slides.Rmd

---
title: "R for spatial vector data"
author: "Loïc Dutrieux, Jan Verbesselt, Sytze de Bruyn, Benjamin Brede"
date: "`r format(Sys.time(), '%d %B, %Y')`"
output: slidy_presentation
---

## Sp classes

| Geometry      | class                     | attribute     |
|-------------  |-----------------          |-------------  |
| points        | *SpatialPoints*           | No            |
| points        | *SpatialPointsDataFrame*  | data.frame   |
| line          | *Line*                    | No           |
| lines         | *Lines*                   | No           |
| lines         | *SpatailLines*            | No           |
| lines         | *SpatialLinesDataFrame*   | data.frame      |
| rings         | *Polygon*                    | No           |
| rings         | *Polygons*                   | No           |
| rings         | *SpatailPolygons*            | No           |
| rings         | *SpatialPolygonsDataFrame*   | data.frame   |


---

## Structure of a Spatial*DataFrame object

```{r}
library(sp)
data(meuse)
coordinates(meuse) = c("x", "y")
str(meuse)
```

---

## Structure of a Spatial*DataFrame object (2)

* A SpatialPoints object
    * Coordinates
    * A coordinate reference system
    * A spatial extent
* A dataframe

### Implications

*Manipulating sp objects is very similar to data.frames manipulation*

### Subsetting

```{r}
meuseClean <- meuse[!is.na(meuse$landuse),]
meuseSub <- meuseClean[meuseClean$landuse == 'Ah',]
```

---

```{r}
plot(meuse, pch = 20, col = 'red')
plot(meuseSub, pch = 20, col = 'green', add = TRUE)
```

---

## Manipulating geometries

**rgeos** package

* `gBuffer()`: Create buffer around geometries
* `gDistance()`: Distance between a set of *projected* points
* `gDifference()`: Difference between 2 geometries
* `gArea()`: Calculate area of given geometry
* `gIntersection()`: Intersection between two geometries