finish capitalization file refactoring

.gitignore

@@ -46,6 +46,6 @@ global_vars.R
 
 .DS_Store
 
-*/data/*.rda
-*/data/*.log
+*/*/data/*.rda
+*/*/data/*.log
 

2024/State_AQC_Tables/Makefile

@@ -0,0 +1,8 @@
+################################################################################
+# Makefile for publishing html content
+#
+
+publish:
+	- mkdir -p ../../docs/2024/state_AQC_Tables
+	mv Rmd/*.html ../../docs/2024/state_AQC_Tables
+

2024/State_AQC_Tables/R/02_prepare_monitoring_data.R

@@ -0,0 +1,60 @@
+# Step 2) Save annual AirNow files locally
+#
+# Download annual files in advance
+
+# ----- Setup ------------------------------------------------------------------
+
+VERSION = "1.0.0"
+
+# Check that the working directory is set properly
+if ( !stringr::str_detect(getwd(), "state_AQC_Tables$") ) {
+  stop("WD_ERROR:  Please set the working directory to 'state_AQC_Tables/'")
+}
+
+if ( packageVersion("AirMonitor") < "0.4.0" ) {
+  stop("VERSION_ERROR:  Please upgrade to AirMonitor 0.4.0 or later.")
+}
+
+library(MazamaCoreUtils)
+library(AirMonitor)
+
+# Use MazamaCoreUtils logging functionality to save logs in the data/ directory
+MazamaCoreUtils::initializeLogging("./data")
+logger.setLevel(TRACE)
+
+logger.info("Running prepare_monitoring_data.R version %s", VERSION)
+
+# Log session info to see package versions
+logger.debug(capture.output(sessionInfo()))
+
+# ----- Process data -----------------------------------------------------------
+
+# All processed data will live in the data/ directory.
+
+# Configurable parameters
+years <- 2014:2023
+
+for ( year in years ) {
+
+  logger.info("\n===== %d =====", year)
+
+  # * AirNow -------------------------------------------------------------------
+
+  logger.info("Processing AirNow %d data...", year)
+
+  # Create an AirNow monitor object 
+  monitor <-
+    airnow_loadAnnual(year)
+
+  # Assign a unique object name so that we can load more than one at a time
+  objectName <- sprintf("airnow_%d", year)
+  assign(objectName, monitor)
+
+  # Save to the data/ directory
+  fileName <- sprintf("%s.rda", objectName)
+  filePath <- file.path("./data", fileName)
+  logger.info("Saving %s", filePath)
+  save(list = objectName, file = filePath)
+
+}
+

2024/State_AQC_Tables/R/03_render_AQC_tables.R

@@ -0,0 +1,30 @@
+# Step 3) Render the daily_AQC_tables.Rmd report for months and states
+#
+#
+# NOTE:  Data archives begin in 2014
+
+# ----- Setup ------------------------------------------------------------------
+
+# Check that the working directory is set properly
+if ( !stringr::str_detect(getwd(), "state_AQC_tables$") ) {
+  stop("WD_ERROR:  Please set the working directory to 'state_AQC_tables/'")
+}
+
+if ( packageVersion("AirMonitor") < "0.4.0" ) {
+  stop("VERSION_ERROR:  Please upgrade to AirMonitor 0.4.0 or later.")
+}
+
+# ----- Render daily_AQC_tables reports ----------------------------------------
+
+for ( year in 2014:2023 ) {
+
+  params <- list(year = year)
+
+  # This path is relative to the Rmd/ directory
+  htmlPath <- sprintf("./html/monitoring_data_%d.html", year)
+
+  rmarkdown::render(input = 'Rmd/monitoring_data.Rmd',
+                    params = params,
+                    output_file = htmlPath)
+
+}

2024/State_AQC_Tables/README.md

@@ -0,0 +1,14 @@
+# State Air Quality Category Tables
+
+_Last updated on February 26, 2024_
+
+_Disclaimer: Data presented here come from a variety of sources and do not 
+represent official EPA measurements. Please visit
+[AirNow](https://www.airnow.gov) for official measurements._
+
+---
+
+## Background
+
+
+

2024/State_AQC_Tables/Rmd/daily_AQC_tables.Rmd

@@ -0,0 +1,183 @@
+---
+title: "Air Quality Category Tables"
+output:
+  html_document: 
+    css: report_styles.css
+---
+
+Processed on `r paste(lubridate::now(),format(lubridate::now(), "%Z"))`.
+
+----
+
+```{r setup, include = FALSE}
+knitr::opts_chunk$set(echo = FALSE, error = FALSE)
+
+suppressPackageStartupMessages({
+  library(AirMonitor)
+})
+
+stateCode <- "GA"
+timezone <- "America/New_York"
+month <- 4
+
+monthName <- 
+  MazamaCoreUtils::parseDatetime(
+    sprintf("2024%02d15", month), 
+    timezone = "America/New_York"
+  ) %>% 
+  strftime("%B", tzone = timezone)
+
+stateName <-
+  MazamaSpatialUtils::US_stateCodeToName(stateCode)
+
+```
+
+_Disclaimer: Data presented here come from a variety of sources and do not 
+represent official EPA measurements. Please visit
+[AirNow](https://www.airnow.gov) for official measurements._
+
+----
+
+```{r state_name, results = "asis"}
+cat(sprintf("\n\n# %s in %s\n", monthName, stateName))
+```
+
+
+The following tables summarize the cumulative number of days air quality monitors spent in 
+each AQI category during the the month of `r monthName`. Tables include only
+permanent monitors.
+
+The 'Old' row is calculated using the 2023 PM~2.5~ NAAQS while the 'New' row is
+calculated using the 2024 PM~2.5~ NAAQS.
+
+
+```{r year_loop, results = "asis", fig.width = 10, fig.height = 3, fig.align = "left"}
+
+# NOTE:  No monitoring data in Georgia before 2016
+for ( year in 2016:2023 ) {
+  
+  startString <- sprintf("%d%02d01", year, month)
+  startdate <- MazamaCoreUtils::parseDatetime(startString, timezone = timezone)
+  enddate <- startdate + lubridate::ddays(40)
+  lubridate::day(enddate) <- 1
+  
+  # ----- Assemble data --------------------------------------------------------
+  
+  # NOTE:  These paths are relative to the Rmd/ directory
+  airnow <- get(load(sprintf("../data/airnow_%d.rda", year)))
+  
+  monitor <-
+    airnow %>%
+    monitor_filter(stateCode == "GA") %>%
+    monitor_filterDate(startdate, enddate) %>%
+    monitor_filter(deploymentType == "Permanent") %>%
+    monitor_dropEmpty()
+  
+  #cat(sprintf("\n\n## %d\n", year))
+  cat(sprintf("\n\n## %s %d -- %d permanent monitors\n", monthName, year, nrow(monitor$meta)))
+  
+  #cat(sprintf("\n\n**Air Quality Categories under Old (2023) and New (2024) NAAQS**\n\n"))
+  
+  # ----- Create table with rownames for printing ------------------------------
+  
+  oldAQCSummary <-
+    monitor %>%
+    monitor_dailyStatistic(
+      FUN = mean,
+      na.rm = TRUE,
+      minHours = 18,
+      dayBoundary = c("LST")
+    ) %>%
+    monitor_dropEmpty() %>%
+    monitor_toAQCTable(NAAQS = "PM2.5") %>%
+    colSums()
+  
+  newAQCSummary <-
+    monitor %>%
+    monitor_dailyStatistic(
+      FUN = mean,
+      na.rm = TRUE,
+      minHours = 18,
+      dayBoundary = c("LST")
+    ) %>%
+    monitor_dropEmpty() %>%
+    monitor_toAQCTable(NAAQS = "PM2.5_2024") %>%
+    colSums()
+  
+  dailyAQCTable <- 
+    dplyr::bind_rows(oldAQCSummary, newAQCSummary) %>%
+    as.data.frame()
+  
+  rownames(dailyAQCTable) <- c("Old", "New")
+  
+  # ----- Print table ----------------------------------------------------------
+  
+  print(knitr::kable(
+    dailyAQCTable,
+    format = "html", table.attr = "style='width:95%; font-size:1.5rem; margin: 20px 60px 0px 15px; line-height:2em'"
+  ))
+  
+  # ----- Create barplot -------------------------------------------------------
+  
+  # Reorder so that the horizontal barplot ends up with Old on top 
+  df_barplot <-
+    dplyr::bind_rows(newAQCSummary, oldAQCSummary) %>%
+    as.data.frame()
+  
+  rownames(df_barplot) <- c("New", "Old")
+    
+  par(mar = c(1,3,3,1) + 0.1)
+  
+  barplot(
+    as.matrix(t(df_barplot)),
+    col = c(US_AQI$colors_EPA, "gray90"),
+    horiz = TRUE,
+    las = 1,
+    ylim = c(0, 1.5),
+    width = 0.5,
+    axes = FALSE
+  )
+  
+  par(mar = c(5,4,4,2) + 0.1)
+  
+   
+  # title("Title goes here")
+  # 
+  legend(
+    "top",
+    legend = c(US_AQI$names_eng, "Missing"),
+    fill = c(US_AQI$colors_EPA, "gray90"),
+    horiz = TRUE,
+    cex = 0.8,
+    bty = 'n'
+  )
+  
+    
+  # # ----- Create ggplot barplot ------------------------------------------------
+  # 
+  # df <-
+  #   dailyAQCTable %>%
+  #   tibble::rownames_to_column(var = "NAAQS") %>% 
+  #   tidyr::pivot_longer(-"NAAQS", names_to = "AQC", values_to = "count")
+  # 
+  # # ----- Plot table -----------------------------------------------------------
+  # 
+  # gg <- 
+  #   ggplot(df, aes(x = NAAQS, y = count))+
+  #   geom_col(aes(fill = AQC), width = 0.7)
+  # 
+  # 
+  # gg <- 
+  #   ggplot(df, aes(x = count, y = NAAQS))+
+  #   geom_col(
+  #     orientation = "y",
+  #     aes(fill = AQC), 
+  #     width = 0.7
+  #   )
+  
+  
+  
+}
+```
+
+