Gustavo Carrio (Department of Atmospheric Science, Colorado State University, CO, USA)
On bridging the gaps between scales: Potential uses of optimal estimation
Abstract
The performance of climate, general circulation, and even weather forecast models is limited by a variety of smaller-scale phenomena. Among them, clouds as well as surface-atmosphere interactions, including complex topographies, sea-ice, and urban areas. For instance the poor representation of clouds in climate models is among their greatest uncertainties.
The impacts of natural and anthropogenic aerosols on weather, precipitation, extreme events, and cloud radiative/optical properties appear to be highly dependent on atmospheric conditions, and therefore, reluctant to generalizations. In order to illustrate this, results of recent study focused on convective storms will be briefly summarized. An idealized examination focused on the response of orographic snow precipitation will then be presented in more detail. A rather large number of numerical experiments explored the “environmental space" varying both the concentration of cloud condensation nuclei (CCN) and low level moisture. Results indicate that the concept that enhanced CCN reduces precipitation cannot be generalized for warmer-based winter orographic clouds.
The presentation ends exploring the use of Data Assimilation (D.A.) as tool to retain both the observational truth and the ability of capturing the physical relationships, which only a numerical model can offer. Enhancing the aforementioned ability cannot only contribute to the understanding of various processes but also aim to improve their treatment in regional and global climate and weather prediction models.Instead of focusing on the spatial location and intensity of individual flow features, DA algorithms can be linked to a more statistical perspective of observations (e.g., representing the grid-cell of a “parent” scale). These modeling framework have a wide variety of potential uses, from the optimal selection of parameters in specific schemes (or the model itself) to the development of stochastic treatments (e.g. sub-grid cloud distributions)
.On the other extreme, if we focused on interactions between sea-ice and boundary-layer clouds, the scarcity of observations with high spatial/temporal resolution frequently makes purely observational approaches quite complex. In these cases, the use of such modeling frameworks can complement extensive observations from field experiments. The optimization model state vector can lead to optimal estimations of physical quantities difficult to measure/observe and to the development of parameterizations.
Carrio is one of four top ranked candidates for our open position as associate professor in Meteorology. We will inform about the upcoming seminars of the remaining two candidates early next week.