Splash dispersal multi-binomial in a pixel

Generate a two dimensional probability distribution representing the probability of spore dispersal from an inoculum source. Generates the probability distribution with each entry representing a single plant. Pixel size represents the whole spatial size

Usage

multi_var_binom(
  row_spacing = 0.3,
  stdev = NULL,
  px = 1,
  seeding_r = 40,
  IO = 1,
  m
)

Arguments

row_spacing

Crop row spacing in meters

stdev

Standard deviation of splash distribution. defaults to pixel size

px

pixel size, length and width of square pixel

seeding_r

number of plants sown in a pixel (px)

IO

number of infective sites within the pixel

m

(optional) a matrix representing each plant and the number of infective sites for each plant

Value

probability matrix where each entry gives the probability of splash dispersal from infective sites to the respective entry

Examples

set.seed(7)
# obtain matrix of probabilities
mvb <- multi_var_binom()
image(t(mvb))

# sample the plant (matrix cell) where the spore may drop
sample(seq_along(mvb), size = 10, prob = mvb, replace = TRUE)
#>  [1] 20 20 19  7 21 44 20 21 20 19
# sample the cardinal coordinates (matrix cell) where the spore may drop
sam_mvb <- sample(seq_along(mvb), size = 10, prob = mvb, replace = TRUE)
data.frame(x = sapply(sam_mvb,FUN = function(x) ceiling(x/ nrow(mvb))),
           y = sapply(sam_mvb,FUN = function(x) x %% nrow(mvb)))
#>    x y
#> 1  1 7
#> 2  2 8
#> 3  2 9
#> 4  2 8
#> 5  1 8
#> 6  2 8
#> 7  4 9
#> 8  2 7
#> 9  1 8
#> 10 2 7

# simulate spore dispersal over 40 spread events
for(sp in 1:40){
if(sp == 1){
  m1 <- multi_var_binom()
  m0 <- matrix(0,nrow = nrow(m1),ncol = ncol(m1))
  ind1 <- sample(seq_along(m1), size = 1, prob = m1, replace = TRUE)
  m0[ind1] <- 1
}else{
  m1 <- multi_var_binom(m = m0)
  ind1 <- sample(seq_along(m1), size = 1, prob = m1, replace = TRUE)
  m0[ind1] <- m0[ind1] + 1
}
}
image(t(m1))