data-wrangling.qmd

---
title: "neotomalakes-data+wrangling"
author: "Florencia D'Andrea"
format: html
---

# NeotomaLakes app data wrangling

```{r echo = FALSE}
library(tidyverse) #not a good idea for renv
library(sf)
```


## Countries

```{r}
# I select some countries to try the app.
mdg <- st_read('data/gadm41_MDG.gpkg') |>
  st_simplify(preserveTopology = TRUE)
prt <- st_read('data/gadm41_PRT.gpkg') |>
  st_simplify(preserveTopology = TRUE)
arg <- st_read('data/gadm41_ARG.gpkg') |>
  st_simplify(preserveTopology = TRUE)
jpn <- st_read('data/gadm41_JPN.gpkg') |>
  st_simplify(preserveTopology = TRUE)

countries =  countries_sf = rbind(jpn, prt, mdg)

# Save country polygons
write_sf(countries_sf, 'data/countries.gpkg')
```


```{r}
st_geometry(countries_sf) <- NULL
```

## NEOTOMADB SITES

```{r}
sites <- read.csv('../Dropbox/neotoma_project/sites_20231101_.csv')
#Some sites are points and other ones are polygons
pointsites <-
  sites[stringr::str_detect(sites$geog, pattern = "POINT"),] |>
  sf::st_as_sf(wkt = 'geog') |>
  select(siteid, sitename, sitedescription, notes, geog) |>
  mutate(lonc = sf::st_coordinates(geog)[, 1],
         # need this to setView in leaflet
         latc = sf::st_coordinates(geog)[, 2])
polysites <-
  sites[stringr::str_detect(sites$geog, pattern = "POLYGON"),] |>
  sf::st_as_sf(wkt = 'geog') |>
  select(siteid, sitename, sitedescription, notes, geog) |>
  st_simplify(preserveTopology = TRUE,
              dTolerance = 100) |>
  mutate(
    centroid = st_point_on_surface(geog),
    # need this to setView in leaflet
    lonc = sf::st_coordinates(centroid)[, 1],
    latc = sf::st_coordinates(centroid)[, 2]
  ) |>
  select(-centroid)
datasf <- rbind(pointsites, polysites) |> st_set_crs(4326)
#extra step= removing sites of countries that aren't in the demo
datasf_country = st_filter(datasf, countries_sf)
```


## HYDROLAKESDB

```{r}
lakes <-
  st_read('../Dropbox/neotoma_project/HydroLAKES_polys_v10_shp/HydroLAKES_polys_v10_shp') |>
  select(
    Hylak_id,
    Lake_name,
    Country,
    Vol_total,
    Lake_area,
    Shore_len,
    Depth_avg,
    Elevation,
    geometry
  )
lakes_country <- countries_sf |>
  left_join(lakes, by = c('COUNTRY' = 'Country'))

write_sf(lakes_country, "data/lakes_country.gpkg")
```


```{r}
sites_country <-
  datasf |> st_join(countries_sf, left = TRUE) |> 
  tidyr::drop_na(COUNTRY)
write_sf(datasf_country, "data/sites_country.gpkg")
```