I think that it wouldn't work without redesigning the cards somewhat. The point is that co-adapted predator-prey links are safer against both invasives and certain events; but without further information on the cards, you can't determine what's co-adapted to what.
OK, I'm going to try to strip this down to the bare essentials, the simplest possible computation. In order to do that, I'm moving the whole concept of size; it's no longer part of initial compatibility/playability of a card, as climate and terrain are, but rather a part of predator/prey coadaption, and thus important only for invasive attacks and starvation/overpredation events.
First off, there's an issue of terminology. In order to discuss this, you need to draw a distinction between predation attack/defense, and invasion attack/defense. Perhaps by judicious use of a thesaurus, you could find 4 distinct terms, but I'm just going to use the modifiers.
The upper left corner would have foodchain level, as currently. But size would be replaced by movement - have some color or icon code for spread, conditional spread (explained in text), move, or flight. You'd have a third number next to these two, which would be a "breeding speed/growth" value from 0-4 (4 used just for plants, 3 mostly plants and the most rapacious animals). Then, to the left of these three values, you'd have 0-3 "predation defense" icons inscribed in shields, from the set:
small, midsized, big, yucky, armor, fast, flies, burrows, hides, fights, flowers
Duplicated (extra-strong) defenses are possible. So are x-ed out defenses (special vulnerabilities, they count as "overmatched" in the presence of the corresponding attack). Anything from mouse to rabbit size is midsized. Bugs have 1 "small", bacteria have 2. Deer have 1 "big", oxen to elephants have 2. Moles would have 2 "burrows" defenses, rabbits and potatoes 1, and foxes 0 (they only sleep there, it's not their main defense from mountain lions). Thorns would be armor. Flowers is a special "defense" which only counts to help protect pollinators/fructivores against invasives.
Then, under the picture, you'd have any number of "attack" icons, the same as the defense icons without the shields.
When initially playing a card, predation attack/defense would not matter. You could play a spider eating an elephant or an orca eating a giant clam, whatever. Compatibility would be solely a matter of food-chain level, habitat, and climate.
Later addition: So, the predation attack/defense icons would ONLY matter in resolving event cards. There would be 3 basic kinds of event cards which use those icons: invasive attacks, overpredation events, and predator failure/starvation events.
When resolving an "invasive" attack event, the base stat is the breeding speed/growth number. Invasive defender wins ties; if the invasive attacker is superior by one you flip a coin. The invasive defender gets +1 for each predator or prey beyond the first, and -1 for each parasite. The invasive defender can also choose a single predator or prey to "help defend". If the "defense helper" is a prey, then both invasive attacker and invasive defender get +1 for each matched predatory defense of the prey, and +2 for each overmatched defense. If the "defense helper" a predator, both invasive attacker and invasive defender get +2 for each unmatched predatory defense they have except "flowers", +1 for each equally-matched predatory defense, and 0 for each overmatched predatory defense.
The "starvation" event would be reusable by either player as long as it were on top of the discard pile. It would remove a predator if it breeds faster than the prey. The prey breeding (that is, the amount of "catchable" prey) would get -2 for each unmatched predator defense of the prey.
The "overpredation" event would also be reusable. It would remove a prey if its predator breeds as fast or faster. The prey's breeding would get +1 for each unmatched predator defense it has, and -1 for each overmatched defense it has.
There could also be special "invasion" event cards, which give one-time bonuses to an invasion. That would be fun: invasive pandas! No, all invasive attacks would be event cards.
Some results of these rules:
- A stable ecosystem would have slower-breeding predators and faster-breeding prey. Duh. (Later edit: actually, this is a very important learning goal; it is really part of the ABCs of ecosystems. I believe this is one of the major advantages of my proposal; kids do not need to be taught that bigger animals eat smaller animals, but they do need to be taught this.) Thus slow-breeding plants and fast-breeding predators would be high-point cards.
- a fast-breeding predator could "overpredate" and kill your opponents' cards. But it would also be vulnerable to "starvation". Similarly, fast-breeding animals would be effective "invasive" attackers, but might then die from overpredation or starvation.
- Species which currently have the "invasive" characteristic would either have a high breeding number or they'd have, in the text, a bonus to breeding for the purposes of invasion.
- A card with many predators and/or prey on the board would be essentially immune from invasion.
- Anything that could eat bacteria could eat bugs. Oh well. (I don't think that the game should even strive for perfect realism. It's impossible anyway, and unreal combinations are part of the fun.)
- Predator-prey combinations which aren't co-adapted are likely to be vulnerable to at least one of starvation or overpredation. Thus, while you could play an orca feeding on a giant clam, your opponent would have an easy way to attack that link, removing one or both.
- These rules now allow prey vulnerabilities (I edited them). "Cats come to Australia and eat all the kangaroo mice" (special vulnerability against "fast" attack) or "invasive snakes in polynesia eat all the birds' eggs" (special vulnerability against "hides" attack) would be an interesting dynamic to have in the game.
I think that the mechanics of this proposal are manageable. It actually makes initial "compatibility" checking easier, because size becomes a matter for events and invasive attacks, not for initial playability/compatibility. Invasive attack/defense is the most complicated, but even there, it's a single stat each for attacker/defender, and the stat is calculable by looking at the number of neighbors and the actual content of only two cards.
It is also just... fun. It gives the game more of a battle mechanic, but in an eco-consistent way: rabbits don't kill kangaroo mice in paw-to-paw combat, but they can drive them out all the same. I can easily imagine some kind of Pokemon crossover rules with this dynamic - although your average Pokemon would quickly overpredate all its food sources and then die.
Any questions? (I'm sure I haven't been quite clear enough about all the details, but I hope you get the idea.)