constraints: Establishes per pixel probability of fields being converted...

In PrincetonUniversity/agroEcoTradeoff: Landuse Tradeoff analysis model

Description

This function determines the per pixel probability of a field being converted for a crop as a function of that crop's yield, as well as any constraints related to carbon, biodiversity, and travel costs, factoring in any prior yield modifications made (e.g. due to climate change or added irrigation). This version uses data.tables rather than raster and reframes the constraints as per yield potential.

Usage

constraints(inlist, cbetas, silent = TRUE)

Arguments

inlist	A list of data.tables for the four constraints
cbetas	4 element vector (values 0-1, summing to 1) of land use weights
silent	TRUE, otherwise FALSE gives verbose mode
code	Unique simulation code resulting from run_code function
cropnames	Vector of crop names in analysis

Details

For inlist, the input should be a named list, with the first element named "Y, then "C", "BD", and "COST". The weights must sum to 1, regardless of how they are apportioned. For weights of 1/3, write them to at least the third decimal place (code tolerates summing to 0.999)

Value

data.table of conversion probabilities for each crop

Examples

rc <- run_code(input_key = "ZA")
il <- fetch_inputs(input_key = "ZA")  # fetch all necessary inputs
# ybetas <- list(1, 1)
# ybeta <- yield_mod(inlist = il_dt[[2]][c("p_yield")], 
#                   ybetas = ybetas, code = rc, cropnames = il$cropnames)
                   
inlist <- list("y_std" = il$y_std, "C" = il$carbon_p, "bd" = il$cons_p, 
               "cost" = il$cost_p)
CRSobj <- spatial_meta("ZA")$crs

# Checking case where yields are excluded
cbetas <- c("y_std" = 0, "C" = 0.333, "bd" = 0.333, "cost" = 0.333)
con1 <- constraints(inlist = inlist, cbetas = cbetas, code = rc, 
                    cropnames = il$cropnames)
#con1[, lapply(.SD, max, na.rm = TRUE)]

# all four constraints
cbetas <- c("Y" = 0.25, "C" = 0.25, "BD" = 0.25, "COST" = 0.25)
con2 <- constraints(inlist = inlist, cbetas = cbetas, code = rc, 
                    cropnames = il$cropnames)
con2r <- dt_list_to_raster(base = il$mask, inlist = list(con2), 
                           CRSobj = CRSobj)[[1]]
plot(con2r)
# current function versus shorter code (current version much faster)
system.time(con2 <- constraints(inlist = inlist, cbetas = cbetas, code = rc, 
                                cropnames = il$cropnames))

PrincetonUniversity/agroEcoTradeoff documentation built on May 8, 2019, 3:12 p.m.