[ianbicking]

The Miyawaki method [1], which I'm sure has been on HN before, is a very different approach to these projects. When I first read about it, it seemed like a kind of too-good-to-be-true miracle approach, but reading further it's really just a lot of hard work.

Site preparation is a huge part of it. This photo gallery [2] gives some sense. It starts with soil testing and soil amendment, I doubt they ever consider the soil "good enough" at the outset. I'm not sure if they also do any hydrological changes? Then they plant a dense and diverse set of trees. I'm not clear how many trees ultimately survive. There's theories about the set of trees you'd use, but I can only imagine some of the process is just natural selection, and a belief that early density is positive to later growth.

Bringing it back to technology, I do wonder what tools could support this kind of higher-effort higher-impact forestation. It seems like there's work to be done performing soil tests and understand the results and recommended amendments, including some decision trees around tests and results. There's general guidance on the choice of trees, but it requires matching that guidance against local conditions and local plants.

In some ways the process is simpler than landscaping a house: you aren't trying to get a perfect set of plants, and you aren't imposing other requirements. You're really trying to build a mini ecosystem, and the ecosystem is there to do a lot of the work on its own.

I am less sure how this approach translates to more marginal locations. It's a bit easier to rapidly create a lush and vibrant forest in India than at the edge of a desert. Most of the examples are in tropical locations.

[1] https://www.crowdforesting.org/miyawaki-model/forest-kerala

[2] https://www.greenyatra.org/miyawaki.php

[sn41]

Hmm. Pleasantly surprised to find Kerala being mentioned in this context - I wasn't aware of this.

But on a related note, the south west of India has a rich tradition of "Sarpa Kavus", literally, "Serpent Shrine", but which are in reality, sacred groves in some corner of the yard of many traditional homes (see [1] for a typical example) - these are mostly left to themselves for most of the year, except for a couple of festival days. In practice, it is almost a biome within the yard.

[1] https://nandakishorevarma.wordpress.com/2016/12/30/a-sacred-...

[hinkley]

Some of these notions of building a forest have parallels in the still developing field of probiotics versus prebiotics. Setting the dominoes instead of trying and failing to set the scene.

It turns out that building a healthy forest is a long difficult process that can span multiple administrations and in some cases lifetimes. Building the conditions for a second growth successional forest is something most of us can watch in real time.

These things are quality over quantity, which requires some cleverness in order to leverage. Forests (vs tree farms) spread by mycelium, by root, by seed, and by wing, and pretty much in that order. You'll get more success planting the entire perimeter of an intact forest than planting a rectangular area next to it, and more success planting a rectangular area next to an intact forest than planting a random hill in the middle of a clearcut. I have a hypothesis that planting rich islands within line of sight of each other and then letting nature in-fill between them also works better, but I have seen no research supporting or refuting that hypothesis. Nature corridors seem to be pretty close to this model and those have been proven.

One thing I'd like to see us do is move away from square and rectangular clearcuts toward more linear ones. Perhaps on contour, and leaving support species instead of nuking everything before replanting. See also research by Suzanne Simard and her peers on the soil food web.

[sitkack]

I did a fun experiment in my yard, I took a small mixture of grass seeds (store bought bird seed) and planted them in various places in the yard (after sprouting). It was amazing to see what did and did not thrive in various places due to sun, water and soil.

One place supported all 4 kinds, the other three only supported 1 or 2 kinds of grasses.

I would think reforesting a barren land might have to take multiple phases of growth to prepare the soil, ability to hold water, fungal colonies to extract nutrients, etc.

[antisthenes]

The Miyawaki model is incredibly labor intensive and requires far more sophistication in monitoring and planting methods than developing countries are usually willing to commit to mass planting projects.

90%+ of these mass planting "1 million trees in 30 seconds" projects is usually little more than putting sticks in the ground, hoping some of them make it, with little regard for survivability, usefulness, tree species nativity, etc.

[UncleOxidant]

It seems like some of the most successful reforestation projects are driven by a few very committed individuals who live on the land and want to see the forest return. Like the case of India's "Forest Man" https://www.npr.org/sections/parallels/2017/12/26/572421590/...

[antisthenes]

Trees are living organisms that need care and help adapting to a changing environment, so I'm not surprised.

The old and magnificent trees we see today in streets and neighborhoods suffer from extreme survivor bias. Many tree species aren't even that large/long-lived as people tend to think about average trees.

There's also a strong correlation between speed of growth and longevity.

Going for the quantity approach that many of these low-effort projects do is little more than just tossing seeds out of a bag onto the ground. In the end you'll get more trees than you had before, but probably not by much, and with varying success.

[mabbo]

> I do wonder what tools could support this kind of higher-effort higher-impact forestation

What's neat is that the tools already exist. Much of modern farming is a data problem- knowing soil conditions and nutrient levels across a large area and which plants would work best where. They often make use of satellite data, watershed simulation, weather and climate models.

I wonder if anyone has documented using those tools for this purpose.

[atonalfreerider]

Ceres is one that I know about: https://www.ceresimaging.net/customer-stories