DRAFT,  v2.5 29 coast (Johns et al., 1995).  An important component to this flow is a southward-flowing current   from the North Atlantic along the shelf and slope regions.  Cooler, fresher water (more buoyant)  associated with both coastally trapped, buoyant outflows from estuaries and rivers and this slope  current interacts with warm, higher salinity tropical water to form filaments, rips, and eddies in the  oceanic flows (Marmorino and Trump, 1994).  In this domain, coastal surface current patterns     from HF radar reveal a complex time-space continuum of flows across the North Carolina shelf  including 3 to 7 day intrusions of the GS’s North Wall.  This is also a region of extra-tropical  cyclogenesis during the winter months, which can form the Nor’easters that impact the weather in  the MAB (Austin and Lentz, 1999).    1.2 Implications for system layout The physical setting within the SEACOOS domain suggests overall design considerations for the observing system.   There clearly is a need for ocean-side boundary conditions. This is particularly challenging for the SEACOOS domain because of the presence of the Loop Current - Florida  Current - Gulf Stream system.  Known variations in cross and along shelf transports dictate regular  monitoring at the shelf-break, mid-shelf, inner shelf at a minimum; and the complex inshore    waters, which support strong spatial gradients and are highly structured, will require significant  investment to do well.    The monitoring will be accomplished through a combination of fixed and moving in-situ and  remote sensing platforms.  Fixed in-situ sampling platforms are expected to be the principal  observing tool in the system’s first implementation, and ideally the spacing between platforms is  set by correlation scales of the dominant processes.  To take advantage of existing infrastructure,  however, SEACOOS has begun with an initial set of fixed platforms determined by the history of  observing in the Southeast. 1.3 History of Observing in the Southeast– initial conditions for SEACOOS SEACOOS began as a collaboration between existing subregional efforts and components of federal observing systems, each with its own design rationale.  Though not ideal, this sets the starting point (initial conditions) for the system. Main federal assets include the NDBC CMAN and buoy systems, the NOS NWLON, and USGS  water resources monitoring network.  The NDBC network provides relatively sparse but region-  wide spatial coverage.  Specific locations have been based on political priorities in some cases, and  the positioning is driven by atmospheric criteria, not oceanographic. Each measures a number of  meteorological parameters (wind speed and direction, air temperature and humidity, barometric  pressure) with some measuring water temperature and scalar surface waves.