---
title: "`r lubridate::year(lubridate::today())` Sierra Nevada Bioacoustic Monitoring"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    theme: journal
    social: menu
    source_code: embed
    # runtime: shiny
---

```{r setup, include = FALSE}
library(flexdashboard)
library(shiny)
library(plotly)
library(CAbioacoustics)
library(sf)
library(tidyverse)
library(mapview)

# connect to spotted owl database -----------------------------------------

# create connection
# credentials stored in keyring
cb_connect_db()

deployments_df <-
  tbl(conn, "acoustic_field_visits") |>
  filter(survey_year == 2024 & is_invalid == 0) |>
  glimpse() |>
  select(
    id, study_type, deploy_by, deploy_date, recover_date, group_id, visit_id, survey_year, db_cell_id = cell_id,
    unit_number, deployment_name, swift_id, cardin, utm_zone, utme, utmn) |>
  collect() |>
  mutate(
    zone_utm = utm_zone, 
    northing_utm = utmn, 
    easting_utm = utme,
    status = case_when(
      !is.na(recover_date) ~ 'Retrieved',
      TRUE ~ 'Deployed'
    )
  )

# disconnect
cb_disconnect_db()

# load hexes
hexes_sf <-
  cb_get_spatial('hexes') |>
  select(cell_id, ownership)

# convert to sf
deployments_sf <-
  deployments_df |>
  dplyr::group_split(utm_zone) |>
  purrr::map_dfr(cb_make_aru_sf) |>
  rename(utm_zone = zone_utm, utme = easting_utm, utmn = northing_utm)

# do point-in-polygon
deployments_sf <-
  st_join(deployments_sf, hexes_sf, join = st_within) |>
  # this is the cell ID from the CAbioacoustics layer
  rename(shp_cell_id = cell_id) |> 
  select(cell_id = db_cell_id, deployment_name, deploy_date, recover_date, status, ownership)

cb_disconnect_db()

total_deployments <- nrow(deployments_sf)
retrieved_arus <- nrow(deployments_sf |> filter(status == 'Retrieved'))
cells_surveyed <- length(unique(deployments_sf$cell_id))
updated <- lubridate::today()

```

## About {.sidebar}

This is a data dashboard to view the status of University of Wisconsin - Madison passive acoustic monitoring efforts in the Sierra Nevada, California. More information on the project can be found [here](https://peery.russell.wisc.edu/bioacoustics/).

Note that data are preliminary and may contain errors. Autonomous recording unit (ARU) deployments are also jittered to keep locations confidential.

<br>

**ARUs deployed**: `r total_deployments`

**ARUs retrieved**: `r retrieved_arus`

**Cells surveyed**: `r cells_surveyed`

**Last updated**: `r updated`

<br>
<br>
<br>
<br>
<br>
<br>
<br>

For questions, please contact Jay Winiarski, Ecological Data Analyst/Manager (jwiniarski@wisc.edu)

Column {data-width=600}
-----------------------------------------------------------------------

### Current ARU deployments

```{r}
m1 <-
  cb_get_spatial('sierra_study_area') |>
  mapview(
    layer.name = 'Study area',
    alpha.regions = 0,
    lwd = 2,
    map.types = c('CartoDB.Positron', 'OpenStreetMap', 'OpenTopoMap', 'Esri.WorldImagery'),
    legend = FALSE
  )

m2 <-
  deployments_sf |>
  st_transform(3310) |> 
  st_jitter(amount = 200) |> 
  mapview(layer.name = 'ARU status', zcol = 'status', col.regions = c('#0072B2', '#D55E00'))

m3 <-
  cb_get_spatial('usfs_boundaries') |>
  mutate(
    frst_nm = as.factor(str_remove(frst_nm, ' National Forest')),
    frst_nm = fct_relevel(frst_nm, c('Lassen', 'Plumas', 'Tahoe', 'Eldorado', 'Stanislaus', 'Sierra', 'Sequoia'))
  ) |> 
  mapview(zcol = 'frst_nm', layer.name = 'National Forest')

m4 <-
  hexes_sf |> 
  mapview(layer.name = 'Hexes', col.regions = NA, legend = FALSE)

# make rendered html self-contained
(deployment_map <- m1 + m3 + m4 + m2)
```


Column {data-width=400}
-----------------------------------------------------------------------

### ARU deployments per National Forest

```{r}
p <-
  deployments_sf |> 
  st_drop_geometry() |> 
  select(ownership) |> 
  drop_na() |> 
  group_by(ownership) |> 
  tally() |> 
  mutate(
    ownership = as.factor(str_remove(ownership, ' National Forest')),
    ownership = fct_relevel(ownership, c('Lassen', 'Plumas', 'Tahoe', 'Eldorado', 'Stanislaus', 'Sierra', 'Sequoia'))
  ) |> 
  ggplot() +
  geom_col(aes(ownership, n, fill = ownership)) +
  scale_fill_viridis_d(name = 'National Forest') +
  theme_light() +
  theme(legend.position = 'none', panel.grid.minor = element_blank()) +
  labs(y = 'Number of ARUs', x = NULL)
ggplotly(p)
```

### Cells surveyed per National Forest

```{r}
p <-
  deployments_sf |> 
  st_drop_geometry() |> 
  select(cell_id, ownership) |> 
  drop_na() |> 
  distinct() |> 
  group_by(ownership) |> 
  tally() |> 
  mutate(
    ownership = as.factor(str_remove(ownership, ' National Forest')),
    ownership = fct_relevel(ownership, c('Lassen', 'Plumas', 'Tahoe', 'Eldorado', 'Stanislaus', 'Sierra', 'Sequoia'))
  ) |> 
  ggplot() +
  geom_col(aes(ownership, n, fill = ownership)) +
  scale_fill_viridis_d(name = 'National Forest') +
  theme_light() +
  theme(legend.position = 'none', panel.grid.minor = element_blank()) +
  labs(y = 'Number of cells', x = NULL)
ggplotly(p)
```

<!-- ### ARU deployments {.value-box} -->

<!-- ```{r} -->
<!-- valueBox( -->
<!--   value = nrow(deployments_sf), -->
<!--   icon = "fa-microphone" -->
<!-- ) -->
<!-- ``` -->

<!-- ### Cells surveyed {.value-box} -->

<!-- ```{r} -->
<!-- valueBox( -->
<!--   value = length(unique(deployments_sf$cell_id)), -->
<!--   icon = "glyphicon-check" -->
<!-- ) -->
<!-- ``` -->

<!-- ### Last updated {.value-box} -->

<!-- ```{r} -->
<!-- valueBox( -->
<!--   value = lubridate::today(), -->
<!--   icon = "fa-calendar" -->
<!-- ) -->
<!-- ``` -->