03_plot-speclibs.Rmd

---
title: "Part 3: Plotting of Speclibs"
author: "Patrick Schratz"
date: "`r format(Sys.time(), '%a %d %b %Y')`"
output: 
  html_document:
    theme: paper
    highlight: haddock 
    # code_folding: show
    toc: yes
    toc_depth: 4
    toc_float: yes
editor_options: 
  chunk_output_type: console
---

```{r setup, include=FALSE, cache=FALSE}
## knitr options
# knitr::opts_chunk$set(cache = FALSE)
knitr::opts_chunk$set(comment = "")

library("raster")
library("hsdar")
library("here")
library("rasterVis", )
library("mapview")
library("magrittr")
library("stars")

# create objects from previous document
file <- here("data") %>%
  list.files(pattern = ".tif$", full.names = TRUE)
raster <- raster::brick(file[1])
wavelength <- c(
  404.08, 408.5, 412.92, 417.36, 
  421.81, 426.27, 430.73, 435.20,
  439.69, 444.18, 448.68, 453.18, 457.69, 462.22, 466.75, 471.29,
  475.83, 480.39, 484.95, 489.52, 494.09, 498.68, 503.26, 507.86,
  512.47, 517.08, 521.70, 526.32, 530.95, 535.58, 540.23, 544.88,
  549.54, 554.20, 558.86, 563.54, 568.22, 572.90, 577.60, 582.29,
  586.99, 591.70, 596.41, 601.13, 605.85, 610.58, 615.31, 620.05,
  624.79, 629.54, 634.29, 639.04, 643.80, 648.56, 653.33, 658.10,
  662.88, 667.66, 672.44, 677.23, 682.02, 686.81, 691.60, 696.40,
  701.21, 706.01, 710.82, 715.64, 720.45, 725.27, 730.09, 734.91,
  739.73, 744.56, 749.39, 754.22, 759.05, 763.89, 768.72, 773.56,
  778.40, 783.24, 788.08, 792.93, 797.77, 802.62, 807.47, 812.32,
  817.17, 822.02, 826.87, 831.72, 836.57, 841.42, 846.28, 851.13,
  855.98, 860.83, 865.69, 870.54, 875.39, 880.24, 885.09, 889.94,
  894.79, 899.64, 904.49, 909.34, 914.18, 919.03, 923.87, 928.71,
  933.55, 938.39, 943.23, 948.07, 952.90, 957.73, 962.56, 967.39,
  972.22, 977.04, 981.87, 986.68, 991.50, 996.31
)
speclib <- hsdar::speclib(raster, wavelength)
mask(speclib) <- c(404, 418)
hyperspecs <- hsdar::HyperSpecRaster(raster, wavelength)
```

Objects of class `Speclib` can be plotted using the generic `plot()` function.  
Additional arguments can be specified to produce detailed plots.

# Default `plot()` call

The default `plot()` call plots the mean values including the standard deviation across all spectra in our `speclib`.

```{r 03-plot-speclibs-1 }
plot(speclib)
```

# Plotting of a specific spectrum 

We have 28350 (162*175) spectra stored in our `speclib` object.

```{r 03-plot-speclibs-2 }
nspectra(speclib)
```

To plot the 100th spectrum, we can do the following:

```{r 03-plot-speclibs-3 }
plot(speclib, FUN = 100, main = "100th spectrum of Speclib")
```

# Combining multiple spectra

In R the `par()` argument specifies the arrangement of multiple plots using the base `plot()` call.

Let´s combine four different plots in a 2x2 matrix and color the single spectrum

```{r 03-plot-speclibs-4 }
par(mfrow = c(2,2))

plot(speclib, main = "All spectra")
plot(speclib, FUN = 100, main = "100th spectrum of Speclib", col = "cyan")
plot(speclib, FUN = "median", main = "Median spectrum")
plot(speclib, FUN = "mean", main = "Mean spectrum")
```

When combining multiple single spectra, you might want to distinguish them by color. This can be done by appending plots using the argument `new = FALSE`. 
This prevents the call of a new plot and appends to the previous one.
Let's take the 1st, the 5th and the 10th spectra.

```{r 03-plot-speclibs-5 }
par(mfrow = c(1,1))
plot(speclib, FUN = 1, col = "red")
plot(speclib, FUN = 5, col = "blue", new = FALSE)
plot(speclib, FUN = 10, col = "orange", new = FALSE)
```