rules.txt

We need the following matrices
1) Nutrient matrix
2) Sunlight matrix
3) Water matrix
4) Age matrix

Values of each element of the matrices range between 0 and 1
Normalization will be done by linear interpolation or sigmoid curve
This function has to be decided experimentally, whichever gives better results

In age matrix 1 means there is no tree in that cell
In age matrix >.8 means the tree is decaying
All matrices will update every iteration
Number of trees we consider "adjacent" has to be set before simulation 

- Nutrient Matrix -
--Soil nutrient will decrease if there are more trees in the vicinity competing for resources
--The further the competitor, the lesser the competition
--Decaying trees increase soil nutrient over time

nutrientConsumption = 0
nutrientSupply = 0
for each tree in adjacent trees:
	distanceFactor = distance from i,j
	// Dead trees don't consume, middle aged trees consume most
	nutrientConsumption += if age of an adjacent tree > .8 0. else gaussian(tree.age * 10/8)/distanceFactor 
	// Nutrients supplied by dead trees decrease over time
	nutrientSupply += (if age of an adjacent tree > .8 then 1-(age-.8 * 5) else 0)/distanceFactor 

nutrient[i][j] = normalize(nutrientSupply - nutrientConsumption) // Limit between 0 and 1

- Sunlight matrix -
-- Sunlight depends on number of adjacent trees, their age and distance

cumulativeFactor = 0
for each tree in adjacent trees
	ageFactor = normalizedGaussian(tree.age)
	// If the gaussian distribution does not work as expected we can try a linear function
	//ageFactor = if tree.age > .8 then (1-tree.age*5) else tree.age * 10/8
	distanceFactor = 1/distance of it from i,j
	cumulativeFactor += ageFactor + distanceFactor

sunlight[i][j] = 1 - normalize(cumulativeFactor)

- Water matrix -
--The further the competitor, the lesser the competition
--Older trees consume more water

cumulativeFactor = 0
for each tree in adjacent trees
	distanceFactor = 1/distance of it from i,j
	// Older trees consume more water but dead trees do not consume water
	ageFactor = if tree.age > .8 then 0 else sigmoid(tree.age*10/8) 
	cumulativeFactor += distanceFactor + ageFactor
	
water[i][j] = 1 - normalize(cumulativeFactor) 

- Age matrix -
--Every iteration increment age by a random value in a range
--If the tree is unhealthy, have a shorter life
health = nutrient[i][j] * water[i][j] * sunlight[i][j]
if age[i][j] is not 1 then age[i][j] += (1/heath) + value between .001 and .005 or something IDK

- Chance of birth -
--If there are no trees in the forest, age[i][j] for all i,j is 1 (totally decayed)
--If a tree is born, we say age[i][j] = 0
--chance depends on reproductive capability of adjacent tree and their distance
cumulativeChance = 0
for each tree in adjacent trees
	// Assuming middle aged trees are more reproductive
	ageFactor = if tree.age > .8 then 0 else gaussian(tree.age * 10/8)
	distanceFactor = 1/(distance of tree from i,j)
	cumulativeChance += ageFactor + distanceFactor

if Math.random() < normalize(cumulativeChance) then age[i][j] = 0