When speaking about the effectiveness of carbon offsets in addressing emissions, two things are critically important - that they’re real, and that they’re permanent. Real means it really happened, and that it was at least a tonne. Planning to plant a tree next year is a worthy aspiration, but it won’t pull down atmospheric CO2 today.
Permanent means that it not only happened, but that it’ll last.
When you take a flight from Christchurch to Sydney, you turn about 55kg of jet fuel into around 180kg of CO2 (and a lot of heat, water, and thrust). That CO2 will stay in the atmosphere for a long time, so if you want to offset it by paying someone to suck some CO2 out of the atmosphere, you’ll want to be confident they’ve sucked out enough, and they’re going to need to keep it out of the atmosphere a long time too.
If you’re the one sucking it out of the air using native forest on your land, and you sell a CarbonCrop Unit (CCU), you’re not just selling rights to the tonne of CO2 your forest removed last year, you’re also selling a promise that you’re going to do your damndest to keep it there, and that you’ll clean up if it goes wrong.
If you sold a tonne without really sucking out a tonne, or without a commitment to keep it locked up in a tree, your removal doesn’t mean much.
It’s that commitment - and the mechanisms which enforce it - that means we can all rest a little cooler.
How do we ensure it’s real?
The ‘real-ness’ of sequestration in forest can be hard to determine - it equates to the additional carbon stored in the biomass of the forest - primarily in the wood of trees. A typical hectare of regenerating native forest might only sequester a few tonnes of CO2 equivalent each year, and it could have hundreds of trees. In the high biodiversity native forest incentivised by CCUs, these trees will be of different species, and ages, and growing conditions. Measuring all of them by hand is out of the question. Historically, this has been a real obstacle to accuracy, and hence real-ness and permanence.
We approach the problem with remote sensing data from satellites and aircraft, state of the art machine learning methods, and targeted field sampling.
CarbonCrop uses the same core AI+remote-sensing enabled technologies to assure both the ‘real’ and ‘permanent’ aspects of offsets are met, together with contractual obligations with landowners and methodology provisions around how and when offsets are issued, and what needs to happen if sequestration permanence (also called ‘durability’) goes awry.
We have trained computer vision models to identify vegetation species mix at sub-meter resolution using aerial imagery, along with canopy density and extent, and regeneration stage. We estimate carbon stock and sequestration rates for this vegetation at the same 1m resolution across a monitored area, by taking these AI-derived profiles and fitting them to environment-adjusted models of forest species growth from independent research.
We’re going to continue calibrating and validating these models with ongoing field spot surveys across our entire portfolio of registered projects - we’re in the very early stages of this at present, but fortunately we can get things started by incorporating thousands of openly available measurements of specific areas of forest across New Zealand - many thanks to Manaaki Whenua’s National Vegetation Survey dataset!
There’s a lot more to this, but we’ll leave the details for our Methodology (take a read), and additional information we’ll release in the future as we complete our IP protection activities, but the key here is that we make the output visible and traceable. When we issue a CCU, it has a unique ID, and is linked to a unique Carbon Tracking Region. We make the carbon stock and sequestration maps of that CTR available for audit and review, and they’re a map that’s referenced to the real world. You can go there in Google maps and see for yourself the forest that’s backing the CCU you’re using.
How do we ensure it’s permanent?
That’s ‘real’ taken care of - but now we need to make sure it’s permanent. We do this with the same core technology: tech that’s good at identifying when a forest is sequestering carbon is also very good at recognising when it’s going backwards.
For all active projects registered with CarbonCrop, we take an ongoing monitoring responsibility along with monitoring and incentivising ongoing sequestration. The durability term landholders commit to for every single CCU issued is 100 years - aligned with the widely used ‘100 year Global Warming Potential’ equivalent term used by the Intergovernmental Panel on Climate Change. For every CarbonCrop project, the landholder enters into a binding contract with us which - along with the methodology - sets out their obligations in connection with CCUs, especially to future holders of their CCUs.
This contract places various obligations on the landholder - our methodology has many details, but in particular:
If the forest suffers an accidental reversal (e.g. gets blown over in a storm), the landholder is obligated to enable the restoration of new forest on that site, and will not receive new CCUs until the new forest achieves the same carbon stock as the old one.
If the landholder deliberately clears the forest, they need to secure and surrender a volume of CCUs equivalent to what has been issued to the area of forest, so they can make good the buyers.
If the landholder ever sells the land, the obligation doesn’t go away. It has to stay with the land, moving to the new owner.
If CarbonCrop ever awards too many CCUs to a forest through an error in modelling, the landholder has to wait as new issuances of CCUs are paused for that Carbon Tracking Region until the forest’s true sequestration catches up.
In addition to these provisions, the landholder also agrees to allow CarbonCrop to direct 10% of all CCUs issued to a CarbonTrackingRegion to a collective ‘buffer’ pool. These CCU aren’t owned by the landholder - nor even for the benefit of CarbonCrop (though CarbonCrop manages the pool at present). They are shared across all registered projects, and can be deployed only as defined under the methodology, to make good any unremedied reversals which may occur.
We consider that these provisions are well aligned with industry norms, and we believe our monitoring technology is state of the art. Nature-based solutions like forest restoration are critical to filling the gap in the climate change response while emissions reductions scale up - and to sorting out the emissions already in the atmosphere. We’re confident that the technological and contractual mechanisms associated with our CCUs will allow them to deliver on the promises being made, by buyers to sellers, and by both buyers and sellers to the climate.