Intro to spatial analysis

class: center, middle, inverse, title-slide

# Intro to spatial analysis
### Jakub Nowosad <br><a href="mailto:nowosad.jakub@gmail.com">nowosad.jakub@gmail.com</a>
### 2017-04-27

---

## Introduction

- R has powerful GIS capabilities due to many additional packages - see [CRAN Task View: Analysis of Spatial Data](https://cran.r-project.org/web/views/Spatial.html) and [CRAN Task View: Handling and Analyzing Spatio-Temporal Data](https://cran.r-project.org/web/views/SpatioTemporal.html)
- **sf** and **sp** are the most important R packages to handle vector data; **sf** is a successor of **sp**, but it's still evolving. Moreover, many other R packages depend on the functions and classes for the **sp** package 
- **raster** is an extension of spatial data classes to work with rasters
- there are many ways to vizualize spatial data in R, for example **ggplot2**, **rasterVis**, **tmap**, **leaflet**, and **mapview** packages
- it is easy to connect R with a GIS software - GRASS GIS (**rgrass7**), SAGA (**RSAGA**), QGIS (**RQGIS**), and ArcGIS (**arcgisbinding**)

---
class: inverse, center, middle
## Vector data

---
## The **sf** package

The **sf** package in an R implementation of [Simple Features](https://en.wikipedia.org/wiki/Simple_Features). This package incorporates:
- a new spatial data class system in R
- functions for reading and writing data
- tools for spatial operations on vectors

Most of the functions in this package starts with prefix `st_`.

```r
devtools::install_github('edzer/sfr') # development version
```
or

```r
install.packages('sf') # stable version
```

You need a recent version of the GDAL, GEOS, Proj.4, and UDUNITS libraries installed for this to work on Mac and Linux. More information on that at https://github.com/edzer/sfr.

```r
library('sf')
```

---
## Reading spatial data

```r
wrld <- st_read('data/wrld.shp')
```

```
## Reading layer `wrld' from data source `/home/jn/Documents/Intro_to_spatial_data/data/wrld.shp' using driver `ESRI Shapefile'
## converted into: POLYGON
## Simple feature collection with 177 features and 10 fields
## geometry type:  MULTIPOLYGON
## dimension:      XY
## bbox:           xmin: -180 ymin: -90 xmax: 180 ymax: 83.64513
## epsg (SRID):    4326
## proj4string:    +proj=longlat +datum=WGS84 +no_defs
```

```r
ham <- st_read('data/hamilton_county.gpkg')
```

```
## Reading layer `hamilton_county.gpkg' from data source `/home/jn/Documents/Intro_to_spatial_data/data/hamilton_county.gpkg' using driver `GPKG'
## converted into: POLYGON
## Simple feature collection with 1 feature and 17 fields
## geometry type:  POLYGON
## dimension:      XY
## bbox:           xmin: -84.8203 ymin: 39.02153 xmax: -84.25651 ymax: 39.31206
## epsg (SRID):    4269
## proj4string:    +proj=longlat +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +no_defs
```

---
## Reading spatial data - text files

```r
ham_cities <- read.csv('data/hamiltion_cities.csv')
```

```r
ham_cities_sf <- st_as_sf(ham_cities, coords = c("X","Y"))
ham_cities_sf
```

```
## Simple feature collection with 19 features and 2 fields
## geometry type:  POINT
## dimension:      XY
## bbox:           xmin: -84.81995 ymin: 39.07862 xmax: -84.26384 ymax: 39.29034
## epsg (SRID):    NA
## proj4string:    NA
##             name place                      geometry
## 1          Delhi  town POINT(-84.6052225 39.0950595)
## 2        Milford  town  POINT(-84.2958988 39.174625)
## 3      Covington  town  POINT(-84.508371 39.0836224)
## 4       Harrison  town POINT(-84.8199516 39.2619993)
## 5     Cincinnati  city POINT(-84.5124602 39.1014537)
## 6     Springdale  town  POINT(-84.4852213 39.287002)
## 7       Sycamore  town  POINT(-84.347427 39.2800883)
## 8        Norwood  town POINT(-84.4596641 39.1556149)
## 9       Loveland  town POINT(-84.2638406 39.2689562)
## 10   Fort Thomas  town POINT(-84.4483432 39.0786242)
## 11 Saint Bernard  town POINT(-84.4985541 39.1670033)
## 12    Montgomery  town POINT(-84.3539662 39.2282432)
## 13   Forest Park  town POINT(-84.5041108 39.2903353)
## 14      Blue Ash  town POINT(-84.3782817 39.2320073)
## 15        Ludlow  town POINT(-84.5474435 39.0925598)
## 16   Sharonville  town POINT(-84.4132779 39.2681145)
## 17       Reading  town POINT(-84.4421641 39.2236694)
## 18      Bellevue  town POINT(-84.4826482 39.1067078)
## 19       Newport  town POINT(-84.4919524 39.0889469)
```

---
## Writing spatial data

```r
st_write(wrld, 'data/new_wrld.shp')
```

```
## Writing layer `new_wrld' to data source `data/new_wrld.shp' using driver `ESRI Shapefile'
## features:       177
## fields:         10
## geometry type:  Multi Polygon
```

```r
st_write(wrld, 'data/new_wrld.gpkg')
```

```
## Writing layer `new_wrld.gpkg' to data source `data/new_wrld.gpkg' using driver `GPKG'
## features:       177
## fields:         10
## geometry type:  Multi Polygon
```

---
## **sf** structure

```r
class(wrld)
```

```
## [1] "sf"         "data.frame"
```

**sf** objects usually have two classes - `sf` and `data.frame`. Two main differences comparing to a regular `data.frame` object are spatial metadata (`geometry type`, `dimension`, `bbox`, `epsg (SRID)`, `proj4string`) and additional column - typically named `geom` or `geometry`.

```r
wrld[1:2, 1:3]
```

```
## Simple feature collection with 2 features and 3 fields
## geometry type:  MULTIPOLYGON
## dimension:      XY
## bbox:           xmin: 11.6401 ymin: -17.93064 xmax: 75.15803 ymax: 38.48628
## epsg (SRID):    4326
## proj4string:    +proj=longlat +datum=WGS84 +no_defs
##   iso_a2   name_long continent                       geometry
## 1     AF Afghanistan      Asia MULTIPOLYGON(((61.210817091...
## 2     AO      Angola    Africa MULTIPOLYGON(((16.326528354...
```

---
## Attributes

- **sf** object can be used as a regular `data.frame` object in many operations

```r
head(wrld)
nrow(wrld)
ncol(wrld)
wrld[, c(1, 3)]
wrld[1:5, 2]
wrld[c(5, 10, 15), ]
```

- The `st_set_geometry` function can be used to remove the geometry column:

```r
wrld_df <- st_set_geometry(wrld, NULL)
class(wrld_df)
```

```
## [1] "data.frame"
```