In addition, the reference scenarios used for the AR4 report have not been updated since 1994. These scenarios are discussed in the 2000 Special Report on Emission Scenarios (SRES). In response, climate scientists state that these details do not effect the results published in the IPCC AR4 report because the scenarios are ran for long periods of time, collecting data on averages not deviations from the mean. Secondly, many scenarios are ran to cover a wide range of future emission situations, which in turn requires policy action from the government. I have included links to some graphs from the IPCC SRES report to help explain the various scenarios.

SRES Scenarios (1.7.2) and Global Carbon Dioxide Emissions (1.7.3)

The first graph is a basic break down of the various IPCC scenarios. Moving the bottom of the tree to the top means the scenarios are changing from low to high economic growth. Moving from the left of the tree to the right means the scenarios are changing low to high population growth. For more details on each scenario please see the SRES report.
The second graph shows projections for the various SRES scenarios. In the IPCC report the CO2 emission drivers include Population, Economic Activity (GDP) per capita, Energy Intensity (energy consumption per unit of GDP) and Carbon Intensity (CO2 emission per unit of energy). These same drivers were used by the authors analysis of assumptions in the IPCC AR4 report.

The authors of this article analyzed the spontaneous technology portion of each scenario to understand how much of the future carbon emissions was already accounted for in the baseline assumptions. They approached this problem by implementing a frozen technology baseline, thus eliminating spontaneous decarbonization. This revealed that a huge amount of emission-reducing technology is already built into the SRES scenarios.

Figure 1 in the article highlights how a large amount of projected decrease in carbon dioxide production is due to spontaneous technological development. The bar graph has three colors, blue, red and yellow. Blue indicates the reduction in carbon dioxide due to spontaneous technological development, Red indicates reduction due to climate policy and Yellow is where atmospheric carbon dioxide would be considered stabilized (500 p.p.m). One can immediately see that the blue section of the bars on the graph are larger than the red section, for all scenarios. For some scenarios this difference is quite large. Figure 2 in the article hits upon our current problem, that many past predictions can not keep up with the current observations. All of the IPCC AR4 scenarios are too conservative, thus predicting changes in Energy Intensity and Carbon Intensity that are lower than observations between 2000-2005.

These IPCC predictions indicate a decline in Energy Intensity exceeding 1.0% per year, this is unrealistic. One can only hope for approximately a 20% (+/- 10%) decrease in Global Energy Intensity due to sectoral shifts over a century. The observed rise in Global Energy and Carbon Intensities is due to developing worlds like China and India. In these countries many of the rural populations are moving to cities, where consumption of energy and energy-intensive materials is higher. This is obviously not accounted for in the IPCC scenarios. For example, China’s carbon dioxide emissions have risen by 11%-13% each year from 2000-2010 but the SRES Scenarios only assume a 2.6%-4.8% increase.

From these results the authors conclude that the IPCC report assumes most technological advances will occur automatically. Instead of pushing an overly optimistic future they should be using their report to shake things up and create conditions for the innovations assumed in the IPCC report to occur. These reports have a large effect on US Science Policy, thus all assumptions need to be stated clearly. Policy makers need to be fully informed, thus they need to know that the assumed technological transformations would take decades to complete, even if we started now. Fortunately the IPCC panel plans on updating the SRES in it’s next report, due in 2013 or later, but in the meantime policy makers need to be aware of how conservative the IPCC report really is.

Hopefully when I return I will get things moving with this blog, but for now it is on a bit of a hiatus.  When I get back I´ll try to post some information on some of the climatically interesting things I´ve observed while travelling (e.g., melting tropical glaciers, the driest desert in the world, etc…).

I didn’t expect to see stories in the news on the Arctic Sea Ice extent this early in the season, since the minimum sea ice extent does not occur until around September. But as of now, it looks as though the melt is tracking last years record loss.

This will be interesting to monitor… For those who are interested, the progress of the sea ice can be monitored here from a webpage set up by the National Snow and Ice Data Center here in Boulder.

Venus has an average atmospheric temperature of 864 degrees Fahrenheit and 90 times the amount of pressure Earth has, therefore it seem plausible that Venus once had just as much water as Earth and over time, it evaporated away. On the spacecraft taken to Venus had a plasma analyzer that detects ions leaving Venus’ atmosphere. The three main ions detected were helium, oxygen, and hydrogen which is very suspicious of evaporated water. When water molecules split, not only is hydrogen and oxygen released, but traces of deuterium are a byproduct of the reaction. Therefore the amount of deuterium found is associated with the amount of water evaporated. During with mission, they found high levels of deuterium further more proving that water was being evaporated into Venus’ atmosphere.

Throughout the mission, they have also found large cloud vortexes over the polar regions that have much higher temperatures than the area around it. From this finding they suspect that like Earth, Venus has circulating warm air traveling from the equatorial region to the polar regions. I think this is very fascinating to find this similar trend in the planets closest to Earth.

For the first time, scientists have evidence of lightning on Venus found from this mission. Many scientists were caught off guard in seeing it. Scientists believed that it wasn’t possible to have lightning on Venus due to the smog-type of cloud found there. Since smog-type clouds generally don’t produce elecrical charges, they assumed it wasn’t possible. They now wonder if they have thought of all the reasons electricity could be generated by the atmosphere.

I was very surprised to see that scientists have found lightning on other planets. Not only that but to also realize how much Venus and Earth really are alike. Advances in technology have made it possible to prove not only our theories about Venus but create new ones. I can only wonder how much more we will find out about how similar Venus and other planets are to our own.

Kathryn Dougherty

As said before, the study of ice crystals has been around for some time. More recently, however, a nuclear physicist named Ukichiro Nakaya delved into observing thousands of ice crystals. He came to categorize them into different morphologies. He concluded that each crystal was the product of the temperature and the saturation of the atmosphere in which it formed. He found that different temperatures brought about various shapes (i.e. plate versus prism) and that the plate and prism grew into more ornate and complicated crystals at higher saturations, such as the dendrite and needle respectively. Nakaya came up with a diagram that summarized his findings. It can be found through the following link: Nakaya Diagram.

The transition between prism to plate to prism is sharp and abrupt with changing temperatures. This brings about confusion and complication and it still not fully understood. What causes such sudden alteration in shape?

At the bottom of the diagram, there are the basic prism and plate shapes. Their facets are “smooth on all scales down to the molecular level.” Some remain smooth up to 0°C while others become rough at lower temperatures. This change from a smooth to rough surface, as well as the liquid present on the crystal and the equilibrium/nonequilibrium shape of the crystal all affect those transitions from prism to plate in the Nakaya diagram.

It is not uncommon for ice crystals to have a small liquid film around them at temperatures below freezing. The molecules at the surface are not held in place as tightly as the ones in the main bulk of the crystal. Thus they have more mobility. This liquid layer has an effect on the roughening transition for the crystal. The roughening transition is just the point where the snow crystal goes from being faceted to being rough. It differs depending on whether the ice is in contact with water or with water vapor. The roughening transition is also a function of temperature. Below the roughness temperature, the crystal is smooth. If the temperature of the atmosphere is greater than that of the roughening transition, the equilibrium shape of the crystal will be rough. The roughness of the crystal in turn helps determine the shape and growth of the crystal.

One can see that all the factors affecting the prism-plate transition in the Nakaya diagram also rely upon each other. Each facet of the crystal has different properties (in other words, it is anisotropic) and thus can take on different forms. Thus understanding the structure or nature of the surface of the crystal is indispensable in comprehending the shape the crystal will grow to. But in the end, there is still much more to be learned about this subject.

This article was an interesting one to read, but in the end for me it raised more questions than were answered. The authors never really explained the roughening transitions or when/why it occurs. Perhaps it was my misunderstanding (or misreading), but the only thing I got from that was that it was affected by the liquid present on the crystal.

Brianna Coté

The comparison of the parameterizations for the five different turbulent collision kernels revealed their strengths and weaknesses. A general kinematic formulation, radial distribution function (RDF), gravitational force, Stokes drag force (related to terminal velocity), and other effects were combined in different ways and used in model outputs. The Ayala or A05 kernel turned out to be the most realistic. The resulting kernels were evaluated through plots of kernel size as determined by collision droplet size. They were also compared with the Hall kernel (a base kernel commonly used in collision-coalescence that does not take turbulence into account) to measure the enhancement effects of turbulence in each parameterization. Moderate enhancement was seen in the Ayala kernel, the ZWW-RW overestimated the enhancement, and the remaining two were similar to that of the Hall kernel.

For the remainder of the paper, the Ayala kernel is understood to be the most realistic, so I will include the effects for only that kernel, unless another proves to be more relevant. In turbulence enriched collision and coalescence, size matters. Droplets were arranged into three classifications (A,B, and C) depending on their size. Group A included drops of radius < 50µm, B was between 50 and 100µm, and C was drops > 100µm. The Ayala kernel showed maximum turbulence effects at small and medium sized droplets colliding with droplets of similar size. This illustrates the concept of the preferential concentration of droplets, an idea closely related to the enhancement of collisions through turbulence.

When looking at the temporal aspect of droplet growth, three phases were identified: the autoconversion phase, accretion phase, and the self-collection phase. All three phases can be fairly easily identified on plots of mass density transfer. The radii for the maxima on the graphs can then be translated into radii verses time graph. Maximum droplet growth occurs during the accretion phase. Turbulence helps to initiate the onset of phase two faster, and therefore increases the likelihood of larger drops. From this, they found that the Ayala kernel can shorten the time for drizzle drop formation by up to 39% compared to the Hall kernel.

While the math and numerical simulations involved in this paper may be beyond my realm of understanding, the concepts and ideas presented seemed relevant. The individual steps to reach their conclusions were somewhat confusing, but the conclusions made perfect sense. In my mind I compared it somewhat to an updraft increasing the final droplet size. Turbulence increases the opportunity for more collisions and therefore decreases the time it takes to create precipitation sized droplets.

Surface temperature changes are also related to sea ice cover. Cavalieri et al (2003) showed a decrease of total Arctic sea ice extent of about 36000 km/yr from 1979-2002. Studies shown by Parkinson et al (1999) and Parkinson and Cavalieri (2002) show similar results. Sea ice becomes important in a discussion of surface temperature because of a high albedo and associated feedback effects. Decreasing sea ice cover creates larger areas of open water and the varying atmospheric circulation changes moisture advection into the Arctic. Wang and Key (2005) show trends that the Arctic has become cloudier in spring and summer but less cloudy in winter. Trends in cloud amount have an impact on trends of surface temperature. Yinghui Liu, Jeffrey R. Key, and Xuanji Wang quantify the influence of changes in Arctic cloud cover on the surface temperature trend in their Journal of Climate article “The Influence of Changes in Cloud Cover on Recent Surface Temperature Trends in the Arctic” Volume 21 Issue 4 (February 2008).

Liu, Key and Wang used the APP-x dataset which extends the APP products(see (http://nsidc.org/data/docs/daac/nsidc0066_avhrr_5km.gd.html) to include cloud optical depth, cloud particle phase and size, cloud top temperature and pressure, all-sky surface temperature and broadband albedo, and radiative fluxes. Retrievals were done with the Cloud and Surface Parameter Retrieval System (CASPR) which was specifically designed for polar AVHRR (Advanced Very High Resolution Radiometer) daytime and nighttime data. The APP-x dataset includes daily composites at both 0400 and 1400 local solar standard time(LST) from January 1982 to December 2004. Monthly means of cloud cover and surface temperature under cloud free and cloudy and all-sky conditions were calculated from twice daily data.

In winter, the surface temperature increases over northern Canada with a maximum over Hudson Bay. The surface temperature decreases over the Arctic Ocean and eastern Arctic. Over the central Arctic Ocean, the decreasing trend is approximately -2.5K/decade. In spring, warming is seen over most of the Arctic with the trend about 2.0K/decade. In summer, surface temperature is generally increasing though with a smaller magnitude than in spring and trends are near 0 over the central Arctic Ocean. In autumn, there are strong increasing trends over the Beaufort Sea, the Chukchi Sea, Alaska and northern Canada with a maximum trend around 2.0K/decade.

The cloud free and cloudy surface temperature trends exhibit similar pattern as the all-sky trends. The differences in surface temperature trends under clear and cloudy conditions are due not only to changes in cloud cover but also to changes in cloud properties. A decreasing trend in APP-x cloud top height may contribute to a stronger cooling trend over the Chukchi Sea and central Arctic in winter under cloudy conditions than under clear conditions. Warmer clouds emit more longwave radiation upward. For regions where both clear and cloudy trends are positive, the magnitude of the warming trends under cloud free conditions is larger than under cloudy conditions. For regions where both trends are negative the magnitude of both trends are comparable. Under cloudy conditions, the trends are not as dramatic as under cloud free conditions. This tends to imply that clouds have a negative feedback on the surface temperature trends.

The article is very straightforward and the authors offer no speculation on what their conclusions may imply. They present some lengthy mathematics to quantify the influence of cloud cover on surface trends. This confirms that the atmosphere is very complex and that there are many things that interact with each other.

Rainfall estimates over the southern U.S. and adjacent waters were made using the Tropical Rainfall Measuring Mission (TRMM) satellite’s Microwave Imager which, due to its orbit, can see from 40 S to 40 N. The analysis showed that there was a highly significant increase in precipitation estimates during mid-week summertime afternoon thunderstorms and that storm heights were higher. Reanalysis winds also showed that the midweek low-level wind convergence, mid-level vertical wind velocity and upper-level wind divergence all increased as would be expected with stronger thunderstorm development. Data from rain gauges also showed an increase in precipitation, but to a lesser extent than was shown in the satellite estimates. The opposite was found over the adjacent ocean waters where precipitation was suppressed during the midweek period but showed a peak over the weekends. This was attributed to the air at the top of the land-based thunderstorms moving out and descending over the ocean thereby suppressing the development of thunderstorms over the ocean waters, especially the Atlantic.

The reason for the midweek peak in precipitation over the southeast U.S. is due to weekly variations in particulate concentrations. Data from the Environmental Protection Agency was analyzed and it was found that the average particulate concentration smaller than 10 microns peaked from Tuesday to Thursday. There was also a Tuesday to Thursday peak for particulates smaller than 2.5 microns, but it was less pronounced than for the 10 micron size. These measurements were taken at the surface so more research is needed into these particulate concentrations in the vertical.

It is pointed out in the article that both pollution levels and types have changed over the years as technology has developed, populations have increased and moved about, and the government regulated industry. Therefore, pollution of yesteryear may not have had the same effects as that of today, and pollution in the future may affect precipitation differently.

The authors do mention that this weekly variation shows up in the afternoon precipitation in most, but not all summers. I wonder why this is so. Also, I wonder how fast pollutants move about in both the vertical and horizontal. If heavy rain “scrubs” some of the pollution from the atmosphere, how fast do the particulates repopulate the air? And, as also pointed out in the article, which particular pollutants are involved in the processes described in the article that invigorate the convection?

While there are several other explanations for these anomalies, along with arguments stating the justification of these issues has nothing to do with absorbing aerosols, a lack of research and limitations involved in previous studies have attributed to the debate of their causations. In the study conducted by Erlick and Schlesinger, large drop clouds (CL) and smaller drop clouds (CS) are modeled with the Intercomparison of Radiation Codes in Climate Models and set to a constant thickness and cloud extinction optical depth. These clouds are then placed at either a low height (between 800 and 900 mb) or a higher height (200 and 800 mb). Low and high CS and CL clouds are also observed at zenith angles of thirty and sixty degrees. Dispersed throughout these clouds are drops containing supermicron absorbing aerosols comprised of dust, soot and a combination of the two. It is improtant to note that within this experiment, the dispersion of aerosols is added to the clean clouds, rather than clean cloud drops being displaced by the contaminated droplets. It is also improtant to understand that while soot and mineral aerosols are observed at high elevations, the microphysical interaction with clouds has yet to be well assessed. The Maxwell-Garnett model (in which the distrubution and size distrubution of the included aerosols is random), the Sihvola and Sharma approximation (in which the distribution of included aerosols is random and the size distribution dictates inclusions larger than dipoels) and a core plus model are used in the calculation of the single scattering behaviors of droplets enclosing absorbing aerosols. An algorithm of Freidenreich and Ramaswamy is also used in the calculation of the solar cloud forcing ratio, scattering behaviors and the configuration of droplet dispersion.

Six scenarios are evaluated regarding the aerosol size distribution within the CL and CS clouds, all of which are analyzed with respect to a cloud of pure water (or without a core). The evaluated concentration categories are as follows: 1) All mineral dust, 2) Soot up to 1.250 micrometers, 3) Soot up to 1.584, 4) Soot up to 2.008 micrometers, 5) 97% dust and 3% soot and 6) 70% dust and 30% soot. Upon the completion of the multiple calculations and collection of vast data samples, arrangements into tables and figures are performed. This allows for easy comparisons to the reference cloud consisting of pure water droplets. The albedos of CL clouds appear to not have exceeded .7 to .8, which corresponds to the initial observation inquiry. Cloud forcing ratios are all significantly greater than 1.0, the accepted pure cloud forcing ratio. The CL clouds containing 70% dust and 30% soot have exhibited particularly high forcing values of 1.41 to 2.60. These results are also in accordance with the second aberration stated earlier. It is noted that the high (200 to 180 mb) clouds did not result in values which vary much from that of a pure cloud. This result was attributed, in this case, to the assumption that high clouds do not amplify absorption of solar radiation as efficiently as lower level clouds. The same general trend of heightened cloud forcing values and depleted albedos are evident among the CS clouds, as in CL clouds.

Through this study, Erlick and Schlesinger are able to show that significant enhancement of cloud forcing ratios can be achieved through the distribution of drops containing supermicron cores. It is indicated in the conclusion of the study that two commonitions may have influenced the archived results. Only one supermicron aerosol distribution method was used for lack of additional data, and second the two clean cloud size distributions used for comparison were static. While this study is fairly inconclusive, it is an important step in expressing the potential for absorbing aerosols in cloud droplets to be held accountable for the lower albedos and higher cloud forcing ratios observed.

While this article is very annotative, it would aid the reader if the methods used in the calculations were more clearly explained. Understandably so, the explanations of different methodologies would most likely result in a novel by themselves, the assistance in the comprehension of the ways in which the distributions were portrayed would be greatly improved under the assumption that the reader is unfamiliar with the named formulas. Overall the article was indepth and the results of the calculations were portrayed in such a way that made them easily understandable. While the issue regarding the effect of absorbing aerosols on cloud absorption and forcing ratios is far from being conclusive, the study conducted by Carynelisa Erlick and Dana Schlesinger is a step in the right direction to uncovering an accepted cause for cloud observations that deviate from one’s expected values.