---
title: "First Steps in RGeostats"
author: "RGeostats Team"
date: "September 2019"
output:
  slidy_presentation: default
  ioslides_presentation: default
  beamer_presentation:
    colortheme: beaver
    fig_height: 3.5
    fig_width: 3.5
    theme: Boadilla
fontsize: 8pt
theme: Boadilla
colortheme: beaver
---

\frametitle{Loading Library}

```{r setup,echo=FALSE}
library(knitr)
```

In each R session, you must load the **RGeostats** library

```{r Load_library, include=FALSE}
library(RGeostats)
```

Clean the contents of your workspace

```{r Cleaning_WorkSpace}
rm(list=ls())
```

---

\frametitle{Main Classes}

This is the (non-exhaustive) list of classes (of objects) in RGeostats:

* db: numerical data base
* vario: experimental variograms (vardir)
* model: variogram model (melem)
* neigh: neighborhood
* anam: gaussian anamorphosis
* polygon: 2-D polygonal shapes (polyset)
* rule: lithotype rule for thresholds (truncated plurigaussian models)

---

\frametitle{Loading_Data_File}

You should download the ASCII file called *Scotland_Temperatures.csv* (organized as a CSV file) and store it on your disk in the current working directory, then convert it into a *Db*. 

Instead, you can download it (from the standard distribution) using the *rg.load* facility. It is named **dat**: the target variable is **January_temp**

```{r Loading_Data}
rg.load("Exdemo_Scotland_Temperatures","dat",flag.overwrite=TRUE)
dat = db.locate(dat,"January*","z")
```

---

\frametitle{Db class}

Typing the name of the (R) object automatically launches the **print** action (i.e. launching **db.print** function):

```{r Other_printouts, eval=FALSE}
print(dat)
db.print(dat)
```

Check the description of the function (i.e. dbplot) by typing the following command. Note that the line has been commented in order to allow processing in batch.

```{r Interactive_Help, eval=FALSE}
cat("Chunk is suppressed for batch processing")
#?db.print
```

---

\frametitle{Db class}

The following sentences are equivalent and demonstrate the different ways to designate one or several attributes of a Db:

```{r Prints,eval=FALSE}
print(dat,flag.stats=T,names=4)
print(dat,flag.stats=T,names="Elevation")
print(dat,flag.stats=T,names="Elev*")
```

---

\frametitle{Assessors for Db class}

Using the data frame internal organization:

```{r Using_bracket_assessors,eval=FALSE}
dat[]
```

Looking for the fourth field specifically:

```{r Using_bracket_assessor_selective}
dat[,4]
```

```{r Using_assesors_with_names, eval=FALSE}
dat[,"Elevation"]
dat[,"*lev*"]
```

---

\frametitle{Assessors for Db class}

List of all assessors available (valid for any class)

```{r Dumping_assessors}
dat$all
```

---

\frametitle{Assessors for Db class}

```{r Local_assessors}
dat$nsamples
dat$natt
```

Check the geographical extension of the data set:

```{r Getting_Space_Limits}
dat$limits
dat$ndim
```

---

\frametitle{Locators for Db}

```{r Help_on_locators}
cat("Chunk is suppressed for batch processing")
#?db.locate
```

Set the variables 2 and 3 as coordinates :

```{r Setting_X_locators}
dat = db.locate(dat,2:3,"x")
```

Set the variables 4 and 5 as the target regionalized variables :

```{r Setting_Z_locator}
dat = db.locate(dat,4:5,"z")
```

---

\frametitle{Locators for Db}

Check the locator modification:
 
```{r Printout}
print(dat,flag.resume=FALSE)
```

Unset the variable 4 (**Elevation**):

```{r Deleting_locator}
dat=db.locate(dat,4)
```

Which promotes **January_temp** as the first regionalized variable:

```{r Print_again,eval=FALSE}
print(dat,flag.resume=FALSE)
```

---

\frametitle {Plotting a Db}

Plot the contents of a Db using **plot** function (alias for **db.plot**). The proportional (defaulted) option applies to **z1** variable

```{r Plotting,fig.align="center"}
plot(dat)
```

---

\frametitle {Plotting a Db}

Use the same scales on both axes: *isometric* option

```{r Plotting_in_scale,fig.align="center"}
plot(dat,asp=1)
```

---

\frametitle {Plotting a Db}

Set the *isometric* option in the environment. Define other options for the plot.

```{r Setting_isometric_as_constant,fig.align="center"}
constant.define("asp",1)
plot(dat,title="Temperature",pch=21,col="blue",bg="yellow")
```

---

\frametitle {Plotting a Db}

Use the **name** option to modify the displayed variable.

```{r Plotting_with_decoration,fig.align="center"}
plot(dat,pch=21,col="blue",bg="yellow",name="Elevation")
```

---

\frametitle{Digital Elevation Model}

Load the Digital Elevation Model into another Db. Its name within the distribution kit is **Exdemo_Scotland_Elevations**. In your working space, it will be named **db.grid**

```{r Loading_Elevation}
rg.load("Exdemo_Scotland_Elevations","db.grid")
```

The Db is organized as a regular 2-D grid (checked using assessors):

```{r Asking_from_Map_limits}
db.grid$ndim
db.grid$flag.grid
```

---

\frametitle{Digital Elevation Model}


```{r Printing_grid_contents}
print(db.grid,flag.extend=TRUE)
```

---

\frametitle{Selections in Db}

The file **db.grid** contains the field **inshore** which corresponds to a **selection**. Only 3094 (out of 11097) samples (nodes) are **active** (represented in green in the plot of the grid)

```{r Plotting_inshore,fig.align="center"}
plot(db.grid,name="inshore")
```

---

\frametitle{Selections in Db}

We create a selection on the file **dat**, by supressing samples located beyond latitude 1100: active samples in yellow, masked samples in black. 

```{r Display_using_selection, fig.align="center"}
plot(db.grid,name="inshore")
plot(dat,pch=19,col="black",add=T)
dat=db.sel(dat,Latitude<1100)
plot(dat,pch=21,col="blue",bg="yellow",add=T)
```