’t really do without in assessing global climate change. I would therefore argue that for the global mean the well-mixed GHGs and the counterbalancing reflecitve aerosol effects are ‘first-order’ – without GHGs there is no appreciable warming signal, and without the aerosols, the warming from GHGs is excessive and important changes in the diurnal cycle and cloudiness are not captured. Everything else (apart from volcanos, which are a special case) is in the noise. If we were to break it down even further, I would argue that CO2, CH4 and sulphates (the main non-soot aerosol) were the only ‘first order’ forcings. It is curious to note that this is the combination of forcings that were predominantly used in the simulations discussed in IPCC (1995) where the conclusion was made that the ‘balance of evidence’ supported the notion of ongoing human-caused climate change.
Before the emails come streaming in, let me make it clear that this isn’t to say that ‘second order’ forcings are unimportant. On the contrary, many of these effects have very specific signatures in the climate system (in the stratosphere, in the Arctic and in the tropics) that need to be understood much better – however they are unlikely to have a big impact on the global mean temperature. Thus when it comes to global warming, neither land use change/vegetation type, nor for instance, the biogeochemical effect of increased CO2 are ‘first order’. The first example is clearly important locally (impacts of deforestation, urban development etc.) while the second effect is as yet inadequately unquantified but there doesn’t appear to be any a priori reason to think it is globally important. It doesn’t therefore make much sense to claim that some of the smaller forcings are ‘first-order’ despite their importance, and conceivably dominance, at smaller scales.
To be sure, some of these effects (such as the impact of irrigation on surface water vapour, or land use changes on evapotranspiration) are not easily dealt with in terms of the tropospheric radiative forcing – a point that was well made in the National Academies report on radiative forcing (on which Dr. Pielke was an author). However, the dominance of well-mixed greenhouse gases on the anthropogenic forcing over the last few decades is robust to almost any estimate of the uncertainty in the other forcings. This is clearly a different opinion to that held by Dr. Pielke. However, this is probably due to our different perspectives in what we feel are important questions (local vs. global), rather than a disagreement over fundamentals.
Hansen, J., Mki. Sato, R. Ruedy, L. Nazarenko, A. Lacis, G.A. Schmidt, G. Russell, I. Aleinov, M. Bauer, S. Bauer, N. Bell, B. Cairns, V. Canuto, M. Chandler, Y. Cheng, A. Del Genio, G. Faluvegi, E. Fleming, A. Friend, T. Hall, C. Jackman, M. Kelley, N. Kiang, D. Koch, J. Lean, J. Lerner, K. Lo, S. Menon, R. Miller, P. Minnis, T. Novakov, V. Oinas, Ja. Perlwitz, Ju. Perlwitz, D. Rind, A. Romanou, D. Shindell, P. Stone, S. Sun, N. Tausnev, D. Thresher, B. Wielicki, T. Wong, M. Yao, and S. Zhang 2005. Efficacy of climate forcings. J. Geophys. Res., in press.
Myhre, G. and A. Myhre, Uncertainties in radiative forcing due to surface albedo changes caused by land use changes,
Ramankutty, N., and J.A. Foley (1999). Estimating historical changes in global land cover: croplands from 1700 to 1992.