inst/doc/spatial-integration.R

In geospatialsuite: Comprehensive Geospatiotemporal Analysis and Multimodal Integration Toolkit

## ----include = FALSE----------------------------------------------------------
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  warning = FALSE,
  message = FALSE,
  fig.width = 8,
  fig.height = 6
)

## ----eval=FALSE---------------------------------------------------------------
# # Load required packages
# library(geospatialsuite)
# library(terra)
# library(sf)
# 
# # Verify package functionality
# test_function_availability()

## ----eval=FALSE---------------------------------------------------------------
# # Create sample point data (field sites)
# field_sites <- data.frame(
#   site_id = paste0("Site_", 1:20),
#   lon = runif(20, -83.5, -83.0),
#   lat = runif(20, 40.2, 40.7),
#   crop_type = sample(c("corn", "soybeans", "wheat"), 20, replace = TRUE)
# )
# 
# # Create sample raster data (satellite imagery)
# satellite_raster <- rast(nrows = 50, ncols = 50,
#                         xmin = -83.5, xmax = -83.0,
#                         ymin = 40.2, ymax = 40.7)
# values(satellite_raster) <- runif(2500, 0.2, 0.9)
# names(satellite_raster) <- "ndvi"
# 
# # Extract raster values to points (automatic method detection)
# extracted_results <- universal_spatial_join(
#   source_data = field_sites,
#   target_data = satellite_raster,
#   method = "auto",  # Automatically detects "extract"
#   verbose = TRUE
# )
# 
# # View results - original data plus extracted values
# head(extracted_results)
# print(names(extracted_results))

## ----eval=FALSE---------------------------------------------------------------
# # Extract values with 100m buffer around each point
# buffered_extraction <- universal_spatial_join(
#   source_data = field_sites,
#   target_data = satellite_raster,
#   method = "extract",
#   buffer_distance = 0.001,  # ~100m in degrees
#   summary_function = "mean",
#   verbose = TRUE
# )
# 
# # Compare point vs buffered extraction
# comparison_data <- data.frame(
#   site_id = extracted_results$site_id,
#   point_extraction = extracted_results$extracted_ndvi,
#   buffer_extraction = buffered_extraction$extracted_ndvi
# )
# 
# # Calculate differences
# comparison_data$difference <- abs(comparison_data$point_extraction -
#                                 comparison_data$buffer_extraction)
# print(summary(comparison_data$difference))

## ----eval=FALSE---------------------------------------------------------------
# # Create sample polygon data (management zones)
# management_zones <- data.frame(
#   zone_id = 1:5,
#   x_center = runif(5, -83.4, -83.1),
#   y_center = runif(5, 40.3, 40.6),
#   zone_type = sample(c("irrigated", "dryland"), 5, replace = TRUE)
# )
# 
# # Convert to polygons with buffers
# zones_sf <- st_as_sf(management_zones,
#                      coords = c("x_center", "y_center"),
#                      crs = 4326)
# zones_polygons <- st_buffer(zones_sf, dist = 0.02)  # ~2km buffer
# 
# # Calculate zonal statistics
# zonal_results <- universal_spatial_join(
#   source_data = satellite_raster,  # Raster first for zonal
#   target_data = zones_polygons,    # Polygons second
#   method = "zonal",
#   summary_function = "mean",
#   verbose = TRUE
# )
# 
# # View zonal statistics
# head(zonal_results)

## ----eval=FALSE---------------------------------------------------------------
# # Calculate multiple statistics for each zone
# summary_functions <- c("mean", "median", "max", "min", "sd")
# 
# zonal_multi_stats <- list()
# for (func in summary_functions) {
#   result <- universal_spatial_join(
#     source_data = satellite_raster,
#     target_data = zones_polygons,
#     method = "zonal",
#     summary_function = func,
#     verbose = FALSE
#   )
# 
#   # Extract the new column
#   stat_col <- names(result)[!names(result) %in% names(zones_polygons)]
#   zonal_multi_stats[[func]] <- result[[stat_col[1]]]
# }
# 
# # Combine into summary data frame
# zone_summary <- data.frame(
#   zone_id = zones_polygons$zone_id,
#   zone_type = zones_polygons$zone_type,
#   mean_ndvi = zonal_multi_stats$mean,
#   median_ndvi = zonal_multi_stats$median,
#   max_ndvi = zonal_multi_stats$max,
#   min_ndvi = zonal_multi_stats$min,
#   sd_ndvi = zonal_multi_stats$sd
# )
# 
# print(zone_summary)

## ----eval=FALSE---------------------------------------------------------------
# # Create rasters with different resolutions
# high_res_raster <- rast(nrows = 100, ncols = 100,
#                        xmin = -83.5, xmax = -83.0,
#                        ymin = 40.2, ymax = 40.7)
# values(high_res_raster) <- runif(10000, 0.3, 0.8)
# names(high_res_raster) <- "detailed_ndvi"
# 
# low_res_raster <- rast(nrows = 20, ncols = 20,
#                       xmin = -83.5, xmax = -83.0,
#                       ymin = 40.2, ymax = 40.7)
# values(low_res_raster) <- runif(400, 0.1, 0.6)
# names(low_res_raster) <- "coarse_data"
# 
# # Resample high resolution to match low resolution
# resampled_result <- universal_spatial_join(
#   source_data = high_res_raster,
#   target_data = low_res_raster,
#   method = "resample",
#   summary_function = "mean",
#   verbose = TRUE
# )
# 
# # Check resolution change
# cat("Original resolution:", res(high_res_raster), "\n")
# cat("Resampled resolution:", res(resampled_result), "\n")

## ----eval=FALSE---------------------------------------------------------------
# # Aggregate by scale factor (coarser resolution)
# aggregated_raster <- universal_spatial_join(
#   source_data = high_res_raster,
#   target_data = NULL,  # No target needed for scaling
#   method = "resample",
#   scale_factor = 5,    # 5x coarser resolution
#   summary_function = "mean",
#   verbose = TRUE
# )
# 
# # Disaggregate (finer resolution)
# disaggregated_raster <- universal_spatial_join(
#   source_data = low_res_raster,
#   target_data = NULL,
#   method = "resample",
#   scale_factor = 0.5,  # 2x finer resolution
#   verbose = TRUE
# )
# 
# cat("Original low res:", res(low_res_raster), "\n")
# cat("Disaggregated res:", res(disaggregated_raster), "\n")

## ----eval=FALSE---------------------------------------------------------------
# # Create county boundaries (simplified)
# counties <- data.frame(
#   county = c("Franklin", "Delaware", "Union"),
#   x_center = c(-83.0, -83.1, -83.3),
#   y_center = c(40.0, 40.4, 40.2)
# )
# 
# counties_sf <- st_as_sf(counties, coords = c("x_center", "y_center"), crs = 4326)
# counties_polygons <- st_buffer(counties_sf, dist = 0.15)  # Large county-like areas
# 
# # Find which county each field site is in
# sites_with_counties <- universal_spatial_join(
#   source_data = field_sites,
#   target_data = counties_polygons,
#   method = "overlay",
#   verbose = TRUE
# )
# 
# # View results
# head(sites_with_counties)

## ----eval=FALSE---------------------------------------------------------------
# # Find nearest weather station for each field site
# weather_stations <- data.frame(
#   station_id = paste0("WX_", 1:8),
#   longitude = runif(8, -83.6, -82.9),
#   latitude = runif(8, 40.1, 40.8),
#   elevation = runif(8, 200, 400),
#   avg_temp = runif(8, 10, 15)
# )
# 
# nearest_results <- universal_spatial_join(
#   source_data = field_sites,
#   target_data = weather_stations,
#   method = "nearest",
#   verbose = TRUE
# )
# 
# # Check distances to nearest stations
# # The function automatically calculates spatial relationships
# head(nearest_results)

## ----eval=FALSE---------------------------------------------------------------
# # Integrate multiple datasets to field sites
# raster_datasets <- list(
#   elevation = rast(nrows = 30, ncols = 30,
#                   xmin = -83.5, xmax = -83.0,
#                   ymin = 40.2, ymax = 40.7),
#   temperature = rast(nrows = 30, ncols = 30,
#                     xmin = -83.5, xmax = -83.0,
#                     ymin = 40.2, ymax = 40.7),
#   precipitation = rast(nrows = 30, ncols = 30,
#                       xmin = -83.5, xmax = -83.0,
#                       ymin = 40.2, ymax = 40.7)
# )
# 
# # Add random values
# values(raster_datasets$elevation) <- runif(900, 200, 400)
# values(raster_datasets$temperature) <- runif(900, 8, 18)
# values(raster_datasets$precipitation) <- runif(900, 800, 1200)
# 
# # Extract all environmental variables to field sites
# environmental_data <- field_sites
# for (var_name in names(raster_datasets)) {
#   extraction_result <- universal_spatial_join(
#     source_data = environmental_data,
#     target_data = raster_datasets[[var_name]],
#     method = "extract",
#     verbose = FALSE
#   )
# 
#   # Add extracted values to main dataset
#   new_col <- names(extraction_result)[!names(extraction_result) %in% names(environmental_data)]
#   environmental_data[[var_name]] <- extraction_result[[new_col[1]]]
# }
# 
# # View integrated dataset
# head(environmental_data)

## ----eval=FALSE---------------------------------------------------------------
# # Create sample DEM
# dem_raster <- rast(nrows = 60, ncols = 60,
#                   xmin = -83.5, xmax = -83.0,
#                   ymin = 40.2, ymax = 40.7)
# values(dem_raster) <- runif(3600, 200, 500)
# names(dem_raster) <- "elevation"
# 
# # Use the integrated terrain analysis function
# terrain_results <- integrate_terrain_analysis(
#   vector_data = field_sites,
#   elevation_raster = dem_raster,
#   terrain_vars = c("slope", "aspect", "TRI", "TPI"),
#   extraction_method = "extract"
# )
# 
# # View terrain-enhanced field sites
# head(terrain_results)

## ----eval=FALSE---------------------------------------------------------------
# # Create data in different coordinate systems
# utm_points <- data.frame(
#   id = 1:10,
#   x_utm = runif(10, 300000, 350000),  # UTM coordinates
#   y_utm = runif(10, 4450000, 4480000)
# )
# 
# # Convert to UTM Zone 17N
# utm_sf <- st_as_sf(utm_points, coords = c("x_utm", "y_utm"), crs = 32617)
# 
# # Our raster is in WGS84 (EPSG:4326)
# wgs84_raster <- satellite_raster
# 
# # Universal spatial join handles CRS conversion automatically
# utm_extraction <- universal_spatial_join(
#   source_data = utm_sf,
#   target_data = wgs84_raster,
#   method = "extract",
#   verbose = TRUE  # Shows CRS conversion messages
# )
# 
# # Check that extraction worked despite different CRS
# head(utm_extraction)

## ----eval=FALSE---------------------------------------------------------------
# # Force specific output CRS
# projected_result <- universal_spatial_join(
#   source_data = field_sites,
#   target_data = satellite_raster,
#   method = "extract",
#   crs_target = 32617,  # UTM Zone 17N
#   verbose = TRUE
# )
# 
# # Check output CRS
# st_crs(projected_result)

## ----eval=FALSE---------------------------------------------------------------
# # Create raster with some NA values
# sparse_raster <- satellite_raster
# values(sparse_raster)[sample(1:ncell(sparse_raster), 500)] <- NA
# names(sparse_raster) <- "sparse_data"
# 
# # Different strategies for handling NAs
# strategies <- c("remove", "nearest", "zero")
# 
# na_handling_results <- list()
# for (strategy in strategies) {
#   result <- universal_spatial_join(
#     source_data = field_sites,
#     target_data = sparse_raster,
#     method = "extract",
#     na_strategy = strategy,
#     verbose = FALSE
#   )
# 
#   extracted_col <- names(result)[grepl("extracted", names(result))]
#   na_handling_results[[strategy]] <- result[[extracted_col[1]]]
# }
# 
# # Compare different NA handling approaches
# na_comparison <- data.frame(
#   site_id = field_sites$site_id,
#   remove_na = na_handling_results$remove,
#   nearest_fill = na_handling_results$nearest,
#   zero_fill = na_handling_results$zero
# )
# 
# # Count NAs in each approach
# sapply(na_comparison[-1], function(x) sum(is.na(x)))

## ----eval=FALSE---------------------------------------------------------------
# # Create rasters at different scales
# scales <- c(1, 2, 4, 8)  # Different aggregation levels
# multi_scale_data <- list()
# 
# base_raster <- satellite_raster
# for (scale in scales) {
#   if (scale == 1) {
#     multi_scale_data[[paste0("scale_", scale)]] <- base_raster
#   } else {
#     aggregated <- universal_spatial_join(
#       source_data = base_raster,
#       target_data = NULL,
#       method = "resample",
#       scale_factor = scale,
#       summary_function = "mean"
#     )
#     multi_scale_data[[paste0("scale_", scale)]] <- aggregated
#   }
# }
# 
# # Extract values at different scales
# multi_scale_extraction <- field_sites
# for (scale_name in names(multi_scale_data)) {
#   result <- universal_spatial_join(
#     source_data = multi_scale_extraction,
#     target_data = multi_scale_data[[scale_name]],
#     method = "extract",
#     verbose = FALSE
#   )
# 
#   # Add to main dataset
#   new_col <- names(result)[!names(result) %in% names(multi_scale_extraction)]
#   multi_scale_extraction[[scale_name]] <- result[[new_col[1]]]
# }
# 
# # Analyze scale effects
# scale_columns <- names(multi_scale_extraction)[grepl("scale_", names(multi_scale_extraction))]
# scale_analysis <- multi_scale_extraction[c("site_id", scale_columns)]
# head(scale_analysis)

## ----eval=FALSE---------------------------------------------------------------
# # Create sparse monitoring data
# sparse_monitoring <- data.frame(
#   monitor_id = 1:8,
#   longitude = runif(8, -83.4, -83.1),
#   latitude = runif(8, 40.3, 40.6),
#   soil_ph = runif(8, 6.0, 7.5),
#   organic_matter = runif(8, 2, 6)
# )
# 
# # Some missing values
# sparse_monitoring$soil_ph[c(2, 5)] <- NA
# sparse_monitoring$organic_matter[c(3, 7)] <- NA
# 
# # Interpolate missing values
# interpolated_data <- spatial_interpolation_comprehensive(
#   spatial_data = sparse_monitoring,
#   target_variables = c("soil_ph", "organic_matter"),
#   method = "NN",  # Nearest neighbor
#   verbose = TRUE
# )
# 
# # Compare before and after interpolation
# comparison <- data.frame(
#   original_ph = sparse_monitoring$soil_ph,
#   interpolated_ph = interpolated_data$soil_ph,
#   original_om = sparse_monitoring$organic_matter,
#   interpolated_om = interpolated_data$organic_matter
# )
# 
# print(comparison)

## ----eval=FALSE---------------------------------------------------------------
# # Simulate large point dataset
# large_dataset <- data.frame(
#   point_id = 1:5000,
#   x = runif(5000, -83.5, -83.0),
#   y = runif(5000, 40.2, 40.7),
#   measurement = runif(5000, 0, 100)
# )
# 
# # Process in chunks for memory efficiency
# chunked_extraction <- universal_spatial_join(
#   source_data = large_dataset,
#   target_data = satellite_raster,
#   method = "extract",
#   chunk_size = 1000,  # Process 1000 points at a time
#   verbose = TRUE
# )
# 
# # Check processing efficiency
# cat("Processed", nrow(chunked_extraction), "points successfully\n")

## ----eval=FALSE---------------------------------------------------------------
# # Create multiple large rasters
# raster_list <- list()
# for (i in 1:3) {
#   r <- rast(nrows = 200, ncols = 200,
#            xmin = -84, xmax = -82, ymin = 40, ymax = 42)
#   values(r) <- runif(40000, 0, 1)
#   names(r) <- paste0("band_", i)
#   raster_list[[i]] <- r
# }
# 
# # Efficiently combine rasters
# combined_raster <- raster_to_raster_ops(
#   raster1 = raster_list[[1]],
#   raster2 = raster_list[[2]],
#   operation = "add",
#   handle_na = "ignore",
#   verbose = TRUE
# )
# 
# # Multi-raster operations
# mean_composite <- universal_spatial_join(
#   source_data = raster_list[[1]],
#   target_data = raster_list[[2]],
#   method = "resample",
#   summary_function = "mean",
#   verbose = TRUE
# )

## ----eval=FALSE---------------------------------------------------------------
# # Simulate complete agricultural analysis workflow
# farm_fields <- data.frame(
#   field_id = paste0("Field_", LETTERS[1:15]),
#   longitude = runif(15, -83.4, -83.1),
#   latitude = runif(15, 40.3, 40.6),
#   crop_type = sample(c("corn", "soybeans"), 15, replace = TRUE),
#   planting_date = as.Date("2023-05-01") + sample(0:20, 15, replace = TRUE)
# )
# 
# # Convert to polygons (field boundaries)
# farm_sf <- st_as_sf(farm_fields, coords = c("longitude", "latitude"), crs = 4326)
# field_polygons <- st_buffer(farm_sf, dist = 0.008)  # ~800m field size
# 
# # Environmental datasets
# environmental_rasters <- list(
#   ndvi = satellite_raster,
#   elevation = dem_raster,
#   soil_moisture = rast(nrows = 40, ncols = 40,
#                       xmin = -83.5, xmax = -83.0,
#                       ymin = 40.2, ymax = 40.7)
# )
# values(environmental_rasters$soil_moisture) <- runif(1600, 0.1, 0.4)
# 
# # Comprehensive field characterization
# field_analysis <- farm_fields
# for (env_var in names(environmental_rasters)) {
#   # Calculate field averages using zonal statistics
#   zonal_result <- universal_spatial_join(
#     source_data = environmental_rasters[[env_var]],
#     target_data = field_polygons,
#     method = "zonal",
#     summary_function = "mean",
#     verbose = FALSE
#   )
# 
#   # Extract the zonal statistic
#   stat_col <- names(zonal_result)[grepl("zonal", names(zonal_result))]
#   field_analysis[[paste0("avg_", env_var)]] <- zonal_result[[stat_col[1]]]
# }
# 
# # View comprehensive field analysis
# head(field_analysis)

## ----eval=FALSE---------------------------------------------------------------
# # Create watershed polygons
# watersheds <- data.frame(
#   watershed_id = paste0("WS_", 1:6),
#   outlet_lon = runif(6, -83.4, -83.1),
#   outlet_lat = runif(6, 40.3, 40.6),
#   area_km2 = runif(6, 10, 100)
# )
# 
# watersheds_sf <- st_as_sf(watersheds, coords = c("outlet_lon", "outlet_lat"), crs = 4326)
# # Create watershed polygons proportional to area
# watersheds_polygons <- st_buffer(watersheds_sf, dist = sqrt(watersheds$area_km2) * 0.002)
# 
# # Calculate watershed statistics from multiple sources
# watershed_stats <- watersheds
# raster_sources <- list(
#   mean_elevation = dem_raster,
#   mean_ndvi = satellite_raster,
#   vegetation_variability = satellite_raster  # Will calculate SD
# )
# 
# summary_functions <- list(
#   mean_elevation = "mean",
#   mean_ndvi = "mean",
#   vegetation_variability = "sd"
# )
# 
# for (var_name in names(raster_sources)) {
#   result <- universal_spatial_join(
#     source_data = raster_sources[[var_name]],
#     target_data = watersheds_polygons,
#     method = "zonal",
#     summary_function = summary_functions[[var_name]],
#     verbose = FALSE
#   )
# 
#   stat_col <- names(result)[grepl("zonal", names(result))]
#   watershed_stats[[var_name]] <- result[[stat_col[1]]]
# }
# 
# print(watershed_stats)

## ----eval=FALSE---------------------------------------------------------------
# # Handle geometry mismatches gracefully
# mismatched_raster <- rast(nrows = 25, ncols = 30,  # Different aspect ratio
#                          xmin = -83.6, xmax = -82.9,  # Slightly different extent
#                          ymin = 40.1, ymax = 40.8)
# values(mismatched_raster) <- runif(750, 0, 1)
# 
# # Function automatically handles alignment
# robust_extraction <- universal_spatial_join(
#   source_data = field_sites,
#   target_data = mismatched_raster,
#   method = "extract",
#   verbose = TRUE  # Shows alignment messages
# )
# 
# # Handle empty results gracefully
# empty_region <- data.frame(
#   x = c(-90, -89),  # Far from our data
#   y = c(35, 36)
# )
# 
# empty_sf <- st_as_sf(empty_region, coords = c("x", "y"), crs = 4326)
# 
# # This will work but return NA values where appropriate
# empty_result <- universal_spatial_join(
#   source_data = empty_sf,
#   target_data = satellite_raster,
#   method = "extract",
#   verbose = TRUE
# )
# 
# print(empty_result)

## ----eval=FALSE---------------------------------------------------------------
# # Monitor performance for different data sizes
# data_sizes <- c(10, 50, 100, 500)
# performance_results <- data.frame(
#   n_points = data_sizes,
#   processing_time = numeric(length(data_sizes))
# )
# 
# for (i in seq_along(data_sizes)) {
#   n <- data_sizes[i]
#   test_points <- data.frame(
#     id = 1:n,
#     lon = runif(n, -83.4, -83.1),
#     lat = runif(n, 40.3, 40.6)
#   )
# 
#   start_time <- Sys.time()
#   result <- universal_spatial_join(
#     source_data = test_points,
#     target_data = satellite_raster,
#     method = "extract",
#     verbose = FALSE
#   )
#   end_time <- Sys.time()
# 
#   performance_results$processing_time[i] <- as.numeric(end_time - start_time)
# }
# 
# print(performance_results)

## ----eval=FALSE---------------------------------------------------------------
# # Extract vegetation indices to management zones
# vegetation_stack <- calculate_multiple_indices(
#   red = satellite_raster * 0.7,  # Simulate red band
#   nir = satellite_raster,
#   indices = c("NDVI", "DVI", "RVI"),
#   output_stack = TRUE
# )
# 
# # Extract all vegetation indices to zones
# vegetation_by_zone <- universal_spatial_join(
#   source_data = zones_polygons,
#   target_data = vegetation_stack,
#   method = "extract",
#   buffer_distance = 0,  # No buffer for polygon extraction
#   summary_function = "mean",
#   verbose = TRUE
# )
# 
# # Analyze vegetation patterns by zone type
# zone_veg_summary <- aggregate(
#   vegetation_by_zone[grepl("extracted", names(vegetation_by_zone))],
#   by = list(zone_type = vegetation_by_zone$zone_type),
#   FUN = mean,
#   na.rm = TRUE
# )
# 
# print(zone_veg_summary)

## ----eval=FALSE---------------------------------------------------------------
# # Combine water indices with field water quality measurements
# water_index_stack <- calculate_multiple_water_indices(
#   green = satellite_raster * 0.8,  # Simulate green band
#   nir = satellite_raster,
#   swir1 = satellite_raster * 0.5,  # Simulate SWIR1
#   indices = c("NDWI", "MNDWI", "NDMI"),
#   output_stack = TRUE
# )
# 
# # Extract to water quality monitoring sites
# water_sites <- data.frame(
#   site_id = paste0("WQ_", 1:12),
#   longitude = runif(12, -83.4, -83.1),
#   latitude = runif(12, 40.3, 40.6),
#   measured_turbidity = runif(12, 5, 25),
#   measured_chlorophyll = runif(12, 8, 35)
# )
# 
# remote_vs_field <- universal_spatial_join(
#   source_data = water_sites,
#   target_data = water_index_stack,
#   method = "extract",
#   buffer_distance = 0.002,  # 200m buffer
#   summary_function = "mean",
#   verbose = TRUE
# )
# 
# # Analyze relationships between remote sensing and field data
# correlations <- cor(
#   remote_vs_field[c("measured_turbidity", "measured_chlorophyll")],
#   remote_vs_field[grepl("extracted", names(remote_vs_field))],
#   use = "complete.obs"
# )
# 
# print(correlations)

## ----eval=FALSE---------------------------------------------------------------
# # Use the specialized multi-scale function
# multi_scale_analysis <- multiscale_operations(
#   spatial_data = satellite_raster,
#   target_scales = c(1, 2, 4, 8),
#   operation = "mean",
#   pyramid = TRUE
# )
# 
# # Extract at multiple scales to compare spatial patterns
# scale_comparison <- field_sites
# for (scale_name in names(multi_scale_analysis)) {
#   result <- universal_spatial_join(
#     source_data = scale_comparison,
#     target_data = multi_scale_analysis[[scale_name]],
#     method = "extract",
#     verbose = FALSE
#   )
# 
#   new_col <- names(result)[!names(result) %in% names(scale_comparison)]
#   scale_comparison[[scale_name]] <- result[[new_col[1]]]
# }
# 
# # Analyze scale effects
# scale_vars <- names(scale_comparison)[grepl("scale_", names(scale_comparison))]
# scale_effects <- scale_comparison[c("site_id", scale_vars)]
# head(scale_effects)

## ----eval=FALSE---------------------------------------------------------------
# # Create two related rasters for mathematical operations
# raster_a <- satellite_raster
# raster_b <- rast(nrows = 50, ncols = 50,
#                 xmin = -83.5, xmax = -83.0,
#                 ymin = 40.2, ymax = 40.7)
# values(raster_b) <- runif(2500, 0.1, 0.7)
# 
# # Mathematical operations between rasters
# addition_result <- raster_to_raster_ops(
#   raster1 = raster_a,
#   raster2 = raster_b,
#   operation = "add",
#   align_method = "resample",
#   verbose = TRUE
# )
# 
# difference_result <- raster_to_raster_ops(
#   raster1 = raster_a,
#   raster2 = raster_b,
#   operation = "subtract",
#   handle_na = "ignore",
#   verbose = TRUE
# )
# 
# ratio_result <- raster_to_raster_ops(
#   raster1 = raster_a,
#   raster2 = raster_b,
#   operation = "ratio",
#   verbose = TRUE
# )
# 
# # Extract mathematical results to points
# math_results <- universal_spatial_join(
#   source_data = field_sites,
#   target_data = c(addition_result, difference_result, ratio_result),
#   method = "extract",
#   verbose = TRUE
# )

## ----eval=FALSE---------------------------------------------------------------
# # Create comprehensive visualization of integrated results
# integrated_map <- create_spatial_map(
#   spatial_data = environmental_data,
#   fill_variable = "elevation",
#   color_scheme = "terrain",
#   title = "Field Sites with Environmental Data",
#   point_size = 5
# )
# 
# # Quick visualization of zonal results
# zonal_map <- create_spatial_map(
#   spatial_data = zonal_results,
#   fill_variable = names(zonal_results)[grepl("zonal", names(zonal_results))][1],
#   map_type = "polygons",
#   color_scheme = "viridis",
#   title = "Management Zones - Mean NDVI"
# )

## ----eval=FALSE---------------------------------------------------------------
# # Compare extraction methods
# comparison_data <- data.frame(
#   site_id = field_sites$site_id,
#   point_extraction = extracted_results$extracted_ndvi,
#   buffer_extraction = buffered_extraction$extracted_ndvi,
#   difference = abs(extracted_results$extracted_ndvi - buffered_extraction$extracted_ndvi)
# )
# 
# # Plot comparison
# plot(comparison_data$point_extraction, comparison_data$buffer_extraction,
#      xlab = "Point Extraction", ylab = "Buffer Extraction",
#      main = "Point vs Buffer Extraction Comparison")
# abline(0, 1, col = "red", lty = 2)  # 1:1 line

## ----eval=FALSE---------------------------------------------------------------
# # Always validate inputs before processing
# validate_spatial_data <- function(data) {
#   if (inherits(data, "sf")) {
#     # Check for valid geometries
#     invalid_geoms <- !st_is_valid(data)
#     if (any(invalid_geoms)) {
#       warning(paste("Found", sum(invalid_geoms), "invalid geometries"))
#       data <- st_make_valid(data)
#     }
#   }
# 
#   if (inherits(data, "SpatRaster")) {
#     # Check for data coverage
#     valid_cells <- sum(!is.na(values(data, mat = FALSE)))
#     total_cells <- ncell(data)
#     coverage <- (valid_cells / total_cells) * 100
# 
#     if (coverage < 10) {
#       warning(paste("Low data coverage:", round(coverage, 1), "%"))
#     }
#   }
# 
#   return(data)
# }
# 
# # Use validation in workflows
# validated_sites <- validate_spatial_data(
#   st_as_sf(field_sites, coords = c("lon", "lat"), crs = 4326)
# )
# validated_raster <- validate_spatial_data(satellite_raster)

## ----eval=FALSE---------------------------------------------------------------
# # Guidelines for choosing spatial join methods
# method_guide <- data.frame(
#   Source_Type = c("Points", "Polygons", "Raster", "Raster", "Points"),
#   Target_Type = c("Raster", "Raster", "Polygons", "Raster", "Points"),
#   Recommended_Method = c("extract", "extract", "zonal", "resample", "nearest"),
#   Use_Case = c("Sample at locations", "Area averages", "Regional stats", "Align data", "Find closest"),
#   stringsAsFactors = FALSE
# )
# 
# print(method_guide)

## ----eval=FALSE---------------------------------------------------------------
# # Optimize for different scenarios
# optimization_tips <- list(
#   large_datasets = "Use chunk_size parameter to control memory usage",
#   different_crs = "Let the function handle CRS conversion automatically",
#   missing_data = "Choose appropriate na_strategy for your analysis needs",
#   multiple_variables = "Process variables separately for better error handling",
#   visualization = "Use quick_map() for fast preliminary visualization"
# )
# 
# for (tip_name in names(optimization_tips)) {
#   cat(tip_name, ":", optimization_tips[[tip_name]], "\n")
# }