Test presentation.Rmd

---
title: "Marcuson - Technical Presentation"
author: "Christopher Martin"
date: "`r Sys.Date()`"
output: powerpoint_presentation
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = F)

library(knitr)
```

# Introduction

The aim of this report is to detail to extensions made to the R package dsims through the inclusion of density surface modelling (dsm) simulations.

# Plot sampling
```{r crop = TRUE, fig.show='hold', out.width='50%'}
plot_crop(include_graphics(paste0(getwd(),'/Reports/Plots/Plot sampling example.jpg')))
```
# Distance sampling

- Line transects
- Point transects

# Line transects
```{r crop = TRUE, fig.show='hold', out.width='50%'}
plot_crop(include_graphics(paste0(getwd(),'/Reports/Plots/DS line example.jpg')))
```

# Line transects - Detection function
```{r crop = TRUE, fig.show='hold', out.width='50%'}
plot_crop(include_graphics(paste0(getwd(),'/Reports/Plots/DS detect.jpg')))
```

# Density surface modelling

- 1 or 2 stage approaches
- Count per segment model
$$ \mathbb{E}(n_j) = \hat{p_j}A_jexp[ \, \beta_0 + \sum_k f_k(z_{jk} ) ]\,$$

# Simulation profile

- Region and design specification.
- Prediction grid generated.
- For each run of the simulation:
- - run a survey
- -  coerce survey data into correct format
- -  fit a distance sampling model
- -  fit a density surface model
- -  extract results
  
# Region and design specification

- Region profile
- Point or Line transect options.
- Construct density surface
- Describe population
- Define detection function


# Region and Density Surface
```{r crop = TRUE, fig.show='hold', out.width='50%'}
plot_crop(include_graphics(paste0(getwd(),'/Reports/Plots/NS region.jpg')))
plot_crop(include_graphics(paste0(getwd(),'/Reports/Plots/North_Sea_density.jpg')))
```

# Example survey designs
```{r fig.show = 'hold', out.width='33%'}
knitr::include_graphics(paste0(getwd(),'/Reports/Plots/North Sea survey point.jpg'))
knitr::include_graphics(paste0(getwd(),'/Reports/Plots/North Sea survey line.jpg'))
knitr::include_graphics(paste0(getwd(),'/Reports/Plots/North Sea survey zigzag.jpg'))
```

# Prediction Grid Generation

```{r fig.show = 'hold', out.width='33%'}
include_graphics(paste0(getwd(),'/Reports/Plots/Prediction grid.jpg'))
```

# Survey
```{r fig.show = 'hold', out.width='33%'}
include_graphics(paste0(getwd(),'/Reports/Plots/Montrave examp survey.jpg'))
```

# Data coercion

- transects into segments
- link observations to segments

# Fit Distance sampling model

 - ds( 
 - - survey$dist.data,
 - - truncation = design.trunc,
 - - transect = transect.type,
 - - formula = ~1,
 - - key = key.function,
 - - adjustment = NULL)

# Fit DSM model

 - dsm( 
 - - count ~ s(x, y, k = sum(transects),
 - - ddf.obj = ds.mod,
 - - segment.data = segdata,
 - - segment.area = segdata$Area,
 - - observation.data = obsdata,
 - - family = tw(),
 - - transect = transect.type)

# Results - Point transect
```{r fig.show = 'hold', out.width='50%'}
Mont.rand <-  read.csv('Estimates/Montrave5000linerandom.csv')

hist(Mont.rand$dsm.est,breaks = 50,
     main = 'Histogram of DSM estimates',
     xlab = 'DSM Abundance estimates')
abline(v = mean(Mont.rand$dsm.est), col = 'red')

hist(Mont.rand$ds.est,breaks = 50, 
     main = 'Histogram of DS estimates',
     xlab = 'DS Abundance estimates')
abline(v = mean(Mont.rand$ds.est), col = 'red')
```

# Results - Point transects


# Results - Parallel Line
```{r fig.show = 'hold', out.width='50%'}
Mont.para <-  read.csv('Estimates/Montrave5000lineparallel.csv')

hist(Mont.para$dsm.est,breaks = 50,
     main = 'Histogram of DSM estimates',
     xlab = 'DSM Abundance estimates')
abline(v = mean(Mont.para$dsm.est), col = 'red')

hist(Mont.para$ds.est,breaks = 50, 
     main = 'Histogram of DS estimates',
     xlab = 'DS Abundance estimates')
abline(v = mean(Mont.para$ds.est), col = 'red')
```

# Results - Parallel Line

# Results - Zigzag Line
```{r fig.show = 'hold', out.width='50%'}
Mont.zig <-  read.csv('Estimates/Montrave5000linezigzag.csv')

hist(Mont.zig$dsm.est,breaks = 50,
     main = 'Histogram of DSM estimates',
     xlab = 'DSM Abundance estimates')
abline(v = mean(Mont.zig$dsm.est), col = 'red')

hist(Mont.zig$ds.est,breaks = 50, 
     main = 'Histogram of DS estimates',
     xlab = 'DS Abundance estimates')
abline(v = mean(Mont.zig$ds.est), col = 'red')
```

# Results - Zigzag Line

# Further investigations

# Conclusions