---
title: "Intro to fractalforest"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{intro-to-fractal-forest}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

The `fractalforest` package allows the simulation of fractal trees and, consequently, fractal forests. The trees are created using the Lindenmayer system (L-system).  

It is possible to use forest inventory, floristic surveys, and phytosociological studies databases to represent the schematic profile of the forest, as shown in the following example.

```{r, message=FALSE, warning=FALSE, results='hide', out.width='100%', fig.width=7, fig.height=5, dpi = 200}
library(fractalforest)
library(forestmangr)
library(dplyr)

data("exfm20")

dados <- exfm20 %>%
  filter(!dead)

fracflor <- build_forest_profile(data = dados,
                                 height = Htot,
                                 diameter = dbh,
                                 label = scientific.name,
                                 leaf_size = 50,
                                 tree_model = 2,
                                 sample = TRUE,
                                 n_trees = 30,
)

plot_forest_profile(fracflor,
                    label = scientific.name)
```

The `build_forest_profile` function creates a fractal tree database from a dataset where each row represents a tree. When the argument `sample = TRUE`, the trees to be plotted are randomly selected from the dataset, and the number of trees is defined by the `n_trees` argument.  
The `plot_forest_profile` function uses the `ggplot2` package to plot the fractal forest based on the dataset created by the `build_forest_profile` function.  

Some parameters make it possible to customize the visualization of fractal trees, including adding colors, and diameters from the base to the top of the trees.  

```{r, results='hide', out.width='100%', fig.width=7, fig.height=5, dpi = 200}
fracflor_colors <- fracflor %>%
  select(tree_id) %>%
  distinct() %>%
  rowwise() %>%
  mutate(cor_folhas = sample(c('darkgreen','darkolivegreen','darkolivegreen4','darkseagreen4'), 1),
         cor_ramos = sample(c('firebrick4','brown4','chocolate4','bisque4'),1))

fracflor2 <- fracflor %>%
  left_join(fracflor_colors, by = 'tree_id')

plot_forest_profile(fracflor2,
                    d_col = diameter,
                    branch_color = cor_ramos,
                    leaf_color = cor_folhas,
                    label = scientific.name)
```

There are several ways to introduce randomness when constructing the fractal trees and fractal forests. The `build_forest_profile` function includes arguments that control the variability of branch angles, lengths, and even the distance between trees. The parameters are detailed in the function documentation. 


```{r, results='hide', out.width='100%', fig.width=7, fig.height=5, dpi = 200}

fracflor <- build_forest_profile(data = dados,
                                 height = Htot,
                                 diameter = dbh,
                                 label = scientific.name,
                                 leaf_size = 50,
                                 randomness = T,
                                 angle_cv = .1,
                                 length_cv = .3,
                                 dist_cv = .2,
                                 tree_model = 6,
                                 sample = TRUE,
                                 n_trees = 30,
)

fracflor2 <- fracflor %>%
  left_join(fracflor_colors, by = 'tree_id')

plot_forest_profile(fracflor2,
                    d_col = diameter,
                    branch_color = cor_ramos,
                    leaf_color = cor_folhas,
                    label = scientific.name)
```