yb_maps_Pseudorca.Rmd

---
title: "Whale Maps for Yvonne"
output: html_notebook
---
```{text}
# Hi Maia,
# Here are the spreadsheets with locations for sperm whales and Pseudorca. I'd like the maps to look similar to this one (from Bradford 2015 tech memo 47). Is it possible to have the map on this scale where it is zoomed into the archipelago more? I like how the latitude and longitude are listed on the outsides of the axes to show where everything is.
# Inline image 1
# 
# 
# 
# 
# The Pseudorca map will be more colorful. 
# Each population (pel, nwhi, mhi) will get a color. pel = pink, nwhi = blue, mhi = green (or if a different color combo looks better, I trust your judgment)
# I'd also like the recorder for each detection labeled using a shape. 
# towed array = medium dot
# fostex = filled square
# harp = filled triangle
# pmrf = X
# So for 'pel' detections, they are all going to be pink dots.
# 'nwhi' will have blue dots, blue Xs
# 'mhi' will have green dots, green squares, and green triangles
# Are there any details I missed? Let me know. I hope the map can be scaled to be mostly the archipelago instead of the central North Pacific. The EEZ boundaries don't need to be included right now. I'll ask Amanda how to plot those for later.
#   
# thanks a million!
```



# Load Packages
```{r messages=F warning=F}
library(geoR)
library(fields)
library(sp)
library(maptools)
library(rgdal)
library(classInt)
library(lattice)
library(zoo)
library(ggplot2)
library(Rmisc)
library(MuMIn)
library(matrixStats)
```

# Sperm Whale Map

The sperm whale map will be simpler. There are more sperm whale detections overall because I added all acoustic detections that were visually sighted into the dataset. The 'sighted' column in 'Sperm_Whale_AcLocations_EEZonly.csv'  indicates whether the acoustic detection was sighted (= '1' ) or not sighted (='0'). That will help determine the symbol for each detection.
Label all sighted acoustic detections with a black 'X'
Label all non-sighted acoustic detection with a filled black dot.  (I thought about coloring the X's and dots also by cruise like before, but I think that would be too busy. What do you think? That would make the legend complicated.)
The legend can go wherever is easiest.

## Load and Clean Data
```{r}
## read in data
#swdf0 = read.csv("E:/PHD\\sperm whales\\Sperm_Whale_AcLocations_EEZonly.csv")
swdata = read.csv("C:\\Users\\Yvonne\\Documents\\PHD\\CHP2&3-Sperm\\data\\Sperm_Whale_Acoustic_Locations.csv")
swdf0 <- swdata[, c(1,2,5,8:13)]
plot(1, type = 'n', xlim = c(150,180), ylim = c(15,32))
with(swdf0, plot(latitude~longitude))

## recenter coordinates ###YB Aug 14, this turns E coordinates west, which is wrong...
swdf0$longitude2 = ifelse(swdf0$E_W == 'E', swdf0$longitude*-1, swdf0$longitude)
swdf0$sighted <- NA
swdf0$sighted <- ifelse(swdf0$visual_id != 999, "1", "0")   #for encoutners with no visual
#swdf0$sighted <- ifelse(swdf0$acoustic_id == 999, "2", "1") DOESNT WORK



#modified by YB on Aug 12
## make a new column with text describing sighted, unsighted
swdf0$sighted2[swdf0$sighted =="0"]<-"Acoustics Only (n=264)"
swdf0$sighted2[swdf0$sighted =="1"]<-"Acoustics and Visuals (n=80)"
swdf0$sighted2[swdf0$sighted =="2"]<-"Visuals Only (n=3)"

swdf0 <- subset(swdf0, (longitude > 177 | longitude < -151))
#swdf0 <- subset(swdf0, (latitude > 15 & latitude < 32 ) & (longitude > 177 | longitude < -151))

#count number of entries for acoustics only, acoustics and visuals, and visuals only
swdf0$sighted <-as.numeric(swdf0$sighted)
test <- subset(swdf0, swdf0$sighted2 == 'Acoustics Only (n=264)')
test <- subset(swdf0, swdf0$sighted2 == "Acoustics and Visuals (n=80)")
test <- subset(swdf0, swdf0$sighted2 == 'Visuals Only (n=16)')

## make a new dataframe with only useful information - may not use cruise number
 swdf = with(swdf0, data.frame('lat' = latitude, 'lon' = longitude, 'lon2' = longitude2, 'sighted' = sighted2, 'cruisenum' = Cruise.Number))

#swdf = with(swdf0, data.frame('lat' = latitude, 'lon' = longitude, 'sighted' = sighted, 'cruisenum' = Cruise.Number))
```

## Get Hawaii Polygon

You will need to download the data from the US Census Bureau Tiger/Line Shapefiles web interface: https://www.census.gov/geo/maps-data/data/tiger-line.html Select year = 2016 (or the most recent year available) and Layertype = States (and equivalent).  Metadata are also available in the .xml files included in the download. Technical details on the files in general are available in the technical documentation on the website.
 Once you have downloaded and extracted the shapefile to your working directory
 Load the file.  You will only need the .shp file for R plotting.

```{r}

#us<-readShapePoly("C:\\Users\\yvers\\Documents\\sperm whale\\yb_maps\\tl_2016_us_state.shp")
#us<-readShapePoly("E:\\PHD\\sperm whales\\yb_maps\\tl_2016_us_state.shp")
us<-readShapePoly("C:\\Users\\Yvonne\\Documents\\PHD\\CHP2&3-Sperm\\code\\yb_maps\\shp\\tl_2016_us_state.shp")
hawaii<-subset(us,NAME=="Hawaii")
hawaiiMap<-fortify(hawaii) #converts shape file in a readable format for ggplot
```

## Plot the map with ggplot
```{r}

png(file = "C:\\Users\\Yvonne\\Documents\\PHD\\CHP2&3-Sperm\\data\\MAPtest.png", width = 8, height = 5, res = 200, units = "in")

ggplot() +  ## this just makes the background white
  theme_bw() +
  ## remove the legend title, x and y labels, and gridlines in "theme"
  theme(
  legend.title = element_blank(),
  legend.key = element_blank(),
  legend.text = element_text(size=12),
  axis.title.x = element_blank(),
  axis.title.y = element_blank(),
  panel.grid = element_blank(),
  plot.margin = unit(c(1,1,1,1), "cm"),
  legend.position = c(.2,.2)
  ) +
  ## customize the x labels to have the degree symbol
  scale_x_continuous(
  # limits = c(-179,140),
  breaks = seq(-180, -150, 10),
  labels = c(expression('180' *  ~ degree), 
             expression('170' *  ~ degree * W), 
             expression('160' *  ~ degree * W), 
             expression('150' *  ~ degree * W))
             
  ) +
  
  scale_y_continuous(
  limits = c(15,33),
  breaks = seq(15, 30, 5),
  labels = c(expression('15' *  ~ degree * N), 
             expression('20' *  ~ degree * N), 
             expression('25' *  ~ degree * N), 
             expression('30' *  ~ degree * N ))
  ) +
  ## indicate which shapes you want (do ?pch to see options, they are coded by number), ONE FOR EACH FACTOR IN SIGHTED
  scale_shape_manual(values = c(19,19,17))+
  scale_color_manual(values = c('darkred','navy','darkviolet'))+
  
  ## add the map downloaded above
  geom_polygon(data = hawaiiMap, aes(x = long, y = lat, group = group), fill = 'gray', color = 'dimgray') +
  coord_fixed(1) +
  
  ## now add the data points
  geom_point(data = swdf, aes(
  x = lon2,
  y = lat,
  pch = factor(sighted),
  col = factor(sighted)
  ))
dev.off() ## so you don't overwrite the map png it later
```

# Pseudorca map
The Pseudorca map will be more colorful. 
Each population (pel, nwhi, mhi) will get a color. pel = pink, nwhi = blue, mhi = green (or if a different color combo looks better, I trust your judgment)
I'd also like the recorder for each detection labeled using a shape. 

fostex = filled square
harp = filled triangle
pmrf = X
towed array = medium dot
So for 'pel' detections, they are all going to be pink dots.
'nwhi' will have blue dots, blue Xs
'mhi' will have green dots, green squares, and green triangles

## Load and clean data
```{r}
psdf = read.csv("C:\\Users\\Yvonne\\OneDrive\\PHD\\CHP1-FKW\\data\\Pseudorca_AcLocations.csv")
levels(psdf$population)[3] <- 'Pelagic'
levels(psdf$population)[2] <- 'NWHI'

# ## recenter coordinates
psdf$longitude2 = ifelse(psdf$longitude > 0, psdf$longitude*-1, psdf$longitude)

psdf$population = with(psdf, factor(population, levels = rev(levels(population))))



## make a new column with text describing sighted, unsighted
# swdf0$sighted2 = ifelse(swdf0$sighted == '0', "Not Sighted", "Sighted")

## make a new dataframe with only useful information - may not use cruise number
# swdf = with(swdf0, data.frame('lat' = latitude, 'lon' = longitude, 'lon2' = longitude2, 'sighted' = sighted2, 'cruisenum' = Cruise.Number))
```

```{r}
png(file = "C:\\Users\\Yvonne\\OneDrive\\PHD\\CHP1-FKW\\data\\PseudMap.png", width = 9, height = 5, res = 200, units = "in", bg="transparent")


L = ggplot(data = psdf, aes(x = longitude2, y = latitude)) +  ## this just makes the background white
  theme_bw() +
  ## remove the legend title, x and y labels, and gridlines
  theme(
  legend.title = element_blank(),
  legend.key = element_blank(),
  axis.title.x = element_blank(),
  axis.title.y = element_blank(),
  panel.grid = element_blank(),
  legend.text = element_text(size=12),
  legend.position = c(0.1,0.25),
  plot.margin = unit(c(1,1,1,1), "cm"),
  legend.background = element_rect(fill = NA),
  panel.background = element_rect(fill = "transparent"),
  plot.background = element_rect(fill = "transparent")

  
  ) +
  
  scale_x_continuous(
  breaks = seq(-180, -150, 10),
  labels = c(expression('180' *  ~ degree * W), 
             expression('170' *  ~ degree * W), 
             expression('160' *  ~ degree * W), 
             expression('150' *  ~ degree * W))
  ) +
  
  scale_y_continuous(
  limits = c(15,32),
  breaks = seq(15, 30, 5),
  labels = c(expression('15' *  ~ degree * N), 
             expression('20' *  ~ degree * N), 
             expression('25' *  ~ degree * N), 
             expression('30' *  ~ degree * N ))
  ) +
  
  ## indicate which shapes you want (do ?pch to see options)
  scale_shape_manual(values = c(3, 4, 8))+
  scale_color_manual(values = c('red','dodgerblue','darkorange2')) +
  
  ## add the map downloaded above
  geom_polygon(data = hawaiiMap, aes(x = long, y = lat, group = group), fill = 'darkgrey', color = 'black') +
  coord_fixed(1) 
  


  ## now add the data points
L +  geom_point(size=1) +
     geom_text_repel(data = psdf, size=2.5, aes(x = longitude2, y = latitude, label = factor(encounter.ID))) 
  #+
     geom_text_repel(psdf, aes(label=psdf$tmp))#aes(color=factor(psdf$encounter.ID), hjust=0, vjust=0))  
  
  #geom_point(data = psdf, aes(x = longitude2, y = latitude))
  #guides(colour = guide_legend(override.aes = list(size=2))) 
  


dev.off() ## so you don't overwrite the map png it later
```

##Zoom in on Main Hawaiian Islands (Does not work)
```{r}

MHImap <- subset(hawaiiMap, long > -163  )
fortify(MHImap)

png(file = "C:\\Users\\Yvonne\\Documents\\PHD\\CHP1-FKW\\data\\results/PseudMapMHI.png", width = 8, height = 5, res = 200, units = "in")
ggplot() +  ## this just makes the background white
  theme_bw() +
  ## remove the legend title, x and y labels, and gridlines
  theme(
  legend.title = element_blank(),
  axis.title.x = element_blank(),
  axis.title.y = element_blank(),
  panel.grid = element_blank(),
  legend.text = element_text(size=12),
  legend.position = c(.5,.4),
  plot.margin = unit(c(1,1,1,1), "cm"),
  legend.background = element_rect(fill = NA)
  ) +
  
  # scale_x_continuous(
  # breaks = seq(-180, -150, 10),
  # labels = c(expression('180' *  ~ degree),
  #            expression('170' *  ~ degree * W),
  #            expression('160' *  ~ degree * W),
  #            expression('150' *  ~ degree * W ))
  
  scale_x_continuous(
  breaks = seq(-162, -150, 4),
  labels = c(expression('162' *  ~ degree * W), 
             expression('158' *  ~ degree * W), 
             expression('154' *  ~ degree * W),
             expression('150' *  ~ degree * W ))
  ) +
  
  scale_y_continuous(
  limits = c(15,25),
  breaks = seq(15, 25, 5),
  labels = c(expression('15' *  ~ degree * N), 
             expression('20' *  ~ degree * N), 
             expression('25' *  ~ degree * N )) 
  ) +
  
  ## indicate which shapes you want (do ?pch to see options)
  scale_shape_manual(values = c(4,2,0,18))+
  scale_color_manual(values = c('navy','coral1','lightslateblue')) +
  ## add the map downloaded above
  geom_polygon(data = MHImap, aes(x = long, y = lat, group = group), fill = 'gray', color = 'darkgray') +
  coord_fixed(1) +
  
  ## now add the data points
  geom_point(data = psdf, aes(
  x = longitude2,
  y = latitude,
  pch = factor(recorder),
  col = factor(population)
  ))
dev.off() ## so you don't overwrite the map png it later
```