From b6d18a1fa85b0df89efdb4ad10448261882ca99e Mon Sep 17 00:00:00 2001
From: ayushman1210 <ayushmangupta892@gmail.com>
Date: Tue, 30 Jun 2026 14:58:16 +0530
Subject: [PATCH 1/5] feat: implement automated reporting and visualization
 layer

---
 .../benchmark/R/generate_validation_report.R  | 62 +++++++++++++++
 modules/benchmark/R/metric_residual_plot.R    | 28 +++++--
 modules/benchmark/R/metric_scatter_plot.R     | 19 +++--
 modules/benchmark/R/metric_timeseries_plot.R  | 39 ++++++----
 .../inst/reports/Validation_report.qmd        | 75 +++++++++++++++++++
 .../tests/testthat/test-visualization.R       | 53 +++++++++++++
 6 files changed, 245 insertions(+), 31 deletions(-)
 create mode 100644 modules/benchmark/R/generate_validation_report.R
 create mode 100644 modules/benchmark/inst/reports/Validation_report.qmd
 create mode 100644 modules/benchmark/tests/testthat/test-visualization.R

diff --git a/modules/benchmark/R/generate_validation_report.R b/modules/benchmark/R/generate_validation_report.R
new file mode 100644
index 00000000000..6399cbcc16f
--- /dev/null
+++ b/modules/benchmark/R/generate_validation_report.R
@@ -0,0 +1,62 @@
+##' Generate Validation Benchmark Report
+##'
+##' @param benchmark_results A list containing `metrics` (data.frame), `aligned_data` (data.frame), and `plots` (list of ggplot objects) returned by the validation pipeline.
+##' @param output_file The path where the compiled report should be saved (e.g., "validation_report.html").
+##' @param template The path to the Quarto template. Defaults to the one provided in the package `inst/reports/Validation_report.qmd`.
+##' 
+##' @author PEcAn Project
+##' @export
+generate_validation_report <- function(benchmark_results, output_file = "Validation_report.html", template = NULL) {
+  PEcAn.logger::logger.info("Generating Validation Benchmark Report...")
+  
+  if (is.null(template)) {
+    template <- system.file("reports", "Validation_report.qmd", package = "PEcAn.benchmark")
+    if (template == "") {
+      # Fallback for development mode
+      template <- file.path(getwd(), "inst", "reports", "Validation_report.qmd")
+    }
+  }
+  
+  if (!file.exists(template)) {
+    PEcAn.logger::logger.severe("Template file not found:", template)
+    stop("Quarto template not found.")
+  }
+  
+  if (!requireNamespace("quarto", quietly = TRUE)) {
+    PEcAn.logger::logger.severe("The 'quarto' package is required to generate the report.")
+    stop("Please install the 'quarto' R package.")
+  }
+  
+  # Ensure absolute paths
+  output_file <- normalizePath(output_file, mustWork = FALSE)
+  output_dir <- dirname(output_file)
+  
+  if (!dir.exists(output_dir)) {
+    dir.create(output_dir, recursive = TRUE)
+  }
+  
+  # Copy template to output directory to avoid permission issues in system folders
+  temp_qmd <- file.path(output_dir, basename(template))
+  file.copy(template, temp_qmd, overwrite = TRUE)
+  
+  # Render the document
+  tryCatch({
+    quarto::quarto_render(
+      input = temp_qmd,
+      output_file = basename(output_file),
+      execute_params = list(benchmark_results = benchmark_results)
+    )
+    
+    PEcAn.logger::logger.info("Validation report successfully generated at:", output_file)
+  }, error = function(e) {
+    PEcAn.logger::logger.severe("Failed to render validation report:", e$message)
+    stop(e)
+  }, finally = {
+    # Clean up the temporary template file
+    if (file.exists(temp_qmd)) {
+      file.remove(temp_qmd)
+    }
+  })
+  
+  return(invisible(output_file))
+}
diff --git a/modules/benchmark/R/metric_residual_plot.R b/modules/benchmark/R/metric_residual_plot.R
index 38a20a65125..4678d29903b 100644
--- a/modules/benchmark/R/metric_residual_plot.R
+++ b/modules/benchmark/R/metric_residual_plot.R
@@ -10,22 +10,34 @@
 metric_residual_plot <- function(metric_dat, var, filename = NA, draw.plot = is.na(filename)) {
   PEcAn.logger::logger.info("Metric: Residual Plot")
   
-  metric_dat$time <- lubridate::year(as.Date(as.character(metric_dat$time), format = "%Y"))
-  metric_dat$diff <- abs(metric_dat$model - metric_dat$obvs)
-  metric_dat$zeros <- rep(0, length(metric_dat$time))
+  metric_dat <- as.data.frame(metric_dat)
   
-  p <- ggplot2::ggplot(data = metric_dat, ggplot2::aes(x = .data$time)) 
-  p <- p + ggplot2::geom_path(ggplot2::aes(y = .data$zeros), colour = "#666666", size = 2, linetype = 2, lineend = "round") 
-  p <- p + ggplot2::geom_point(ggplot2::aes(y = .data$diff), size = 4, colour = "#619CFF") 
-  p <- p + ggplot2::labs(title = var, x = "years", y = "abs(model - observation)")
+  if (!"time" %in% colnames(metric_dat)) {
+    metric_dat$time <- seq_len(nrow(metric_dat))
+  } else {
+    date.time <- try(as.Date(as.character(metric_dat$time)), silent = TRUE)
+    if (!inherits(date.time, "try-error") && !all(is.na(date.time))) {
+      metric_dat$time <- date.time
+    }
+  }
+  
+  # Calculate residuals (Model - Observation)
+  metric_dat$diff <- metric_dat$model - metric_dat$obvs
+  
+  p <- ggplot2::ggplot(data = metric_dat, ggplot2::aes(x = .data$time, y = .data$diff)) +
+    ggplot2::geom_hline(yintercept = 0, colour = "#666666", linewidth = 1, linetype = 2) +
+    ggplot2::geom_point(size = 2, alpha = 0.7, colour = "#619CFF") +
+    ggplot2::labs(title = var, x = "Time", y = "Residual (Model - Obs)") +
+    ggplot2::theme_minimal(base_size = 14)
   
   if (!is.na(filename)) {
     grDevices::pdf(filename, width = 10, height = 6)
-    plot(p)
+    print(p)
     grDevices::dev.off()
   }
   
   if (draw.plot) {
     return(p)
   }
+  invisible(p)
 } # metric_residual_plot
\ No newline at end of file
diff --git a/modules/benchmark/R/metric_scatter_plot.R b/modules/benchmark/R/metric_scatter_plot.R
index 751d08e53b4..82e63b1903d 100644
--- a/modules/benchmark/R/metric_scatter_plot.R
+++ b/modules/benchmark/R/metric_scatter_plot.R
@@ -1,29 +1,32 @@
 ##' Scatter Plot
 ##'
 ##' @param metric_dat dataframe to plot, with at least columns `model` and `obvs`
-##' @param var ignored
+##' @param var title for the plot
 ##' @param filename path to save plot, or NA to not save
 ##' @param draw.plot logical: Return the plot object?
 ##' 
 ##' @author Betsy Cowdery
 ##' @export
-
 metric_scatter_plot <- function(metric_dat, var, filename = NA, draw.plot = is.na(filename)) {
   PEcAn.logger::logger.info("Metric: Scatter Plot")
   
-  p <- ggplot2::ggplot(data = metric_dat) 
-  p <- p + ggplot2::geom_point(ggplot2::aes(x = .data$model, y = .data$obvs), size = 4) 
-  p <- p + ggplot2::geom_abline(slope = 1, intercept = 0, colour = "#666666", 
-                       size = 2, linetype = 2)
+  metric_dat <- as.data.frame(metric_dat)
+  
+  p <- ggplot2::ggplot(data = metric_dat, ggplot2::aes(x = .data$model, y = .data$obvs)) +
+    ggplot2::geom_point(size = 2, alpha = 0.7, colour = "#619CFF") +
+    ggplot2::geom_abline(slope = 1, intercept = 0, colour = "#666666", 
+                         linewidth = 1, linetype = 2) +
+    ggplot2::labs(title = var, x = "Modeled", y = "Observed") +
+    ggplot2::theme_minimal(base_size = 14)
   
   if (!is.na(filename)) {
     grDevices::pdf(filename, width = 10, height = 6)
-    plot(p)
+    print(p)
     grDevices::dev.off()
   }
   
   if (draw.plot) {
     return(p)
   }
-
+  invisible(p)
 } # metric_scatter_plot
diff --git a/modules/benchmark/R/metric_timeseries_plot.R b/modules/benchmark/R/metric_timeseries_plot.R
index ca879b95648..93049f8e8f5 100644
--- a/modules/benchmark/R/metric_timeseries_plot.R
+++ b/modules/benchmark/R/metric_timeseries_plot.R
@@ -6,34 +6,43 @@
 ##' @param draw.plot logical: Return the plot object?
 ##'
 ##' @author Betsy Cowdery
-##' @importFrom ggplot2 ggplot labs geom_path geom_point
 ##' @export
-
 metric_timeseries_plot <- function(metric_dat, var, filename = NA, draw.plot = is.na(filename)) {
   PEcAn.logger::logger.info("Metric: Timeseries Plot")
   
-  # Attempt at getting around the fact that time can be annual and thus as.Date won't work
-  date.time <- try(as.Date(metric_dat$time), silent = TRUE)
-  if (inherits(date.time, "try-error")) {
-    PEcAn.logger::logger.warn("Can't coerce time column to Date format, attempting plot anyway")
-  }else{
-    metric_dat$time <- date.time
+  # Ensure metric_dat is a data.frame for ggplot2
+  metric_dat <- as.data.frame(metric_dat)
+  
+  if (!"time" %in% colnames(metric_dat)) {
+    PEcAn.logger::logger.warn("Missing 'time' column in metric_dat, using row index instead.")
+    metric_dat$time <- seq_len(nrow(metric_dat))
+  } else {
+    date.time <- try(as.Date(as.character(metric_dat$time)), silent = TRUE)
+    if (!inherits(date.time, "try-error") && !all(is.na(date.time))) {
+      metric_dat$time <- date.time
+    } else {
+      PEcAn.logger::logger.warn("Can't coerce time column to Date format, using original format.")
+    }
   }
   
-  p <- ggplot(data = metric_dat, ggplot2::aes(x = .data$time)) 
-  p <- p + labs(title = var, y = "") 
-  p <- p + geom_path(ggplot2::aes(y = .data$model, colour = "Model"), size = 2) 
-  p <- p + geom_point(ggplot2::aes(y = .data$model, colour = "Model"), size = 4) 
-  p <- p + geom_path(ggplot2::aes(y = .data$obvs, colour = "Observed"), size = 2) 
-  p <- p + geom_point(ggplot2::aes(y = .data$obvs, colour = "Observed"), size = 4)
+  p <- ggplot2::ggplot(data = metric_dat, ggplot2::aes(x = .data$time)) +
+    ggplot2::geom_line(ggplot2::aes(y = .data$model, colour = "Model"), linewidth = 1) +
+    ggplot2::geom_point(ggplot2::aes(y = .data$model, colour = "Model"), size = 2, alpha = 0.7) +
+    ggplot2::geom_line(ggplot2::aes(y = .data$obvs, colour = "Observed"), linewidth = 1) +
+    ggplot2::geom_point(ggplot2::aes(y = .data$obvs, colour = "Observed"), size = 2, alpha = 0.7) +
+    ggplot2::scale_colour_manual(values = c("Model" = "#619CFF", "Observed" = "#F8766D")) +
+    ggplot2::labs(title = var, x = "Time", y = "Value", color = "Source") +
+    ggplot2::theme_minimal(base_size = 14) +
+    ggplot2::theme(legend.position = "bottom")
   
   if (!is.na(filename)) {
     grDevices::pdf(filename, width = 10, height = 6)
-    plot(p)
+    print(p)
     grDevices::dev.off()
   }
   
   if (draw.plot) {
     return(p)
   }
+  invisible(p)
 } # metric_timeseries_plot
diff --git a/modules/benchmark/inst/reports/Validation_report.qmd b/modules/benchmark/inst/reports/Validation_report.qmd
new file mode 100644
index 00000000000..d8b08067185
--- /dev/null
+++ b/modules/benchmark/inst/reports/Validation_report.qmd
@@ -0,0 +1,75 @@
+---
+title: "PEcAn Validation Benchmark Report"
+author: "PEcAn Validation Toolkit"
+format:
+  html:
+    theme: cosmo
+    toc: true
+    toc-depth: 3
+    embed-resources: true
+params:
+  benchmark_results: null
+---
+
+```{r setup, include=FALSE}
+knitr::opts_chunk$set(echo = FALSE, warning = FALSE, message = FALSE)
+library(ggplot2)
+```
+
+## Executive Summary
+
+The following scorecard summarizes the validation metrics computed during the benchmark run.
+
+```{r scorecard}
+results <- params$benchmark_results
+
+if (!is.null(results) && !is.null(results$metrics)) {
+  # Check if knitr is available to use kable
+  if (requireNamespace("knitr", quietly = TRUE)) {
+    knitr::kable(results$metrics, digits = 3)
+  } else {
+    print(results$metrics)
+  }
+} else {
+  cat("No metrics data provided.\n")
+}
+```
+
+## Visual Diagnostics
+
+```{r plots, results='asis', fig.width=10, fig.height=6}
+if (!is.null(results) && !is.null(results$plots)) {
+  plots <- results$plots
+  
+  # if plots is a flat list of ggplot objects
+  if (inherits(plots[[1]], "ggplot")) {
+    for (p_name in names(plots)) {
+      cat(sprintf("\n### %s\n\n", p_name))
+      print(plots[[p_name]])
+      cat("\n\n")
+    }
+  } else {
+    # If plots are nested by variable: plots[[var_name]]$timeseries
+    for (var in names(plots)) {
+      cat(sprintf("\n### Variable: %s\n\n", var))
+      
+      var_plots <- plots[[var]]
+      
+      if (!is.null(var_plots$timeseries)) {
+        print(var_plots$timeseries)
+        cat("\n\n")
+      }
+      if (!is.null(var_plots$scatter)) {
+        print(var_plots$scatter)
+        cat("\n\n")
+      }
+      if (!is.null(var_plots$residual)) {
+        print(var_plots$residual)
+        cat("\n\n")
+      }
+    }
+  }
+} else {
+  cat("No plots provided.\n")
+}
+```
diff --git a/modules/benchmark/tests/testthat/test-visualization.R b/modules/benchmark/tests/testthat/test-visualization.R
new file mode 100644
index 00000000000..b8fac28f3b2
--- /dev/null
+++ b/modules/benchmark/tests/testthat/test-visualization.R
@@ -0,0 +1,53 @@
+test_that("metric_timeseries_plot generates a ggplot", {
+  mock_data <- data.frame(
+    time = as.Date("2020-01-01") + 0:9,
+    model = runif(10, 10, 20),
+    obvs = runif(10, 10, 20)
+  )
+  
+  p <- metric_timeseries_plot(mock_data, var = "Test Variable", draw.plot = TRUE)
+  
+  expect_true(inherits(p, "ggplot"))
+})
+
+test_that("metric_scatter_plot generates a ggplot", {
+  mock_data <- data.frame(
+    model = runif(10, 10, 20),
+    obvs = runif(10, 10, 20)
+  )
+  
+  p <- metric_scatter_plot(mock_data, var = "Test Variable", draw.plot = TRUE)
+  
+  expect_true(inherits(p, "ggplot"))
+})
+
+test_that("metric_residual_plot generates a ggplot", {
+  mock_data <- data.frame(
+    time = as.Date("2020-01-01") + 0:9,
+    model = runif(10, 10, 20),
+    obvs = runif(10, 10, 20)
+  )
+  
+  p <- metric_residual_plot(mock_data, var = "Test Variable", draw.plot = TRUE)
+  
+  expect_true(inherits(p, "ggplot"))
+})
+
+test_that("generate_validation_report fails gracefully if no template", {
+  # Mock benchmark results
+  mock_results <- list(
+    metrics = data.frame(Metric = "RMSE", Value = 1.2),
+    plots = list(
+      "Var1" = list(
+        timeseries = metric_timeseries_plot(
+          data.frame(time = 1:5, model = 1:5, obvs = 1:5), "Var1", draw.plot = TRUE
+        )
+      )
+    )
+  )
+  
+  expect_error(
+    generate_validation_report(mock_results, template = "non_existent.qmd"),
+    "Quarto template not found"
+  )
+})

From 1f2609fbfc7b58b2bd28e85553c294a054b56e20 Mon Sep 17 00:00:00 2001
From: ayushman1210 <ayushmangupta892@gmail.com>
Date: Fri, 3 Jul 2026 10:53:43 +0530
Subject: [PATCH 2/5] Fix: Serialize benchmark_results to RDS to prevent Quarto
 YAML parsing crash

---
 modules/benchmark/R/generate_validation_report.R     | 7 ++++++-
 modules/benchmark/inst/reports/Validation_report.qmd | 7 ++++++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/modules/benchmark/R/generate_validation_report.R b/modules/benchmark/R/generate_validation_report.R
index 6399cbcc16f..838722846ce 100644
--- a/modules/benchmark/R/generate_validation_report.R
+++ b/modules/benchmark/R/generate_validation_report.R
@@ -39,12 +39,17 @@ generate_validation_report <- function(benchmark_results, output_file = "Validat
   temp_qmd <- file.path(output_dir, basename(template))
   file.copy(template, temp_qmd, overwrite = TRUE)
   
+  # Quarto execute_params are converted to YAML. Complex R objects like ggplots
+  # cannot be passed via YAML. We must save them to an RDS and pass the path.
+  results_rds <- file.path(output_dir, "benchmark_results.rds")
+  saveRDS(benchmark_results, results_rds)
+  
   # Render the document
   tryCatch({
     quarto::quarto_render(
       input = temp_qmd,
       output_file = basename(output_file),
-      execute_params = list(benchmark_results = benchmark_results)
+      execute_params = list(benchmark_results = results_rds)
     )
     
     PEcAn.logger::logger.info("Validation report successfully generated at:", output_file)
diff --git a/modules/benchmark/inst/reports/Validation_report.qmd b/modules/benchmark/inst/reports/Validation_report.qmd
index d8b08067185..36dc039d1cc 100644
--- a/modules/benchmark/inst/reports/Validation_report.qmd
+++ b/modules/benchmark/inst/reports/Validation_report.qmd
@@ -21,7 +21,12 @@ library(ggplot2)
 The following scorecard summarizes the validation metrics computed during the benchmark run.
 
 ```{r scorecard}
-results <- params$benchmark_results
+# Quarto execute_params passes strings. If it's a file path, load the RDS.
+if (is.character(params$benchmark_results) && file.exists(params$benchmark_results)) {
+  results <- readRDS(params$benchmark_results)
+} else {
+  results <- params$benchmark_results
+}
 
 if (!is.null(results) && !is.null(results$metrics)) {
   # Check if knitr is available to use kable

From 4da8557455b5153c5f9e0df2011b76d20148abfc Mon Sep 17 00:00:00 2001
From: ayushman1210 <ayushmangupta892@gmail.com>
Date: Mon, 6 Jul 2026 14:09:51 +0530
Subject: [PATCH 3/5] Fix CI: Add quarto dependency, update tests, and sync
 Roxygen docs

---
 modules/benchmark/DESCRIPTION                 |  1 +
 modules/benchmark/NAMESPACE                   |  5 +---
 .../man/generate_validation_report.Rd         | 25 +++++++++++++++++++
 modules/benchmark/man/metric_scatter_plot.Rd  |  2 +-
 .../tests/testthat/test-visualization.R       |  2 +-
 5 files changed, 29 insertions(+), 6 deletions(-)
 create mode 100644 modules/benchmark/man/generate_validation_report.Rd

diff --git a/modules/benchmark/DESCRIPTION b/modules/benchmark/DESCRIPTION
index f8ada65d1d7..7606f26d1dd 100644
--- a/modules/benchmark/DESCRIPTION
+++ b/modules/benchmark/DESCRIPTION
@@ -46,6 +46,7 @@ Imports:
     zoo,
     yaml
 Suggests:
+    quarto,
     PEcAn.data.land,
     testthat (>= 2.0.0)
 License: BSD_3_clause + file LICENSE
diff --git a/modules/benchmark/NAMESPACE b/modules/benchmark/NAMESPACE
index 07067c6c4c0..24c97d902b2 100644
--- a/modules/benchmark/NAMESPACE
+++ b/modules/benchmark/NAMESPACE
@@ -13,6 +13,7 @@ export(clean_settings_BRR)
 export(create_BRR)
 export(define_benchmark)
 export(format_wide2long)
+export(generate_validation_report)
 export(load_and_map_data)
 export(load_csv)
 export(load_data)
@@ -42,10 +43,6 @@ export(read_settings_BRR)
 export(register_metric)
 export(run_benchmark)
 importFrom(dplyr,rename)
-importFrom(ggplot2,geom_path)
-importFrom(ggplot2,geom_point)
-importFrom(ggplot2,ggplot)
-importFrom(ggplot2,labs)
 importFrom(magrittr,"%>%")
 importFrom(rlang,.data)
 importFrom(yaml,read_yaml)
diff --git a/modules/benchmark/man/generate_validation_report.Rd b/modules/benchmark/man/generate_validation_report.Rd
new file mode 100644
index 00000000000..d7a8b767cf5
--- /dev/null
+++ b/modules/benchmark/man/generate_validation_report.Rd
@@ -0,0 +1,25 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/generate_validation_report.R
+\name{generate_validation_report}
+\alias{generate_validation_report}
+\title{Generate Validation Benchmark Report}
+\usage{
+generate_validation_report(
+  benchmark_results,
+  output_file = "Validation_report.html",
+  template = NULL
+)
+}
+\arguments{
+\item{benchmark_results}{A list containing `metrics` (data.frame), `aligned_data` (data.frame), and `plots` (list of ggplot objects) returned by the validation pipeline.}
+
+\item{output_file}{The path where the compiled report should be saved (e.g., "validation_report.html").}
+
+\item{template}{The path to the Quarto template. Defaults to the one provided in the package `inst/reports/Validation_report.qmd`.}
+}
+\description{
+Generate Validation Benchmark Report
+}
+\author{
+PEcAn Project
+}
diff --git a/modules/benchmark/man/metric_scatter_plot.Rd b/modules/benchmark/man/metric_scatter_plot.Rd
index 93da2a3db5e..cc1b94749bf 100644
--- a/modules/benchmark/man/metric_scatter_plot.Rd
+++ b/modules/benchmark/man/metric_scatter_plot.Rd
@@ -14,7 +14,7 @@ metric_scatter_plot(
 \arguments{
 \item{metric_dat}{dataframe to plot, with at least columns `model` and `obvs`}
 
-\item{var}{ignored}
+\item{var}{title for the plot}
 
 \item{filename}{path to save plot, or NA to not save}
 
diff --git a/modules/benchmark/tests/testthat/test-visualization.R b/modules/benchmark/tests/testthat/test-visualization.R
index b8fac28f3b2..e738bb7dd51 100644
--- a/modules/benchmark/tests/testthat/test-visualization.R
+++ b/modules/benchmark/tests/testthat/test-visualization.R
@@ -48,6 +48,6 @@ test_that("generate_validation_report fails gracefully if no template", {
   
   expect_error(
     generate_validation_report(mock_results, template = "non_existent.qmd"),
-    "Quarto template not found"
+    "Template file not found: non_existent.qmd"
   )
 })

From 8efa2108faa7eff39f81dbd9b1a429c5fdf0cb13 Mon Sep 17 00:00:00 2001
From: ayushman1210 <ayushmangupta892@gmail.com>
Date: Mon, 6 Jul 2026 14:24:35 +0530
Subject: [PATCH 4/5] Feat: Add uncertainty ribbon and error bars to timeseries
 plot

---
 modules/benchmark/R/metric_timeseries_plot.R | 28 +++++--
 test_ameri.R                                 | 83 ++++++++++++++++++++
 2 files changed, 106 insertions(+), 5 deletions(-)
 create mode 100644 test_ameri.R

diff --git a/modules/benchmark/R/metric_timeseries_plot.R b/modules/benchmark/R/metric_timeseries_plot.R
index 93049f8e8f5..0e4966f8c56 100644
--- a/modules/benchmark/R/metric_timeseries_plot.R
+++ b/modules/benchmark/R/metric_timeseries_plot.R
@@ -25,12 +25,30 @@ metric_timeseries_plot <- function(metric_dat, var, filename = NA, draw.plot = i
     }
   }
   
-  p <- ggplot2::ggplot(data = metric_dat, ggplot2::aes(x = .data$time)) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$model, colour = "Model"), linewidth = 1) +
-    ggplot2::geom_point(ggplot2::aes(y = .data$model, colour = "Model"), size = 2, alpha = 0.7) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$obvs, colour = "Observed"), linewidth = 1) +
-    ggplot2::geom_point(ggplot2::aes(y = .data$obvs, colour = "Observed"), size = 2, alpha = 0.7) +
+  p <- ggplot2::ggplot(data = metric_dat, ggplot2::aes(x = .data$time))
+  
+  # 1. Model Ribbon (if available)
+  if (all(c("model_q05", "model_q95") %in% colnames(metric_dat))) {
+    p <- p + ggplot2::geom_ribbon(ggplot2::aes(ymin = .data$model_q05, ymax = .data$model_q95, fill = "Model 90% CI"), alpha = 0.3)
+  }
+  
+  # 2. Model Mean Line
+  p <- p + ggplot2::geom_line(ggplot2::aes(y = .data$model, colour = "Model"), linewidth = 1)
+  
+  # 3. Observational Points & Error Bars
+  if ("obvs_sd" %in% colnames(metric_dat)) {
+    p <- p + ggplot2::geom_pointrange(
+      ggplot2::aes(y = .data$obvs, ymin = .data$obvs - .data$obvs_sd, ymax = .data$obvs + .data$obvs_sd, colour = "Observed"), 
+      size = 0.5, alpha = 0.7
+    )
+  } else {
+    p <- p + ggplot2::geom_point(ggplot2::aes(y = .data$obvs, colour = "Observed"), size = 2, alpha = 0.7)
+  }
+  
+  # Adjust Scales
+  p <- p +
     ggplot2::scale_colour_manual(values = c("Model" = "#619CFF", "Observed" = "#F8766D")) +
+    ggplot2::scale_fill_manual(values = c("Model 90% CI" = "#619CFF"), name = "Intervals") +
     ggplot2::labs(title = var, x = "Time", y = "Value", color = "Source") +
     ggplot2::theme_minimal(base_size = 14) +
     ggplot2::theme(legend.position = "bottom")
diff --git a/test_ameri.R b/test_ameri.R
new file mode 100644
index 00000000000..4d86ea66221
--- /dev/null
+++ b/test_ameri.R
@@ -0,0 +1,83 @@
+# --- DUMMY LOGGER TO BYPASS DEPENDENCIES ---
+PEcAn.logger <- new.env()
+PEcAn.logger$logger.info <- function(...) cat("INFO:", ..., "\n")
+PEcAn.logger$logger.warn <- function(...) cat("WARN:", ..., "\n")
+PEcAn.logger$logger.severe <- function(...) stop("SEVERE:", ...)
+`::` <- function(pkg, name) {
+  pkg_name <- as.character(substitute(pkg))
+  name_str <- as.character(substitute(name))
+  if (pkg_name == "PEcAn.logger") return(get(name_str, envir = PEcAn.logger))
+  
+  # Safely call original ::
+  orig_colon <- get("::", envir = asNamespace("base"))
+  eval(substitute(orig_colon(pkg, name)))
+}
+# -------------------------------------------
+
+# Source all R files in the benchmark module locally
+lapply(list.files("modules/benchmark/R", pattern="\\.R$", full.names=TRUE), source)
+
+# 1. LOAD YOUR CSV
+csv_path <- "/home/ayushman1210/Downloads/observations.csv" # Replace with your actual path!
+
+# Load the raw CSV
+raw_df <- read.csv(csv_path, check.names = FALSE)
+
+# Fix the unnamed 7th column (which holds the variable name)
+names(raw_df)[7] <- "variable"
+
+# 2. EXTRACT AND FORMAT THE OBSERVATIONS
+obvs_df <- raw_df[raw_df$variable == "NEE", ]
+obvs_df$time <- as.POSIXct(as.Date(obvs_df$max_date, format="%m/%d/%Y"))
+obvs_df$obvs <- as.numeric(obvs_df$value)
+obvs_df <- obvs_df[, c("time", "obvs")]
+
+# 3. CREATE A FAKE "MODEL" DATASET (For testing purposes today)
+set.seed(42)
+model_df <- obvs_df
+model_df$model <- model_df$obvs + rnorm(nrow(model_df), mean = 0, sd = 0.5)
+
+# 4. ALIGN THE DATA
+# Bypassing align_data to avoid lubridate requirement
+aligned_df <- data.frame(
+  time = obvs_df$time,
+  model = model_df$model,
+  obvs = obvs_df$obvs,
+  model_q05 = model_df$model - runif(nrow(model_df), 0.5, 1.5),
+  model_q95 = model_df$model + runif(nrow(model_df), 0.5, 1.5),
+  obvs_sd = runif(nrow(obvs_df), 0.2, 0.8)
+)
+
+# 5. CALCULATE METRICS
+metrics_list <- list(
+  RMSE = metric_RMSE(aligned_df),
+  R2   = metric_R2(aligned_df),
+  MAE  = metric_MAE(aligned_df)
+)
+
+metrics_df <- data.frame(
+  Metric = names(metrics_list),
+  Value = as.numeric(metrics_list)
+)
+
+# 6. GENERATE PLOTS
+p_timeseries <- metric_timeseries_plot(aligned_df, var = "NEE")
+p_scatter <- metric_scatter_plot(aligned_df, var = "NEE")
+p_residual <- metric_residual_plot(aligned_df, var = "NEE")
+
+# 7. BUNDLE AND RENDER REPORT
+benchmark_results <- list(
+  metrics = metrics_df,
+  aligned_data = aligned_df,
+  plots = list(
+    "NEE Timeseries" = p_timeseries,
+    "NEE Scatter" = p_scatter,
+    "NEE Residuals" = p_residual
+  )
+)
+
+generate_validation_report(
+  benchmark_results = benchmark_results,
+  output_file = "AmeriFlux_Validation_Report.html",
+  template = "modules/benchmark/inst/reports/Validation_report.qmd"
+)

From 9635b362f88fb7ad1757ebf4602371efd332b74c Mon Sep 17 00:00:00 2001
From: ayushman1210 <ayushmangupta892@gmail.com>
Date: Mon, 6 Jul 2026 14:28:19 +0530
Subject: [PATCH 5/5] chore: remove local scratch script from tracking

---
 test_ameri.R | 83 ----------------------------------------------------
 1 file changed, 83 deletions(-)
 delete mode 100644 test_ameri.R

diff --git a/test_ameri.R b/test_ameri.R
deleted file mode 100644
index 4d86ea66221..00000000000
--- a/test_ameri.R
+++ /dev/null
@@ -1,83 +0,0 @@
-# --- DUMMY LOGGER TO BYPASS DEPENDENCIES ---
-PEcAn.logger <- new.env()
-PEcAn.logger$logger.info <- function(...) cat("INFO:", ..., "\n")
-PEcAn.logger$logger.warn <- function(...) cat("WARN:", ..., "\n")
-PEcAn.logger$logger.severe <- function(...) stop("SEVERE:", ...)
-`::` <- function(pkg, name) {
-  pkg_name <- as.character(substitute(pkg))
-  name_str <- as.character(substitute(name))
-  if (pkg_name == "PEcAn.logger") return(get(name_str, envir = PEcAn.logger))
-  
-  # Safely call original ::
-  orig_colon <- get("::", envir = asNamespace("base"))
-  eval(substitute(orig_colon(pkg, name)))
-}
-# -------------------------------------------
-
-# Source all R files in the benchmark module locally
-lapply(list.files("modules/benchmark/R", pattern="\\.R$", full.names=TRUE), source)
-
-# 1. LOAD YOUR CSV
-csv_path <- "/home/ayushman1210/Downloads/observations.csv" # Replace with your actual path!
-
-# Load the raw CSV
-raw_df <- read.csv(csv_path, check.names = FALSE)
-
-# Fix the unnamed 7th column (which holds the variable name)
-names(raw_df)[7] <- "variable"
-
-# 2. EXTRACT AND FORMAT THE OBSERVATIONS
-obvs_df <- raw_df[raw_df$variable == "NEE", ]
-obvs_df$time <- as.POSIXct(as.Date(obvs_df$max_date, format="%m/%d/%Y"))
-obvs_df$obvs <- as.numeric(obvs_df$value)
-obvs_df <- obvs_df[, c("time", "obvs")]
-
-# 3. CREATE A FAKE "MODEL" DATASET (For testing purposes today)
-set.seed(42)
-model_df <- obvs_df
-model_df$model <- model_df$obvs + rnorm(nrow(model_df), mean = 0, sd = 0.5)
-
-# 4. ALIGN THE DATA
-# Bypassing align_data to avoid lubridate requirement
-aligned_df <- data.frame(
-  time = obvs_df$time,
-  model = model_df$model,
-  obvs = obvs_df$obvs,
-  model_q05 = model_df$model - runif(nrow(model_df), 0.5, 1.5),
-  model_q95 = model_df$model + runif(nrow(model_df), 0.5, 1.5),
-  obvs_sd = runif(nrow(obvs_df), 0.2, 0.8)
-)
-
-# 5. CALCULATE METRICS
-metrics_list <- list(
-  RMSE = metric_RMSE(aligned_df),
-  R2   = metric_R2(aligned_df),
-  MAE  = metric_MAE(aligned_df)
-)
-
-metrics_df <- data.frame(
-  Metric = names(metrics_list),
-  Value = as.numeric(metrics_list)
-)
-
-# 6. GENERATE PLOTS
-p_timeseries <- metric_timeseries_plot(aligned_df, var = "NEE")
-p_scatter <- metric_scatter_plot(aligned_df, var = "NEE")
-p_residual <- metric_residual_plot(aligned_df, var = "NEE")
-
-# 7. BUNDLE AND RENDER REPORT
-benchmark_results <- list(
-  metrics = metrics_df,
-  aligned_data = aligned_df,
-  plots = list(
-    "NEE Timeseries" = p_timeseries,
-    "NEE Scatter" = p_scatter,
-    "NEE Residuals" = p_residual
-  )
-)
-
-generate_validation_report(
-  benchmark_results = benchmark_results,
-  output_file = "AmeriFlux_Validation_Report.html",
-  template = "modules/benchmark/inst/reports/Validation_report.qmd"
-)