Changed the pipe to not using assigments --> like this it prevents from using...

Changed the pipe to not using assigments --> like this it prevents from using too much memory and writes it directly to the drive. I also added the physical land and assumed the tnuts share for the harvested area to be the same for phys area. Yields are excluded.

SPAM_Data_Transformation/SPAM_no_agg_spatial.R

@@ -282,24 +282,30 @@ sph.harv = read_sf(dsn = wd, layer = "GADM_treenut_rest_ha")
 sph.all = st_join(sph.prod, sph.harv, join = st_within)
 # write_sf(sph.all, dsn = here("Input_data", driver = "ESRI shapefile"))
 sph.prod.tib = as_tibble(sph.prod)
+sph.prod.tib = dplyr::select(sph.prod.tib, UID, GID_0, GID_1, trnt_r_)
+
 sph.harv.tib = as_tibble(sph.harv)
+sph.harv.tib = dplyr::select(sph.harv.tib, UID, GID_0, GID_1, trnt_r_)
+
 # loading SPAM-GADM join
 
 spam.gadm.link = read_sf(dsn = here("Input_data", "GADMtoSPAM", "GADM_to_SPAM_within"), 
                          layer = "GADM_to_SPAM_within")
 
-# Join spam.gadm.link with sph.harv/prd 
 
 # Production
 # rbenchmark::benchmark(
-dat_join_prod = spam.gadm.link %>% 
+spam.gadm.link %>% 
   as_tibble() %>% 
-  left_join(., prd_dat, by="alloc_key") %>%
   left_join(.,sph.prod.tib, by="UID") %>% 
+  left_join(.,prd_dat, by="alloc_key") %>% 
   mutate(tnuts = trnt_r_*rest) %>% 
-    mutate(rest.1 = (1-trnt_r_)*rest) %>% 
+  mutate(rest.1 = (1-trnt_r_)*rest) %>% 
   dplyr::select(alloc_key, UID, GID_0.x, GID_1.x, rec_type, tech_type, unit,
-         whea:rest, tnuts, rest.1, geometry.x, geometry.y, x, y)
+         whea:rest, tnuts, rest.1) %>% 
+  write.csv(here("Input_data", "prod_tnuts_spam_gadm"))
+
+  
 
 # Approach validation > do the numbers make sense?
 xy=dat_join_prod %>% 
@@ -312,24 +318,36 @@ write_csv(dat_join_prod, here("Input_data", "dat_treenut_prod.csv"))
 rm(prd_dat, spam.gadm.link, sph.prod, sph.prod.tib,x,xy)  
 # Harvested area 
 # rbenchmark::benchmark(
-dat_join_harv = spam.gadm.link %>% 
+spam.gadm.link %>% 
     as_tibble() %>% 
     left_join(., har_dat, by="alloc_key") %>%
     left_join(.,sph.harv.tib, by="UID") %>% 
-    mutate(tnuts = trnt_pd*rest) %>% 
-    mutate(rest.1 = rest_pd*rest) %>% 
-    mutate(rest.2 = rest-tnuts) #%>% 
-    select(alloc_key, UID, GID_0.x, GID_1.x, rec_type, tech_type, unit, whea:rest, 
-           tnuts, rest.2, geometry.x, geometry.y, x,y)
+  mutate(tnuts = trnt_r_*rest) %>% 
+  mutate(rest.1 = (1-trnt_r_)*rest) %>% 
+  dplyr::select(alloc_key, UID, GID_0.x, GID_1.x, rec_type, tech_type, unit,
+                whea:rest, tnuts, rest.1) %>% 
+  write.csv(here("Input_data", "harv_tnuts_spam_gadm"))
 # )
 
-write.csv(dat_join_harv, here("Input_data", "dat_treenut_harv.csv"), row.names=T)
-save.image("dat_join_prod")
-# Join harvested area (sph) with SPAM 
+# Physical land
+# rbenchmark::benchmark(
+spam.gadm.link %>% 
+  as_tibble() %>% 
+  left_join(., phy_dat, by="alloc_key") %>%
+  left_join(.,sph.harv.tib, by="UID") %>% 
+  mutate(tnuts = trnt_r_*rest) %>% 
+  mutate(rest.1 = (1-trnt_r_)*rest) %>% 
+  dplyr::select(alloc_key, UID, GID_0.x, GID_1.x, rec_type, tech_type, unit,
+                whea:rest, tnuts, rest.1) %>% 
+  write.csv(here("Input_data", "phys_tnuts_spam_gadm"))
+
+
+# # Calculate yields 
+# 1. join the two dfs
+# 2. Gather crops 
+# 3. Spread harv. area and production >> ha | Prod | ha*prd(conversion from mt to kg) = yield
 
-# Join production files caluclate ratio treenuts/adjust units
 
-# Calculate yields 
 
 # Validate results 
 
@@ -348,7 +366,7 @@ food_cat <- read_csv('foodnonfood.csv')
 nam_cat <- read_csv('crop_names.csv')
 # loading FAO country codes
 FAOcntr <- read_csv('FAO_countries.csv')
-FAOcntr <- select(FAOcntr, 'ISO3', 'FAOSTAT_CODE', "GAUL_CODE", 'Continent', 'Subcontinent')
+FAOcntr <- dplyr::select(FAOcntr, 'ISO3', 'FAOSTAT_CODE', "GAUL_CODE", 'Continent', 'Subcontinent')
 
 # joining all together
 names(dat.comb)[8] <- "Crop_short"