---
title: "Manipulate a Simple Feature Data Frame"
output: 
  html_notebook:
    toc: yes
    toc_float: yes
---

## Introduction

In this Notebook you'll practice importing some tabular data in to R, converting it to a simple feature data frame, projecting it, and saving it to disk as a GeoJSON file.

\

## 1) Import the missing persons CSV file:

```{r chunk01}
miss_pips_df <- read.csv("./data/yosemite_missing_people.csv", stringsAsFactors = FALSE)

head(miss_pips_df)
# View(miss_pips_df)
```

\

Let's reduce the number of columns:

```{r chunk02}
library(dplyr)

miss_pips_thin_df <- miss_pips_df |> 
  select(Long, Lat, Type, CaseNumber, IncidYear, ContactMet, IncidType, NumberofSu, GroupDynam, 
         SubjectCat, SubSex, SubAge, IncidContr, Scenario, RescueMeth, Incident_N, Found_GR_N)

head(miss_pips_thin_df)
```

\

## 2) Convert the data frame to a simple feature data frame

For convenience, here are some EPSG numbers:

```{r chunk03}
epsg_geo_wgs84 <- 4326       ## General long-lat
epsg_utm11n_nad83 <- 26911   ## use this one for YNP
```

Convert a data frame to a spatial object with `st_as_sf()`. You have to specify the names of the columns that contain the coordinates, and the EPSG number of the coordinates.

```{r chunk04}
library(sf)

miss_pips_sf <- st_as_sf(miss_pips_thin_df, 
                         coords = c("Long", "Lat"), 
                         crs = epsg_geo_wgs84) 

miss_pips_sf
```

\

## 3) Project the missing persons layer to UTM 11N (NAD83)

[Hint](https://bit.ly/3wJ46lJ). [Answer](https://bit.ly/3wmhGv6)

```{r chunk05}
## Your answer here

```

\

## 4) Save as GeoJSON

You can save a spatial object to disk with `st_write()`. R will figure out what format to use based on the file name extension. 

Save the Missing Persons layer as GeoJSON: [Answer](https://bit.ly/3wb8KcM)

```{r chunk06}
## Your answer here

```