- American Physical Society Sites
- Meetings & Events
- Policy & Advocacy
- Careers In Physics
- About APS
- Become a Member
Kyle G. Pressel
The University of California, Berkeley
William D. Collins
Lawrence Berkeley National Laboratory
Water vapor is the most significant greenhouse gas in the Earth's atmosphere and its spatial distribution plays a role in determining the location and structure of clouds. Despite the importance of water vapor the processes determining the distribution of water vapor remain poorly understood. The difficulty in understanding the distribution of water vapor is due in large part to the complexity of atmospheric motions which transport water vapor throughout the atmosphere.
Understanding the distribution of water vapor is of central importance to the representation of clouds in global climate models (GCMs). Clouds exert a cooling effect on the climate system by reflecting light from the sun and a warming effect through the trapping of heat within the climate system. The combination of the two effects represents the net effect of clouds on climate. It remains uncertain how these effects may change under climate change scenarios to serve as a feedback, either enhancing or suppressing human induced changes in the climate system. Furthermore, clouds exhibit variability and dependence on processes across a range of scales which cannot be resolved in current GCMs. Therefore, a model must be used to represent the effects of small-scale variability of clouds within a GCM grid box. The importance of cloud feedback and the difficulty of modeling the effects of small-scale variability has led to the identification of the representation of clouds in GCMs as the largest source of uncertainty in climate predictions.
GCMs break up the atmosphere into a three dimensional grid of boxes. Within each grid box atmospheric properties are assumed to be uniform. Statistical cloud models (SCMs) have been developed to represent the effects of small-scale variability across a single GCM grid box. In the SCM framework cloud properties can be determined if the distribution of total water substance (the sum of water vapor, liquid water, and ice) is know. A minimum requirement for a SCM is that it represent fractional cloudiness across a single GCM grid box. The relationship between the distribution of total water substance and fractional cloudiness has been shown to be simple to derive, however the form of the distribution remains hard to isolate. Our research focuses on using measurements from NASA's Earth Observing System satellites to constrain the distributions of water vapor upon which SCMs are built.
The Atmospheric Infrared Sounder (AIRS) on board NASA's Aqua satellite provides vertical profiles of water vapor throughout the lower portions of the clear and partially cloudy atmosphere with global coverage. The global coverage of satellite based measurements comes at the expense of not being able to measure water vapor at cloud scale, which is likely the relevant scale for constraining the distributions underlying SCMs. In both the clear and partially cloudy atmosphere the vast majority of all water substance exists in the vapor state. Therefore, we seek to justify the use of measurements of water vapor from AIRS in the clear and partially cloudy atmosphere to constrain the distributions used by SCMs. Central to this justification is showing that the measurements of water vapor at the 45 kilometer resolution of AIRS can be used to say something about the statistics of smaller scales which are used to define SCM distributions. To do this we propose to show that under certain conditions scale similarity can be used to write down simple relationships between measurements from AIRS and the statistics of small-scale water vapor variability.
Many processes which exhibit complicated nonlinear behavior also exhibit scale similarity, in which statistics computed over areas of varying size can be related through quite simple expressions. Others have already shown that statistics of water vapor as measured by AIRS exhibit scale similarity across scales larger than the resolution of AIRS and that water vapor measurements from aircraft exhibit scale similarity across scales smaller than those measured by AIRS. Our research seeks to use ground based sensors to identify conditions under which the large-scale similarity as observed by AIRS can be connected to small-scale scale similarity as observed by aircraft. If a set of conditions exists for which this connection can be made, then we suggest there is support for using AIRS measurements to constrain water vapor distributions under similar conditions.