Guest Blog – Richard McNider, University of Alabama in Huntsville
We have just had a paper published in JGR entitled
McNider, R. T., G.J. Steeneveld, B. Holtslag, R. Pielke Sr, S. Mackaro, A. Pour Biazar, J. T. Walters, U. S. Nair, and J. R. Christy (2012). Response and sensitivity of the nocturnal boundary layer over land to added longwave radiative forcing, J. Geophys. Res.,doi:10.1029/2012JD017578, in press. [for the complete paper, click here]
The paper addresses the diurnal asymmetry in warming that has occurred in the observed temperature trends in the last century in which minimum temperatures have warmed at a substantially greater rate than maximum temperatures. While the paper goes into considerable detail on the response of the stable boundary layer to radiative forcing that perhaps only a stable boundary layer junkie can appreciate, the implications of the paper ,I believe, are critical to interpreting both the historical temperature data set and global modeling over the last century. For those who do not want to be overwhelmed with details, I believe the introduction and conclusions are tractable for non-boundary layer specialists.
Here let me summarize and at the end editorialize on the key points of the paper. In the last century minimum temperatures have warmed nearly three times more than maximum temperatures as captured by the NOAA Global Historical Climate Network. In fact this asymmetry is one of the most significant signals in the climate record and has been the subject of many papers. Our paper shows that the CMIP3 climate models only capture about 20% of this trend difference. This is consistent with other studies. Because climate models have not captured this asymmetry, many investigators have looked to forcing or processes that models have not included such as jet contrails, cloud trends, aerosols, and land use change to explain the lack of fidelity of models. However, our paper takes an alternative approach that explores the role of nonlinear dynamics of the stable nocturnal boundary layer that may provide a general explanation of the asymmetry. This was first postulated in a nonlinear analysis of a simple two layer model we carried out a few years ago (Walters et al. 2007) that indicated that slight changes in incoming longwave radiation from greenhouse gases might result in large changes in the near surface temperature as the boundary is destabilized slightly due to the added downward radiation. This produced a mixing of warmer temperatures from aloft to the surface as the turbulent mixing was enhanced just as an increase in wind speed can destabilize the nighttime boundary and mix warm air from aloft to the surface.
The purpose of the present paper was to see whether this behavior in the simple two layer model was retained in a more complete multi-layer column model for the stable boundary layer. Basically, we subjected a nocturnal boundary layer to an added increment of downward radiation (4.8 W m -2 ) then looked at the difference in the model solution without this forcing. We also carried out detail budget calculations to see where the added energy ended up being deposited. Increased downward radiation from CO2 and water vapor feedbacks has been part of the global forcing in climate studies. However, aerosols can also add downward longwave radiation (Nair et al 2011).
The results of these experiments showed that indeed the stable boundary layer grew slightly and was less stable due to the added longwave radiation. The result of the growth of the boundary layer and destabilization was that warm air was entrained from aloft down to the surface by the added turbulence. The model showed that the 1.5 m air temperature (that is near the standard shelter height) warmed substantially due to this destabilization. Moreover, the budget calculations showed that only about 20% of the warming was due to the added longwave energy. Most of the warming at shelter height was due to the warm air mixed from aloft. This is illustrated in figure 10 in the paper. Thus, this process is a highly sensitive positive feedback to surface warming.
Figure 10: (top) Expanded view of the difference in potential temperature profile between the case of added GHG energy and base case for a geostrophic wind of 8 m s-1(top). (bottom) Expanded view of profile difference.
Our budget calculations in the paper also showed that the ultimate fate of the added input of longwave energy was highly sensitive to boundary layer parameters and turbulent parameterizations. In our simple model, the added radiation could go to heating the atmosphere, heating the near surface ground temperature, heating the deep ground temperature or lost to radiative emission from the skin surface. The model showed that at light winds (with weak turbulence) the atmosphere was not able to effectively lift this energy off the surface and into the atmosphere. Thus, more radiation was emitted from the surface. If soil conductivity and /or heat capacity were large then more of the energy would go to heating the ground. When we tested boundary layer parameterizations of the type employed in large scale models, we found they generally added much more sensible heat to the atmosphere as opposed to being lost by radiation or to the ground.
To capture the type sensitivity we found in our model, climate models would need very fine vertical high resolution and also stable boundary layer parameterizations that don’t have large background mixing such as is often added to large scale models with coarse resolution. Our paper also showed that the stable nocturnal boundary layer was very sensitive to the turbulent parameterization and surface characteristics such as roughness, and land surface heat capacity and conductivity. In fact because current coarse resolution global models do not capture the asymmetry in warming in minimum temperatures and likely do not represent the stable boundary layer very well, we further suggested that truthful replication of the night-time warming may be out of the reach of current models.
Thus, it may be better for current climate models, when they test replication of past climates and to project future global warming, to only use maximum temperatures rather than the current metric of using the mean daily temperature, which contains the minimum temperature. Of course, changes in night-time temperatures represent real changes and possible impacts to the climate system (e.g., melting ice), to society (agricultural productivity) and to ecosystems. Thus, ultimately we need to develop climate models that do have the resolution and sensitivity to capture changes in minimum temperatures.
In this blog I would like to now editorialize on the implications of this work which were not explicitly stated in the peer reviewed paper. While the asymmetrical warming of the nighttime temperatures and the lack of fidelity of models in capturing the asymmetry that we discuss has also been the subject of other papers, it seems that no one has looked at the implications of this to the general ability of models to forecast climate change. But, consider the following is a thought experiment.
Model credibility in the IPCC has been based on the ability to replicate the last 130 years of the global instrumental temperature record with anthropogenic forcing. But, remember that the global temperature record in such comparisons is based on the daily Tmean (the average of Tmax and Tmin). If models are replicating Tmean but are not capturing the trend in Tmin, then this must mean that the model Tmax is warming faster than the actual Tmax. Also, if most of the warming in the instrumental record is warming in the nighttime boundary then by its very nature this is warming of a very thin layer of order 200m or so. In fact, if our results are correct, we show that it is only the lowest part of the nighttime boundary layer that is being warmed or a thin layer than of no more 20-50 meters. Maximum temperature observations made in daytime boundary layers which are 1- 2 km in depth, reflect a measure of a much deeper layer temperature. Thus, the instrumental observational data when viewed in light of boundary layer theory is showing that most of the warming is occurring in a very thin layer and the deeper atmosphere, as captured by Tmax, is not warming as much as models.
However, one of the largest positive feedbacks in climate simulations is the accumulation of additional water vapor as the deep atmosphere warms and this adds an additional greenhouse effect. In fact, the added water vapor effect depends on a deep layer of added water vapor. If the deep atmosphere is not warming then this water vapor feedback will not be nearly as strong. Thus, models may be overstating the water vapor feedback. This in turn begs revisiting the continued controversy of the difference in warming of lower troposphere as measured by satellite and balloon borne data and the warming predicted by models since 1979.
Because of the disagreement between the satellites and balloon data and models, the consensus of most of the climate change community has been that the warming of the surface data is consistent with model so that there must be problems in method or sampling with the satellite or balloon data sets. However, if you throw out Tmin as related to the heat content of deeper atmosphere, it is likely that trends in observed Tmax may be more consistent with the satellite and balloon data. Of course, here we are talking about land processes and consideration of the total warming over land and water would have to be considered. However, I believe it deserves further detailed investigation by both the observational and modeling community to determine whether this thought experiment is valid.
In regards to the oceans (since I started my career as an ocean modeler), I think we should also be careful about similar turbulent processes connecting the atmosphere and ocean surface. Just as for the land surface, the ultimate fate of added energy may be tied to the details of how efficiently and quickly turbulence in the atmosphere and the ocean can remove this added energy from the skin surface. Any errors in this near surface turbulence will impact the fate of the added energy. I am not certain at all that coupled ocean-atmospheric models get these details right.
Nair, U. S., R. McNider, F. Patadia, S. A. Christopher, and K. Fuller (2011), Sensitivity of nocturnal boundary layer temperature to tropospheric aerosol surface radiative forcing under clear‐sky conditions, J. Geophys. Res., 116, D02205, doi:10.1029/2010JD014068.
Walters, J. T., R. T. McNider, X. Shi, and W. B. Norris (2007), Positive surface temperature feedback in the stable nocturnal boundary layer, Geophys. Res. Lett., 34, L12709, doi:10.1029/2007/GL029505.