My simplified suggestion was meant to include water availability as a variable. I admit that soil type would also place additional constraints on gameplay, but it needn't get into such detail as the parts potasium to parts sodium in the soil. If that were the case, the only real solution is to import clays from different regions, and radically overhaul the clay types. (Some clays are volcanic, others are sedimentary, for example. They have very different soil properties, but both are usually just fine for pottery. If the local clay is deficient, which is why parts potasium is low, you would need a way to import a buttload of a different clay, and till it in, thus changing the soil tile type. As is, this would be rife for abuse, because you could simply import a bunch of fire clay, and keep tilling it into a 1x1 tile farm, until it is basically pure fireclay, unbuild the farm, designate it as a clay source, and totally break imports as the means of getting more. Same with sand, etc.)
Well, I don't think I've ever suggested that we actually change a soil layer type.
Soil in this suggestion is more like the contaminant of mud that's on the floor of caverns (in fact, I even suggested just making it a contaminant at one point), where the sandy silt soil merely directs a starting point for drainage. In fact, since there is a biome variable for drainage, I think it better to use the fact that one soil layer may be sand or clay as merely a modifier of biome drainage. (Sand in a swamp will offer less drainage than sand in a hill biome. In fact, if proximity to a slope could be factored into starting drainage, it would be even more ideal.)
Hence, importing a lot of clay would only affect what sort of plants could grow there.
Of course, there could be something abusable about making your soil very acidic in a desert or significantly reducing drainage in that, since part of the idea is to make weeds just sprout naturally like grass, but make grass and the like play by the same rules as crops, if you make your soil significantly different from what the nearby grasses like, you can keep out the most likely weeds.
(Because it is a complex topic, the notion of tracking in seeds of invasive species - or young of pests - is kind of sketchy, and just based on invisible off-board population models, with possibly a division into different population counts based upon layer.)
I could see alkaline<->acid as a variable, because that would influence weather or not you grow wheat (prefers alkaline) or strawberries (prefers acid), but going too far is a problem.
I can agree with this - aside from soil nutrients and pollutants, it's just drainage and acidity for the long-term biome-based variables.
Soil nutrients are short-term things based upon composting, as I already said (and honestly, if we want to play up carbon more, then H
20, C, N, P would be fine instead of H
2O, C, N, P, K and just fudge K into P, although that eliminates potash, but where it might be somewhat easier for players to understand) and have a value for its complexity I've already argued for.
Beyond that, the pollutants exist as penalties for specific types of actions. Biological pollution is basically the penalty for hot fertilizer - "nuking the soil", as you put it earlier, with uncomposted chicken manure as opposed to allowing it to lose a good portion of its N and taking up time, space, and labor composting it. Heavy Metal is for composting certain creature types, especially apex predators and sentients, in excess, and also for magma-irrigation, which can boost fertility tremendously, but render many plants incapable of dealing with heavy metals.
Salinity is something that borders a pollutant and a biome trait, however, but it can be used for penalizing brackish/salt water irrigation as well as reducing some lands to salt rushes until you can seep enough salt from the soil.
Rather than be "soil building" directly, was more implying that it was soil building, *if* you did not harvest the crop, tilled it under, and left the plot fallow for one full year, because that is really what you have to do with soil building crops to re-establish biomass.
Instead, the desired protocol I had in mind, was to make agriculture expensive, because it is. The system I proposed (or thought I proposed anyway...) would have required soil neutral plants that have a low nutritional yeild be the primary route to soil building, via a very inefficient composting process on the side.
That's largely consistent with what I suggested, as well - only tiny few plants do well without biomass, and there are some "starter grasses/molds" that exist just for building up the soil and functionally converting the biome from a desert to a marginal grassland.
Plain dumping compost and working on getting some grasses going with water before you work on getting real crops going is the idea for turning a patch of desert into a garden. (Short of tundra/glacier, deserts would be agricultural hard mode for good reason.)
Eg, you grow a boat load of ratweeds, because it is all your rotten marginal embark can grow to start with. You use some of it to make alcohol, so you don't get the game over screen, and rely on imported food and possibly poultry to survive starvation. The rest of it, you directly shove into the composter.
That's actually what I was suggesting plump helmets do. They're basically an edible composter, capable of being grown anywhere dark with nothing more than a log and water, and where the end result is (partial?) compost and a mushroom. You'd have to keep supplying logs to keep growing plump helmets, though, so it's not long-term sustainable, but it's how you can jump-start composting.
Of course, that's just a suggestion of a specific means of implementing the broader idea I was already talking about, especially with regards to making an easy start-up.