The Assessment of Skill for Coastal Ocean Transients (ASCOT) project is a series of real-time Coastal Predictive Skill Experimentation/Rapid Environmental Assessment (CPSE/REA) experiments and simulations focused on quantitative skill evaluation and cost-effective forecast system development. ASCOT-01 is to be carried out in Massachusetts Bay/Gulf of Maine in June 2001. The ASCOT overall goal is to enhance the efficiency, improve the accuracy and extend the scope of nowcasting and forecasting of oceanic fields for CPSE and for REA in the coastal ocean and to quantify such CPSE and REA capabilities.

REA requires multiscale capabilities for different kinds of warfare (e.g. anti-submarine (ASW), mine warfare (MW), etc.). An experiment which is to assess the predictive skill of a forecast system must therefore measure and evaluate on multiple scales. Knowledge of the multiscale dynamics is essential. For ASCOT-01, the coupling extends from Massachusetts Bay, through the Gulf of Maine, out to the northwest Atlantic. Skill metrics will be designed to take the coupling of scales into account. As a predictive skill experiment, ASCOT-01 will include oversampling, in order that sources of error can be tracked. During the verification survey a significant fraction of the initialization survey will be repeated. Adaptive sampling survey patterns will be designed to address: 1) the interactions of Massachusetts Bay and the Gulf of Maine (inflow updates, exchanges, etc.); 2) response to storms or air-sea exchanges (upwelling, structures of currents and gyres, bifurcation structures in the Gulf of Maine, etc.); 3) coupling of wind-response and buoyancy currents; reduction of multivariate forecast errors; 4) the assessment of key biogeochemical-ecological processes and, 5) update of information for feature model parameters. Such scenarios will be designed in advance through OSSEs.

Many field campaigns in coastal and oceanic waters have attempted to set biological investigations in the context of physical studies, but few of these have been at the cutting edge of modern physical oceanography. The opportunity to embed an inquiry regarding the regulation of coastal system biological production into a physical study as sophisticated as ASCOT-01 is unprecedented. After an initial survey, physical, biological, and chemical data will be assimilated to guide adaptive sampling each day, which will maintain a bay-wide scale synoptic assessment (with sub-mesoscale resolution) and will facilitate the siting of subsequent sampling observations in dynamically interesting places. Guided by and armed with these tools, strategic sampling for the assessment of key biogeochemical-ecological processes can be efficiently deployed. The survey aspect of this project will lend itself to data assimilation and forecast methodologies with a level of sophistication that has never before been attained in ecosystem studies. The more nimble strategic sampling effort will enable assessment of near instantaneous response of biological systems to stimulating and enhancing perturbations at an equally unprecedented level.

1. obtain data sets adequate for: 1) definitive real-time verification of regional coastal ocean predictive skills, with and without REA constraints; 2) CPSE and REA Observational System Simulation Experiments (OSSEs), both for ASCOT design and more generally; and, 3) definitive knowledge of dynamics
2. define skill metrics and real-time forecast validation and verification procedures for REA
3. assemble, calibrate, exercise in real-time, evaluate and improve a generic, portable, scalable advanced ocean forecast system (dynamical models and data analysis, management and assimilation schemes) applicable for CPSE in general and NATO REA in particular.

1. carry out and quantitatively evaluate in Massachusetts Bay (MB) and the Gulf of Maine (GOM) a coupled multiscale interdisciplinary real-time forecast experiment
2. obtain an intensive data set adequate for definitive quantitative skill assessment and suitable for the design of minimal data requirements for both REA and for an efficient regional monitoring and prediction system
3. obtain a data set adequate to define multiscale physical dynamical processes (sub-meso-, meso-, bay-, gulf- scales) that govern the formation and evolution of structures and events, including generic processes and the coupling of wind-forced events and buoyancy currents and capable of providing a context for interdisciplinary forecasting and process studies
4. maintaining a continuous synoptic description of Mass. Bay with mesoscale resolution throughout and sub-mesoscale resolution as dynamically required via surveys and adaptive sampling

To include a biological component that will enable testing of hypotheses that link the physical and biological event scales in the coastal ocean and provide the basis for coupled forecasting.

1. obtain a companion suite of data for biological state variables, and fully embed these in the real-time forecast experiment
2. using forecasts for the evolution of physical processes and events, and for the biological responses to them, strategically assess these biological reaction rates for the further refinement of models.

The primary platform from which data will be collected will be the NRV Alliance. Plans are underway to include MIT Autonomous Underwater Vehicles (AUVs) as a component of the main experiment. Additional vessels (e.g. R/V Lucky Lady (UMass.-Dartmouth), R/V Neritic (UMass.-Boston), R/V Aquamonitor (MWRA-Battelle)) will be utilized for: fine-scale resolution of bay features, extended measurements in the Gulf of Maine; coupled and interdisciplinary experimentation (biogeochemical/ecosystem dynamics, acoustical dynamics, etc.); relating ocean variability to acoustic variability and coherence; and maintenance of a synoptic picture of Massachusetts Bay. The ASCOT-01 operational center will be aboard the NRV Alliance.

Data analysis, data assimilation and numerical simulations will be carried out on a daily basis in real-time throughout the duration of the exercise. In situ data will be acquired by the NRV Alliance as well as by the other chartered vessels or ships of opportunity. Remotely sensed data will be available via SACLANTCEN or other sites. Data will be analyzed, quality controlled and processed as it is received and made available for assimilation into the Harvard Ocean Prediction System (HOPS).

The ASCOT-01 simulation and operational system will consist of a set of three two-way nested domains: the Northwest Atlantic (NWA), the Gulf of Maine (GOM) and Massachusetts Bay (MB). In the operational context, there will be two-way nesting between the NWA and GOM (NWA/GOM) domains and the GOM and MB (GOM/MB) domains. The NWA/GOM nested run will provide boundary conditions for the GOM during the GOM/MB nested run.

Forecasts will be available on a daily basis after the initialization survey in order to provide adaptive sampling patterns for the subsequent day's sampling. Products will be available via the experiment web site. Example products might include (for both the Gulf of Maine and Massachusetts Bay modeling domains): synoptic maps and forecasts of temperature or salinity with superimposed velocity vectors for levels of interest, vertical sections of chosen quantities at locations of interest, profiles of temperature or sound speed at locations of interest, etc. It is desirable to have the forecasts carried out in two modes: in Predictive Skill Assessment mode - i.e. using all data as acquired in order to most accurately predict future states; and in REA mode - i.e. using a reduced data set in order to mimic REA conditions and demonstrate the ability to utilize minimal data.