---
title: "Applications"
output: 
  html_document:
    theme: flatly
    keep_md: yes
    number_sections: true
    highlighted: default 
    toc: yes
    toc_float:
      collapsed: no
      smooth_scroll: no
    toc_depth: 2
vignette: >
  %\VignetteIndexEntry{Applications withs StreamCatTools}
  %\VignetteEncoding{UTF-8}{inputenc}
  %\VignetteEngine{knitr::rmarkdown}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>", 
  warning = FALSE, 
  message = FALSE
)
```

### Plotting watershed data
In this example we access a single variable for the Calapooia River using `sc_get_data` function. We then use the `nhdplusTools` library to grab flowlines and watershed for the Calapooia, plot the selected StreamCat metric for the Calapooia River and show the watershed.
```{r wshd, results='hide'}
library(StreamCatTools)
start_comid = 23763517
nldi_feature <- list(featureSource = "comid", featureID = start_comid)

flowline_nldi <- nhdplusTools::navigate_nldi(nldi_feature, mode = "UT", data_source = "flowlines", distance=5000)

# get StreamCat metrics
df <- sc_get_data(metric='pctimp2011', aoi='cat', comid=flowline_nldi$UT_flowlines$nhdplus_comid)

# We can also pull out comids the following way:
# comids <- paste(as.integer(flowline_nldi$UT_flowlines$nhdplus_comid), collapse=",",sep="")
# df <- sc_get_data(metric='pctimp2011', aoi='cat', comid=comids)

flowline_nldi <- flowline_nldi$UT_flowlines
flowline_nldi$PCTIMP2011CAT <- df$pctimp2011cat[match(flowline_nldi$nhdplus_comid, df$comid)]

basin <- nhdplusTools::get_nldi_basin(nldi_feature = nldi_feature)
```

```{r wshd pt2}
library(mapview)
mapview::mapviewOptions(fgb=FALSE)
mapview::mapview(basin, alpha.regions=.08) + mapview::mapview(flowline_nldi, zcol = "PCTIMP2011CAT", legend = TRUE)
```

### Working with NARS data
In this example we demonstrate a data 'mashup' by grabbing NRSA data from the EPA National Aquatic Resource Surveys (NARS) website directly in R, pull particular StreamCat metrics for sites using `sc_get_data`, and compare landscape metrics with other NRSA metrics
```{r nars, results='hide'}
nrsa <- readr::read_csv("https://www.epa.gov/sites/production/files/2015-09/siteinfo_0.csv")

dplyr::glimpse(nrsa)

# Promote data frame to sf spatial points data frame
nrsa_sf <- sf::st_as_sf(nrsa, coords = c("LON_DD83", "LAT_DD83"), crs = 4269)

# Get COMIDs using nhdplusTools package
# nrsa$COMID<- NA
# for (i in 1:nrow(nrsa_sf)){
#   print (i)
#   nrsa_sf[i,'COMID'] <- discover_nhdplus_id(nrsa_sf[i,c('geometry')])
# }
load(system.file("extdata", "sample_nrsa_data.rda", package="StreamCatTools"))

# get particular StreamCat data for all these NRSA sites
# nrsa_sf$COMID <- as.character(nrsa_sf$COMID)
comids <- nrsa_sf$COMID
comids <- comids[!is.na(comids)]
comids <- comids[c(1:700)]
comids <- paste(comids,collapse=',')
df <- sc_get_data(metric='pctcrop2006', aoi='ws', comid=comids)

# glimpse(df)
df$COMID <- as.integer(df$comid)
nrsa_sf <- dplyr::left_join(nrsa_sf, df, by='COMID')
```

```{r nars_ggplot, warning=FALSE}
#| fig.alt: >
#|   NRSA Benthic MMI versus  % Crop in Watershed from 2006 NLCD.
# download mmi from NARS web page
library(dplyr)
library(ggplot2)
mmi <- readr::read_csv("https://www.epa.gov/sites/production/files/2015-09/bentcond.csv")
# dplyr::glimpse(mmi)

# join mmi to NARS info data frame with StreamCat PctCrop metric
nrsa_sf <- dplyr::left_join(nrsa_sf, mmi[,c('SITE_ID','BENT_MMI_COND')], by='SITE_ID')
bxplt <- nrsa_sf %>% 
  tidyr::drop_na(BENT_MMI_COND) %>%
  ggplot2::ggplot(aes(x=pctcrop2006ws, y=BENT_MMI_COND))+
  ggplot2::geom_boxplot()+
  ggplot2::ggtitle('NRSA Benthic MMI versus % Crop in Watershed from 2006 NLCD')
suppressWarnings(print(bxplt))
```