There is a strong scientific consensus that the primary driver of climate change is the release of previously sequestered carbon locked away in fossil fuels into the environment. But a new study reminds us that there is another contributor that must be accounted for in climate models – changes in land use.

The core claim of climate change is actually quite simple, and has not been successfully refuted by climate change deniers. So-called “greenhouse gases” in the atmosphere warm the planet because they reflect more infrared radiation back down to the surface, so that less of it escapes the Earth. Without this effect the Earth would be a snowball.

It is irrefutable at this point that adding more greenhouse gases to the atmosphere will increase this effect, resulting in more warming. The only real question is – how much warming? This is where things become ridiculously complex. Climate scientists use models to predict what will happen as more CO2 is released into the atmosphere, but it is very difficult to model a complex system. This is why there are large error bars on projections of future warming.

As a quick aside, carbon dioxide (CO2) is not the only greenhouse gas, but it is the major one. Methane is a more potent greenhouse gas (GHG) than CO2, but it does not last as long in the atmosphere. Within the first 20 years after its release, methane is 84 times as potent a GHG than CO2, but only 34 times if you consider its effects over 100 years. Much less methane is released into the atmosphere than CO2, but it is not negligible and needs to be considered in climate models.

CO2, however, is the primary driver of climate change. What is important in terms of CO2’s effect on climate is the amount of CO2 in the atmosphere. CO2 in the oceans decreases the Ph (acidifying the ocean) and this causes other problems, but does not cause warming. So in order to calculate how much additional CO2 there will be in the atmosphere we not only need to know how much is being released, we need to know where all that carbon is going. Essentially released carbon is going into the environmental carbon cycle, which spends only part of its time in the atmosphere.

This is where climate models run into extreme complexity. Not only do scientists need to figure out how much carbon is in all the various places that environmental carbon goes, then need to figure out how this will change with higher overall carbon in the cycle. For example, high CO2 levels in the atmosphere allow plants to grow more vigorously, increasing their potential to store carbon. But there are limits to this phenomenon, as most plants are not optimally adapted to higher CO2 levels. But perhaps that will change – there are lots of unknowns when trying to extrapolate into the future.

Also, the ability of the oceans to absorb CO2 is likely not constant, so we have to monitor how that changes with higher CO2 levels.

The new study adds another factor that must be included in climate models – changes to the ecosystem by changing land use. When we cut down a forest and plant crops, how does that change the carbon cycle? They found that:

Projected global greening and land C (carbon) storage are dampened, in all models, by 22% and 24% on average and projected C loss by disturbances enhanced by +49% when LULCC (land-use and land cover changes) are taken into account.

So their models indicate that the amount of the carbon we will release into the atmosphere over the next century that will be bound in plants, and therefore not in the atmosphere, is reduced by 22-24%. This is an important, if modest, adjustment to climate models, but every such adjustment matters. Unfortunately, this one indicates that we need to adjust estimates of future climate change up when we consider likely changes to land use.

This also means it is vital to consider the overall environmental effects of changes in land use, which include impacts on climate models. The bottom line is that trees are great at capturing CO2 from the atmosphere and keeping it locked away for a long time. It doesn’t matter that the carbon is eventually released. What’s important is the percentage of carbon in the environment that, at any point in time, exists in plants vs as CO2 in the atmosphere. Even though carbon may be slowly turning over in a forest, the forest is still locking up a lot of carbon.

Cutting down forests for farmland, therefore, has a significant negative impact on the climate by reducing an important carbon store, and reducing the environment’s ability to buffer released carbon.

I have discussed before that perhaps the most significant effect on the ecosystem from agriculture is land use. Improving efficiency, the ability to grow more food on less land, is therefore critical. Even if our population stabilizes, and everyone has food security, we still want to increase agricultural efficiency and return farmland to natural ecosystems. Anyone who lives in New England and has seen a stone wall in the middle of a dense forest has see the results of this. As farming moved from New England to other parts of the country, much former farmland returned to forest.

The current study adds to the evidence that land use is a dominant factor when considering agriculture’s effects. The study also adds evidence to the notion that we should protect the world’s forests as an important natural resource. In addition to many things, they are an important store of environmental carbon.