---
title: "Predictions - Results"
author: ""
date: ""
output: 
  html_document:
    toc: true
    toc_float: true
    theme: cerulean
---

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

#main packages #delete/add as needed
library(dplyr)
library(ggplot2)
library(knitr)
library(ggpubr)
library(plotly)
library(pracma) #for standard error
library(reactable)
library(sf)
library(rgdal)
library(sp)
library(tidyterra)
library(terra)
library(wesanderson)


theme_set(theme_bw())
```

## About

File for results from predictions, specifically:

* maps of predicted spatial usage
* how many tracks come within 10 m of fish pass

For the simulations that began in the lower 100 m of the river.


## Barbel

### Maps 

```{r all barbel maps}

dat1 <- read.csv("PREDICTIONS - downstream start/all_barbel_up_to_agent_100_disch20.csv")
dat2 <- read.csv("PREDICTIONS - downstream start/all_barbel_up_to_agent_100_disch50.csv")
dat3 <- read.csv("PREDICTIONS - downstream start/all_barbel_up_to_agent_100_disch80.csv")


barb_dat1 <- bind_rows(dat1,dat2,dat3)
barb_dat <- barb_dat1

barb_dat <- barb_dat %>% filter(y!=0) #to catch a handful of errors


#remove points in lower 150m
shp <- st_read("lower150m/lower150m.shp")

start_loc_pol <- st_read("start_sections/start_sections.shp")


barb_dat2 <- st_as_sf(barb_dat,coords = c("x","y"),crs=st_crs(32632)) %>% st_transform(crs=25832)


barb_dat2$in_lower_zone <- st_intersects(barb_dat2$geometry,shp) %>% as.integer()

barb_dat <- barb_dat[is.na(barb_dat2$in_lower_zone),]



barb_dat3 <- barb_dat1 %>% group_by(fish_id,approach,discharge) %>% summarise(x=first(x),y=first(y))

barb_dat4 <- st_as_sf(barb_dat3,coords = c("x","y"),crs=st_crs(32632)) %>% st_transform(crs=25832)

barb_dat4$start_loc <- st_intersects(barb_dat4$geometry,start_loc_pol) %>% as.integer() 

barb_dat4 <- barb_dat4 %>% st_drop_geometry()

barb_dat4$start_loc[barb_dat4$start_loc==3] <- "centre"
barb_dat4$start_loc[barb_dat4$start_loc==2] <- "right"
barb_dat4$start_loc[barb_dat4$start_loc==1] <- "left"
barb_dat4$start_loc[is.na(barb_dat4$start_loc)] <- "left"


barb_dat5 <- barb_dat4 %>% select(c(fish_id,approach,start_loc,discharge))

barb_dat <- merge(barb_dat,barb_dat5,by=c("fish_id","approach","discharge"))

#

barb_plts <- ggplot(barb_dat,aes(x=x,y=y))+
  geom_bin2d()+
  facet_wrap(~discharge)+
  geom_point(x=591683.2,y=5296867.5,col="red")+
    scale_fill_gradientn(colours = terrain.colors(7))
barb_plts

ggplot(barb_dat,aes(x=x,y=y))+
  geom_bin2d()+
  facet_grid(state~discharge)+
  geom_point(x=591683.2,y=5296867.5,col="red")+
    scale_fill_gradientn(colours = terrain.colors(7))

```






## Grayling

### Maps 

```{r all grayling maps}

dat1 <- read.csv("PREDICTIONS - downstream start/all_grayling_up_to_agent_100_disch20.csv")
dat2 <- read.csv("PREDICTIONS - downstream start/all_grayling_up_to_agent_100_disch50.csv")
dat3 <- read.csv("PREDICTIONS - downstream start/all_grayling_up_to_agent_100_disch80.csv")

#dat3 <- read.csv("PREDICTIONS - downstream start/all_grayling_up_to_agent_78.csv")


#gray_dat <- dat1
#gray_dat <- bind_rows(dat1,dat2)
gray_dat1 <- bind_rows(dat1,dat2,dat3)

gray_dat <- gray_dat1 %>% filter(y!=0)# & y<5296950)

#remove points

shp <- st_read("lower150m/lower150m.shp")
start_loc_pol <- st_read("start_sections/start_sections.shp")


gray_dat2 <- st_as_sf(gray_dat,coords = c("x","y"),crs=st_crs(32632)) %>% st_transform(crs=25832)

gray_dat2$in_lower_zone <- st_intersects(gray_dat2$geometry,shp) %>% as.integer()
gray_dat2$in_lower_zone[is.na(gray_dat2$in_lower_zone)] <- 2
gray_dat2 <- st_drop_geometry(gray_dat2)

gray_dat2 %>% group_by(discharge,in_lower_zone) %>% summarise(n=n())

#gray_dat <- gray_dat[is.na(gray_dat2$in_lower_zone),]
gray_dat <- gray_dat[gray_dat2$in_lower_zone==2,]


gray_dat3 <- gray_dat1 %>% group_by(fish_id,approach,discharge) %>% summarise(x=first(x),y=first(y))

gray_dat4 <- st_as_sf(gray_dat3,coords = c("x","y"),crs=st_crs(32632)) %>% st_transform(crs=25832)

gray_dat4$start_loc <- st_intersects(gray_dat4$geometry,start_loc_pol) %>% as.integer() 

gray_dat4 <- gray_dat4 %>% st_drop_geometry()

gray_dat4$start_loc[gray_dat4$start_loc==3] <- "center"
gray_dat4$start_loc[gray_dat4$start_loc==2] <- "right"
gray_dat4$start_loc[gray_dat4$start_loc==1] <- "left"
gray_dat4$start_loc[is.na(gray_dat4$start_loc)] <- "left"


#gray_dat4$start_loc[gray_dat4$start_loc==3] <- "right"
#gray_dat4$start_loc[gray_dat4$start_loc==2] <- "center"
#gray_dat4$start_loc[gray_dat4$start_loc==1] <- "left"

gray_dat <- merge(gray_dat,gray_dat4,by=c("fish_id","approach","discharge"))

shp2 <- vect("lower150m/lower150m.shp")

#ggplot(data=shp2)+
 # geom_point(data=gray_dat,aes(x=x,y=y))+
#  facet_wrap(~discharge)+
 # geom_point(x=591683.2,y=5296867.5,col="red")+
  #  geom_spatvector(fill=NA)
  

gray_plts <- ggplot(gray_dat,aes(x=x,y=y))+
  geom_bin2d()+
  facet_wrap(~discharge)+
  geom_point(x=591683.2,y=5296867.5,col="red")+
  scale_fill_gradientn(colours = terrain.colors(7))+
  labs(title="Grayling predicted usage where tracks began downstream")

#ggsave(filename="figures/gray_plts_dwnstrm.png",height=3,width=8,plot=gray_plts)
gray_plts


#select some agents to plot tracks for

p <- ggplot(filter(gray_dat,fish_id<=7 &discharge==20),aes(x=x,y=y,col=as.factor(approach)))+
  geom_point()+
  geom_path()+
  #facet_grid(fish_id~discharge)+
  facet_wrap(~fish_id,ncol=2)+
  scale_color_manual(values=wes_palette("Darjeeling1",n=5,type="discrete"),name="Approach ID")+
  geom_point(x=591683.2,y=5296867.5,col="black",pch=17)+
  scale_x_continuous(breaks=c(591650,591850))

#ggsave(filename = "figures/grayling_sim_tracks_FINAL.png",plot=p,width=5,height=5)
  
#fish id 1, 76, look particularly nice for discharge 20


pp <- gray_dat %>% filter(fish_id==76)

#ggplot(pp,aes(x=x,y=y,col=as.factor(approach)))+
 # geom_point()+
  #geom_path()+
  #facet_wrap(~discharge)+
  #geom_point(x=591683.2,y=5296867.5,col="red")+
  #scale_fill_gradientn(colours = terrain.colors(7))
  #facet_wrap(~fish_id)

```


## Looking at % of tracks and fish that did not enter the array

Can check if consistent among start locations too

```{r barbel tracks per discharge}
barb1 <- barb_dat1 %>% group_by(discharge,fish_id) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge) %>% summarise(n_tracks_all=sum(n_tracks))

barb2 <- barb_dat %>% group_by(discharge,fish_id) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge) %>% summarise(n_tracks_up=sum(n_tracks))

barb3 <- merge(barb1,barb2,by="discharge")

barb3$perc <- barb3$n_tracks_up/barb3$n_tracks_all*100

reactable(barb3)

```


```{r barbel tracks per discharge ands tart loc}
barb1 <- barb_dat4 %>% group_by(discharge,fish_id,start_loc) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge,start_loc) %>% summarise(n_tracks_all=sum(n_tracks))

barb2 <- barb_dat %>% group_by(discharge,fish_id,start_loc) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge,start_loc) %>% summarise(n_tracks_up=sum(n_tracks))

barb3 <- merge(barb1,barb2,by=c("discharge","start_loc"))

barb3$perc <- barb3$n_tracks_up/barb3$n_tracks_all*100

reactable(barb3)

```

```{r barbel fish per discharge}

barb2 <- barb_dat %>% group_by(discharge) %>% 
  summarise(n_fish=length(unique(fish_id)))



barb2$perc <- barb2$n_tracks_up

reactable(barb2)

```


```{r barbel tracks per discharge}
barb1 <- barb_dat1 %>% group_by(discharge,fish_id) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge) %>% summarise(n_tracks_all=sum(n_tracks))

barb2 <- barb_dat %>% group_by(discharge,fish_id) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge) %>% summarise(n_tracks_up=sum(n_tracks))

barb3 <- merge(barb1,barb2,by="discharge")

barb3$perc <- barb3$n_tracks_up/barb3$n_tracks_all*100

reactable(barb3)

```

```{r barbel fish per discharge}

barb2 <- barb_dat %>% group_by(discharge) %>% 
  summarise(n_fish=length(unique(fish_id)))



barb2$perc <- barb2$n_tracks_up

reactable(barb2)

#per start location
#this needs to be dones as a comparison to total number fish per start loc

barb2 <- barb_dat %>% group_by(discharge,start_loc) %>% 
  summarise(n_fish=length(unique(fish_id)))

barb3 <- barb_dat4 %>% group_by(discharge,start_loc) %>% #get all that started
  summarise(n_fish_tot=length(unique(fish_id)))

barb4 <- merge(barb2,barb3,by=c("discharge","start_loc"))

barb4$perc <- barb4$n_fish/barb4$n_fish_tot*100
reactable(barb4)

```


```{r graylng tracks per discharge}
gray1 <- gray_dat1 %>% group_by(discharge,fish_id) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge) %>% summarise(n_tracks_all=sum(n_tracks))

gray2 <- gray_dat %>% group_by(discharge,fish_id) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge) %>% summarise(n_tracks_up=sum(n_tracks))

gray3 <- merge(gray1,gray2,by="discharge")

gray3$perc <- gray3$n_tracks_up/gray3$n_tracks_all*100

reactable(gray3)

```

```{r grayling tracks per discharge ands tart loc}
gray1 <- gray_dat4 %>% group_by(discharge,fish_id,start_loc) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge,start_loc) %>% summarise(n_tracks_all=sum(n_tracks))

gray2 <- gray_dat %>% group_by(discharge,fish_id,start_loc) %>% 
  summarise(n_tracks=length(unique(approach))) %>% ungroup() %>%
  group_by(discharge,start_loc) %>% summarise(n_tracks_up=sum(n_tracks))

gray3 <- merge(gray1,gray2,by=c("discharge","start_loc"))

gray3$perc <- gray3$n_tracks_up/gray3$n_tracks_all*100

reactable(gray3)

```

```{r grayling fish per discharge}

gray2 <- gray_dat %>% group_by(discharge) %>% 
  summarise(n_fish=length(unique(fish_id)))



gray2$perc <- gray2$n_tracks_up

reactable(gray2)

#per start location
#this needs to be dones as a comparison to total number fish per start loc

gray2 <- gray_dat %>% group_by(discharge,start_loc) %>% 
  summarise(n_fish=length(unique(fish_id)))

gray3 <- gray_dat4 %>% group_by(discharge,start_loc) %>% #get all that started
  summarise(n_fish_tot=length(unique(fish_id)))

gray4 <- merge(gray2,gray3,by=c("discharge","start_loc"))

gray4$perc <- gray4$n_fish/gray4$n_fish_tot*100
reactable(gray4)

```