This is a past event.

Environmental Engineering Graduate Seminar

Ara Hakim, Environmental Engineering Ph.D. student, Graduate Research Assistant

Bio:
Ara Hakim is a graduate research assistant at the Dept. of Civil, Environmental, and Geospatial Engineering in Michigan Tech. He has a BS degree in Oceanography (2006) from Bandung Institute of Technology (Indonesia), and MS degrees in Coastal Geoscience & Engineering (2009) and Marine Science & Technology (2015) from University of Kiel (Germany) and UMass Dartmouth respectively. He served as a consultant for various government agencies and private companies in Indonesia, being involved in various works such as marine policy, marine spatial planning, and ocean operational forecast system development. Two years prior to joining Michigan Tech, easing his way back to academia Ara worked as a research fellow at Hydrography Research Group in Bandung Institute of Technology and as an adjunct lecturer at the School of Fisheries and Marine Science in Padjadjaran University, both located at his hometown. He currently works under Dr. Pengfei Xue doing implementation of Data Assimilation techniques in the Great Lakes, and receives CIGLR student fellowship for the year 2021.

Abstract: 
Lake surface temperature (LST) is one of the most important physical variables in the Great Lakes. It plays a major role and acts as a crucial proxy for understanding atmosphere-lake interactions, ecosystem dynamics, and future projection due to the changing climate. Hence, accurate estimation of LST within the Great Lakes Operational Forecast System (GLOFS) has been under continuous development since the 90s. This research uses ensemble-based data assimilation approach to improve the accuracy of short-term LST forecast for Lake Erie. The Finite Volume Community Ocean Model (FVCOM) is the basis of the operational forecast system. Daily LST from Great Lakes Surface Environmental Analysis (GLSEA) database is assumed to be the true state, and it is also used to correct initial conditions during the Data Assimilation (DA) phase utilizing Universal Model Domain Local Ensemble Transform Kalman Filter (UMD-LETKF). In this investigation we are using 20-member ensemble models, which are generated using atmospheric forcing from NOAA’s High Resolution Ensemble Forecast (HREF). Our current work demonstrates that UMD-LETKF has been successfully implemented in improving the initial conditions for FVCOM forecast model run. Preliminary results from evaluation of 7-day forecast in the mid of August 2021 show improvement for forecast simulation with DA compared to forecast simulation without DA.