These days everyone seems to thinks that "planting trees" is an important solution to the climate crisis. They're mostly wrong, and in this paper we explain why. Instead of planting trees, we need to talk about people managing landscapes. 1/x

ok - any news on who I should vote for in the St Paul ward 4 city council race? My absentee ballot is arriving today.

I hope our modeling out scenarios using best available data helps highlight limitations of our datasets and assumptions, as well as broader limitations of global models, and helps stir discussions about where and how to best restore forests. This was my goal in engaging in this.

We know our models are flawed (which is one reason we put out many scenarios rather than a single preferred model), and we probably disagree about how useful these models are (I’m pretty skeptical of their practical utility).

Since the paper has been out for a few weeks already, and has gotten some news coverage, I’ve seen some critics pointing to places where our data is poor or particular modeling assumptions we made don’t play out on the ground. This, I think, is a good thing!.

I’m particularly angry about this this week because we’ve heard that there will be no new funding for NASA’s Land Cover and Land Use Change program - which supports integration of social scientific and remote sensing based understanding of land systems - in the next few years.

If anything close to those resources was devoted to studying the actual causes of forest degradation and loss and restoration and gain - all of which lie in the social scientific realm - we might make some real progress on understanding these issues and developing good policy.

Spending decades building an infrastructure for earth observing satellites, developing complicated new data processing algorithms for satellite data and incredibly detailed sensors and building vast networks of computers to analyze all this data was *really* expensive.

In my work on this topic, I’ve sometimes been told that this is because its too difficult - too expensive - to have detailed social scientific data. Do you know what’s expensive?.

I also want to note that the quality of social scientific data we have relevant to restoration is *dramatically* poorer than the quality of biophysical data, and in particular relative to the vast amount of data about tree characteristics being generated by remote sensing.

However I do think its a big deal that the opportunity to mitigate climate change through forest restoration is probably only about a tenth as big as previous analyses have shown - and that tenth is beset by a bunch of difficult tradeoffs with other values.

I’m sure that if we had better data on decision-making, land tenure security, and livelihoods, we’d find places in South Asia which have really ripe conditions for beneficial rewtoration - eg parts of Northeast India, where there is strong land tenure and local decision-making.

Its possible I disagree with some of my coauthors here, but to me the data is too coarse on nearly all fronts, and the numbers too dependent on minor variations in assumptions.

I’m intentionally *not* emphasizing the topline numbers of how much land is available for forest restoration, or how much carbon it would store, nor am I suggesting that anyone use this to locate specific restoration sites.

Restoration research and policy-making based on European, North American, or even Brazilian models may not apply well to the rest of the world because social, economic, and political conditions are very different.

Rather, we are pointing to the fact that in most of the world it will be necessary to engage closely with questions of land tenure, local decision-making, and rural livelihoods in order to give restoration any chance of success.

Also, we are *not* saying restoration “should” happen in certain places. We recognize that our data is too coarse to make local decisions. From my view, the only way that decisions can be made about where to put restoration is through working with locally affected communities.

Restoration has benefits and it would be wrong to deprive people of restoration benefits when they are already deprived of the benefits of secure land tenure and democratic decision-making.

We are *not* saying that restoration should not happen anywhere in Africa or Asia (and much of Latin America too).

Interestingly, these regions of the world are where most restoration ecology research happens, but its important to note that they are profoundly unrepresentative of the social and political conditions of most of the world.

Basically areas in the US, Canada, Europe, Colombia, Brazil, and a few smaller countries meet these broad criteria. Most of the rest of the world does not.