IPCC tier 1 vs tier 2 vs tier 3 for forest carbon reporting

Anyone who has sat across from a registry reviewer knows the tier question comes up before the methodology question does. Before anyone argues about activity data or emission factors, the first thing on the table is: what tier are you reporting at, and can you defend it.
The IPCC's tiered structure for AFOLU inventories states how much of the uncertainty in your biomass numbers comes from global defaults versus your own measurements. Each step up the ladder swaps a borrowed number for one you can trace back to your own landscape.
Tier 1: defaults and broad strokes
Tier 1 runs on the IPCC's global and regional default values for biomass expansion factors, carbon fractions, and growth rates, stratified by broad climate and forest type. No national forest inventory required, no plot network, no biomass model calibrated to your species mix. You plug in activity data (hectares of forest, hectares converted) and let the defaults do the rest.
It's defensible for a first submission or a jurisdiction without an established forest monitoring program, and the IPCC Guidelines explicitly allow it as a starting point. But the uncertainty bands are wide, because a default growth curve for "tropical moist forest" doesn't know your watershed has unusually dense stocking or a disturbance history that pulls the real number well off the regional average. For a national communication under the UNFCCC it can be enough. For a jurisdictional REDD+ program or a project under a voluntary standard, a reviewer will ask why you're still here.
Tier 2 and Tier 3: where the country's own data starts doing the work
Tier 2 keeps the same basic equations but replaces the IPCC defaults with country-specific or region-specific emission factors and activity data, usually built from a national forest inventory, permanent sample plots, or species-level allometric equations developed for your forest types. It's the tier most mature NFI programs land on by default, because the plot network already exists for timber or biodiversity purposes and the carbon accounting piggybacks on it.
Tier 3 goes further: repeated measurement of the same plots over time, or model-based estimation systems that integrate growth, yield, and disturbance at a resolution fine enough to track change at the management-unit level rather than the national average. Good Practice Guidance treats Tier 3 as the appropriate level for categories the inventory flags as "key" (large or rapidly changing pools), because that's where a coarse estimate does the most damage to the overall inventory's credibility.
The catch with both Tier 2 and Tier 3 is the same catch that's been limiting forestry agencies for decades: a permanent sample plot network is expensive to install and more expensive to keep remeasuring on a five- or ten-year cycle, especially across remote or contested terrain. Budgets get cut, cycles slip, and the carbon stock change estimate drifts back toward something closer to Tier 2 defaults dressed up as Tier 3.
Where a remote-sensing biomass layer fits
This is the gap wide-swath multispectral imagery was built to close, not by replacing your plot network but by giving it something to regress against between remeasurement cycles. A biomass layer built from canopy structure and spectral response, refreshed on an annual rather than decadal cadence, lets you hold a Tier 3-style spatial resolution on stock change without waiting for the next full field campaign to catch a harvest, a fire, or an unexpected growth pulse. Biomass Mapping builds exactly this: a landscape-scale biomass surface derived from your existing imagery and whatever plot data you already have calibrated against it, so the tier you report at doesn't quietly slide backward between inventory years.
If your program is eyeing a move from Tier 2 to Tier 3, or trying to hold a Tier 3 inventory steady without another field campaign this cycle, it's worth seeing what an annual biomass layer looks like over your own landscape.