MEOPAR Relocatable Ocean Modeling: Strategy, Status, Progress Youyu Lu With inputs from Keith Thompson, Hal Ritchie Collaborators in MEOPAR, CONCEPTS & Mercator-Ocean

Goals of Relocatable Ocean Modelling A data assimilative coupled atmosphere-wave-ocean forecast system that can be set-up within hours of a marine emergency will be developed in order to provide short-term forecasts (hours to days) of the physical properties of the ocean and atmosphere to help guide the response to the emergency. The system will also include the capacity to track plumes of hazardous materials that can evolve based on the physical conditions. Our research will focus on new capabilities including (i) the ability to relocate the system anywhere in Canadian waters within hours of an emergency, and provide high resolution forecasts of variables such as wind, sea fog, sea level, waves and ocean currents (ii) the assimilation of physical properties of the ocean and atmosphere in order to improve the initial conditions and thus the forecasts (iii) off-line modules for tracking the movement of passive surface, and subsurface plumes, that can take into account the non- conservative properties of spilled materials such as radioactive decay or degrading hydrocarbons and the associated oxygen depletion, and (iv) the explicit integration of knowledge and modelling approaches related to biophysical effects with those dealing with anticipated socioeconomic vulnerabilities, risks and impacts.

Elements of a Relocatable Ocean Model Domain definition Grid creation Bathymetry Surface forcing Lateral boundary condition Code adjustment & compile Model parameters Running, validation, analysis, product

Strategy Closely coordinate/collaborate with CONCEPTS -Adopt CONCEPTS releases of NEMO code (1.1.0, soon) -Assemble model preparation/analysis software developed/cumulated over a decade through GOAPP, CONCEPTS, etc. -Tests high-resolution regional models (& study) -Automate software for relocatable capability Collaborate with Mercator-Ocean Collaborate within MEOPAR: 1.1, 1.2 & Prediction Core, sharing resources

Project 1.1 A Relocatable Coupled Atmosphere-Ocean Prediction System Project 1.2 Building a Network of Fixed Coastal Observing and Forecast Systems Project 2.1 Climate Change and Extreme Events in the Marine Environment Project 2.2 Biogeochemical Projections Under a Changing Climate Prediction Core Observation Core Ocean Modelling Components in MEOPAR

Collaboration with Mercator-Ocean MERCATOR relocatable model: -Leader in region NEMO code development (e.g., OBC) -Java interface, high level automation -Inflexibility grids – only follow ORCA MEOPAR different & complimentary aspects: -Flexible grids, bathymetry from Canadian sources -Canadian atmosphere-Ocean forcing (GEM & CONCEPTS) -Lower level automation Co-development and sharing agreed

Status: NEMO Team in CONCEPTS Model developers & users: F. Roy (code keeper), F. Dupont, Y. Lu, L. Zhai, M. Casey Data assimilation: G. Smith, Y. Liu, Z. He Validation & Product F. Davidson, S. Higginson, J. Xu

Status: NEMO Applications in Canada CONCEPTS EC, DFO Academics Dalhousie, U Alberta, MUN, UQAR, U Manitoba, UBC, … Climate Groups CCCma, UQAM, …

Status: Existing NEMO Configurations Global: 1/4° for operational ocean & coupled weather-ocean forecasting 1° for coupled seasonal forecasting Regional: CREG (Arctic + N Atlantic, plan to include N Pacific) 1/12°, 1/4° Newfoundland shelf 1/36° Gulf of St Lawrence, 5 km, plan 1 km Great Lakes, 2 km, plan 1 km Strait of Georgia, 0.5 km

Relocatable capability applied to the Strait of Georgia Curvilinear grid definition using Seagrid* Grid definition using AGRIF AGRIF refinement - conserve the ORCA grid orientation One out of 10 grid point is showed Grid size: 898x398 Resolution ~450m Example: Grid Generation Follow ORCA grid Create curvilinear grid using Seagrid*

Example: Use of in situ Observed Bathymetry Uncertainty in Hudson Bay/Strait NABD NABD – ETOPO1

Diff in M2 magnitude Improve M2 Tide Solution by Refining Bathymetry CREG12 solution (T1)

Example: High-Resolution Regional Modelling Newfound Shelf, 1/36° lat/lon (~2.5 km), 50 vertical levels Take initial and boundary conditions from CREG12, GEM atmospheric forcing Work started in April 2013 (L. Zhai) Solved a puzzle in setting lateral viscosity Stared validation & analysis Animation of SSH

Progress & Plan Prediction Core PDF, J-P Paquin, started may 2013: -Learned NEMO CONCEPTS v Grid generation with Seagrid -Improved documents for model preparation -Created 0.5 km Strait of Georgia model and tested simulation of tides -Package transferred to UBC Plan: Automate software packages for “relocatable model” development by PDF F. Chegini, expected to start in early 2014.