---
title: "Background"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Background}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

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

# Introduction
...\
We provide methodology for computing the three main product of the *impIndicator*.

# Impact map

The impact risk map shows the impact score for each site, where multiple species can be present. To compute the impact risk per site, aggregated scores across species at each site are needed. The 
`site_impact()` uses *max*, *sum* and *mean* metrics to aggregate impact scores across species as proposed by Boulesnane-Guengant et al., (in preparation). The combinations of within species aggregation metrics for each species and across species for each site leads to five methods of calculating an impact indicator, namely, **precautionary**, **precautionary cumulative**, **mean**, **mean cumulative** and **cumulative**.\


- **precautionary**: maximum score across species' max in each site\
- **precautionary cumulative**: cumulative score across species' max in each site\
- **mean**: mean score across species' mean in each site\
- **mean cumulative**: cumulative score across species' mean in each site\
- **cumulative**: cumulative score across species' sum of maximum score per mechanism\

# Impact indicator

To compute the impact indicator of alien taxa, we sum all the yearly impact scores of each site of the study region. To correct for sampling effort we divide the yearly impact scores by number of sites in the study region with at least a single occurrence throughout the whole year.\
$$I_i = \frac{\sum{S_i}}{N}$$\
- $I_i$ is impact score at year $i$.\
- $S_i$ is the sum of risk map value, where $S=\{s_1,s_2,...,s_n\}$ and  $s_n$ is the site score for site $n$\
- $N$ is number of sites occupied through out the study years of the region.

# Species impact

We compute the impact indicator per species by summing the impact risk map per species and correct for sampling effort by dividing by $N$.