OceanTEFFH O icial MAGAZINEog OF the OCEANOGRAPHYraphy SOCIETY CITATION May, P.W., J.D. Doyle, J.D. Pullen, and L.T. David. 2011. Two-way coupled atmosphere-ocean modeling of the PhilEx Intensive Observational Periods. Oceanography 24(1):48–57, doi:10.5670/ oceanog.2011.03. COPYRIGHT This article has been published inOceanography , Volume 24, Number 1, a quarterly journal of The Oceanography Society. Copyright 2011 by The Oceanography Society. All rights reserved. USAGE Permission is granted to copy this article for use in teaching and research. Republication, systematic reproduction, or collective redistribution of any portion of this article by photocopy machine, reposting, or other means is permitted only with the approval of The Oceanography Society. Send all correspondence to: [email protected] or The Oceanography Society, PO Box 1931, Rockville, MD 20849-1931, USA. downloaded FROM www.tos.org/oceanography PHILIppINE STRAITS DYNAMICS EXPERIMENT BY PAUL W. MAY, JAMES D. DOYLE, JULIE D. PULLEN, And LAURA T. DAVID Two-Way Coupled Atmosphere-Ocean Modeling of the PhilEx Intensive Observational Periods ABSTRACT. High-resolution coupled atmosphere-ocean simulations of the primarily controlled by topography and Philippines show the regional and local nature of atmospheric patterns and ocean geometry, and they act to complicate response during Intensive Observational Period cruises in January–February 2008 and obscure an emerging understanding (IOP-08) and February–March 2009 (IOP-09) for the Philippine Straits Dynamics of the interisland circulation. Exploring Experiment. Winds were stronger and more variable during IOP-08 because the time the 10–100 km circulation patterns period covered was near the peak of the northeast monsoon season. Distinct wind jets of the Philippines, particularly of the between islands and wakes behind islands are common northeast monsoon features surface winds and upper-ocean currents, that are controlled by winds interacting with island topography. The modeled upper- requires many observations and the ocean flow associated with Philippine straits during IOP-08 exhibits large (> 1 m -1s ) interpretive assistance of high-resolution westward mean flow through Suriago Strait and highly variable flow through Mindoro numerical modeling. Strait. The model shows prominent eddies in the Bohol Sea and Cuyo East Pass that The Naval Research Laboratory were also observed during the field experiment. A high-resolution nested simulation (NRL) contributed to the Philippine of the Verde Island Passage finds local wind-driven upwelling that is confirmed Straits Dynamics Experiment (PhilEx) by shipboard sea surface temperature measurements and satellite observations of Intensive Observational Period research chlorophyll concentration. cruises in 2008 and 2009 (IOP-08 and IOP-09) by supplying the shipboard InTRodUCTION western Pacific Ocean to the east, with team with real-time atmospheric and The Philippine Archipelago is an intri- its nearby North Equatorial, Kuroshio, oceanic forecasting, and with addi- cate array of islands, straits, and semi- and Mindanao currents. Observational tional retrospective two-way coupled enclosed seas whose broad atmospheric and numerical studies of these global- numerical hindcasts. The forecasts circulation is governed by the rhythm scale phenomena abound (Qu et al., provided information to scientists at of the East Asian monsoon and steady 2009), but within the island archipelago, sea both for general operations and Pacific trade winds. Philippine oceanic there are finer scales at play in both as guidance for developing underway circulation is dominated by the posi- the atmosphere and ocean that have sampling strategies. The more carefully tion of the island chain between the not received as much attention. These controlled, a posteriori simulations offer South China Sea to the west and the smaller-scale winds and currents are important insights into complex regional 48 Oceanography | Vol.24, No.1 kinematics and dynamics. Although graupel (snow pellets). In the two-way oceanic and atmospheric COAMPS there are other numerical studies of coupling scheme, the atmospheric forecasts were supplied to researchers on the region (e.g., Metzger and Hurlburt, model passes parameterized surface the cruises in real time (see Pullen et al., 1996; Han et al., 2009), the NRL coupled momentum, heat, and moisture fluxes 2011, for details of how the model guid- atmosphere-ocean model is a unique through ESMF to the ocean model. ance was used in ship-track planning). tool for simultaneous high-resolution The ocean, which also uses a Mellor- Additional simulations, using two-way studies of both the meteorology and Yamada level 2.5 turbulent boundary coupled COAMPS with higher ocean oceanography of a region. layer mixing scheme, completes the resolutions were run later for the time two-way coupling by returning the sea period that encompassed the IOP-08. CoUPLED AIR-SEA ModEL surface temperature through ESMF to From these simulations, a 60-day period NRL numerical simulations for the act as a lower boundary condition to covering January through February 2008 Philippines used the Coupled Ocean/ the atmosphere. The coupled model and a 40-day period covering February Atmosphere Mesoscale Prediction supports grid nesting, which permits through March 2009 were selected for System (COAMPS), which consists progressively higher resolution with comparison here. For the 2008 time of an atmospheric model coupled each grid refinement—typically imple- period, nested atmospheric grid meshes with the Navy Coastal Ocean Model mented at a 3:1 ratio. Further details of 27, 9, and 3 km ran above nested (NCOM). The atmospheric model is of the parameterizations, nesting, and ocean grids of 3 and 1 km. There were nonhydrostatic with a terrain-following coupling in COAMPS may be found 40 vertical levels in both the atmosphere vertical coordinate (Hodur, 1997), while in Chen et al. (2010). and ocean models. After initializing the ocean model features hydrostatic Regional forecast models require with NOGAPS and gNCOM fields on dynamics, generalized curvilinear external information in the form of January 1, 2008, the coupled model horizontal coordinates, a bathymetry- initial conditions, lateral boundary was run for 60 days with atmospheric following vertical coordinate, and a free conditions, and timely observations. The data assimilation cycles every 12 hours. surface (Barron et al., 2006). Air-to- atmospheric data assimilation of quality- For 2009, nested atmospheric grids of ocean two-way coupling is accomplished controlled upper-air sounding, surface, 27 and 9 km were used with a 9-km through the community-based Earth commercial aircraft, and satellite data is ocean grid. In the 2009 model, hori- System Modeling Framework (ESMF; achieved with the use of a Multivariate zontal spatial domains differed slightly Earth System Modeling, 2010), which Optimum Interpolation (MVOI) from those in 2008, reflecting a ship allows each model to use flexible, scheme. Ocean observations, which course that was further south. Also, generalized grids and a configurable include floats, drifters, moorings, satellite because of the coarser resolution (9 km) air-sea coupling interval, typically every sea surface temperatures (SST), conduc- of the ocean model run during the 2009 10 to 15 minutes. Atmospheric physical tivity-temperature-depth/expendable field experiment, it is not considered parameterizations for surface fluxes use bathythermograph (CTD/XBT) profiles, in the comparisons. In all the cases bulk aerodynamic formulae based on and satellite altimetry, are assimilated by presented here, NCOM ran without tidal the TOGA-COARE (Tropical Ocean the MVOI-based NRL Coupled Ocean boundary forcing. Global Atmosphere-Coupled Ocean Data Assimilation System (Cummings, Atmosphere Response Experiment) 2005). The initial state and lateral ATMospHERIC FORCING algorithms (Fairall et al., 1996), a boundary conditions for the outermost DURING IOP-08 And IOP-09 level 2.5 Mellor-Yamada boundary atmospheric and oceanic nests are from Surface wind is an important forcing layer representation, short- and the 1° Navy Operational Global Analysis mechanism for the upper ocean on both long-wave radiation parameteriza- and Prediction System (NOGAPS) and global and local scales, and is particu- tion, and moist processes, including the operational 1/8° Global NCOM larly evident in circulation patterns in explicit microphysical equations for (gNCOM) forecast fields, respectively. and around the Philippine Archipelago. rain, snow, cloud water, cloud ice, and During IOP-08 and IOP-09, daily In nearby and surrounding waters, Oceanography | March 2011 49 large-scale stresses from the Pacific trade Sea (7°N, 120°E). A series of wind-speed consequence of the partial blocking of winds are responsible for the North maxima and minima are noticeable cross-island monsoon flow by the moun- Equatorial Current and the resulting sea features on the leeward or downwind tains (e.g., Gabersek and Durran, 2004). level differences between the western side of the Philippine Archipelago (see Downstream wakes form from air Pacific and the South China Sea. Figure 1a in Gordon et al., 2011, for descent, acceleration, wave breaking, and Seasonal monsoon winds are also known locations of major islands and passages). potential vorticity generation that occur to modulate circulation in the adjacent The most prominent wind jets occur over the islands (Smith et al., 1997). South China Sea (Metzger