Data

Read data

The raw data were organized in excel. Let’s load it using the read_excel function.

dat_survey <- read_excel("data/survey-data-farm-level5.xlsx") 

Transform data

The latitude and longitude from fields in West Wellega are in different format (decimal) than the rest of the fiels. We will first exclude the West Wellega zone from the dataset, convert to decimal and then combine everything again.

Non West Wellega

survey2 <- dat_survey %>%
  filter(zone != "West Wellega") %>%
  separate(lon, c("lon_d", "lon_m", "lon_s", "lon_mili"), extra = "merge") %>%
  separate(lat, c("lat_d", "lat_m", "lat_s", "lat_mili"), extra = "merge") %>%
  mutate(
    lon_d = as.numeric(lon_d),
    lon_m = as.numeric(lon_m),
    lon_s = as.numeric(lon_s),
    lon_mili = as.numeric(lon_mili),
    lat_d = as.numeric(lat_d),
    lat_m = as.numeric(lat_m),
    lat_s = as.numeric(lat_s),
    lat_mili = as.numeric(lat_mili)
  ) %>%
  mutate(
    lat = lat_d + (lat_m / 60) + (lat_s / 3600),
    lon = lon_d + (lon_m / 60) + (lon_s / 3600)
  ) %>%
  mutate(elevation_class = case_when(
    altitude < 1500 ~ "<1500m",
    altitude < 1800 ~ "1500 to 1800m",
    TRUE ~ ">1800m"
  )) %>%
  mutate(sev2 = ((sev / 100) * (inc / 100)) * 100)

## Warning: Expected 4 pieces. Missing pieces filled with `NA` in 174 rows [1, 2,
## 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, ...].

## Warning: Expected 4 pieces. Missing pieces filled with `NA` in 173 rows [1, 2,
## 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, ...].

## Select only the variables that are in the other dataset

survey2 <- survey2 %>%
  dplyr::select(farm, region, zone, district, lon, lat, altitude, cultivar, shade, cropping_system, farm_management, inc, sev, lesion_count,  sev2)

West Wellega fields

The lat and lon are in decimal format, but not numeric, which needs transformation.

survey3 <- dat_survey %>%
  filter(zone == "West Wellega") %>%
  mutate(
    lon = as.numeric(lon),
    lat = as.numeric(lat),
    sev2 = ((sev / 100) * (inc / 100)) * 100
  )

Join dataframes

survey <- rbind(survey2, survey3)

Rename zone names

survey$zone <- plyr::revalue(survey$zone, c(
  "Gedeo" = "Gedio"
))
survey$region <- plyr::revalue(survey$region, c(
  "South" = "SNNPR"
))

Create a CSV file

survey <- survey %>%
  dplyr::select(
    farm, region, zone, district, lon, lat, altitude,
    cultivar, shade, cropping_system, farm_management, inc, sev2
  )

write_csv(survey, "data/survey_clean.csv")

Copyright 2020 Emerson Del Ponte