library(DataflowR)

library(sp)

template <- 201308

zero_cutoff <- 120

percent_cutoff <- 0.80

good_years <- read.csv("data/goodyears.csv")[,1]

good_years <- good_years[-which(good_years %in% c(201002, 201102, 201209))] #rm small spatial extent

dt <- lapply(good_years, function(x) coordinatize(streamget(as.character(x)), latname = "lat_dd", lonname = "lon_dd"))

loop_ind <- 1:length(dt)

template_ind <- which(good_years == template)

template_dt <- dt[[template_ind]]

loop_ind <- loop_ind[-template_ind]

res_ind <- 1:nrow(dt[[template_ind]])

res <- list()

for(i in loop_ind){

cur_sub <- dt[[i]]

cur_sub <- remove.duplicates(cur_sub)

cur_zdist <- zerodist2(template_dt, cur_sub, zero = zero_cutoff)

length(unique(cur_zdist[,1])) #res should be equal

res[[i]] <- res_ind[res_ind %in% cur_zdist[,1]]

# print(i)

# print(length(res[[i]]))

}

res_table <- data.frame(table(do.call("c", res)))

res_table$Freq <- res_table$Freq / length(res)

res_table <- res_table[res_table$Freq > percent_cutoff,]

plot(template_dt[res_table[,1],], main = paste(percent_cutoff, zero_cutoff))