added load_data for shiny app for Zimbabwe

drought/trigger-model/Drought_trigger_dashboard/load_data_zwe.R

@@ -1,88 +1,47 @@
 
-countries <- c("Kenya" = 1, "Ethiopia" = 2,"Mozambique"=3)
+countries <- c("Zimbabwe" = 1)
 levels <- c("LEVEL 2" = 2, "LEVEL 3" = 3)
 
 
 
 ###########3read admin boundaries 
+zwe <- sf::read_sf("./shapefiles/zwe/zwe_admbnda_adm2_zimstat_ocha_20180911.shp") %>%
+  dplyr::select(ADM1_PCODE, ADM2_PCODE, ADM0_EN)
 
-Ethiopia <- sf::read_sf("./shapefiles/Ethiopia/eth-administrative-divisions-shapefiles/eth_admbnda_adm2_csa_bofed_20201008.shp")%>%
-  dplyr::select(ADM1_PCODE,ADM2_PCODE,ADM0_EN)
-
-SA_admin2_region <-sf::read_sf("./shapefiles/SA_admin2_region.shp")%>%
-  dplyr::select(ADM1_PCODE,ADM2_PCODE,ADM0_EN)
-
-Mozambique <-SA_admin2_region%>%filter(ADM0_EN %in% c('Mozambique'))
-
-Lesotho <-SA_admin2_region%>%filter(ADM0_EN %in% c('Lesotho'))
-Namibia <-SA_admin2_region%>%filter(ADM0_EN %in% c('Namibia'))
-
-
-kenya<- sf::read_sf("./shapefiles/kenya/KEN_Adm2.shp")%>%
-  dplyr::mutate(ADM1_PCODE=DSCodeAdm1,ADM2_PCODE=DSCodeAdm2,ADM0_EN='KENYA')%>%
-  dplyr::select(ADM1_PCODE,ADM2_PCODE,ADM0_EN)
-
-Mozambique_lhz <-sf::read_sf("./shapefiles/MZ_LHZ_2013.shp")%>%
-  dplyr::mutate(ADM1_PCODE=LZCODE,ADM2_PCODE=LZCODE,ADM0_EN=COUNTRY)%>%
-  dplyr::select(ADM1_PCODE,ADM2_PCODE,ADM0_EN)
-Lesotho_lhz <-sf::read_sf("./shapefiles/LS_LHZ_2011.shp")%>%
-  dplyr::mutate(ADM1_PCODE=LZCODE,ADM2_PCODE=LZCODE,ADM0_EN=COUNTRY)%>%
-  dplyr::select(ADM1_PCODE,ADM2_PCODE,ADM0_EN)
-
-Ethiopia_lhz <-sf::read_sf("./shapefiles/Ethiopia/ET_LHZ_2018/ET_LHZ_2018.shp")%>%
-  dplyr::mutate(ADM1_PCODE=LZCODE,
-                meher=ifelse(grepl("Cropping-Meher", LZTYPE),1,0),
-                belge=ifelse(grepl("Cropping-Belg", LZTYPE),2,0),
-                Agropastoral=ifelse(grepl("Agropastoral", LZTYPE),3,0),
-                pastoral=ifelse(grepl("Pastoral", LZTYPE),4,0),
-                crop_calander=meher+belge+Agropastoral+pastoral,
-                ADM2_PCODE=LZCODE,
-                ADM0_EN=COUNTRY)%>%
-  dplyr::select(ADM1_PCODE,ADM2_PCODE,ADM0_EN)#,LZTYPE,crop_calander)
-
-kenya_lhz <- sf::read_sf("./shapefiles/kenya/KE_LHZ_2011.shp")%>%
-  dplyr::mutate(ADM1_PCODE=LZCODE,ADM2_PCODE=LZCODE,ADM0_EN=COUNTRY)%>%
-  dplyr::select(ADM1_PCODE,ADM2_PCODE,ADM0_EN)
+zwe_lhz <-sf::read_sf("./shapefiles/zwe/ZW_LHZ_2011_fixed.shp") %>%
+  dplyr::mutate(ADM1_PCODE=LZCODE, ADM2_PCODE=LZCODE, ADM0_EN=COUNTRY) %>%
+  dplyr::select(ADM1_PCODE, ADM2_PCODE, ADM0_EN)
 
 # create a data frame with admin boundaries 
 
-Admin_all<-rbind(Ethiopia,kenya,SA_admin2_region,Mozambique_lhz,Lesotho_lhz,Ethiopia_lhz,kenya_lhz)
-
-All_df<-read.csv("./data/all_indicators.csv")%>%
-  dplyr::mutate(date=ymd(date1))
+Admin_all<-rbind(zwe,zwe_lhz)
+admin <- list(zwe)
 
-admin <- list(Ethiopia,kenya,Mozambique,Lesotho,Namibia)
 
 
+Admin_all1 <- Admin_all
+st_geometry(Admin_all1) <- NULL 
 
-Admin_all1<-Admin_all
-st_geometry(Admin_all1)<-NULL 
-
-vci_df<-read.csv("./data/vci.csv")%>%
-  dplyr::mutate(date=ymd(date))%>%
-  dplyr::mutate(ADM2_PCODE=factor(pcode))%>%
-  dplyr::select(ADM2_PCODE,date,vci)%>%
+vci_df <- read.csv("./data/zwe_vci.csv") %>%
+  dplyr::mutate(date = ymd(sprintf("%s0101", Year))) %>%
+  dplyr::mutate(ADM0_EN = "zwe") %>%
+  dplyr::mutate(vci = VCI_0) %>% 
+  dplyr::select(ADM2_PCODE, date, vci) %>%
   left_join(Admin_all1,by='ADM2_PCODE')#%>%filter(ADM0_EN %in% c('KENYA','ETHIOPIA'))
-#group_by(ADM1_PCODE,date)%>%  dplyr::summarise(vci=mean(vci),ADM0_EN=first(ADM0_EN))%>%ungroup()%>%filter(ADM0_EN %in% c('Mozambique','Namibia','Lesotho'))
 
+#====================
 rain_df_daily<-read.csv("./data/Daily_rainfall.csv")%>%
   dplyr::mutate(date=ymd(date))%>%
   dplyr::mutate(ADM2_PCODE=factor(pcode))%>%
   dplyr::select(ADM2_PCODE,date,rain)%>%
   left_join(Admin_all1,by='ADM2_PCODE')#%>%  filter(ADM0_EN %in% c('KENYA','ETHIOPIA'))
 #group_by(ADM1_PCODE,date)%>%  dplyr::summarise(rain=mean(rain),ADM0_EN=first(ADM0_EN))%>%ungroup()%>%filter(ADM0_EN %in% c('Mozambique','Namibia','Lesotho'))
+#====================
 
 
+ipc_zwe <- read.csv("./data/zwe_ipc.csv") %>% mutate(date = ymd(as.Date(Date)))
 
-ipc_KE<-read.csv("./data/ipc_admin_KE.csv")%>%mutate(date=ymd(as.Date(Date)))
-ipc_eth<-read.csv("./data/ipc_admin_ETH.csv")%>%mutate(date=ymd(as.Date(Date)))
-ipc_lhz_ETH<-read.csv("./data/ipc_lhz_ETH.csv") %>%rename(ADM2_PCODE='LZCODE')%>%mutate(date=ymd(as.Date(Date)))
-ipc_lhz_KE<-read.csv("./data/ipc_lhz_KE.csv") %>%rename(ADM2_PCODE='LZCODE')%>%mutate(date=ymd(as.Date(Date)))
-
-ipc_MZ<-read.csv("./data/ipc_admin_SAP.csv") %>%mutate(date=ymd(as.Date(Date)))
-ipc_lhz_MZ<-read.csv("./data/ipc_lhz_SAP.csv") %>%rename(ADM2_PCODE='LZCODE')%>%mutate(date=ymd(as.Date(Date)))
-
-ipc_data<-rbind(ipc_KE,ipc_eth,ipc_lhz_ETH,ipc_lhz_KE,ipc_MZ,ipc_lhz_MZ) 
+ipc_data <- rbind(ipc_zwe) 
 
 
 all_days <- tibble(date = seq(min(ipc_data$date),max(ipc_data$date) , by="months"))
@@ -96,17 +55,14 @@ ipc_filled <- as.data.frame(merge(all_days, tibble(ADM2_PCODE = unique(ipc_data$
   dplyr::ungroup()%>%left_join(Admin_all1,by='ADM2_PCODE')%>%dplyr::mutate(ipc_class=CS)
 
 
-vci_rain_df <- vci_df %>% full_join(ipc_filled%>%dplyr::select(ADM2_PCODE,date,ipc_class), by=c("ADM2_PCODE","date"))
-
-
-ENSO1<-read.csv("./data/ENSO.csv") %>% gather("MON",'ENSO',-Year)%>% 
+ENSO1 <- read.csv("./data/ENSO.csv") %>% gather("MON",'ENSO',-Year)%>% 
   arrange(Year) %>%
   dplyr::mutate(date=seq(as.Date("1950/01/01"), by = "month", length.out = 852))%>%
   filter(date>= as.Date("1980/01/01"))
 
 #Dipole Mode Index (DMI)
 
-IOD<-read.csv("./data/IOD_DMI_standard.csv") %>%
+IOD <- read.csv("./data/IOD_DMI_standard.csv") %>%
   gather("MON",'IOD',-Year) %>% arrange(Year)%>%
   dplyr::mutate(date=seq(as.Date("1870/01/01"), by = "month", length.out = 1812),IOD=as.numeric(IOD))%>%
   filter(date >= as.Date("1980/01/01"))
@@ -118,54 +74,27 @@ SST_var<-ENSO1%>%
 
 all_days <- tibble(date = seq(min(vci_df$date),max(vci_df$date) , by="months"))
 
-dmp_filled <- merge(all_days, tibble(ADM2_PCODE = unique(vci_df$ADM2_PCODE)))
+df_filled <- merge(all_days, tibble(ADM2_PCODE = unique(vci_df$ADM2_PCODE)))
 
-dmp_filled<-as.data.frame(dmp_filled) %>% dplyr::mutate(ADM2_PCODE=factor(ADM2_PCODE))
+df_filled <- as.data.frame(df_filled) %>% dplyr::mutate(ADM2_PCODE=factor(ADM2_PCODE))
 
-dmp_filled2 <- dmp_filled %>%
+df_filled2 <- df_filled %>%
   left_join(vci_df%>% dplyr::mutate(ADM2_PCODE=factor(ADM2_PCODE)), by = c("ADM2_PCODE", "date"))
 
 
 
-All_df_filled <- dmp_filled2 %>%dplyr::select(ADM2_PCODE,date,vci)%>%
-  dplyr::select(ADM2_PCODE,date,vci)%>%left_join(Admin_all1,by='ADM2_PCODE')%>%
-  left_join(SST_var,by='date')%>%
-  dplyr::group_by(ADM2_PCODE) %>%arrange(date)%>%fill(vci, .direction = "downup")%>%
+All_df_filled <- df_filled2 %>% dplyr::select(ADM2_PCODE,date,vci) %>%
+  dplyr::select(ADM2_PCODE,date,vci) %>% left_join(Admin_all1,by='ADM2_PCODE') %>%
+  left_join(SST_var,by='date') %>%
+  dplyr::group_by(ADM2_PCODE) %>% arrange(date) %>% 
+  fill(vci, .direction = "downup") %>%
   dplyr::ungroup()
 
-# All_df_filled_admin1 <- dmp_filled2%>%dplyr::select(ADM2_PCODE,date,vci)%>%
-#   left_join(rain_df, by = c("ADM2_PCODE", "date"))%>%
-#   dplyr::select(ADM2_PCODE,date,vci,rain_mon)%>%left_join(Admin_all1,by='ADM2_PCODE')%>%
-#   group_by(ADM1_PCODE,date)%>%
-#   dplyr::summarise(rain_mon=mean(rain_mon), 
-#                    ADM0_EN=first(ADM0_EN),
-#                    vci=mean(vci))%>%
-#   ungroup()%>%left_join(SST_var,by='date')%>% 
-#   dplyr::group_by(ADM1_PCODE) %>%fill(vci, .direction = "downup")%>%
-#   dplyr::ungroup()
-
-
-
-
-Emdat_impact_sff<-sf::read_sf("./data/Emdat_impact.shp")%>% 
-  dplyr::mutate(total_affected=TtlAffc,no_affected=NAffctd)
-
-
-
-
 
-Kenya_impact<-st_join(Emdat_impact_sff %>%  filter(Country=='Kenya'), kenya) %>% st_set_geometry(NULL)
-Ethiopia_impact<-st_join(Emdat_impact_sff %>%  filter(Country=='Ethiopia') , Ethiopia) %>% st_set_geometry(NULL)
-#Mozambique_impact<-st_join(Emdat_impact_sff %>%  filter(Country=='Mozambique') , Mozambique) %>% st_set_geometry(NULL)
+Emdat_impact_sff <- read.csv("./data/impact_zwe.csv") %>% 
+  dplyr::mutate(total_affected = TtlAffc, no_affected = NAffctd)
 
-Mozambique_impact<-Mozambique%>%left_join(read.csv("./data/impact_moz.csv",sep=';'),by='ADM1_PCODE')%>%
-  dplyr::select(ADM1_EN,ADM1_PCODE,ADM2_PCODE,StartYr,EndYear,Year,Origin,total_affected,no_affected,date)%>%
-  st_set_geometry(NULL)%>%
-  filter(StartYr!='NA')
-
-Lesotho_impact<-st_join(Emdat_impact_sff %>%  filter(Country=='Lesotho') , Lesotho) %>% st_set_geometry(NULL)
-
-Namibia_impact<-st_join(Emdat_impact_sff %>%  filter(Country=='Namibia') , Namibia) %>% st_set_geometry(NULL)
+zwe_impact <- Emdat_impact_sff
 
 
 
@@ -177,39 +106,13 @@ for (n in range(1,length(admin))){
 # Clean impact and keep relevant columns
 df_impact_raw <- list()
 
-df_impact_raw[[1]] <- Ethiopia_impact %>%
-  clean_names() %>% dplyr::mutate(date = dmy(date),pcode = adm2_pcode) %>% 
-  dplyr::select(pcode, date,total_affected,no_affected) %>%
-  unique() %>%  arrange(pcode, date)
-
-df_impact_raw[[2]] <- Kenya_impact %>%
-  clean_names() %>% dplyr::mutate(date = dmy(date),pcode = adm2_pcode) %>% 
-  dplyr::select(pcode, date,total_affected,no_affected) %>%
-  unique() %>%  arrange(pcode, date)
-
-df_impact_raw[[3]] <- Mozambique_impact %>%
-  clean_names() %>% dplyr::mutate(date = dmy(date),pcode = adm2_pcode) %>% 
-  dplyr::select(pcode, date,total_affected,no_affected) %>%
+df_impact_raw[[1]] <- zwe_impact %>%
+  clean_names() %>% dplyr::mutate(date = dmy(substr(date_event, 1,10)), pcode = adm1_pcode) %>% 
+  dplyr::select(pcode, date, total_affected, no_affected) %>%
   unique() %>%  arrange(pcode, date)
 
-df_impact_raw[[4]] <- Lesotho_impact %>%
-  clean_names() %>% dplyr::mutate(date = dmy(date),pcode = adm2_pcode) %>% 
-  dplyr::select(pcode, date,total_affected,no_affected) %>%
-  unique() %>%  arrange(pcode, date)
-df_impact_raw[[5]] <- Namibia_impact %>%
-  clean_names() %>% dplyr::mutate(date = dmy(date),pcode = adm2_pcode) %>% 
-  dplyr::select(pcode, date,total_affected,no_affected) %>%
-  unique() %>%  arrange(pcode, date)
-
-
-
-
 
-df_impact_raw[[10]] <- df_impact_raw[[1]] 
-df_impact_raw[[20]] <- df_impact_raw[[2]] 
-df_impact_raw[[30]] <- df_impact_raw[[3]] 
-df_impact_raw[[40]] <- df_impact_raw[[4]] 
-df_impact_raw[[50]] <- df_impact_raw[[5]] 
+df_impact_raw[[10]] <- df_impact_raw[[1]] # admin lvl's 1 and 2
 
 
 # Used to join against
@@ -223,76 +126,23 @@ summarize_events <- function(df) {
 }
 
 
-#summarize_events(df_impact_raw[[1]]) %>% dplyr::select(pcode, n_events)
-
-
-
 admin[[1]] <- admin[[1]] %>%
   left_join(summarize_events(df_impact_raw[[1]]) %>%
               dplyr::select(pcode, n_events),   by = c("ADM2_PCODE" = "pcode")) %>%   dplyr::filter(!is.na(n_events))
 
-admin[[2]] <- admin[[2]] %>%
-  left_join(summarize_events(df_impact_raw[[2]]) %>%
-              dplyr::select(pcode, n_events), by = c("ADM2_PCODE" = "pcode")) %>% dplyr::filter(!is.na(n_events))
-
-admin[[3]] <- admin[[3]] %>%
-  left_join(summarize_events(df_impact_raw[[3]]) %>%
-              dplyr::select(pcode, n_events), by = c("ADM2_PCODE" = "pcode")) %>% dplyr::filter(!is.na(n_events))
-
-admin[[4]] <- admin[[4]] %>%
-  left_join(summarize_events(df_impact_raw[[4]]) %>%
-              dplyr::select(pcode, n_events), by = c("ADM2_PCODE" = "pcode")) %>% dplyr::filter(!is.na(n_events))
-
-admin[[5]] <- admin[[5]] %>%
-  left_join(summarize_events(df_impact_raw[[5]]) %>%
-              dplyr::select(pcode, n_events), by = c("ADM2_PCODE" = "pcode")) %>% dplyr::filter(!is.na(n_events))
-
-
-
 #change wih livelyhodzone
 admin[[10]] <- admin[[1]]
-admin[[20]] <- admin[[2]]
-admin[[30]] <- admin[[3]]
-admin[[40]] <- admin[[4]]
-admin[[50]] <- admin[[5]]
-
 
 df_indicators <- list()
 
-df_indicators[[1]] <- All_df_filled %>% dplyr::filter( ADM0_EN %in% c("Ethiopia","ETHIOPIA"))
-df_indicators[[2]] <- All_df_filled %>% dplyr::filter( ADM0_EN %in% c("KENYA","Kenya"))
-df_indicators[[3]] <- All_df_filled %>% dplyr::filter( ADM0_EN %in% c("MOZAMBIQUE","Mozambique"))
-df_indicators[[4]] <- All_df_filled %>% dplyr::filter( ADM0_EN %in% c("LESOTHO","Lesotho"))
-df_indicators[[5]] <- All_df_filled %>% dplyr::filter( ADM0_EN %in% c("NAMIBIA","Namibia"))
-
-
-df_indicators[[10]] <-All_df_filled %>% dplyr::filter( ADM0_EN %in% c("ET"))
-df_indicators[[20]] <-All_df_filled %>% dplyr::filter( ADM0_EN %in% c("KE"))
-df_indicators[[30]] <-All_df_filled %>% dplyr::filter( ADM0_EN %in% c("MZ"))
-df_indicators[[40]] <-All_df_filled %>% dplyr::filter( ADM0_EN %in% c("LOT"))
-df_indicators[[50]] <-All_df_filled %>% dplyr::filter( ADM0_EN %in% c("NAM"))
+df_indicators[[1]] <- All_df_filled %>% dplyr::filter( ADM0_EN %in% c("Zimbabwe"))
+df_indicators[[10]] <- df_indicators[[1]]
 
 label <- list()
 label[[1]] <- "ADM2_PCODE"
-label[[2]] <- "ADM2_PCODE"
-label[[3]] <- "ADM2_PCODE"
-label[[4]] <- "ADM2_PCODE"
-label[[5]] <- "ADM2_PCODE"
 label[[10]] <- "ADM2_PCODE"
-label[[20]] <- "ADM2_PCODE"
-label[[30]] <- "ADM2_PCODE"
-label[[40]] <- "ADM2_PCODE"
-label[[50]] <- "ADM2_PCODE"
 
 layerId <- list()
 layerId[[1]] <- "ADM2_PCODE"
-layerId[[2]] <- "ADM2_PCODE"
-layerId[[3]] <- "ADM2_PCODE"
-layerId[[4]] <- "ADM2_PCODE"
-layerId[[5]] <- "ADM2_PCODE"
 
 layerId[[10]] <- "ADM2_PCODE"
-layerId[[20]] <- "ADM2_PCODE"
-layerId[[30]] <- "ADM2_PCODE"
-layerId[[40]] <- "ADM2_PCODE"
-layerId[[50]] <- "ADM2_PCODE"

drought/trigger-model/Drought_trigger_dashboard/ui.R

@@ -53,7 +53,7 @@ ui <- dashboardPage(
                      label = 'Select date range:',
                      start = min(df_impact_raw[[1]]$date, na.rm=T), end = max(df_impact_raw[[1]]$date, na.rm=T)),
 
-      radioButtons("country", "Country:", c("Ethiopia" = 1, "Kenya" = 2)),
+      radioButtons("country", "Country:", c("Zimbabwe" = 1)),
       selectInput("Level", "Select aggregation  Level(for now only Admin ):", c("Provinces"=1, "LHZ"=10),selected=1),#"LEVEL 3"),
       sliderInput("spi_index", "Choose SPI index(months): ",max = 12, step=1,min = 1,round=TRUE,value=1),
       sliderInput("spi_threshold", "Select a Threshold for The SPI: ",