Submetering

You need data gathered by each energy end use, so yes you must cover all of it. How you get the data depends on what is the most cost-effective means to gather it. So sub-metering is not specifically required. You can gather this data with spot measurements, short term trending meters, or even derive it by subtraction (i.e. you measure all electrical use except plug loads so you can derive plug loads by subtracting from the whole building electric meter).

Your focus in Option D is to gather the data you need to calibrate the energy model. So your plan should detail how you are going to gather the data. The data may not even be direct energy use. For example, a measurement of run time on a constant volume fan will allow you to derive energy use.

The level of monitoring varies depending on the project. Some projects may be small enough to lump all HVAC together and still enable a reasonable M&V approach. Some larger projects would probably require data gathering from system components.

M&V is pretty simple - gather the data you need to calibrate your energy model; calibrate your energy model to the utility bills (monthly and annually); create a revised baseline model from your calibrated model; the difference is your verified savings. Basically measure - verify.

You could use the sub-metered room data to verify spot or short-term trend measurements in a sampling of rooms. Assume you are sub-metering electric only right? For all the plug in devices you could use Kill-a-watt meters. For the hard wired you could use clamp on meters for spot measurements or trending meters like the ones from Hobo to gather the data you need. Add it up and use the whole room meter to verify your measurements. So you could gather data in say 10% of the rooms, apply occupancy data for the rooms and extrapolate for the whole building.