Using the Interactive Dashboard

Overview

The manureshed interactive dashboard provides a user-friendly interface for exploring nutrient balance data without writing R code. This is ideal for:

• Policy makers and decision makers
• Extension agents and educators
• Students learning about nutrient management
• Quick exploratory analysis
• Presentations and demonstrations

Installation Requirements

The dashboard requires several additional packages beyond the core manureshed package:

# Install dashboard dependencies
install.packages(c(
  "shiny",
  "shinydashboard",
  "leaflet",
  "plotly",
  "DT"
))

These packages are optional - manureshed works fine without them if you only need the programmatic interface. They’re only needed for the interactive dashboard.

Launching the Dashboard

Once the dependencies are installed, launch the dashboard with:

library(manureshed)

# Launch the dashboard
launch_dashboard()

The dashboard will open in your default web browser. By default, it uses a random available port. You can specify a port:

# Launch on specific port
launch_dashboard(port = 3838)

Dashboard Interface

Main Components

The dashboard has four main sections:

1. Sidebar: Analysis parameters and controls
2. Map Tab: Interactive spatial visualization
3. Statistics Tab: Summary metrics and plots
4. Data Table Tab: Detailed results table
5. Help Tab: Documentation and information

Sidebar Controls

Analysis Parameters

Spatial Scale: - County: ~3,000 units, good for policy analysis - HUC8 Watershed: ~2,000 units, good for watershed management - HUC2 Region: 18 units, good for regional overview

Year: - Slider from 2007-2016 - 2007-2016: WWTP data available - Earlier years: Agricultural data only

Nutrient: - Nitrogen - Phosphorus

Include Wastewater Treatment Plants: - Checked: Shows integrated agricultural + WWTP analysis - Unchecked: Shows agricultural-only analysis

Advanced Options (click “Show advanced options”): - Cropland Threshold: Minimum cropland area for inclusion (default: 1,234 acres)

Action Buttons

Run Analysis: - Click to execute the analysis with selected parameters - Shows progress indicator - Displays notification when complete

Download Data: - Exports results as CSV file - Includes all analysis columns

Using the Map Tab

Interactive Map Features

The map tab displays a color-coded interactive map of nutrient classifications:

Color Legend: - Red: Nutrient sources (surplus) - Blue: Sinks with deficit (no fertilizer use) - Purple: Sinks with deficit (despite fertilizer use) - Green: Within watershed/county (balanced with WWTP) - Light blue: Excluded (below cropland threshold)

Interaction: - Zoom: Scroll wheel or +/- buttons - Pan: Click and drag - Click: See detailed information popup - Hover: Highlight individual units

Popup Information: When you click a map area, you’ll see: - Unit ID (FIPS code, HUC8, or HUC2) - Classification - Surplus/deficit amount (kg) - Cropland area (acres)

Map Navigation Tips

Finding Specific Areas: 1. Zoom in to your region of interest 2. Use the search function in your browser (Ctrl+F / Cmd+F) 3. Click on areas to identify them

Comparing Regions: 1. Run analysis with WWTP included 2. Note colors and patterns 3. Uncheck WWTP box 4. Run analysis again 5. Compare how classifications change

Using the Statistics Tab

Value Boxes

Four key metrics displayed at the top:

1. Sources: Red box - nutrient surplus areas
2. Sink (Deficit): Blue box - deficit without fertilizer
3. Sink (Fertilizer): Purple box - deficit despite fertilizer
4. Within Watershed/County: Green box - balanced (WWTP included)
5. Excluded: Light blue box - below threshold

Classification Distribution

Pie Chart: - Shows proportions of each classification - Interactive: Hover to see exact counts - Click legend items to show/hide categories

Surplus/Deficit Distribution

Histogram: - Shows distribution of surplus (positive) and deficit (negative) values - Interactive: Hover to see bin counts - Zoom by clicking and dragging

Summary Statistics

Text output showing: - Analysis parameters used - Total units analyzed - Classification counts - Statistical summary of surplus/deficit values

Using the Data Table Tab

Interactive Table Features

The data table provides detailed results with:

Search: - Global search box (top right) - Column-specific search (top of each column)

Sorting: - Click column headers to sort - Click again to reverse sort - Multi-column sort: Shift+click

Filtering: - Use column search boxes - Type partial matches - Filter by multiple columns

Pagination: - Choose rows per page (10, 25, 50, 100) - Navigate between pages

Useful Filters

Find All Sources:

In classification column filter: type "Source"

Find High Surplus Areas:

Sort by surplus column (descending)
Look at top entries

Find Specific State:

In ID column filter (for county scale): type state FIPS code
Example: "19" for Iowa

Help Tab

The Help tab provides: - About manureshed information - How to use instructions - Classification definitions - Package version - Link to documentation

Workflow Examples

Example 1: Basic State Analysis

Goal: Analyze Iowa watersheds in 2016

Steps: 1. Launch dashboard: launch_dashboard() 2. Set Scale: HUC8 Watershed 3. Set Year: 2016 4. Set Nutrient: Nitrogen 5. Check: Include Wastewater Treatment Plants 6. Click: Run Analysis 7. Wait for completion notification 8. Explore results in Map tab 9. Click Download Data to save results

Example 2: Compare Agricultural vs. Integrated

Goal: See how WWTP changes classifications

Steps: 1. Set parameters (e.g., HUC8, 2016, Nitrogen) 2. Check: Include WWTP 3. Click: Run Analysis 4. Note the Statistics tab value boxes 5. Uncheck: Include WWTP 6. Observe how value boxes change 7. Compare the differences

What to Look For: - How many sources remain? - How many sinks become “Within Watershed”? - How does total surplus/deficit change?

Example 3: Year-over-Year Exploration

Goal: Explore temporal trends

Steps: 1. Set Scale: County 2. Set Nutrient: Nitrogen 3. Set Year: 2010 4. Check: Include WWTP 5. Run Analysis 6. Note key statistics 7. Change Year: 2013 8. Run Analysis again 9. Compare with 2010 results 10. Repeat for 2016

Document Your Findings: - Screenshot the Statistics tab for each year - Download data files for each year - Use scenario comparison function for formal analysis

Example 4: Threshold Sensitivity

Goal: See how threshold affects excluded areas

Steps: 1. Click: Show advanced options 2. Set Threshold: 500 acres 3. Run Analysis 4. Note number in “Excluded” box 5. Set Threshold: 1234 acres (default) 6. Run Analysis 7. Note change in “Excluded” 8. Set Threshold: 2000 acres 9. Run Analysis 10. Compare all three

Example 5: Create Presentation Materials

Goal: Generate maps for a presentation

Steps: 1. Set desired parameters 2. Run Analysis 3. Map tab: Take screenshot (Print Screen / Cmd+Shift+4) 4. Statistics tab: Take screenshot of charts 5. Arrange screenshots in presentation software

Tips: - Zoom map to desired extent before screenshot - Use full-screen browser mode for cleaner screenshots - Consider running multiple scenarios

Downloading Results

Data Download

Click “Download Data” button to export a CSV file containing:

Columns Include: - Unit ID (FIPS, HUC8, or HUC2) - Cropland area - Surplus/deficit values - Classifications - If WWTP included: combined classifications and surplus

File Name Format:

manureshed_[scale]_[year]_[nutrient].csv

Example: manureshed_huc8_2016_nitrogen.csv

Using Downloaded Data

Import into spreadsheet software:

# In Excel or Google Sheets:
# File > Import > CSV
# Select the downloaded file

Further analysis in R:

# Read downloaded data
data <- read.csv("manureshed_huc8_2016_nitrogen.csv")

# Filter to sources only
sources <- data[data$N_class == "Source", ]

# Calculate statistics
mean(sources$N_surplus)

Performance Tips

For Faster Analysis

Choose Appropriate Scale: - HUC2: Fastest (~18 units) - HUC8: Moderate (~2,000 units) - County: Slowest (~3,000 units)

Processing Times (approximate): - County: 30-60 seconds - HUC8: 20-40 seconds - HUC2: 5-10 seconds

Note: First run for a year is slower (downloads data), subsequent runs are faster (uses cached data).

Browser Performance

Recommended Browsers: - Chrome (best performance) - Firefox (good performance) - Safari (good on Mac) - Edge (good on Windows)

If Dashboard is Slow: - Close unused browser tabs - Try a different browser - Restart R session - Clear browser cache

Troubleshooting

Dashboard Won’t Launch

Error: “Package ‘shiny’ is required”

Solution:

install.packages(c("shiny", "shinydashboard", "leaflet", "plotly", "DT"))

Error: “Could not find dashboard files”

Solution:

devtools::install()  # Reinstall package

Analysis Fails

Error during analysis:

Possible causes: - Network issue (downloading data) - Invalid year selection - Insufficient memory

Solutions: 1. Check internet connection 2. Try a different year 3. Use smaller scale (HUC2) 4. Restart R session

Map Not Displaying

Map shows base layer only, no colored regions:

Possible causes: - Analysis not run yet - Analysis failed silently

Solutions: 1. Click “Run Analysis” again 2. Check browser console (F12) for errors 3. Try different parameters

Value Boxes Show Zero

Green box shows 0:

This is normal when: - Agricultural-only analysis (no “Within Watershed” category) - Area genuinely has no balanced units

To see non-zero values: - Check “Include WWTP” - Run Analysis

Comparison with Programmatic Interface

When to Use Dashboard

Dashboard is better for: - Quick exploration - Demonstrations - Non-R users - Visual presentations - Teaching

When to Use R Code

R code is better for: - Batch processing multiple years - Custom analyses - Reproducible research - Integration with other analyses - Automated workflows

Hybrid Approach

Use both together:

# 1. Explore with dashboard
launch_dashboard()
# Identify interesting patterns

# 2. Reproduce in code for publication
results <- run_builtin_analysis(
  scale = "huc8",
  year = 2016,
  nutrients = "nitrogen",
  include_wwtp = TRUE
)

# 3. Do advanced analysis
comparison <- compare_scenarios(...)

Sharing Results

For Colleagues Without R

1. Take screenshots of dashboard
2. Download data CSV files
3. Email files with brief explanation

For R Users

Share the code instead:

# Colleagues can reproduce with:
library(manureshed)
results <- run_builtin_analysis(
  scale = "huc8",
  year = 2016,
  nutrients = "nitrogen",
  include_wwtp = TRUE
)

For Presentations

1. Use dashboard in presentation mode
2. Navigate to pre-analyzed results
3. Use zoom to highlight areas of interest
4. Switch between tabs to show different views

Accessibility Features

The dashboard includes: - Keyboard navigation support - Screen reader compatibility - High contrast color schemes - Resizable text - Alternative text for graphics

Privacy and Security

Data Privacy: - All analysis runs locally on your computer - No data sent to external servers - Downloads are saved to your local machine

Security: - Dashboard runs on localhost (your computer only) - Not accessible from internet - No authentication required (local use only)

Advanced: Deploying for Teams

If you want to share the dashboard with your team:

Option 1: Shinyapps.io

Deploy to cloud hosting:

library(rsconnect)

# Configure account (one time)
setAccountInfo(name="your-account", token="...", secret="...")

# Deploy
deployApp(
  appDir = system.file("shiny", "dashboard", package = "manureshed"),
  appName = "manureshed-dashboard"
)

Option 2: RStudio Connect

Deploy to institutional server:

# Follow RStudio Connect deployment guide
# Requires RStudio Connect server

Option 3: Shiny Server

Install on shared server: - Requires system administration access - Best for long-term team use - See Shiny Server documentation

Getting Help

Within Dashboard: - Click “Help” tab for basic information

Package Documentation:

?launch_dashboard
?run_builtin_analysis

Online Resources: - Package website - GitHub repository - Vignettes: browseVignettes("manureshed")

Report Issues: - Use GitHub issues for bug reports - Include error messages and screenshots - Describe steps to reproduce

Related Vignettes

• vignette("getting-started") - Basic package usage
• vignette("scenario-comparison") - Compare multiple scenarios
• vignette("advanced-features") - Advanced analysis techniques
• vignette("data-integration") - Using custom data

Summary

The manureshed interactive dashboard provides:

Pros: - No R coding required - Visual and intuitive - Great for exploration - Easy to demonstrate - Interactive maps and plots

Cons: - Requires additional packages - Limited to one analysis at a time - Not suitable for batch processing - Less flexible than coding

Best For: - Initial data exploration - Presentations and teaching - Non-programmer collaborators - Quick “what if” scenarios - Visual communication

For reproducible research and complex workflows, use the programmatic interface alongside the dashboard.