• Characterize the convective stage of the Cb using ground-based instruments, ELDORA, EDOP, and possibly the Citation.

• Characterize the tracer distribution, aerosols (CCN, composition), and thermodynamic state in the lower-tropospheric region where the convection is initiated using Twin Otter and P-3.

• Sample the anvil microphysical properties with as much vertical and horizontal coverage as possible and through as much of the anvil lifecycle as possible using the WB57, Citation, and ER-2/ELDORA/ground-based remote sensing instruments.

• Measure the radiative fluxes above and below the anvil using primarily the ER-2, Twin Otter, and ground-based radiometers.

• Characterize the tracer concentrations and aerosol composition in and around the anvil through as much of its lifecycle as possible.

• If possible, we should also try to coordinate these flights with Aqua overpasses.

Hopefully, we will be able to forecast the locations of relatively small, isolated Cbs. Assume the forecasts give us a location (pcon) and time (tcon) where deep convection is expected to pop up. As the system forms and moves, pcon will change.

ER-2: Launch about 30-60 min. before tcon and fly to pcon. (This location may change before or during the transit. NPOL should provide this information.) First, fly a box pattern around pcon with relatively long legs (perhaps 200 km), and launch a dropsonde along each leg. (Note: dropsonde launches over land probably won't possible.) Then fly a simple box pattern along the axis of the anvil outflow (predicted by forecast winds or mesoscale model runs). The length of the along-wind legs should be long enough to take the aircraft a few minutes beyond the edges of the cloud system (see the plan view diagram). These legs will likely lengthen as the anvil matures. The length of the cross-wind legs should be long enough such that a significantly different part of the anvil is sampled on the two along-wind legs. This pattern should be continued until the system dissipates or the ER-2 time-on-station is exhausted.

Proteus: Same as ER-2, but without the initial survey pattern.

WB57: Launch at about tcon and fly to pcon. (This location should have been updated by this time given NPOL reports and observations from the P-3 and ER-2.) First, fly a leg over the cloud system along the axis (i.e., below the ER-2 flight leg) extending into the air upwind of the cloud system. Use visual observations and reports from NPOL/ELDORA to select a cloud-top altitude, and fly back into the anvil top downwind of the convective cores. Then execute a box pattern similar to the ER-2 pattern, except that it is entirely downwind of the convective cores. On each of the cross-wind legs outside the anvil, descend about 1000-3000 feet (depending on the cloud-top altitude). Continue stairstepping down until FL450 is reached, then stairstep back up. The legs should extend about 5 minutes out of the anvil boundary. Continue until time-on-station is exhausted (even if anvil dissipates). The pilot may be unwilling to fly toward the convective core inside the cloud. An alternative would be to fly toward the core above cloud, then turn and dive into the cloud deck for the outbound leg.

P-3: Launch about 1 hour before tcon and fly to pcon (again, using updated location from NPOL.) During the convective stage of the system, fly legs in the boundary layer parallel to the tropospheric shear at least 10 km off to the side of the of the convection. As the anvil forms, these legs will be extended to cover the full extent of the anvil. If the anvil is laterally extensive, a box pattern may be necessary for complete coverage. Once the convection is uncoupled from the boundary layer, the aircraft will ascend out of the boundary layer to 10,000 ft and continue the anvil measurements. If the convection is aligned perpendicular to the shear vector, then fly legs perpendicular to the shear at least 10 km from the convection in the up-shear or down-shear direction, depending on which direction the anvil extends. When the anvil extends beyond 40 km from the convection, the aircraft will begin flying a T-pattern or box pattern to cover the full extent of the anvil.

CIT: We should wait as long as possible to launch the Citation given its short duration. Once the convection is clearly identified by the radars, launch and fly to pcon. Fly stairstep patterns similar to WB57 pattern starting at cloud-base and going up to Citation ceiling. The Citation will probably be able to fly further into the convection that the WB57. When ceiling is reached, spiral down along edge of convection. If anvil persists, the Citation will probably need to land and refuel.

TWO: Launch 1 hour before tcon, fly to pcon and fly a cloverleaf pattern around pcon to sample the regional environment. Sample lower tropospheric aerosols in the vicinity of convective system as long as possible. Include vertical profiling from the surface up to the aircraft ceiling as well as legs upwind of the forecast convection location. If a persistent anvil forms, refuel and fly legs underneath the anvil and the ER-2 flight path. Continue this pattern as long as possible.

• If we can reliably select an ideal line, we might want to just fly back and forth on the same along-wind line instead of using the box pattern.

• If the outlfow anvil consists of thin layers, the stairstep pattern may miss them. In this case, we might need to porpoise the WB-57 in order to sample the thin layers.