R-GIS_tutorial.Rmd

Spatial data in R: Using R as a GIS     
========================================================


A tutorial to perform basic operations with spatial data in R, such as importing and exporting data (both vectorial and raster), plotting, analysing and making maps.


[Francisco Rodriguez-Sanchez](http://sites.google.com/site/rodriguezsanchezf) 


v 2.2  

27-01-2015 

Licence: [CC BY 4.0](http://creativecommons.org/licenses/by/4.0/)


Check out code and latest version at [GitHub](https://github.com/Pakillo/R-GIS-tutorial/blob/master/R-GIS_tutorial.md)



<br>
<br>
<br>

CONTENTS <a name="contents"></a>
=========

<br>
<br>

[1. INTRODUCTION](#intro)
<br>
<br>


[2. GENERIC MAPPING](#mapping)
<br>
  
  * [Retrieving base maps from Google with `gmap` function in package `dismo`](#gmap)
  
  * [`RgoogleMaps`: Map your data onto Google Map tiles](#rgooglemaps)
  
  * [`googleVis`: visualise data in a web browser using Google Visualisation API](#googlevis)
  
  * [`RWorldMap`: mapping global data](#rworldmap)
<br>
<br>


[3. SPATIAL VECTOR DATA (points, lines, polygons)](#vector)
<br>

  * [Example dataset: retrieve point occurrence data from GBIF](#gbif)
  
  * [Making data 'spatial'](#spatial)
  
  * [Define spatial projection](#projection)
  
  * [Quickly plotting point data on a map](#plot)
  
  * [Subsetting and mapping again](#subset)
  
  * [Mapping vectorial data (points, polygons, polylines)](#mapvector)
  
  * [Drawing polygons and polylines (e.g. for digitising)](#digitise)
  
  * [Converting between formats, reading in, and saving spatial vector data](#iovec)
  
  * [Changing projection of spatial vector data](#changeproj)
<br>
<br>
  
  
[4. USING RASTER (GRID) DATA](#raster)
<br>

  * [Downloading raster climate data from internet](#getdata)
  
  * [Loading a raster layer](#loadraster)
  
  * [Creating a raster stack](#rasterstack)
  
  * [Raster bricks](#rasterbrick)
  
  * [Crop rasters](#cropraster)
  
  * [Define spatial projection of the rasters](#projectionraster)
  
  * [Changing projection](#changeprojraster)
  
  * [Plotting raster data](#plotraster)
  
  * [Spatial autocorrelation](#autocorrelation)
  
  * [Extract values from raster](#extract)
  
  * [Rasterize vector data (points, lines or polygons)](#rasterize)
  
  * [Changing raster resolution](#resolution)
  
  * [Spline interpolation](#interpolation)
  
  * [Setting all rasters to the same extent, projection and resolution all in one](#spatialsync)
  
  * [Elevations, slope, aspect, etc](#elevation)
  
  * [Saving and exporting raster data](#saveraster)
<br>
<br>




[5. SPATIAL STATISTICS](#spatstats)
<br>

  * [Point pattern analysis](#pointpatterns)
  
  * [Geostatistics](#geostatistics)
<br> 
<br>

  
  
[6. INTERACTING WITH OTHER GIS](#othergis)
<br>
<br>

[7. OTHER USEFUL PACKAGES](#otherpackages)
<br>
<br>

[8. TO LEARN MORE](#tolearnmore)


<br>
<br>
<br>
<br>





1. INTRODUCTION <a name="intro"></a>
===============

<br>

R is great not only for doing statistics, but also for many other tasks, including GIS analysis and working with spatial data. For instance, R is capable of doing wonderful maps such as [this](http://spatialanalysis.co.uk/wp-content/uploads/2012/02/bike_ggplot.png) or [this](http://oscarperpinan.github.io/spacetime-vis/images/airMadrid_stamen.png). In this tutorial I will show some basic GIS functionality in R.



#### Basic packages

```{r message=FALSE}

library(sp)         # classes for spatial data
library(raster)     # grids, rasters
library(rasterVis)  # raster visualisation
library(maptools)   
library(rgeos)
    # and their dependencies
```

There are many other useful packages, e.g. check [CRAN Spatial Task View](http://cran.r-project.org/web/views/Spatial.html). Some of them will be used below.

<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>


     
2. GENERIC MAPPING <a name="mapping"></a>
==================

<br>


Retrieving base maps from Google with `gmap` function in package `dismo` <a name="gmap"></a>
------------------------------------------------------------------------

Some examples:

Getting maps for countries:

```{r gmap1, message=FALSE}

library(dismo)
  
mymap <- gmap("France")   # choose whatever country
plot(mymap)
```

Choose map type:
```{r gmap2, message=FALSE}
mymap <- gmap("France", type="satellite")  
plot(mymap)
```

Choose zoom level:
```{r gmap3, message=FALSE}
mymap <- gmap("France", type="satellite", exp=3)
plot(mymap)
```

Save the map as a file in your working directory for future use
```{r message=FALSE}
mymap <- gmap("France", type="satellite", filename="France.gmap")    
```


Now get a map for a region drawn at hand

```{r eval=FALSE}

mymap <- gmap("Europe")
plot(mymap)

select.area <- drawExtent()   
# now click 2 times on the map to select your region
mymap <- gmap(select.area)
plot(mymap)
# See ?gmap for many other possibilities
```



<br>
<br>








`RgoogleMaps`: Map your data onto Google Map tiles <a name="rgooglemaps"></a>
------------------------------------------------

```{r message=FALSE, results='hide'}
library(RgoogleMaps) 
```


Get base maps from Google (a file will be saved in your working directory)
```{r message=FALSE, results='hide'}
newmap <- GetMap(center=c(36.7,-5.9), zoom =10, destfile = "newmap.png", maptype = "satellite")   

# Now using bounding box instead of center coordinates:
newmap2 <- GetMap.bbox(lonR=c(-5, -6), latR=c(36, 37), destfile = "newmap2.png", maptype="terrain")   

# Try different maptypes
newmap3 <- GetMap.bbox(lonR=c(-5, -6), latR=c(36, 37), destfile = "newmap3.png", maptype="satellite")
```


Now plot data onto these maps, e.g. these 3 points
```{r}
PlotOnStaticMap(lat = c(36.3, 35.8, 36.4), lon = c(-5.5, -5.6, -5.8), zoom= 10, 
                cex=4, pch= 19, col="red", FUN = points, add=F)

```






<br>
<br>





`googleVis`: visualise data in a web browser using Google Visualisation API <a name="googlevis"></a>
---------------------------------------------------------------------------

```{r message=FALSE}
library(googleVis) 
```

Run `demo(googleVis)` to see all the possibilities


```{r setOptions, echo=FALSE}
op <- options(gvis.plot.tag = "chart")  
# necessary so that googleVis works with knitr, see http://lamages.blogspot.co.uk/2012/10/googlevis-032-is-released-better.html
```

<br>

### Example: plot country-level data
```{r results='asis', tidy=FALSE, eval=TRUE}
data(Exports)    # a simple data frame
Geo <- gvisGeoMap(Exports, locationvar="Country", numvar="Profit", 
                  options=list(height=400, dataMode='regions'))
plot(Geo)
```
Using `print(Geo)` we can get the HTML code to embed the map in a web page!

<br>

### Example: Plotting point data onto a google map (internet)
```{r results='asis', tidy=FALSE, eval=TRUE}
data(Andrew)
M1 <- gvisMap(Andrew, "LatLong", "Tip", 
              options=list(showTip=TRUE, showLine=F, enableScrollWheel=TRUE, 
                           mapType='satellite', useMapTypeControl=TRUE, width=800,height=400))
plot(M1)
```


  
    
<br>
<br>





`RWorldMap`: mapping global data <a name="rworldmap"></a>
--------------------------------

Some examples

```{r message=FALSE, warning=FALSE}

library(rworldmap)

newmap <- getMap(resolution="coarse")    # different resolutions available
plot(newmap)
```

```{r message=FALSE} 
mapCountryData()
```
```{r message=FALSE}
mapCountryData(mapRegion="europe")
```
```{r message=FALSE}
mapGriddedData()
```
```{r message=FALSE}
mapGriddedData(mapRegion="europe")
  
```




<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>
<br>
    






  
   


3. SPATIAL VECTOR DATA (points, lines, polygons) <a name="vector"></a>
================================================

<br>
<br>

### Example dataset: retrieve point occurrence data from GBIF <a name="gbif"></a>

Let's create an example dataset: retrieve occurrence data 
for the laurel tree (Laurus nobilis) from the 
[Global Biodiversity Information Facility (GBIF)](http://gbif.org)
```{r message=FALSE}
library(dismo)      # check also the nice "rgbif" package! 
laurus <- gbif("Laurus", "nobilis")      
# get data frame with spatial coordinates (points)
locs <- subset(laurus, select=c("country", "lat", "lon"))
head(locs)    # a simple data frame with coordinates

# Discard data with errors in coordinates:
locs <- subset(locs, locs$lat<90)
```

<br>

### Making data 'spatial' <a name="spatial"></a>

So we have got a simple dataframe containing spatial coordinates. 
Let's make these data explicitly *spatial*
```{r}
coordinates(locs) <- c("lon", "lat")    # set spatial coordinates
plot(locs)
```


### Define spatial projection <a name="projection"></a>

Important: define geographical projection. 
Consult the appropriate PROJ.4 description here: 
[http://www.spatialreference.org/](http://www.spatialreference.org/)
```{r}
crs.geo <- CRS("+proj=longlat +ellps=WGS84 +datum=WGS84")    # geographical, datum WGS84
proj4string(locs) <- crs.geo     # define projection system of our data
summary(locs)
```

<br>

### Quickly plotting point data on a map <a name="plot"></a>
```{r}
plot(locs, pch=20, col="steelblue")
library(rworldmap)
# library rworldmap provides different types of global maps, e.g:
data(coastsCoarse) 
data(countriesLow)
plot(coastsCoarse, add=T)
```


### Subsetting and mapping again <a name="subset"></a>
```{r}
table(locs$country)     # see localities of Laurus nobilis by country

locs.gb <- subset(locs, locs$country=="United Kingdom")    # select only locs in UK
plot(locs.gb, pch=20, cex=2, col="steelblue")
title("Laurus nobilis occurrences in UK")
plot(countriesLow, add=T)
summary(locs.gb)

```


<br>

Mapping vectorial data (points, polygons, polylines) <a name="mapvector"></a>
---------------------------------------------------------------------

<br>

### Mapping vectorial data using `gmap` from `dismo`


```{r}
gbmap <- gmap(locs.gb, type="satellite")
locs.gb.merc <- Mercator(locs.gb)    # Google Maps are in Mercator projection. 
  # This function projects the points to that projection to enable mapping
plot(gbmap)
points(locs.gb.merc, pch=20, col="red")
```





<br>

### Mapping vectorial data with `RgoogleMaps` 


```{r message=FALSE}

require(RgoogleMaps)

locs.gb.coords <- as.data.frame(coordinates(locs.gb))    # retrieves coordinates 
  # (1st column for longitude, 2nd column for latitude)
PlotOnStaticMap(lat = locs.gb.coords$lat, lon = locs.gb.coords$lon, 
                zoom= 5, cex=1.4, pch= 19, col="red", FUN = points, add=F)
```

Download base map from Google Maps and plot onto it
```{r message=FALSE}
map.lim <- qbbox (locs.gb.coords$lat, locs.gb.coords$lon, TYPE="all")    # define region 
  # of interest (bounding box)
mymap <- GetMap.bbox(map.lim$lonR, map.lim$latR, destfile = "gmap.png", maptype="satellite")
# see the file in the wd
PlotOnStaticMap(mymap, lat = locs.gb.coords$lat, lon = locs.gb.coords$lon, 
                zoom= NULL, cex=1.3, pch= 19, col="red", FUN = points, add=F)
```

<br>
<br>

Using different background (base map)
```{r message=FALSE}
mymap <- GetMap.bbox(map.lim$lonR, map.lim$latR, destfile = "gmap.png", maptype="hybrid")
PlotOnStaticMap(mymap, lat = locs.gb.coords$lat, lon = locs.gb.coords$lon, 
                zoom= NULL, cex=1.3, pch= 19, col="red", FUN = points, add=F)

```








<br>
<br>

### Map vectorial data with `googleVis` (internet) 


```{r results='asis', tidy=FALSE, eval=TRUE}
points.gb <- as.data.frame(locs.gb)
points.gb$latlon <- paste(points.gb$lat, points.gb$lon, sep=":")
map.gb <- gvisMap(points.gb, locationvar="latlon", tipvar="country", 
                  options = list(showTip=T, showLine=F, enableScrollWheel=TRUE,
                           useMapTypeControl=T, width=1400,height=800))
plot(map.gb)
#print(map.gb)    # get HTML suitable for a webpage
```




<br>
<br>
<br>

### Drawing polygons and polylines (e.g. for digitising) <a name="digitise"></a>


```{r eval=FALSE}
plot(gbmap)
mypolygon <- drawPoly()    # click on the map to draw a polygon and press ESC when finished
summary(mypolygon)    # now you have a spatial polygon! Easy, isn't it?
```


<br>
<br>
<br>
<br>

Converting between formats, reading in, and saving spatial vector data <a name="iovec"></a>
-------------------------------------------------------------------

<br>

### Exporting KML (Google Earth)
```{r}
writeOGR(locs.gb, dsn="locsgb.kml", layer="locs.gb", driver="KML")
```

### Reading KML
```{r}
newmap <- readOGR("locsgb.kml", layer="locs.gb")
```

### Save as shapefile
```{r}
writePointsShape(locs.gb, "locsgb")
```

### Reading shapefiles
```{r}
gb.shape <- readShapePoints("locsgb.shp")
plot(gb.shape)
```
Use `readShapePoly` to read polygon shapefiles, and `readShapeLines` to read polylines.
See also `shapefile` in `raster` package.   



<br>
<br>
<br>

Changing projection of spatial vector data <a name="changeproj"></a>
-------------------------------------------

`spTransform` (package `sp`) will do the projection as long as the original and new projection are correctly specified.

<br>

### Projecting point dataset

To illustrate, let's project the dataframe with Laurus nobilis coordinates that we obtained above:
```{r}
summary(locs)
```
The original coordinates are in lat lon format. Let's define the new desired projection:
Lambert Azimuthal Equal Area in this case 
(look up parameters at [http://spatialreference.org](http://spatialreference.org))
```{r}
crs.laea <- CRS("+proj=laea +lat_0=52 +lon_0=10 +x_0=4321000 +y_0=3210000 +ellps=GRS80 +units=m +no_defs") # Lambert Azimuthal Equal Area
locs.laea <- spTransform(locs, crs.laea)    # spTransform makes the projection
```

<br>

### Projecting shapefile of countries
```{r}
plot(countriesLow)    # countries map in geographical projection
country.laea <- spTransform(countriesLow, crs.laea)  # project
```

Let's plot this:
```{r}
plot(locs.laea, pch=20, col="steelblue")
plot(country.laea, add=T)

# define spatial limits for plotting
plot(locs.laea, pch=20, col="steelblue", xlim=c(1800000, 3900000), ylim=c(1000000, 3000000))
plot(country.laea, add=T)
```




<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>
<br>
<br>







4. USING RASTER (GRID) DATA <a name="raster"></a>
===========================


<br>

### Downloading raster climate data from internet <a name="getdata"></a>
The `getData` function from the `dismo` package will easily retrieve climate data, elevation, administrative boundaries, etc. Check also the excellent [rWBclimate package](http://ropensci.org/packages/rwbclimate.html) by rOpenSci with additional functionality. 

```{r}
tmin <- getData("worldclim", var="tmin", res=10)   # this will download 
  # global data on minimum temperature at 10' resolution
```

<br>

### Loading a raster layer <a name="loadraster"></a>

```{r}
tmin1 <- raster(paste(getwd(), "/wc10/tmin1.bil", sep=""))   # Tmin for January
```
Easy! The `raster` function reads many different formats, including Arc ASCII grids or netcdf files (see raster help). And values are stored on disk instead of memory! (useful for large rasters) 
```{r}
fromDisk(tmin1)
```

Let's examine the raster layer:
```{r}
tmin1 <- tmin1/10    # Worldclim temperature data come in decimal degrees 
tmin1    # look at the info
plot(tmin1)
```

<br>

### Creating a raster stack <a name="rasterstack"></a>

A raster stack is collection of many raster layers with the same projection, spatial extent and resolution.
Let's collect several raster files from disk and read them as a single raster stack:

```{r message=FALSE, warning=FALSE}

library(gtools)
file.remove(paste(getwd(), "/wc10/", "tmin_10m_bil.zip", sep=""))
list.ras <- mixedsort(list.files(paste(getwd(), "/wc10/", sep=""), full.names=T, pattern=".bil"))
list.ras   # I have just collected a list of the files containing monthly temperature values
tmin.all <- stack(list.ras)
tmin.all
tmin.all <- tmin.all/10
plot(tmin.all)
```


<br>

### Raster bricks <a name="rasterbrick"></a>

A rasterbrick is similar to a raster stack (i.e. multiple layers with the same extent and resolution), but all the data must be stored in a single file on disk.

```{r}
tmin.brick <- brick(tmin.all)   # creates rasterbrick
```

<br>

### Crop rasters <a name="cropraster"></a>

Crop raster manually (drawing region of interest):

```{r eval=FALSE}
plot(tmin1)
newext <- drawExtent()    # click twice on the map to select the region of interest
tmin1.c <- crop(tmin1, newext)
plot(tmin1.c)
```

Alternatively, provide coordinates for the limits of the region of interest:
```{r}
newext <- c(-10, 10, 30, 50)   
tmin1.c <- crop(tmin1, newext)
plot(tmin1.c)

tmin.all.c <- crop(tmin.all, newext)
plot(tmin.all.c)
```

<br>

### Define spatial projection of the rasters <a name="projectionraster"></a>

```{r}
crs.geo    # defined above
projection(tmin1.c) <- crs.geo
projection(tmin.all.c) <- crs.geo
tmin1.c    # notice info at coord.ref.
```

<br>

### Changing projection <a name="changeprojraster"></a>

Use `projectRaster` function:
```{r}
tmin1.proj <- projectRaster(tmin1.c, crs="+proj=merc +lon_0=0 +k=1 +x_0=0 +y_0=0 +a=6378137 +b=6378137 +units=m +no_defs")   # can also use a template raster, see ?projectRaster
tmin1.proj   # notice info at coord.ref.
plot(tmin1.proj)
```


<br>

### Plotting raster data <a name="plotraster"></a>

Different plotting functions:
```{r}
histogram(tmin1.c)
pairs(tmin.all.c)
persp(tmin1.c)
contour(tmin1.c)
contourplot(tmin1.c)
levelplot(tmin1.c)
#plot3D(tmin1.c)
bwplot(tmin.all.c)
densityplot(tmin1.c)
```


### Spatial autocorrelation <a name="autocorrelation"></a>

```{r}
Moran(tmin1.c)    # global Moran's I
tmin1.Moran <- MoranLocal(tmin1.c)
plot(tmin1.Moran)
```


### Extract values from raster <a name="extract"></a>

Use `extract` function:
```{r}
head(locs)    # we'll obtain tmin values for our points
projection(tmin1) <- crs.geo
locs$tmin1 <- extract(tmin1, locs)    # raster values 
  # are incorporated to the dataframe
head(locs)
```

You can also extract values for a given region instead of the whole raster:
```{r eval=FALSE}
plot(tmin1.c)
reg.clim <- extract(tmin1.c, drawExtent())  # click twice to 
  # draw extent of the region of interest 
summary(reg.clim)
```

Using `rasterToPoints`:
```{r}
# rasterToPoints
tminvals <- rasterToPoints(tmin1.c)
head(tminvals)
```

And also, the `click` function will get values from particular locations in the map
```{r eval=FALSE}
plot(tmin1.c)
click(tmin1.c, n=3)   # click n times in the map to get values
```

<br>

### Rasterize points, lines or polygons <a name="rasterize"></a>

```{r}
locs2ras <- rasterize(locs.gb, tmin1, field=rep(1,nrow(locs.gb)))
locs2ras
plot(locs2ras, xlim=c(-10,10), ylim=c(45, 60), legend=F)
data(wrld_simpl)
plot(wrld_simpl, add=T)
```


<br>

### Changing raster resolution <a name="resolution"></a>

Use `aggregate` function:
```{r}
tmin1.lowres <- aggregate(tmin1.c, fact=2, fun=mean)
tmin1.lowres
tmin1.c     # compare
par(mfcol=c(1,2))
plot(tmin1.c, main="original")
plot(tmin1.lowres, main="low resolution")
```


### Spline interpolation <a name="interpolation"></a>

```{r message=FALSE, warning=FALSE}
xy <- data.frame(xyFromCell(tmin1.lowres, 1:ncell(tmin1.lowres)))    # get raster cell coordinates
head(xy)
vals <- getValues(tmin1.lowres)
library(fields)
spline <- Tps(xy, vals)    # thin plate spline
intras <- interpolate(tmin1.c, spline)
intras    # note new resolution
plot(intras)  
intras <- mask(intras, tmin1.c)   # mask to land areas only
plot(intras)
title("Interpolated raster")
```

### Setting all rasters to the same extent, projection and resolution all in one <a name="spatialsync"></a>

See `spatial_sync_raster` function from `spatial.tools` package.

<br>


### Elevations, slope, aspect, etc <a name="elevation"></a>

<br>
Download elevation data from internet:
```{r}
elevation <- getData('alt', country='ESP')
```

Some quick maps:
```{r}
x <- terrain(elevation, opt=c('slope', 'aspect'), unit='degrees')
plot(x)

slope <- terrain(elevation, opt='slope')
aspect <- terrain(elevation, opt='aspect')
hill <- hillShade(slope, aspect, 40, 270)
plot(hill, col=grey(0:100/100), legend=FALSE, main='Spain')
plot(elevation, col=rainbow(25, alpha=0.35), add=TRUE)
```


### Saving and exporting raster data <a name="saveraster"></a>


Saving raster to file:

```{r}
writeRaster(tmin1.c, filename="tmin1.c.grd")   
writeRaster(tmin.all.c, filename="tmin.all.grd")
```
`writeRaster` can export to many different file types, see help.

<br>

Exporting to KML (Google Earth)
```{r}
tmin1.c <- raster(tmin.all.c, 1)
KML(tmin1.c, file="tmin1.kml")  
KML(tmin.all.c)     # can export multiple layers

```

<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>

5. SPATIAL STATISTICS (just a glance) <a name="spatstats"></a>
=====================================

<br>

### Point pattern analysis <a name="pointpatterns"></a>

Some useful packages:

```{r message=FALSE}
library(spatial)
library(spatstat)
library(spatgraphs)
library(ecespa)    # ecological focus
```
See [CRAN Spatial Task View](http://cran.r-project.org/web/views/Spatial.html).

Let's do a quick example with Ripley's K function:
```{r}
data(fig1)
plot(fig1)    # point pattern
data(Helianthemum)
cosa12 <- K1K2(Helianthemum, j="deadpl", i="survpl", r=seq(0,200,le=201),
               nsim=99, nrank=1, correction="isotropic")
plot(cosa12$k1k2, lty=c(2, 1, 2), col=c(2, 1, 2), xlim=c(0, 200),
     main= "survival- death",ylab=expression(K[1]-K[2]), legend=FALSE)

```

<br>

### Geostatistics <a name="geostatistics"></a>

Some useful packages:

```{r message=FALSE, eval=FALSE}
library(gstat)
library(geoR)
library(akima)   # for spline interpolation
library(spdep)   # dealing with spatial dependence
```
See [CRAN Spatial Task View](http://cran.r-project.org/web/views/Spatial.html).


<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>

6. INTERACTING WITH OTHER GIS <a name="othergis"></a>
===============================================

```{r message=F, eval=F}
library(spgrass6)   # GRASS
library(RPyGeo)     # ArcGis (Python)
library(RSAGA)      # SAGA
library(spsextante) # Sextante 

```

<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>


7. OTHER USEFUL PACKAGES <a name="otherpackages"></a>
=========================

```{r message=FALSE, eval=FALSE}
library(Metadata)    # automatically collates data from online GIS datasets (land cover, pop density, etc) for a given set of coordinates

#library(GeoXp)    # Interactive exploratory spatial data analysis
example(columbus)
histomap(columbus,"CRIME")

library(maptools)
# readGPS 

library(rangeMapper)    # plotting species distributions, richness and traits


# Species Distribution Modelling
library(dismo)
library(biomod2)
library(SDMTools)

library(BioCalc)   # computes 19 bioclimatic variables from monthly climatic values (tmin, tmax, prec)

```

<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>

8. TO LEARN MORE <a name="tolearnmore"></a>
================

* [ASDAR book](http://www.asdar-book.org/)

* Packages help and vignettes, especially

    http://cran.r-project.org/web/packages/raster/vignettes/Raster.pdf   
    http://cran.r-project.org/web/packages/dismo/vignettes/sdm.pdf   
    http://cran.r-project.org/web/packages/sp/vignettes/sp.pdf   

* [CRAN Task View: Analysis of Spatial Data](http://cran.r-project.org/web/views/Spatial.html)


* [Introduction to Spatial Data and ggplot2](http://rpubs.com/RobinLovelace/intro-spatial)

* [R spatial tips](http://spatial.ly/category/r-spatial-data-hints/)

* [R wiki: tips for spatial data](http://rwiki.sciviews.org/doku.php?id=tips:spatial-data&s=spatial)

* [Spatial analysis in R](http://www.maths.lancs.ac.uk/~rowlings/Teaching/Sheffield2013/index.html)

* [Displaying time series, spatial and space-time data with R](http://oscarperpinan.github.io/spacetime-vis/)

* [Notes on Spatial Data Operations in R](https://dl.dropboxusercontent.com/u/9577903/broomspatial.pdf)

* [Analysing spatial point patterns in R](http://www.csiro.au/resources/pf16h)

* [Spatial data in R](http://science.nature.nps.gov/im/datamgmt/statistics/r/advanced/Spatial.cfm)

* [NCEAS Geospatial use cases](http://www.nceas.ucsb.edu/scicomp/usecases)

* [Spatial Analyst](http://spatial-analyst.net)

* [Making maps with R](http://www.molecularecologist.com/2012/09/making-maps-with-r/)

* [The Visual Raster Cheat Sheet](http://www.rpubs.com/etiennebr/visualraster)

* [R-SIG-Geo mailing list](https://stat.ethz.ch/mailman/listinfo/R-SIG-Geo)

* [Geospatial data processing and analysis in R (slides)](http://rpubs.com/ajlyons/rspatialdata)

* [rMaps](http://rmaps.github.io/)





<br>
[Back to Contents](#contents)
<br>
<br>
<br>
<br>