Adding ground silicates can significantly increase agricultural productivity on various kinds of land. Here's one review article about the subject: "Reminerali

Adding ground silicates can significantly increase agricultural productivity on various kinds of land. Here's one review article about the subject: "Remineralizing soils? The agricultural usage of silicate rock powders: A review" https://www.sciencedirect.com/science/article/pii/S004896972106054X https://www.sciencedirect.com/science/article/pii/S004896972... IIRC the soils that benefit the most include: - Soils depleted in micronutrients - Acidic soil - Silica depleted soil (significant for rice farming and other silica-sensitive crops) Most of the world's acidic and depleted soils that could benefit the most are in warm tropical regions with heavy rainfall. That's also where fresh mineral applications would weather most rapidly to draw down CO2. Unfortunately, it may also be harder (logistically and culturally) to encourage silicate applications on those soils than on the temperate climate soils of North America. After a certain point adding more silicates doesn't help increase soil productivity. The one thing that I haven't found any reporting on is where the "too much" point is reached for soil amendment with crushed silicates. I have seen that yields stop increasing with ever-increasing addition of ground rock. I haven't seen a point where they go down. I would guess that there's some point where yields actually start dropping, and that would mark the upper limit for agricultural soil amendments. Up to the point where yields keep increasing, you could expect farmers to spread crushed silicates on their fields for their own benefit, depending on the cost. In the high-application case where yields flatten out but carbon sequestration continues to increase, you'd need to pay subsidies to get higher application rates. You'd also need to pay subsidies in the lower-application-rate cases if yields increase but not enough to justify the expense on a yield basis.