quickstart.Rmd

marxan quick start guide
============

First, let's load the 'marxan' R package and some example data.

```{r, eval=FALSE}
# load marxan R package
library(marxan)

# load example data
data(taspu, tasinvis)
```

This data comes from the ['Introduction to Marxan'](http://marxan.net/courses.html). The `taspu` object stores data for the planning units, and the `tasinvis` object stores data for 63 vegetation classes--the biodiversity features in this example problem. Let's make some reserve systems and plot them.

```{r, eval=FALSE}
# make reserve systems
results<-marxan(taspu, tasinvis, targets="20%", NUMREPS=100L, BLM=0)

# geoplot for best solution
plot(results, 0)

# geoplot for selection frequencies
plot(results)
```

We can also plot distribution of features in the planning units.

```{r, eval=FALSE}
# plot distribution of vegetation class 5
spplot(results, 5, var='occ')

# plot richness in planning units
spplot(results, var='occ')
```

The solutions in this portfolio are fairly fragmented. Let's generate some more solutions with a higher boundary length modifier (BLM).

```{r, eval=FALSE}
# generate new portfolio of reserve systems
results2<-update(results, ~opt(BLM=500))
```

Now, let's compare solutions in the two portfolios.

```{r, eval=FALSE}
# geoplot comparing best solutions in each portfolio
plot(results, results2, i=0, j=0)

# geoplot comparing selection frequencies
plot(results, results2, basemap='satelliet', alpha=0.4)
```

The solutions in the second portfolio seem to be better. How can we rank the solutions within this portfolio? Let's make some dotcharts.

```{r, eval=FALSE}
# dotchart ranking solutions based on overall score
dotchart(results2, var='score')

# dotchart ranking solutions based on connectivity
dotchart(results2, var='con')
```

Finally, let's use multivariate analyses to visualise and compare the solutions in the second portfolio.

```{r, eval=FALSE}
# dendrogram showing how similar solutions are to each other based on selections
dendrogram(results2, type='dist', var='selections', nbest=10)

# ordination plot show how similar solutions are to each other based on 
# well they represent different features
ordiplot(results2, type='mds', var='amountheld', method='bray')
```

<!--
%\VignetteEngine{knitr::knitr}
%\VignetteIndexEntry{Quick start guide to the marxan R package}
-->