EEWR Brown Bag Seminar with Stimit Shah, Grad Student and Sara Sadri, Postdoctoral Researcher
Speaker: Stimit Shah and Sara Sadri, Civil and Environmental Engineering
Series: EEWR Brown Bag Seminars
Location: Engineering Quad E225
Date/Time: Friday, April 26, 2013, 12:00 p.m. - 1:00 p.m.
Speaker: Stimit Shah
Title: Dynamics of turbulent atmospheric boundary layers with stability and
heterogeneity in surface fluxes
Understanding and parameterizing turbulence in the atmospheric boundary layers (ABL), remains a challenging, yet very important problem in geophysical fluid dynamics. Numerical simulations, with their ability to provide 3D, time-varying information on turbulent structures and dynamics are increasingly used to tackle the problem. However, some uncertainties regarding the performance of models remain, especially with diurnal variation, heterogeneity in surface fluxes and roughness, variation in geostrophic forcing and terrain effects. I would like present a few results on some aspects of the ABL like the influence of stability on coherent structures in the ABL (quantified using proper orthogonal decomposition), its effect in conjunction with heterogeneity and extremely stable conditions leading to collapse of turbulence.
Speaker: Sara Sadri
Title: Time series analysis and validation of VIC model for low flows of the eastern United States
How can we distinguish between stationary and nonstationary processes? How can we know if the VIC model is capable of predicting low flows? In this resentation, I will talk about how we used nonparametric tests to identify abrupt changes in time series, with the main application being to help systematically distinguish the effects of human intervention from those of climate variability. This study has been done on n-day low flows (n=1, 7, 30, 90) of 523 USGS sites in the eastern US. By combining three methods of autocorrelation, the Pettitt test, and the Mann-Kendall test, we designed a recursive decomposition algorithm. The decomposition algorithm measures different kinds of nonstationarity such as abrupt shifts and trends. Examination of the survey notes for a subset of the sites confirmed that many of the largest step changes were associated with dam construction. Furthermore, we used the decomposition algorithm results to filter regulated sites. Currently we are using the remaining sites to find out the usefulness of the current VIC model for low flow predictions.