4 Emission Factors

Overview

This chapter applies IPCC default emission factors (G_ef) to convert fuel consumption estimates from Chapter 3 into greenhouse gas emissions. For forest fires, CH₄ and N₂O are estimated directly using Equation 2.27, while CO₂ is accounted for through carbon stock changes (Equations 2.7-2.14) in the Forest Land chapter. For savanna fires at Tier 1, CO₂ is assumed balanced by regrowth within one year, so only non-CO₂ gases are reported. For organic soil fires, CO₂ is reported separately because peat carbon is not replaced by regrowth on management-relevant timescales.

Environment Setup

easypackages::packages(
  "bslib",
  "cols4all", "covr", "cowplot",
  "dendextend", "digest","DiagrammeR","dtwclust", "downlit",
  "e1071", "exactextractr","elevatr",
  "FNN", "future", "forestdata",
  "gdalcubes", "gdalUtilities", "geojsonsf", "geos", "ggplot2", "ggstats",
  "ggspatial", "ggmap", "ggplotify", "ggpubr", "ggrepel", "giscoR",
  "hdf5r", "httr", "httr2", "htmltools",
  "jsonlite",
  "kohonen", "knitr",
  "leaflet.providers", "leafem", "libgeos","luz","lwgeom", "leaflet", "leafgl",
  "mapedit", "mapview", "maptiles", "methods", "mgcv",
  "ncdf4", "nnet",
  "openxlsx", "parallel", "plotly",
  "randomForest", "rasterVis", "raster", "Rcpp", "RcppArmadillo",
  "RcppCensSpatial","rayshader", "RcppEigen", "RcppParallel",
  "RColorBrewer", "reactable", "rgl", "rsconnect","RStoolbox", "rts", "reticulate",
  "s2", "sf", "scales", "sits","spdep", "stars", "stringr","supercells",
  "terra", "testthat", "tidyverse", "tidyterra","tools",
  "tmap", "tmaptools", "terrainr",
  "xgboost",
  prompt = F)

#mapviewOptions(fgb = FALSE)
sf::sf_use_s2(use_s2 = FALSE)

4.1 IPCC Emission Factors

4.1.1 Completing Equation 2.27

Emission factors represent the final component of the IPCC fire emissions equation:

\[ L_{fire} = A \times M_B \times C_f \times G_{ef} \times 10^{-3} \]

Where:

G_ef = Emission factor (g gas per kg dry matter burned)
L_fire = Total emissions (tonnes of gas)
10^-3 = Unit conversion factor (g to kg, kg to tonnes)

Expanded form: \[ L_{fire} \text{ (tonnes)} = \text{Area (ha)} \times \frac{\text{Biomass (t DM)}}{\text{ha}} \times \frac{\text{g gas}}{\text{kg DM}} \times 10^{-3} \]

4.1.2 IPCC Emission Factor Tables

Key IPCC sources:

Table 2.5 (2019 Refinement): Emission factors for CO₂, CH₄, N₂O, CO, and NO_x from fires in forests, savanna, and other vegetation types
Wetlands Supplement Table 2.6: CO₂ and CH₄ emission factors for fires on organic soils
Chapter 4.2 (2006 Guidelines): Carbon stock change methods for forest CO₂ accounting

4.1.3 IPCC Default Emission Factors

Forest Fires (IPCC 2019 Refinement, Table 2.5):

Forest Type	CO₂ (g kg^-1)	CH₄ (g kg^-1)	N₂O (g kg^-1)	CO (g kg^-1)*	NO_x (g kg^-1)*
Tropical Forest	1580 ± 90	6.8 ± 2.0	0.20 ± 0.10	104 ± 20	3.9 ± 2.4
Extratropical Forest	1569 ± 131	4.7 ± 1.9	0.26 ± 0.15	107 ± 37	3.0 ± 1.4

*Optional precursor gases for Tier 1

Savanna Fires (IPCC 2019 Refinement, Table 2.5):

Vegetation Type	CO₂ (g kg^-1)	CH₄ (g kg^-1)	N₂O (g kg^-1)	CO (g kg^-1)*	NO_x (g kg^-1)*
Savanna/Grassland	1613 ± 95	2.3 ± 0.9	0.21 ± 0.10	65 ± 20	3.9 ± 2.4

Note: Savanna CO₂ emission factor is listed for completeness but is not applied at Tier 1 due to the synchrony assumption (see Section 4.5.1).

Organic Soil Fires (IPCC 2013 Wetlands Supplement, Table 2.6):

Soil Type	CO₂ (g kg^-1)	CH₄ (g kg^-1)	N₂O (g kg^-1)
Tropical Peatland	1703 ± 8	5.7 ± 5.4	Not estimated

4.2 GEE Initialisation

# Point environment to your configured python
reticulate::use_python("/opt/local/bin/python", required=T)

# Initialise Earth Engine
ee <- reticulate::import("ee")
ee$Initialize(project = "murphys-deforisk")
cat("GEE initialised\n")
## GEE initialised

# Helper: visualize EE tile in tmap/leaflet
ee_tile_url <- function(ee_image, vis_params) {
  ee_image$getMapId(vis_params)$tile_fetcher$url_format
}

# Helper: EE FeatureCollection => sf
ee_to_sf <- function(fc) {
  geojsonsf::geojson_sf(
    jsonlite::toJSON(fc$getInfo(), auto_unbox = TRUE))
}

# Load AOI boundaries from FAO GAUL
aoi_country_ee <- ee$FeatureCollection("FAO/GAUL/2015/level0")$
  filter(ee$Filter$eq("ADM0_NAME", "Honduras"))
aoi_states_ee  <- ee$FeatureCollection("FAO/GAUL/2015/level1")$
  filter(ee$Filter$eq("ADM0_NAME", "Honduras"))
aoi_country_sf <- ee_to_sf(aoi_country_ee)
aoi_states_sf  <- ee_to_sf(aoi_states_ee)

cat("AOI loaded:", aoi_country_ee$size()$getInfo(), "country,",
    aoi_states_ee$size()$getInfo(), "states\n")
## AOI loaded: 1 country, 19 states

4.2.1 Upstream Inputs from Chapters 1-3

# --- Chapter 1: Burned area (MCD64A1) ---
mcd64 <- ee$ImageCollection("MODIS/061/MCD64A1")
burned_2010 <- mcd64$
  filterBounds(aoi_country_ee$geometry())$
  filterDate("2010-01-01", "2010-12-31")

# FAOSTAT quality filter (matching Ch1)
faostat_filter <- function(image) {
  unc  <- image$select("Uncertainty")
  burn <- image$select("BurnDate")
  mask <- unc$lt(20L)$And(burn$gt(0L))
  image$updateMask(mask)
}

# Annual burned mask for 2010
annual_burned_2010 <- burned_2010$
  map(faostat_filter)$
  select("BurnDate")$max()$gt(0L)$selfMask()

# --- Chapter 2: Land cover & vegetation masks ---
lc_2010 <- ee$ImageCollection("MODIS/061/MCD12Q1")$
  select("LC_Type1")$
  filterDate("2010-01-01", "2010-12-31")$
  first()$
  clip(aoi_country_ee$geometry())

# Forest mask (IGBP 1-5)
forest_mask <- lc_2010$gte(1L)$And(lc_2010$lte(5L))

# Savanna/grassland mask (IGBP 6-10)
savanna_mask <- lc_2010$gte(6L)$And(lc_2010$lte(10L))

# Climate-based humid tropical delineation
bio <- ee$Image("WORLDCLIM/V1/BIO")
precip <- bio$select("bio12")
temp   <- bio$select("bio01")

tropical <- temp$gt(180L)
moist    <- precip$gt(1500L)
humid_tropical_mask <- tropical$And(moist)

# Forest sub-types
humid_tropical_forest <- forest_mask$And(humid_tropical_mask)
other_forest          <- forest_mask$And(humid_tropical_mask$Not())

# Climate zones
boreal    <- temp$lt(0L)
temperate <- temp$gte(0L)$And(temp$lte(180L))

# --- Chapter 3: Fuel consumption raster ---
fuel_consumption <- ee$Image(0)$rename("fuel_consumption")
htf_mask <- forest_mask$And(humid_tropical_mask)
fuel_consumption <- fuel_consumption$where(htf_mask, 43.1)
forest_boreal <- forest_mask$And(boreal)$And(humid_tropical_mask$Not())
fuel_consumption <- fuel_consumption$where(forest_boreal, 18.5)
forest_temperate <- forest_mask$And(temperate)$And(humid_tropical_mask$Not())
fuel_consumption <- fuel_consumption$where(forest_temperate, 23.7)
forest_tropical <- forest_mask$And(tropical)$And(humid_tropical_mask$Not())
fuel_consumption <- fuel_consumption$where(forest_tropical, 32.6)
fuel_consumption <- fuel_consumption$where(savanna_mask, 7.0)

# Apply to burned areas
fuel_burned <- fuel_consumption$
  updateMask(annual_burned_2010)$
  clip(aoi_country_ee$geometry())

# Pixel area helper
pixelArea_ha <- ee$Image$pixelArea()$divide(10000L)

# Biomass burned per pixel (t DM)
biomass_per_pixel <- fuel_burned$multiply(pixelArea_ha)

cat("Upstream inputs loaded: MCD64A1 + MCD12Q1 + WorldClim + fuel raster\n")
## Upstream inputs loaded: MCD64A1 + MCD12Q1 + WorldClim + fuel raster

4.3 Methane (CH₄) Emissions

4.3.1 CH₄ Emission Mechanisms

Formation process: Incomplete combustion under oxygen-limited conditions

\[ \text{Biomass} \xrightarrow{\text{Incomplete combustion}} \text{CH}_4 + \text{CO}_2 + \text{CO} + \text{char} \]

Key factors affecting CH₄ emissions:

Fire intensity:
- Low-intensity fires = More smoldering = Higher CH₄
- High-intensity fires = More flaming = Lower CH₄
Fuel moisture:
- Wet fuels = Lower combustion temperature = Higher CH₄
- Dry fuels = Higher temperature = Lower CH₄
Fuel type:
- Forest: High CH₄ (dense, moist fuels)
- Savanna: Low CH₄ (grass cures completely)
Combustion phase:
- Flaming: Minimal CH₄ (>90% CO₂)
- Smoldering: High CH₄ (20-40% of carbon as CH₄)

4.3.2 Forest CH₄ Emissions

4.3.2.1 Emission Factors by Forest Type

Tropical Forests (6.8 g CH₄ kg^-1):

Highest emission factors
Dense canopy >> poor ventilation
High fuel moisture >> low combustion efficiency
Long smoldering phase (especially coarse woody debris)

Temperate/Boreal Forests (4.7 g CH₄ kg^-1):

Moderate emission factors
More complete combustion in conifers
Resinous fuels burn hotter
Less smoldering than tropical

Uncertainty: ±30-40% (IPCC Table 2.5)

4.3.2.2 Calculation Example

Honduras Tropical Forest Fires (2010):

Given:

Biomass burned from Chapter 3 (GEE-derived)
Emission factor: 6.8 g CH₄ kg^-1 DM (tropical forest)

Calculation: \[ \begin{aligned} L_{CH_4} &= \text{Biomass} \times G_{ef} \times 10^{-3} \\ &= \text{Biomass (t DM)} \times 1000 \text{ (kg/t)} \times 6.8 \text{ (g kg}^{-1}\text{)} \times 10^{-6} \text{ (t/g)} \\ &= \text{Biomass (Gg)} \times 6.8 \text{ Gg CH}_4 \end{aligned} \]

# Compute forest biomass from GEE
forest_biomass <- biomass_per_pixel$
  updateMask(forest_mask)$
  reduceRegion(
    reducer   = ee$Reducer$sum(),
    geometry  = aoi_country_ee$geometry(),
    scale     = 500L,
    maxPixels = 1e9
  )$getInfo()

forest_biomass_t <- forest_biomass[["fuel_consumption"]] %||% 0
forest_biomass_Gg <- forest_biomass_t / 1000

# IPCC emission factor (tropical forest)
EF_CH4_tropical_forest <- 6.8  # g kg⁻¹

# Calculate CH₄ emissions
# Gg DM × g/kg = Gg × g/kg (numerically equivalent)
forest_CH4_Gg <- forest_biomass_Gg * EF_CH4_tropical_forest

cat("Honduras 2010 — Forest CH₄ Emissions\n")
## Honduras 2010 — Forest CH₄ Emissions
cat("  Forest biomass burned:", round(forest_biomass_Gg, 2), "Gg DM\n")
##   Forest biomass burned: 1683 Gg DM
cat("  EF (tropical):       ", EF_CH4_tropical_forest, "g kg⁻¹\n")
##   EF (tropical):        6.8 g kg⁻¹
cat("  CH₄ emissions:       ", round(forest_CH4_Gg, 3), "Gg CH₄\n")
##   CH₄ emissions:        11442 Gg CH₄

4.3.3 Savanna CH₄ Emissions

4.3.3.1 Lower Emission Factors

Savanna/Grassland (2.3 g CH₄ kg^-1):

Lowest emission factors among all vegetation types
Grass fuels cure completely (low moisture)
Predominantly flaming combustion
Minimal smoldering phase - High combustion efficiency

Why savanna CH₄ < forest CH₄:

1. Fuel moisture: 5-10% (savanna) vs. 30-60% (forest)

2. Fuel bed depth: 10-30 cm (savanna) vs. 1-10 m (forest)

3. Combustion mode: 80% flaming (savanna) vs. 40% flaming (forest)

4.3.3.2 Calculation Example

Honduras Savanna Fires (2010):

# Compute savanna biomass from GEE
savanna_biomass <- biomass_per_pixel$
  updateMask(savanna_mask)$
  reduceRegion(
    reducer   = ee$Reducer$sum(),
    geometry  = aoi_country_ee$geometry(),
    scale     = 500L,
    maxPixels = 1e9
  )$getInfo()

savanna_biomass_t <- savanna_biomass[["fuel_consumption"]] %||% 0
savanna_biomass_Gg <- savanna_biomass_t / 1000

# IPCC emission factor
EF_CH4_savanna <- 2.3  # g kg⁻¹

# Calculate CH₄ emissions
savanna_CH4_Gg <- savanna_biomass_Gg * EF_CH4_savanna

cat("Honduras 2010 — Savanna CH₄ Emissions\n")
## Honduras 2010 — Savanna CH₄ Emissions
cat("  Savanna biomass burned:", round(savanna_biomass_Gg, 3), "Gg DM\n")
##   Savanna biomass burned: 815 Gg DM
cat("  EF (savanna):          ", EF_CH4_savanna, "g kg⁻¹\n")
##   EF (savanna):           2.3 g kg⁻¹
cat("  CH₄ emissions:         ", round(savanna_CH4_Gg, 4), "Gg CH₄\n")
##   CH₄ emissions:          1874 Gg CH₄

4.3.4 Organic Soil CH₄ Emissions

Tropical Peatland (5.7 g CH₄ kg^-1):

Intermediate emission factors
Smoldering combustion dominates
Can burn underground for months
High variability (CV ~95%)

Southeast Asia only (per FAOSTAT):

# Example: Indonesia peat fires (reference — not Honduras)
indonesia_peat <- data.frame(
  country = "Indonesia",
  year = 2015,
  fire_category = "Organic_Soil",
  biomass_burned_Gg = 125  # Example
)

EF_CH4_peat <- 5.7  # g kg⁻¹

indonesia_ch4 <- indonesia_peat %>%
  mutate(CH4_Gg = biomass_burned_Gg * EF_CH4_peat)

cat("Reference — Indonesia peat CH₄:", indonesia_ch4$CH4_Gg, "Gg\n")
## Reference — Indonesia peat CH₄: 712 Gg
cat("Note: Honduras has negligible peatland fire activity\n")
## Note: Honduras has negligible peatland fire activity

4.4 Nitrous Oxide (N₂O) Emissions

4.4.1 N₂O Emission Mechanisms

Formation process: Incomplete oxidation of nitrogen in biomass

\[ \text{Biomass-N} \xrightarrow{\text{Combustion}} \text{N}_2\text{O} + \text{NO}_x + \text{N}_2 + \text{NH}_3 \]

Key factors:

Nitrogen content of fuel:
- Legumes: High N >> High N₂O
- Conifers: Low N >> Low N₂O
- Grasses: Variable N (fertilization-dependent)
Combustion temperature:
- Low temp (400-600°C): High N₂O/NO_x ratio
- High temp (>800°C): Low N₂O (converts to N₂)
Fuel moisture:
- Affects temperature = Affects N₂O production
Fire residency time:
- Long smoldering = More N₂O

4.4.2 Forest N₂O Emissions

4.4.2.1 Emission Factors by Forest Type

Tropical Forests (0.20 g N₂O kg^-1):

Lower emission factors (surprising given high N content)
Explanation: High temperature flaming combustion
More N₂ produced instead of N₂O

Temperate/Boreal Forests (0.26 g N₂O kg^-1):

Moderate emission factors
Conifer needles: Low N content (~0.5%)
Moderate combustion temperatures

Uncertainty: ±50-60% (highest among all emission factors)

4.4.2.2 Calculation Example

Honduras Tropical Forest Fires (2010):

# IPCC emission factor (tropical forest)
EF_N2O_tropical_forest <- 0.20  # g kg⁻¹

# Calculate N₂O emissions
forest_N2O_Gg <- forest_biomass_Gg * EF_N2O_tropical_forest

cat("Honduras 2010 — Forest N₂O Emissions\n")
## Honduras 2010 — Forest N₂O Emissions
cat("  Forest biomass burned:", round(forest_biomass_Gg, 2), "Gg DM\n")
##   Forest biomass burned: 1683 Gg DM
cat("  EF (tropical):       ", EF_N2O_tropical_forest, "g kg⁻¹\n")
##   EF (tropical):        0.2 g kg⁻¹
cat("  N₂O emissions:       ", round(forest_N2O_Gg, 4), "Gg N₂O\n")
##   N₂O emissions:        337 Gg N₂O

4.4.3 Savanna N₂O Emissions

Savanna/Grassland (0.21 g N₂O kg^-1):

Similar to forest emission factors
Nitrogen variability depends on:
Grazing intensity (manure inputs)
Proximity to croplands (fertilizer drift)
Legume abundance

# IPCC emission factor
EF_N2O_savanna <- 0.21  # g kg⁻¹

# Calculate N₂O emissions
savanna_N2O_Gg <- savanna_biomass_Gg * EF_N2O_savanna

cat("Honduras 2010 — Savanna N₂O Emissions\n")
## Honduras 2010 — Savanna N₂O Emissions
cat("  Savanna biomass burned:", round(savanna_biomass_Gg, 3), "Gg DM\n")
##   Savanna biomass burned: 815 Gg DM
cat("  EF (savanna):          ", EF_N2O_savanna, "g kg⁻¹\n")
##   EF (savanna):           0.21 g kg⁻¹
cat("  N₂O emissions:         ", round(savanna_N2O_Gg, 5), "Gg N₂O\n")
##   N₂O emissions:          171 Gg N₂O

4.4.4 Organic Soil N₂O Emissions

NOT ESTIMATED for organic soil fires (IPCC 2013 Wetlands Supplement):

“Due to lack of data, no default N₂O emission factor is provided for fires in organic soils.”

Rationale:

Very few measurements available
Peat nitrogen content highly variable
Smoldering combustion >> complex N chemistry
Most N released as N₂ or NH₃ rather than N₂O

FAOSTAT approach: N₂O = 0 for organic soil fires

4.5 Carbon Dioxide (CO₂) Emissions

4.5.1 CO₂ Reporting Logic

The treatment of CO₂ differs by fire type under IPCC methodology. The key distinction is whether fire CO₂ emissions are synchronous with CO₂ removals from regrowth:

Fire Category	CO₂ Reported?	Accounting Method	IPCC Source
Forest Fires	Yes	Carbon stock changes (Eq. 2.7–2.14)	2006 GL §4.2.4
Savanna Fires	No (Tier 1)	Synchrony assumed for non-woody grassland	2019 Ref §2.4
Organic Soil Fires	Yes	Separate G_ef (Eq. 2.27)	2013 WS Ch. 2

4.5.2 Forest CO₂ Treatment

4.5.2.1 Why forest fire CO₂ must be reported

The 2006 IPCC Guidelines (Vol. 4, Ch. 4, §4.2.4) state:

“In Forest Land Remaining Forest Land, emissions of CO₂ from biomass burning also need to be accounted for because they are generally not synchronous with rates of CO₂ uptake.”

The 2019 Refinement (Vol. 4, Ch. 2, §2.4) reinforces:

“For Forest Land, synchrony is unlikely if significant woody biomass is killed (i.e., losses represent several years of growth and C accumulation), and the net emissions should be reported.”

4.5.2.2 Accounting pathway: Carbon stock changes

Forest fire CO₂ is captured through the carbon stock change framework (Gain-Loss Method, Equations 2.7–2.14), not through the fire-specific Equation 2.27:

Biomass loss from fire = immediate CO₂ emission (captured by Equation 2.14 as a disturbance loss)
Regrowth = gradual CO₂ removal (captured by Equation 2.9 as annual biomass increment)
Net change reported annually in the Forest Land carbon stock change category

Non-CO₂ gases (CH₄, N₂O) cannot be captured by stock change accounting and must be estimated separately using Equation 2.27 and the G_ef values from Table 2.5.

4.5.2.3 Cross-check: CO₂ emission factor from Table 2.5

Table 2.5 lists CO₂ emission factors for forests (Tropical: 1580 ± 90 g kg^-1; Extratropical: 1569 ± 131 g kg^-1). These can serve as a cross-check against the stock-change estimate:

# Table 2.5 CO₂ emission factor (tropical forest)
EF_CO2_tropical_forest <- 1580  # g kg⁻¹

# Cross-check estimate
forest_CO2_crosscheck_Gg <- forest_biomass_Gg * EF_CO2_tropical_forest

cat("Honduras 2010 — Forest CO₂ Cross-Check\n")
## Honduras 2010 — Forest CO₂ Cross-Check
cat("  Forest biomass burned:", round(forest_biomass_Gg, 2), "Gg DM\n")
##   Forest biomass burned: 1683 Gg DM
cat("  EF (tropical Table 2.5):", EF_CO2_tropical_forest, "g kg⁻¹\n")
##   EF (tropical Table 2.5): 1580 g kg⁻¹
cat("  CO₂ cross-check:      ", round(forest_CO2_crosscheck_Gg, 2), "Gg CO₂\n")
##   CO₂ cross-check:       2658675 Gg CO₂
cat("\n  Note: This value is for cross-checking only.\n")
## 
##   Note: This value is for cross-checking only.
cat("  Forest CO₂ is reported through carbon stock changes\n")
##   Forest CO₂ is reported through carbon stock changes
cat("  (Equations 2.7–2.14), not the fire emissions table.\n")
##   (Equations 2.7–2.14), not the fire emissions table.

4.5.3 Savanna CO₂ Treatment

4.5.3.1 Tier 1 synchrony assumption

At Tier 1, CO₂ from savanna and grassland fires is not reported separately because regrowth is assumed to reabsorb the emitted CO₂ within one year. The 2019 Refinement (Vol. 4, Ch. 2, §2.4) states:

“For grassland biomass burning and burning of agriculture residues, the assumption of equivalence is generally reasonable.”

4.5.3.2 Caveats for woody savannas

The Guidelines explicitly warn against applying synchrony to woody systems:

“woody vegetation may also burn in these land categories, and greenhouse gas emissions from those sources should be reported using a higher Tier method”

and:

“care must be taken before assuming synchrony in such systems”

For Honduras Tier 1, the synchrony assumption is applied to savanna/grassland pixels (IGBP classes 6-10). Countries with significant woody savanna burning should consider Tier 2 methods.

4.5.4 Organic Soil CO₂ Emissions

4.5.4.1 Why Organic Soil CO₂ is Different

Peat = Non-renewable carbon source:

Peat accumulation: 0.1-1.0 mm year^-1
Fire consumption: 100-500 mm per event
Recovery time: 100-5,000 years
Conclusion: Fire CO₂ NOT balanced by regrowth

IPCC Wetlands Supplement:

“Emissions of CO₂ from peat fires should be estimated separately from other fire types, as the carbon in peat has accumulated over centuries to millennia and is not rapidly replaced by regrowth.”

4.5.4.2 Emission Factor

Tropical Peatland (1703 g CO₂ kg^-1): \[ \begin{aligned} \text{Carbon content of peat} &= 0.50 \text{ kg C per kg DM} \\ \text{Molecular weight ratio} &= \frac{44 \text{ (CO}_2)}{12 \text{ (C)}} = 3.67 \\ G_{ef,CO_2} &= 0.50 \times 3.67 \times 1000 \text{ g kg}^{-1} \\ &= 1835 \text{ g CO}_2 \text{ kg}^{-1} \text{ DM} \end{aligned} \]

IPCC default: 1703 g kg^-1

Lower value accounting for incomplete carbon conversion)
Uncertainty: Very low (±8 g kg^-1)
CO₂/C ratio is stoichiometric

4.6 Reproducible Workflow

4.6.1 Full Script

library(tidyverse)

# 1. Load IPCC Emission Factors
# Non-CO₂ gases estimated via Equation 2.27
# Forest CO₂ estimated via stock changes (not included here)
# Savanna CO₂ excluded per Tier 1 synchrony assumption
emission_factors <- tribble(
  ~fire_category, ~climate_zone, ~gas, ~EF_g_kg,
  ~uncertainty_pct, ~accounting_note,

  # Forest Fires — non-CO₂ via Equation 2.27
  "Forest", "Tropical", "CH4", 6.8, 30,
  "Eq 2.27",
  "Forest", "Tropical", "N2O", 0.20, 50,
  "Eq 2.27",
  "Forest", "Temperate", "CH4", 4.7, 40,
  "Eq 2.27",
  "Forest", "Temperate", "N2O", 0.26, 60,
  "Eq 2.27",
  "Forest", "Boreal", "CH4", 4.7, 40,
  "Eq 2.27",
  "Forest", "Boreal", "N2O", 0.26, 60,
  "Eq 2.27",

  # Forest CO₂ — cross-check only (primary via stock changes)
  "Forest", "Tropical", "CO2", 1580, 6,
  "Cross-check (stock changes primary)",
  "Forest", "Temperate", "CO2", 1569, 8,
  "Cross-check (stock changes primary)",
  "Forest", "Boreal", "CO2", 1569, 8,
  "Cross-check (stock changes primary)",

  # Savanna Fires — non-CO₂ only at Tier 1
  "Savanna", "All", "CH4", 2.3, 40,
  "Eq 2.27",
  "Savanna", "All", "N2O", 0.21, 50,
  "Eq 2.27",

  # Organic Soil Fires — CO₂ reported (non-renewable C)
  "Organic_Soil", "Tropical", "CO2", 1703, 0.5,
  "Eq 2.27 (WS)",
  "Organic_Soil", "Tropical", "CH4", 5.7, 95,
  "Eq 2.27 (WS)"
)

# 2. Build fuel consumption data from GEE results (Ch3)
fuel_consumption_data <- tribble(
  ~country, ~year, ~fire_category, ~vegetation_type,
  ~climate_zone, ~area_ha, ~biomass_burned_Gg,

  "Honduras", 2010, "Forest", "Humid_Tropical_Forest",
  "Tropical", NA_real_, forest_biomass_Gg,

  "Honduras", 2010, "Savanna", "Savanna_Mixed", "All",
  NA_real_, savanna_biomass_Gg
)

# 3. Calculate Emissions by Gas
# Exclude forest CO₂ cross-check rows from primary totals
emissions <- fuel_consumption_data %>%
  left_join(
    emission_factors %>%
      filter(accounting_note == "Eq 2.27"),
    by = c("fire_category", "climate_zone")
  ) %>%
  mutate(
    emissions_Gg = biomass_burned_Gg * EF_g_kg
  ) %>%
  select(country, year, fire_category, vegetation_type,
         climate_zone, biomass_burned_Gg, gas, EF_g_kg,
         emissions_Gg, uncertainty_pct)

# 4. Summarize by GHG & Category
emissions_summary <- emissions %>%
  group_by(country, year, fire_category, gas) %>%
  summarise(
    total_biomass_Gg = sum(biomass_burned_Gg),
    total_emissions_Gg = sum(emissions_Gg),
    .groups = "drop"
  )

print(emissions_summary)
## # A tibble: 4 × 6
##   country   year fire_category gas   total_biomass_Gg
##   <chr>    <dbl> <chr>         <chr>            <dbl>
## 1 Honduras  2010 Forest        CH4              1683.
## 2 Honduras  2010 Forest        N2O              1683.
## 3 Honduras  2010 Savanna       CH4               815.
## 4 Honduras  2010 Savanna       N2O               815.
## # ℹ 1 more variable: total_emissions_Gg <dbl>

# 5. Convert to CO₂-eq using AR5 GWPs (without feedbacks)
GWP_CH4 <- 28
GWP_N2O <- 265

emissions_co2eq <- emissions_summary %>%
  mutate(
    CO2eq_Gg = case_when(
      gas == "CO2" ~ total_emissions_Gg,
      gas == "CH4" ~ total_emissions_Gg * GWP_CH4,
      gas == "N2O" ~ total_emissions_Gg * GWP_N2O,
      TRUE ~ 0
    )
  )

# Total CO₂-equivalent emissions
total_co2eq <- emissions_co2eq %>%
  group_by(country, year) %>%
  summarise(
    total_CO2eq_Gg = sum(CO2eq_Gg),
    total_CO2eq_Mt = sum(CO2eq_Gg) / 1000,
    .groups = "drop"
  )

print(total_co2eq)
## # A tibble: 1 × 4
##   country   year total_CO2eq_Gg total_CO2eq_Mt
##   <chr>    <dbl>          <dbl>          <dbl>
## 1 Honduras  2010        507368.           507.

# 6. Export
write.csv(
  emissions,
  "Honduras_FireEmissions_2010_detailed.csv",
  row.names = FALSE
)

write.csv(
  emissions_co2eq,
  "Honduras_FireEmissions_2010_CO2eq.csv",
  row.names = FALSE
)

library(knitr)

kable(
  emissions_co2eq %>%
    select(fire_category, gas, total_emissions_Gg, CO2eq_Gg),
  caption = "Honduras 2010 Fire Emissions by GHG & Category (Equation 2.27 gases only)",
  digits = 4
)

Honduras 2010 Fire Emissions by GHG & Category (Equation 2.27 gases only)
fire_category	gas	total_emissions_Gg	CO2eq_Gg
Forest	CH4	11442	320387
Forest	N2O	337	89183
Savanna	CH4	1874	52463
Savanna	N2O	171	45335


kable(
  total_co2eq,
  caption = "Honduras 2010 Tier 1 Total Fire Emissions (non-CO₂ gases)",
  digits = 4
)

Honduras 2010 Tier 1 Total Fire Emissions (non-CO₂ gases)
country	year	total_CO2eq_Gg	total_CO2eq_Mt
Honduras	2010	507368	507

4.7 Global Warming Potentials and CO₂-Equivalents

4.7.1 GWP Concept

Global Warming Potential (GWP): Radiative forcing of 1 kg of gas relative to 1 kg of CO₂ over 100-year, GWP100 (IPCC, 2014, p. 87):

IPCC Assessment Report 5 (AR5) - 100-year GWPs:

Gas	GWP (w/o feedback)	GWP (w/ feedback)
CO₂	1 (by definition)	1
CH₄	28	34
N₂O	265	298

FAOSTAT uses AR5 without feedbacks: GWP_CH4 = 28, GWP_N2O = 265

IPCC Assessment Report 6 (AR6):¹

Gas	GWP (100-year)
CH₄	27.9
N₂O	273

Reporting guidance: Use GWP set consistent with national reporting (typically AR5 for UNFCCC submissions through ~2024)

4.7.2 Calculating CO₂-Equivalents

\[ \text{CO}_2\text{-eq (Gg)} = \sum_{i=1}^{n} \text{Emissions}_i \text{ (Gg)} \times \text{GWP}_i \]

library(tidyverse)

# Aggregate emissions by gas
emissions_by_gas <- emissions_co2eq %>%
  group_by(gas) %>%
  summarise(
    emissions_Gg = sum(total_emissions_Gg),
    CO2eq_Gg = sum(CO2eq_Gg),
    .groups = "drop"
  )

cat("Honduras 2010 — CO₂-Equivalents (Eq 2.27 gases)\n")
## Honduras 2010 — CO₂-Equivalents (Eq 2.27 gases)
for (i in seq_len(nrow(emissions_by_gas))) {
  cat(sprintf("  %s: %.4f Gg → %.2f Gg CO₂-eq\n",
    emissions_by_gas$gas[i],
    emissions_by_gas$emissions_Gg[i],
    emissions_by_gas$CO2eq_Gg[i]))
}
##   CH4: 13316.0772 Gg → 372850.16 Gg CO₂-eq
##   N2O: 507.6161 Gg → 134518.27 Gg CO₂-eq
cat("  Total:", round(sum(emissions_by_gas$CO2eq_Gg), 2),
    "Gg CO₂-eq\n")
##   Total: 507368 Gg CO₂-eq
cat("\n  Note: Forest CO₂ is reported via carbon stock\n")
## 
##   Note: Forest CO₂ is reported via carbon stock
cat("  changes (Ch. 4.2), not included in this total.\n")
##   changes (Ch. 4.2), not included in this total.

# Visualize
ggplot(emissions_by_gas, aes(x = gas, y = CO2eq_Gg, fill = gas)) +
  geom_col() +
  geom_text(aes(label = paste0(round(CO2eq_Gg, 1), " Gg")),
            vjust = -0.5, fontface = "bold") +
  scale_fill_manual(values = c("CH4" = "orange", "N2O" = "red")) +
  labs(
    title = "Fire Emissions by Gas (CO₂-equivalents)",
    subtitle = "Honduras 2010 — Forest and Savanna Fires (Eq 2.27 non-CO₂ gases)",
    x = "Greenhouse Gas",
    y = "Emissions (Gg CO₂-eq)",
    fill = "Gas"
  ) +
  theme_minimal() +
  theme(legend.position = "none")

4.8 Uncertainty Analysis

4.8.1 Sources of Uncertainty

1. Emission Factor Uncertainty (IPCC Table 2.5):

Gas	Forest (%)	Savanna (%)	Organic Soil (%)
CO₂	±6	±6	±0.5
CH₄	±30-40	±40	±95
N₂O	±50-60	±50	N/A

Note: Forest CO₂ emission factor uncertainty (±6%) is relevant for cross-checking. The total uncertainty in forest CO₂ reporting is dominated by stock-change uncertainty (biomass estimates, allometric models), not G_ef.

2. Cumulative Uncertainty (from all chapters):

Recall from Chapter 3: \[ U_{total}^2 = U_A^2 + U_{MB}^2 + U_{Cf}^2 + U_{Gef}^2 \]

For Forest CH₄ emissions:

$U_A$ = ±30% (burned area)
$U_{MB}$ = ±50% (fuel load)
$U_{Cf}$ = ±25% (combustion factor)
$U_{Gef}$ = ±35% (CH₄ emission factor)

Combined uncertainty: \[ U_{total} = \sqrt{30^2 + 50^2 + 25^2 + 35^2} = \sqrt{4650} \approx 68\% \]

For Savanna emissions: ~85% uncertainty

For Organic soil CO₂: ~60% uncertainty (dominated by fuel consumption)

4.8.2 Monte Carlo Uncertainty Propagation

# Parameters: Honduras Humid Tropical Forest
n_simulations <- 10000

# Mean values (from Ch1-3 GEE results)
MB_mean <- 119.6    # t DM ha⁻¹ (humid tropical)
Cf_mean <- 0.36
EF_CH4_mean <- 6.8  # g kg⁻¹ (tropical forest)

# Uncertainties (SD = 95% CI / 1.96)
MB_sd <- 52.4 / 1.96
Cf_sd <- 0.09 / 1.96
EF_CH4_sd <- 2.0 / 1.96

# Generate random samples (area fixed from GEE)
set.seed(42)
simulations <- data.frame(
  MB = rnorm(n_simulations, MB_mean, MB_sd),
  Cf = rnorm(n_simulations, Cf_mean, Cf_sd),
  EF_CH4 = rnorm(n_simulations, EF_CH4_mean, EF_CH4_sd)
) %>%
  mutate(
    MB = pmax(0, MB),
    Cf = pmin(1, pmax(0, Cf)),
    EF_CH4 = pmax(0, EF_CH4),
    # MB_Cf = fuel consumed (t DM ha⁻¹)
    # Emissions per ha = MB × Cf × EF × 10⁻³ (kg CH₄ ha⁻¹)
    CH4_kg_per_ha = MB * Cf * EF_CH4,
    # Relative to central estimate
    ratio = CH4_kg_per_ha / (MB_mean * Cf_mean * EF_CH4_mean)
  )

# Summary statistics
ratio_mean <- mean(simulations$ratio)
ratio_ci_lower <- quantile(simulations$ratio, 0.025)
ratio_ci_upper <- quantile(simulations$ratio, 0.975)

cat(sprintf("Honduras HTF CH₄ — Relative uncertainty\n"))
## Honduras HTF CH₄ — Relative uncertainty
cat(sprintf("  Mean ratio: %.2f\n", ratio_mean))
##   Mean ratio: 1.00
cat(sprintf("  95%% CI:    [%.2f, %.2f]\n",
            ratio_ci_lower, ratio_ci_upper))
##   95% CI:    [0.48, 1.66]
cat(sprintf("  i.e. -%d%% to +%d%% around central estimate\n",
            round((1 - ratio_ci_lower) * 100),
            round((ratio_ci_upper - 1) * 100)))
##   i.e. -52% to +66% around central estimate

# Visualize distribution
ggplot(simulations, aes(x = CH4_kg_per_ha)) +
  geom_histogram(bins = 50, fill = "steelblue", alpha = 0.7) +
  geom_vline(xintercept = MB_mean * Cf_mean * EF_CH4_mean,
             color = "red", linewidth = 1, linetype = "dashed") +
  labs(
    title = "Uncertainty Distribution: Tropical Forest CH₄ Emissions",
    subtitle = "Honduras — Monte Carlo (10,000 simulations)",
    x = "CH₄ Emissions (kg per ha burned)",
    y = "Frequency"
  ) +
  theme_minimal()

4.9 Spatial Emissions Mapping in GEE

4.9.1 Pixel-Level Emissions Calculation

# 1. Derive Emission Factor Rasters (non-CO₂ gases via Eq 2.27)
# CH₄ emission factors by vegetation type
EF_CH4_img <- ee$Image(0)$rename("EF_CH4")
EF_CH4_img <- EF_CH4_img$
  where(forest_mask$And(tropical), 6.8)$
  where(forest_mask$And(tropical$Not()), 4.7)$
  where(savanna_mask, 2.3)

# N₂O emission factors
EF_N2O_img <- ee$Image(0)$rename("EF_N2O")
EF_N2O_img <- EF_N2O_img$
  where(forest_mask$And(tropical), 0.20)$
  where(forest_mask$And(tropical$Not()), 0.26)$
  where(savanna_mask, 0.21)

# CO₂ emission factor raster (cross-check for forest only)
EF_CO2_img <- ee$Image(0)$rename("EF_CO2")
EF_CO2_img <- EF_CO2_img$
  where(forest_mask$And(tropical), 1580)$
  where(forest_mask$And(tropical$Not()), 1569)

# 2. Calculate Emissions Per Pixel
# CH₄ emissions (kg): biomass (t) × 1000 (kg/t) × EF (g/kg) / 1000 (g→kg)
CH4_per_pixel <- biomass_per_pixel$
  multiply(EF_CH4_img)$
  rename("CH4_kg")$
  clip(aoi_country_ee$geometry())

# N₂O emissions (kg)
N2O_per_pixel <- biomass_per_pixel$
  multiply(EF_N2O_img)$
  rename("N2O_kg")$
  clip(aoi_country_ee$geometry())

# CO₂ cross-check (forest only, kg)
CO2_crosscheck_per_pixel <- biomass_per_pixel$
  updateMask(forest_mask)$
  multiply(EF_CO2_img)$
  rename("CO2_crosscheck_kg")$
  clip(aoi_country_ee$geometry())

# 3. Visualize
ch4_vis <- list(
  min = 0L, max = 50L,
  palette = c("yellow", "orange", "red", "darkred")
)
n2o_vis <- list(
  min = 0L, max = 5L,
  palette = c("lightblue", "blue", "darkblue", "purple")
)

tmap::tmap_mode("view")
tmap::tm_shape(aoi_country_sf) + tmap::tm_borders(col = "white", lwd = 1.5) +
  tmap::tm_tiles(
    ee_tile_url(CH4_per_pixel$selfMask(), ch4_vis),
    group = "CH₄ Emissions (kg per pixel)"
  ) +
  tmap::tm_tiles(
    ee_tile_url(N2O_per_pixel$selfMask(), n2o_vis),
    group = "N₂O Emissions (kg per pixel)"
  ) +
  tmap::tm_add_legend(
    type = "polygons",
    fill = c("orange", "red", "blue", "purple"),
    labels = c("CH₄ Low", "CH₄ High", "N₂O Low", "N₂O High"),
    title = "GHG Emissions (kg/pixel)"
  ) +
  tmap::tm_scalebar(position = c("RIGHT", "BOTTOM"), text.size = 0.5) +
  tmap::tm_basemap("Esri.WorldImagery")

# Total CH₄ emissions
CH4_total <- CH4_per_pixel$
  reduceRegion(
    reducer   = ee$Reducer$sum(),
    geometry  = aoi_country_ee$geometry(),
    scale     = 500L,
    maxPixels = 1e9
  )$getInfo()

# Total N₂O emissions
N2O_total <- N2O_per_pixel$
  reduceRegion(
    reducer   = ee$Reducer$sum(),
    geometry  = aoi_country_ee$geometry(),
    scale     = 500L,
    maxPixels = 1e9
  )$getInfo()

# CO₂ cross-check (forest only)
CO2_crosscheck_total <- CO2_crosscheck_per_pixel$
  reduceRegion(
    reducer   = ee$Reducer$sum(),
    geometry  = aoi_country_ee$geometry(),
    scale     = 500L,
    maxPixels = 1e9
  )$getInfo()

CH4_Gg <- (CH4_total[["CH4_kg"]] %||% 0) / 1e6
N2O_Gg <- (N2O_total[["N2O_kg"]] %||% 0) / 1e6
CO2_xcheck_Gg <- (CO2_crosscheck_total[["CO2_crosscheck_kg"]] %||% 0) / 1e6

cat("Honduras 2010 — GEE Pixel-Level Emissions\n")
## Honduras 2010 — GEE Pixel-Level Emissions
cat("  Total CH₄:", round(CH4_Gg, 4), "Gg\n")
##   Total CH₄: 13.3 Gg
cat("  Total N₂O:", round(N2O_Gg, 5), "Gg\n")
##   Total N₂O: 0.508 Gg
cat("  CH₄ CO₂-eq:", round(CH4_Gg * 28, 2), "Gg\n")
##   CH₄ CO₂-eq: 372 Gg
cat("  N₂O CO₂-eq:", round(N2O_Gg * 265, 2), "Gg\n")
##   N₂O CO₂-eq: 135 Gg
cat("  Total CO₂-eq (non-CO₂):",
    round(CH4_Gg * 28 + N2O_Gg * 265, 2), "Gg\n")
##   Total CO₂-eq (non-CO₂): 507 Gg
cat("\n  Forest CO₂ cross-check:",
    round(CO2_xcheck_Gg, 2), "Gg\n")
## 
##   Forest CO₂ cross-check: 2659 Gg
cat("  (Reported via stock changes, not fire emissions table)\n")
##   (Reported via stock changes, not fire emissions table)

4.10 NGHGI Reporting Format

4.10.1 IPCC Common Reporting Format (CRF)

Required tables for fire emissions:

Table 4.C - Fires on Forest Land
- Area burned (ha)
- Biomass burned (tonnes DM)
- CH₄ emissions (Gg)
- N₂O emissions (Gg)
- Note: CO₂ from forest fires is reported in the carbon stock change rows of Table 4.A, not here
Table 4.F - Fires on Grassland
- Area burned (ha)
- Biomass burned (tonnes DM)
- CH₄ emissions (Gg)
- N₂O emissions (Gg)
- CO₂: Not reported (Tier 1 synchrony assumption)
Table 4(V) - Fires in Organic Soils (Wetlands Supplement)
- Area burned (ha)
- CO₂ emissions (Gg)
- CH₄ emissions (Gg)

# Format for CRF Table 4.C (Forest Land)
crf_4c <- emissions %>%
  filter(fire_category == "Forest") %>%
  group_by(country, year) %>%
  summarise(
    biomass_burned_t = sum(biomass_burned_Gg) * 1000,
    CH4_Gg = sum(emissions_Gg[gas == "CH4"]),
    N2O_Gg = sum(emissions_Gg[gas == "N2O"]),
    .groups = "drop"
  ) %>%
  mutate(
    IPCC_Category = "4.C.1",
    Subcategory = "Fires - Forest Land",
    Tier = "Tier 1",
    CO2_note = "CO₂ reported via stock changes (Table 4.A)"
  )

# Format for CRF Table 4.F (Grassland)
crf_4f <- emissions %>%
  filter(fire_category == "Savanna") %>%
  group_by(country, year) %>%
  summarise(
    biomass_burned_t = sum(biomass_burned_Gg) * 1000,
    CH4_Gg = sum(emissions_Gg[gas == "CH4"]),
    N2O_Gg = sum(emissions_Gg[gas == "N2O"]),
    .groups = "drop"
  ) %>%
  mutate(
    IPCC_Category = "4.F.1",
    Subcategory = "Fires - Grassland",
    Tier = "Tier 1",
    CO2_note = "CO₂ excluded (Tier 1 synchrony assumption)"
  )

# Combine for full CRF submission
crf_fires <- bind_rows(crf_4c, crf_4f)

# Export
write.csv(
  crf_fires,
  "Honduras_2010_CRF_Fires.csv",
  row.names = FALSE
)

library(knitr)
kable(
  crf_fires %>%
    select(IPCC_Category, Subcategory, biomass_burned_t,
           CH4_Gg, N2O_Gg, Tier, CO2_note),
  caption = "UNFCCC CRF Tables 4.C and 4.F — Honduras 2010 Fire Emissions",
  digits = c(0, 0, 0, 4, 5, 0, 0)
)

UNFCCC CRF Tables 4.C and 4.F — Honduras 2010 Fire Emissions
IPCC_Category	Subcategory	biomass_burned_t	CH4_Gg	N2O_Gg	Tier	CO2_note
4.C.1	Fires - Forest Land	3365412	11442	337	Tier 1	CO₂ reported via stock changes (Table 4.A)
4.F.1	Fires - Grassland	1629285	1874	171	Tier 1	CO₂ excluded (Tier 1 synchrony assumption)

4.10.2 Documentation Requirements

IPCC Good Practice Guidance requires documentation of:

Data sources:
- MODIS MCD64A1 Collection 6.1 (burned area)
- MODIS MCD12Q1 Collection 6.1 (land cover)
- IPCC 2019 Refinement Tables 2.4 and 2.5
Methods:
- Tier 1 approach
- Equation 2.27 for non-CO₂ gases
- Carbon stock changes (Eq. 2.7–2.14) for forest CO₂
- Quality filters applied (uncertainty < 20%)
CO₂ accounting:
- Forest: CO₂ reported through stock changes (2006 GL §4.2.4)
- Savanna: CO₂ excluded per Tier 1 synchrony assumption (2019 Ref §2.4)
- Organic soils: CO₂ reported separately (2013 WS Ch. 2)
Uncertainty:
- Combined uncertainty: ±65-100%
- Major contributors: Fuel loads, combustion factors
- Monte Carlo analysis performed
QA/QC procedures:
- Multi-source validation (GLAD, ESRI)
- Time series consistency checks
- Comparison with independent estimates
Recalculations:
- If methodology changes: Recalculate entire time series
- Document changes in National Inventory Report (NIR)

4.11 Notes and Next Steps

4.11.1 Key Takeaways

Emission factors convert biomass burned to gas emissions:
- CH₄: 2.3-6.8 g kg^-1 (savanna < forest)
- N₂O: 0.20-0.26 g kg^-1 (similar across types)
- CO₂: 1580-1613 g kg^-1 (forest/savanna, Table 2.5)
- CO₂: 1703 g kg^-1 (organic soils, Wetlands Supp.)
CO₂ reporting:
- Forest: CO₂ is reported via carbon stock changes (not the fire emissions table); synchrony cannot be assumed for woody biomass (2006 GL §4.2.4)
- Savanna: CO₂ excluded at Tier 1 (synchrony assumed for non-woody grassland), with caveats for woody systems (2019 Ref §2.4)
- Organic soils: CO₂ reported directly (permanent carbon loss)
CO₂-equivalents:
- GWP_CH4 = 28, GWP_N2O = 265 (IPCC AR5)
- Total warming impact dominated by CH₄ (65-70%)
Uncertainty:
- Forest fires: ±68%
- Savanna fires: ±85%
- Driven by fuel consumption more than emission factors
Reporting:
- UNFCCC CRF Tables 4.C (Forest) and 4.F (Grassland)
- Forest CO₂ in stock change tables (4.A)
- Transparency documentation required

4.11.2 Chapter 5

Chapter 5: Organic Soils and Peatland Fires will expand on:

Detailed peat fire methodology (IPCC Wetlands Supplement)
Southeast Asia case studies (Indonesia, Malaysia)
Drainage impacts on fire risk and emissions
Depth of burn estimation techniques
Smoldering combustion characteristics

Alternative pathway: If organic soils are not relevant to your region, Chapter 5 could cover:

Quality control and validation procedures
Time series analysis and trend detection
Inter-annual variability and climate drivers

To confirm values, paged links to GWP values from AR5 here, AR2 here, and ~~AR6 here~~.

AR6 states: “Under the Paris Rulebook [Decision 18/CMA.1, annex, paragraph 37], parties have agreed to use GWP100 values from the IPCC AR5 or GWP100 values from a subsequent IPCC Assessment Report to reportaggregate emissions and removals of GHGs. In addition, parties may use other metrics to report supplemental information on aggregate emissions and removals of GHGs.”↩︎

--- title: "Emission Factors" execute: echo: true format: html: toc: true toc-location: right toc-depth: 3 toc-title: "**On this page**" highlight-style: pygments page-layout: article editor_options: markdown: wrap: 60 engine: knitr bibliography: ../references/references.bib csl: ../references/apa.csl citeproc: true --- ## Overview {.unnumbered} This chapter applies IPCC default emission factors (G~ef~) to convert fuel consumption estimates from Chapter 3 into greenhouse gas emissions. For forest fires, CH~4~ and N~2~O are estimated directly using Equation 2.27, while CO~2~ is accounted for through carbon stock changes (Equations 2.7-2.14) in the Forest Land chapter. For savanna fires at Tier 1, CO~2~ is assumed balanced by regrowth within one year, so only non-CO~2~ gases are reported. For organic soil fires, CO~2~ is reported separately because peat carbon is not replaced by regrowth on management-relevant timescales. ### Environment Setup {.unnumbered} ```{r} #| warning: false #| message: false #| echo: true #| comment: NA easypackages::packages( "bslib", "cols4all", "covr", "cowplot", "dendextend", "digest","DiagrammeR","dtwclust", "downlit", "e1071", "exactextractr","elevatr", "FNN", "future", "forestdata", "gdalcubes", "gdalUtilities", "geojsonsf", "geos", "ggplot2", "ggstats", "ggspatial", "ggmap", "ggplotify", "ggpubr", "ggrepel", "giscoR", "hdf5r", "httr", "httr2", "htmltools", "jsonlite", "kohonen", "knitr", "leaflet.providers", "leafem", "libgeos","luz","lwgeom", "leaflet", "leafgl", "mapedit", "mapview", "maptiles", "methods", "mgcv", "ncdf4", "nnet", "openxlsx", "parallel", "plotly", "randomForest", "rasterVis", "raster", "Rcpp", "RcppArmadillo", "RcppCensSpatial","rayshader", "RcppEigen", "RcppParallel", "RColorBrewer", "reactable", "rgl", "rsconnect","RStoolbox", "rts", "reticulate", "s2", "sf", "scales", "sits","spdep", "stars", "stringr","supercells", "terra", "testthat", "tidyverse", "tidyterra","tools", "tmap", "tmaptools", "terrainr", "xgboost", prompt = F) #mapviewOptions(fgb = FALSE) sf::sf_use_s2(use_s2 = FALSE) ``` ```{r} #| warning: false #| message: false #| echo: false #| comment: NA options(repos = c(CRAN = "https://cloud.r-project.org"), htmltools.dir.version = FALSE, htmltools.preserve.raw = FALSE ) knitr::opts_chunk$set( echo = TRUE, message = FALSE, warning = FALSE, error = FALSE, comment = NA, tidy.opts = list(width.cutoff = 60) ) ``` ```{css, echo=FALSE, class.source = 'foldable'} div.column { display: inline-block; vertical-align: top; width: 50%; } #TOC::before { content: ""; display: block; height:200px; width: 200px; background-image: url('https://raw.githubusercontent.com/seamusrobertmurphy/IPCC-wildfire-emissions/refs/heads/main/04-emission-factors/assets/ef-uncertainty-aa.png'); background-size: contain; background-position: 50% 50%; padding-top: 80px !important; background-repeat: no-repeat; } ``` ------------------------------------------------------------ ## IPCC Emission Factors {#sec-emissions-overview} ### Completing Equation 2.27 {#sec-emissions-equation} Emission factors represent the final component of the IPCC fire emissions equation: $$ L_{fire} = A \times M_B \times C_f \times G_{ef} \times 10^{-3} $$ Where: - G~ef~ = Emission factor (g gas per kg dry matter burned) - L~fire~ = Total emissions (tonnes of gas) - 10^-3^ = Unit conversion factor (g to kg, kg to tonnes) Expanded form: $$ L_{fire} \text{ (tonnes)} = \text{Area (ha)} \times \frac{\text{Biomass (t DM)}}{\text{ha}} \times \frac{\text{g gas}}{\text{kg DM}} \times 10^{-3} $$ ### IPCC Emission Factor Tables {#sec-emissions-tables} Key IPCC sources: - Table 2.5 (2019 Refinement): Emission factors for CO~2~, CH~4~, N~2~O, CO, and NO~x~ from fires in forests, savanna, and other vegetation types - Wetlands Supplement Table 2.6: CO~2~ and CH~4~ emission factors for fires on organic soils - Chapter 4.2 (2006 Guidelines): Carbon stock change methods for forest CO~2~ accounting ### IPCC Default Emission Factors {#sec-emissions-defaults} Forest Fires (IPCC 2019 Refinement, Table 2.5): | Forest Type | CO~2~ (g kg^-1^) | CH~4~ (g kg^-1^) | N~2~O (g kg^-1^) | CO (g kg^-1^)\* | NO~x~ (g kg^-1^)\* | |----------|----------|----------|----------|----------|----------| | Tropical Forest | 1580 ± 90 | 6.8 ± 2.0 | 0.20 ± 0.10 | 104 ± 20 | 3.9 ± 2.4 | | Extratropical Forest | 1569 ± 131 | 4.7 ± 1.9 | 0.26 ± 0.15 | 107 ± 37 | 3.0 ± 1.4 | \*Optional precursor gases for Tier 1 Savanna Fires (IPCC 2019 Refinement, Table 2.5): | Vegetation Type | CO~2~ (g kg^-1^) | CH~4~ (g kg^-1^) | N~2~O (g kg^-1^) | CO (g kg^-1^)\* | NO~x~ (g kg^-1^)\* | |----------|----------|----------|----------|----------|----------| | Savanna/Grassland | 1613 ± 95 | 2.3 ± 0.9 | 0.21 ± 0.10 | 65 ± 20 | 3.9 ± 2.4 | Note: Savanna CO~2~ emission factor is listed for completeness but is not applied at Tier 1 due to the synchrony assumption (see @sec-co2-logic). Organic Soil Fires (IPCC 2013 Wetlands Supplement, Table 2.6): | Soil Type | CO~2~ (g kg^-1^) | CH~4~ (g kg^-1^) | N~2~O (g kg^-1^) | |------------------|--------------|--------------|--------------| | Tropical Peatland | 1703 ± 8 | 5.7 ± 5.4 | Not estimated | ------------------------------------------------------------ ## GEE Initialisation {#sec-emissions-gee-init} ```{r} #| label: emissions-gee-setup #| warning: false #| message: false #| error: false # Point environment to your configured python reticulate::use_python("/opt/local/bin/python", required=T) # Initialise Earth Engine ee <- reticulate::import("ee") ee$Initialize(project = "murphys-deforisk") cat("GEE initialised\n") # Helper: visualize EE tile in tmap/leaflet ee_tile_url <- function(ee_image, vis_params) { ee_image$getMapId(vis_params)$tile_fetcher$url_format } # Helper: EE FeatureCollection => sf ee_to_sf <- function(fc) { geojsonsf::geojson_sf( jsonlite::toJSON(fc$getInfo(), auto_unbox = TRUE)) } # Load AOI boundaries from FAO GAUL aoi_country_ee <- ee$FeatureCollection("FAO/GAUL/2015/level0")$ filter(ee$Filter$eq("ADM0_NAME", "Honduras")) aoi_states_ee <- ee$FeatureCollection("FAO/GAUL/2015/level1")$ filter(ee$Filter$eq("ADM0_NAME", "Honduras")) aoi_country_sf <- ee_to_sf(aoi_country_ee) aoi_states_sf <- ee_to_sf(aoi_states_ee) cat("AOI loaded:", aoi_country_ee$size()$getInfo(), "country,", aoi_states_ee$size()$getInfo(), "states\n") ``` ### Upstream Inputs from Chapters 1-3 {#sec-emissions-upstream} ```{r} #| label: emissions-upstream-inputs #| code-summary: "Load burned area (Ch1), vegetation masks (Ch2), fuel consumption (Ch3)" # --- Chapter 1: Burned area (MCD64A1) --- mcd64 <- ee$ImageCollection("MODIS/061/MCD64A1") burned_2010 <- mcd64$ filterBounds(aoi_country_ee$geometry())$ filterDate("2010-01-01", "2010-12-31") # FAOSTAT quality filter (matching Ch1) faostat_filter <- function(image) { unc <- image$select("Uncertainty") burn <- image$select("BurnDate") mask <- unc$lt(20L)$And(burn$gt(0L)) image$updateMask(mask) } # Annual burned mask for 2010 annual_burned_2010 <- burned_2010$ map(faostat_filter)$ select("BurnDate")$max()$gt(0L)$selfMask() # --- Chapter 2: Land cover & vegetation masks --- lc_2010 <- ee$ImageCollection("MODIS/061/MCD12Q1")$ select("LC_Type1")$ filterDate("2010-01-01", "2010-12-31")$ first()$ clip(aoi_country_ee$geometry()) # Forest mask (IGBP 1-5) forest_mask <- lc_2010$gte(1L)$And(lc_2010$lte(5L)) # Savanna/grassland mask (IGBP 6-10) savanna_mask <- lc_2010$gte(6L)$And(lc_2010$lte(10L)) # Climate-based humid tropical delineation bio <- ee$Image("WORLDCLIM/V1/BIO") precip <- bio$select("bio12") temp <- bio$select("bio01") tropical <- temp$gt(180L) moist <- precip$gt(1500L) humid_tropical_mask <- tropical$And(moist) # Forest sub-types humid_tropical_forest <- forest_mask$And(humid_tropical_mask) other_forest <- forest_mask$And(humid_tropical_mask$Not()) # Climate zones boreal <- temp$lt(0L) temperate <- temp$gte(0L)$And(temp$lte(180L)) # --- Chapter 3: Fuel consumption raster --- fuel_consumption <- ee$Image(0)$rename("fuel_consumption") htf_mask <- forest_mask$And(humid_tropical_mask) fuel_consumption <- fuel_consumption$where(htf_mask, 43.1) forest_boreal <- forest_mask$And(boreal)$And(humid_tropical_mask$Not()) fuel_consumption <- fuel_consumption$where(forest_boreal, 18.5) forest_temperate <- forest_mask$And(temperate)$And(humid_tropical_mask$Not()) fuel_consumption <- fuel_consumption$where(forest_temperate, 23.7) forest_tropical <- forest_mask$And(tropical)$And(humid_tropical_mask$Not()) fuel_consumption <- fuel_consumption$where(forest_tropical, 32.6) fuel_consumption <- fuel_consumption$where(savanna_mask, 7.0) # Apply to burned areas fuel_burned <- fuel_consumption$ updateMask(annual_burned_2010)$ clip(aoi_country_ee$geometry()) # Pixel area helper pixelArea_ha <- ee$Image$pixelArea()$divide(10000L) # Biomass burned per pixel (t DM) biomass_per_pixel <- fuel_burned$multiply(pixelArea_ha) cat("Upstream inputs loaded: MCD64A1 + MCD12Q1 + WorldClim + fuel raster\n") ``` ------------------------------------------------------------ ## Methane (CH~4~) Emissions {#sec-emissions-ch4} ### CH~4~ Emission Mechanisms {#sec-ch4-mechanisms} Formation process: Incomplete combustion under oxygen-limited conditions $$ \text{Biomass} \xrightarrow{\text{Incomplete combustion}} \text{CH}_4 + \text{CO}_2 + \text{CO} + \text{char} $$ Key factors affecting CH~4~ emissions: 1. Fire intensity: - Low-intensity fires = More smoldering = Higher CH~4~ - High-intensity fires = More flaming = Lower CH~4~ 2. Fuel moisture: - Wet fuels = Lower combustion temperature = Higher CH~4~ - Dry fuels = Higher temperature = Lower CH~4~ 3. Fuel type: - Forest: High CH~4~ (dense, moist fuels) - Savanna: Low CH~4~ (grass cures completely) 4. Combustion phase: - Flaming: Minimal CH~4~ (\>90% CO~2~) - Smoldering: High CH~4~ (20-40% of carbon as CH~4~) ### Forest CH~4~ Emissions {#sec-ch4-forest} #### Emission Factors by Forest Type {#sec-ch4-forest-ef} Tropical Forests (6.8 g CH~4~ kg^-1^): - Highest emission factors - Dense canopy \>\> poor ventilation - High fuel moisture \>\> low combustion efficiency - Long smoldering phase (especially coarse woody debris) Temperate/Boreal Forests (4.7 g CH~4~ kg^-1^): - Moderate emission factors - More complete combustion in conifers - Resinous fuels burn hotter - Less smoldering than tropical Uncertainty: ±30-40% (IPCC Table 2.5) #### Calculation Example {#sec-ch4-forest-calc} Honduras Tropical Forest Fires (2010): Given: - Biomass burned from Chapter 3 (GEE-derived) - Emission factor: 6.8 g CH~4~ kg^-1^ DM (tropical forest) Calculation: $$ \begin{aligned} L_{CH_4} &= \text{Biomass} \times G_{ef} \times 10^{-3} \\ &= \text{Biomass (t DM)} \times 1000 \text{ (kg/t)} \times 6.8 \text{ (g kg}^{-1}\text{)} \times 10^{-6} \text{ (t/g)} \\ &= \text{Biomass (Gg)} \times 6.8 \text{ Gg CH}_4 \end{aligned} $$ ```{r} #| label: ch4-forest-calculation #| code-summary: "Calculate forest CH₄ emissions — Honduras 2010" # Compute forest biomass from GEE forest_biomass <- biomass_per_pixel$ updateMask(forest_mask)$ reduceRegion( reducer = ee$Reducer$sum(), geometry = aoi_country_ee$geometry(), scale = 500L, maxPixels = 1e9 )$getInfo() forest_biomass_t <- forest_biomass[["fuel_consumption"]] %||% 0 forest_biomass_Gg <- forest_biomass_t / 1000 # IPCC emission factor (tropical forest) EF_CH4_tropical_forest <- 6.8 # g kg⁻¹ # Calculate CH₄ emissions # Gg DM × g/kg = Gg × g/kg (numerically equivalent) forest_CH4_Gg <- forest_biomass_Gg * EF_CH4_tropical_forest cat("Honduras 2010 — Forest CH₄ Emissions\n") cat(" Forest biomass burned:", round(forest_biomass_Gg, 2), "Gg DM\n") cat(" EF (tropical): ", EF_CH4_tropical_forest, "g kg⁻¹\n") cat(" CH₄ emissions: ", round(forest_CH4_Gg, 3), "Gg CH₄\n") ``` ### Savanna CH~4~ Emissions {#sec-ch4-savanna} #### Lower Emission Factors {#sec-ch4-savanna-ef} Savanna/Grassland (2.3 g CH~4~ kg^-1^): - Lowest emission factors among all vegetation types - Grass fuels cure completely (low moisture) - Predominantly flaming combustion - Minimal smoldering phase - High combustion efficiency Why savanna CH~4~ \< forest CH~4~: 1\. Fuel moisture: 5-10% (savanna) vs. 30-60% (forest) 2\. Fuel bed depth: 10-30 cm (savanna) vs. 1-10 m (forest) 3\. Combustion mode: 80% flaming (savanna) vs. 40% flaming (forest) #### Calculation Example {#sec-ch4-savanna-calc} Honduras Savanna Fires (2010): ```{r} #| label: ch4-savanna-calculation #| code-summary: "Calculate savanna CH₄ emissions — Honduras 2010" # Compute savanna biomass from GEE savanna_biomass <- biomass_per_pixel$ updateMask(savanna_mask)$ reduceRegion( reducer = ee$Reducer$sum(), geometry = aoi_country_ee$geometry(), scale = 500L, maxPixels = 1e9 )$getInfo() savanna_biomass_t <- savanna_biomass[["fuel_consumption"]] %||% 0 savanna_biomass_Gg <- savanna_biomass_t / 1000 # IPCC emission factor EF_CH4_savanna <- 2.3 # g kg⁻¹ # Calculate CH₄ emissions savanna_CH4_Gg <- savanna_biomass_Gg * EF_CH4_savanna cat("Honduras 2010 — Savanna CH₄ Emissions\n") cat(" Savanna biomass burned:", round(savanna_biomass_Gg, 3), "Gg DM\n") cat(" EF (savanna): ", EF_CH4_savanna, "g kg⁻¹\n") cat(" CH₄ emissions: ", round(savanna_CH4_Gg, 4), "Gg CH₄\n") ``` ### Organic Soil CH~4~ Emissions {#sec-ch4-organic} Tropical Peatland (5.7 g CH~4~ kg^-1^): - Intermediate emission factors - Smoldering combustion dominates - Can burn underground for months - High variability (CV \~95%) Southeast Asia only (per FAOSTAT): ```{r} #| eval: true #| label: ch4-organic-calculation #| code-summary: "Calculate organic soil CH₄ emissions (reference example)" # Example: Indonesia peat fires (reference — not Honduras) indonesia_peat <- data.frame( country = "Indonesia", year = 2015, fire_category = "Organic_Soil", biomass_burned_Gg = 125 # Example ) EF_CH4_peat <- 5.7 # g kg⁻¹ indonesia_ch4 <- indonesia_peat %>% mutate(CH4_Gg = biomass_burned_Gg * EF_CH4_peat) cat("Reference — Indonesia peat CH₄:", indonesia_ch4$CH4_Gg, "Gg\n") cat("Note: Honduras has negligible peatland fire activity\n") ``` ------------------------------------------------------------ ## Nitrous Oxide (N~2~O) Emissions {#sec-emissions-n2o} ### N~2~O Emission Mechanisms {#sec-n2o-mechanisms} Formation process: Incomplete oxidation of nitrogen in biomass $$ \text{Biomass-N} \xrightarrow{\text{Combustion}} \text{N}_2\text{O} + \text{NO}_x + \text{N}_2 + \text{NH}_3 $$ Key factors: 1. Nitrogen content of fuel: - Legumes: High N \>\> High N~2~O - Conifers: Low N \>\> Low N~2~O - Grasses: Variable N (fertilization-dependent) 2. Combustion temperature: - Low temp (400-600°C): High N~2~O/NO~x~ ratio - High temp (\>800°C): Low N~2~O (converts to N~2~) 3. Fuel moisture: - Affects temperature = Affects N~2~O production 4. Fire residency time: - Long smoldering = More N~2~O ### Forest N~2~O Emissions {#sec-n2o-forest} #### Emission Factors by Forest Type {#sec-n2o-forest-ef} Tropical Forests (0.20 g N~2~O kg^-1^): - Lower emission factors (surprising given high N content) - Explanation: High temperature flaming combustion - More N~2~ produced instead of N~2~O Temperate/Boreal Forests (0.26 g N~2~O kg^-1^): - Moderate emission factors - Conifer needles: Low N content (\~0.5%) - Moderate combustion temperatures Uncertainty: ±50-60% (highest among all emission factors) #### Calculation Example {#sec-n2o-forest-calc} Honduras Tropical Forest Fires (2010): ```{r} #| label: n2o-forest-calculation #| code-summary: "Calculate forest N₂O emissions — Honduras 2010" # IPCC emission factor (tropical forest) EF_N2O_tropical_forest <- 0.20 # g kg⁻¹ # Calculate N₂O emissions forest_N2O_Gg <- forest_biomass_Gg * EF_N2O_tropical_forest cat("Honduras 2010 — Forest N₂O Emissions\n") cat(" Forest biomass burned:", round(forest_biomass_Gg, 2), "Gg DM\n") cat(" EF (tropical): ", EF_N2O_tropical_forest, "g kg⁻¹\n") cat(" N₂O emissions: ", round(forest_N2O_Gg, 4), "Gg N₂O\n") ``` ### Savanna N~2~O Emissions {#sec-n2o-savanna} Savanna/Grassland (0.21 g N~2~O kg^-1^): - Similar to forest emission factors - Nitrogen variability depends on: - Grazing intensity (manure inputs) - Proximity to croplands (fertilizer drift) - Legume abundance ```{r} #| label: n2o-savanna-calculation #| code-summary: "Calculate savanna N₂O emissions — Honduras 2010" # IPCC emission factor EF_N2O_savanna <- 0.21 # g kg⁻¹ # Calculate N₂O emissions savanna_N2O_Gg <- savanna_biomass_Gg * EF_N2O_savanna cat("Honduras 2010 — Savanna N₂O Emissions\n") cat(" Savanna biomass burned:", round(savanna_biomass_Gg, 3), "Gg DM\n") cat(" EF (savanna): ", EF_N2O_savanna, "g kg⁻¹\n") cat(" N₂O emissions: ", round(savanna_N2O_Gg, 5), "Gg N₂O\n") ``` ### Organic Soil N~2~O Emissions {#sec-n2o-organic} NOT ESTIMATED for organic soil fires (IPCC 2013 Wetlands Supplement): > "Due to lack of data, no default N₂O emission factor is > provided for fires in organic soils." Rationale: - Very few measurements available - Peat nitrogen content highly variable - Smoldering combustion \>\> complex N chemistry - Most N released as N~2~ or NH~3~ rather than N~2~O FAOSTAT approach: N~2~O = 0 for organic soil fires ------------------------------------------------------------ ## Carbon Dioxide (CO~2~) Emissions {#sec-emissions-co2} ### CO~2~ Reporting Logic {#sec-co2-logic} The treatment of CO~2~ differs by fire type under IPCC methodology. The key distinction is whether fire CO~2~ emissions are synchronous with CO~2~ removals from regrowth: | Fire Category | CO~2~ Reported? | Accounting Method | IPCC Source | |---------------|---------------|---------------|---------------| | Forest Fires | Yes | Carbon stock changes (Eq. 2.7--2.14) | 2006 GL §4.2.4 | | Savanna Fires | No (Tier 1) | Synchrony assumed for non-woody grassland | 2019 Ref §2.4 | | Organic Soil Fires | Yes | Separate G~ef~ (Eq. 2.27) | 2013 WS Ch. 2 | ### Forest CO~2~ Treatment {#sec-co2-forest} #### Why forest fire CO~2~ must be reported The 2006 IPCC Guidelines (Vol. 4, Ch. 4, §4.2.4) state: > "In Forest Land Remaining Forest Land, emissions of CO~2~ > from biomass burning also need to be accounted for because > they are generally not synchronous with rates of CO~2~ > uptake." The 2019 Refinement (Vol. 4, Ch. 2, §2.4) reinforces: > "For Forest Land, synchrony is unlikely if significant > woody biomass is killed (i.e., losses represent several > years of growth and C accumulation), and the net emissions > should be reported." #### Accounting pathway: Carbon stock changes Forest fire CO~2~ is captured through the carbon stock change framework (Gain-Loss Method, Equations 2.7--2.14), not through the fire-specific Equation 2.27: - Biomass loss from fire = immediate CO~2~ emission (captured by Equation 2.14 as a disturbance loss) - Regrowth = gradual CO~2~ removal (captured by Equation 2.9 as annual biomass increment) - Net change reported annually in the Forest Land carbon stock change category Non-CO~2~ gases (CH~4~, N~2~O) cannot be captured by stock change accounting and must be estimated separately using Equation 2.27 and the G~ef~ values from Table 2.5. #### Cross-check: CO~2~ emission factor from Table 2.5 Table 2.5 lists CO~2~ emission factors for forests (Tropical: 1580 ± 90 g kg^-1^; Extratropical: 1569 ± 131 g kg^-1^). These can serve as a cross-check against the stock-change estimate: ```{r} #| label: co2-forest-crosscheck #| code-summary: "Forest CO₂ cross-check using Table 2.5 Gef — Honduras 2010" # Table 2.5 CO₂ emission factor (tropical forest) EF_CO2_tropical_forest <- 1580 # g kg⁻¹ # Cross-check estimate forest_CO2_crosscheck_Gg <- forest_biomass_Gg * EF_CO2_tropical_forest cat("Honduras 2010 — Forest CO₂ Cross-Check\n") cat(" Forest biomass burned:", round(forest_biomass_Gg, 2), "Gg DM\n") cat(" EF (tropical Table 2.5):", EF_CO2_tropical_forest, "g kg⁻¹\n") cat(" CO₂ cross-check: ", round(forest_CO2_crosscheck_Gg, 2), "Gg CO₂\n") cat("\n Note: This value is for cross-checking only.\n") cat(" Forest CO₂ is reported through carbon stock changes\n") cat(" (Equations 2.7–2.14), not the fire emissions table.\n") ``` ### Savanna CO~2~ Treatment {#sec-co2-savanna} #### Tier 1 synchrony assumption At Tier 1, CO~2~ from savanna and grassland fires is not reported separately because regrowth is assumed to reabsorb the emitted CO~2~ within one year. The 2019 Refinement (Vol. 4, Ch. 2, §2.4) states: > "For grassland biomass burning and burning of agriculture > residues, the assumption of equivalence is generally > reasonable." #### Caveats for woody savannas The Guidelines explicitly warn against applying synchrony to woody systems: > "woody vegetation may also burn in these land categories, > and greenhouse gas emissions from those sources should be > reported using a higher Tier method" and: > "care must be taken before assuming synchrony in such > systems" For Honduras Tier 1, the synchrony assumption is applied to savanna/grassland pixels (IGBP classes 6-10). Countries with significant woody savanna burning should consider Tier 2 methods. ### Organic Soil CO~2~ Emissions {#sec-co2-organic} #### Why Organic Soil CO~2~ is Different {#sec-co2-organic-rationale} Peat = Non-renewable carbon source: - Peat accumulation: 0.1-1.0 mm year^-1^ - Fire consumption: 100-500 mm per event - Recovery time: 100-5,000 years - Conclusion: Fire CO~2~ NOT balanced by regrowth IPCC Wetlands Supplement: > "Emissions of CO₂ from peat fires should be estimated > separately from other fire types, as the carbon in peat > has accumulated over centuries to millennia and is not > rapidly replaced by regrowth." #### Emission Factor {#sec-co2-organic-ef} Tropical Peatland (1703 g CO~2~ kg^-1^): $$ \begin{aligned} \text{Carbon content of peat} &= 0.50 \text{ kg C per kg DM} \\ \text{Molecular weight ratio} &= \frac{44 \text{ (CO}_2)}{12 \text{ (C)}} = 3.67 \\ G_{ef,CO_2} &= 0.50 \times 3.67 \times 1000 \text{ g kg}^{-1} \\ &= 1835 \text{ g CO}_2 \text{ kg}^{-1} \text{ DM} \end{aligned} $$ IPCC default: 1703 g kg^-1^ - Lower value accounting for incomplete carbon conversion) - Uncertainty: Very low (±8 g kg^-1^) - CO~2~/C ratio is stoichiometric ------------------------------------------------------------ ## Reproducible Workflow {#sec-emissions-workflow} ### Full Script {#sec-emissions-pipeline} ```{r} #| label: emissions-complete-workflow #| code-summary: "End-to-end emissions calculation — Honduras 2010" library(tidyverse) # 1. Load IPCC Emission Factors # Non-CO₂ gases estimated via Equation 2.27 # Forest CO₂ estimated via stock changes (not included here) # Savanna CO₂ excluded per Tier 1 synchrony assumption emission_factors <- tribble( ~fire_category, ~climate_zone, ~gas, ~EF_g_kg, ~uncertainty_pct, ~accounting_note, # Forest Fires — non-CO₂ via Equation 2.27 "Forest", "Tropical", "CH4", 6.8, 30, "Eq 2.27", "Forest", "Tropical", "N2O", 0.20, 50, "Eq 2.27", "Forest", "Temperate", "CH4", 4.7, 40, "Eq 2.27", "Forest", "Temperate", "N2O", 0.26, 60, "Eq 2.27", "Forest", "Boreal", "CH4", 4.7, 40, "Eq 2.27", "Forest", "Boreal", "N2O", 0.26, 60, "Eq 2.27", # Forest CO₂ — cross-check only (primary via stock changes) "Forest", "Tropical", "CO2", 1580, 6, "Cross-check (stock changes primary)", "Forest", "Temperate", "CO2", 1569, 8, "Cross-check (stock changes primary)", "Forest", "Boreal", "CO2", 1569, 8, "Cross-check (stock changes primary)", # Savanna Fires — non-CO₂ only at Tier 1 "Savanna", "All", "CH4", 2.3, 40, "Eq 2.27", "Savanna", "All", "N2O", 0.21, 50, "Eq 2.27", # Organic Soil Fires — CO₂ reported (non-renewable C) "Organic_Soil", "Tropical", "CO2", 1703, 0.5, "Eq 2.27 (WS)", "Organic_Soil", "Tropical", "CH4", 5.7, 95, "Eq 2.27 (WS)" ) # 2. Build fuel consumption data from GEE results (Ch3) fuel_consumption_data <- tribble( ~country, ~year, ~fire_category, ~vegetation_type, ~climate_zone, ~area_ha, ~biomass_burned_Gg, "Honduras", 2010, "Forest", "Humid_Tropical_Forest", "Tropical", NA_real_, forest_biomass_Gg, "Honduras", 2010, "Savanna", "Savanna_Mixed", "All", NA_real_, savanna_biomass_Gg ) # 3. Calculate Emissions by Gas # Exclude forest CO₂ cross-check rows from primary totals emissions <- fuel_consumption_data %>% left_join( emission_factors %>% filter(accounting_note == "Eq 2.27"), by = c("fire_category", "climate_zone") ) %>% mutate( emissions_Gg = biomass_burned_Gg * EF_g_kg ) %>% select(country, year, fire_category, vegetation_type, climate_zone, biomass_burned_Gg, gas, EF_g_kg, emissions_Gg, uncertainty_pct) # 4. Summarize by GHG & Category emissions_summary <- emissions %>% group_by(country, year, fire_category, gas) %>% summarise( total_biomass_Gg = sum(biomass_burned_Gg), total_emissions_Gg = sum(emissions_Gg), .groups = "drop" ) print(emissions_summary) # 5. Convert to CO₂-eq using AR5 GWPs (without feedbacks) GWP_CH4 <- 28 GWP_N2O <- 265 emissions_co2eq <- emissions_summary %>% mutate( CO2eq_Gg = case_when( gas == "CO2" ~ total_emissions_Gg, gas == "CH4" ~ total_emissions_Gg * GWP_CH4, gas == "N2O" ~ total_emissions_Gg * GWP_N2O, TRUE ~ 0 ) ) # Total CO₂-equivalent emissions total_co2eq <- emissions_co2eq %>% group_by(country, year) %>% summarise( total_CO2eq_Gg = sum(CO2eq_Gg), total_CO2eq_Mt = sum(CO2eq_Gg) / 1000, .groups = "drop" ) print(total_co2eq) # 6. Export write.csv( emissions, "Honduras_FireEmissions_2010_detailed.csv", row.names = FALSE ) write.csv( emissions_co2eq, "Honduras_FireEmissions_2010_CO2eq.csv", row.names = FALSE ) ``` ```{r} #| label: emissions-honduras-summary-table #| code-summary: "Summary tables — Honduras 2010" library(knitr) kable( emissions_co2eq %>% select(fire_category, gas, total_emissions_Gg, CO2eq_Gg), caption = "Honduras 2010 Fire Emissions by GHG & Category (Equation 2.27 gases only)", digits = 4 ) kable( total_co2eq, caption = "Honduras 2010 Tier 1 Total Fire Emissions (non-CO₂ gases)", digits = 4 ) ``` ------------------------------------------------------------ ## Global Warming Potentials and CO₂-Equivalents {#sec-emissions-gwp} ### GWP Concept {#sec-gwp-concept} Global Warming Potential (GWP): Radiative forcing of 1 kg of gas relative to 1 kg of CO₂ over 100-year, GWP100 [@IPCC2014, pp. 87]: IPCC Assessment Report 5 (AR5) - 100-year GWPs: | Gas | GWP (w/o feedback) | GWP (w/ feedback) | |-------|--------------------|-------------------| | CO~2~ | 1 (by definition) | 1 | | CH~4~ | 28 | 34 | | N~2~O | 265 | 298 | FAOSTAT uses AR5 without feedbacks: GWP~CH4~ = 28, GWP~N2O~ = 265 IPCC Assessment Report 6 (AR6):[^index-1] | Gas | GWP (100-year) | |-------|----------------| | CH~4~ | 27.9 | | N~2~O | 273 | Reporting guidance: Use GWP set consistent with national reporting (typically AR5 for UNFCCC submissions through \~2024) ### Calculating CO~2~-Equivalents {#sec-gwp-calculation} $$ \text{CO}_2\text{-eq (Gg)} = \sum_{i=1}^{n} \text{Emissions}_i \text{ (Gg)} \times \text{GWP}_i $$ ```{r} #| label: co2eq-calculation #| code-summary: "Convert emissions to CO₂-equivalents — Honduras 2010" library(tidyverse) # Aggregate emissions by gas emissions_by_gas <- emissions_co2eq %>% group_by(gas) %>% summarise( emissions_Gg = sum(total_emissions_Gg), CO2eq_Gg = sum(CO2eq_Gg), .groups = "drop" ) cat("Honduras 2010 — CO₂-Equivalents (Eq 2.27 gases)\n") for (i in seq_len(nrow(emissions_by_gas))) { cat(sprintf(" %s: %.4f Gg → %.2f Gg CO₂-eq\n", emissions_by_gas$gas[i], emissions_by_gas$emissions_Gg[i], emissions_by_gas$CO2eq_Gg[i])) } cat(" Total:", round(sum(emissions_by_gas$CO2eq_Gg), 2), "Gg CO₂-eq\n") cat("\n Note: Forest CO₂ is reported via carbon stock\n") cat(" changes (Ch. 4.2), not included in this total.\n") # Visualize ggplot(emissions_by_gas, aes(x = gas, y = CO2eq_Gg, fill = gas)) + geom_col() + geom_text(aes(label = paste0(round(CO2eq_Gg, 1), " Gg")), vjust = -0.5, fontface = "bold") + scale_fill_manual(values = c("CH4" = "orange", "N2O" = "red")) + labs( title = "Fire Emissions by Gas (CO₂-equivalents)", subtitle = "Honduras 2010 — Forest and Savanna Fires (Eq 2.27 non-CO₂ gases)", x = "Greenhouse Gas", y = "Emissions (Gg CO₂-eq)", fill = "Gas" ) + theme_minimal() + theme(legend.position = "none") ``` ------------------------------------------------------------ ## Uncertainty Analysis {#sec-emissions-uncertainty} ### Sources of Uncertainty {#sec-uncertainty-sources} 1\. Emission Factor Uncertainty (IPCC Table 2.5): | Gas | Forest (%) | Savanna (%) | Organic Soil (%) | |-------|------------|-------------|------------------| | CO~2~ | ±6 | ±6 | ±0.5 | | CH~4~ | ±30-40 | ±40 | ±95 | | N~2~O | ±50-60 | ±50 | N/A | Note: Forest CO~2~ emission factor uncertainty (±6%) is relevant for cross-checking. The total uncertainty in forest CO~2~ reporting is dominated by stock-change uncertainty (biomass estimates, allometric models), not G~ef~. 2\. Cumulative Uncertainty (from all chapters): Recall from Chapter 3: $$ U_{total}^2 = U_A^2 + U_{MB}^2 + U_{Cf}^2 + U_{Gef}^2 $$ For Forest CH~4~ emissions: - $U_A$ = ±30% (burned area) - $U_{MB}$ = ±50% (fuel load) - $U_{Cf}$ = ±25% (combustion factor) - $U_{Gef}$ = ±35% (CH~4~ emission factor) Combined uncertainty: $$ U_{total} = \sqrt{30^2 + 50^2 + 25^2 + 35^2} = \sqrt{4650} \approx 68\% $$ For Savanna emissions: \~85% uncertainty For Organic soil CO~2~: \~60% uncertainty (dominated by fuel consumption) ### Monte Carlo Uncertainty Propagation {#sec-uncertainty-montecarlo} ```{r} #| label: uncertainty-montecarlo #| code-summary: "Monte Carlo simulation — Honduras tropical forest CH₄" # Parameters: Honduras Humid Tropical Forest n_simulations <- 10000 # Mean values (from Ch1-3 GEE results) MB_mean <- 119.6 # t DM ha⁻¹ (humid tropical) Cf_mean <- 0.36 EF_CH4_mean <- 6.8 # g kg⁻¹ (tropical forest) # Uncertainties (SD = 95% CI / 1.96) MB_sd <- 52.4 / 1.96 Cf_sd <- 0.09 / 1.96 EF_CH4_sd <- 2.0 / 1.96 # Generate random samples (area fixed from GEE) set.seed(42) simulations <- data.frame( MB = rnorm(n_simulations, MB_mean, MB_sd), Cf = rnorm(n_simulations, Cf_mean, Cf_sd), EF_CH4 = rnorm(n_simulations, EF_CH4_mean, EF_CH4_sd) ) %>% mutate( MB = pmax(0, MB), Cf = pmin(1, pmax(0, Cf)), EF_CH4 = pmax(0, EF_CH4), # MB_Cf = fuel consumed (t DM ha⁻¹) # Emissions per ha = MB × Cf × EF × 10⁻³ (kg CH₄ ha⁻¹) CH4_kg_per_ha = MB * Cf * EF_CH4, # Relative to central estimate ratio = CH4_kg_per_ha / (MB_mean * Cf_mean * EF_CH4_mean) ) # Summary statistics ratio_mean <- mean(simulations$ratio) ratio_ci_lower <- quantile(simulations$ratio, 0.025) ratio_ci_upper <- quantile(simulations$ratio, 0.975) cat(sprintf("Honduras HTF CH₄ — Relative uncertainty\n")) cat(sprintf(" Mean ratio: %.2f\n", ratio_mean)) cat(sprintf(" 95%% CI: [%.2f, %.2f]\n", ratio_ci_lower, ratio_ci_upper)) cat(sprintf(" i.e. -%d%% to +%d%% around central estimate\n", round((1 - ratio_ci_lower) * 100), round((ratio_ci_upper - 1) * 100))) # Visualize distribution ggplot(simulations, aes(x = CH4_kg_per_ha)) + geom_histogram(bins = 50, fill = "steelblue", alpha = 0.7) + geom_vline(xintercept = MB_mean * Cf_mean * EF_CH4_mean, color = "red", linewidth = 1, linetype = "dashed") + labs( title = "Uncertainty Distribution: Tropical Forest CH₄ Emissions", subtitle = "Honduras — Monte Carlo (10,000 simulations)", x = "CH₄ Emissions (kg per ha burned)", y = "Frequency" ) + theme_minimal() ``` ------------------------------------------------------------ ## Spatial Emissions Mapping in GEE {#sec-emissions-gee} ### Pixel-Level Emissions Calculation {#sec-emissions-gee-pixel} ```{r} #| label: emissions-gee-spatial #| code-summary: "Create spatially-explicit emissions maps — Honduras 2010" # 1. Derive Emission Factor Rasters (non-CO₂ gases via Eq 2.27) # CH₄ emission factors by vegetation type EF_CH4_img <- ee$Image(0)$rename("EF_CH4") EF_CH4_img <- EF_CH4_img$ where(forest_mask$And(tropical), 6.8)$ where(forest_mask$And(tropical$Not()), 4.7)$ where(savanna_mask, 2.3) # N₂O emission factors EF_N2O_img <- ee$Image(0)$rename("EF_N2O") EF_N2O_img <- EF_N2O_img$ where(forest_mask$And(tropical), 0.20)$ where(forest_mask$And(tropical$Not()), 0.26)$ where(savanna_mask, 0.21) # CO₂ emission factor raster (cross-check for forest only) EF_CO2_img <- ee$Image(0)$rename("EF_CO2") EF_CO2_img <- EF_CO2_img$ where(forest_mask$And(tropical), 1580)$ where(forest_mask$And(tropical$Not()), 1569) # 2. Calculate Emissions Per Pixel # CH₄ emissions (kg): biomass (t) × 1000 (kg/t) × EF (g/kg) / 1000 (g→kg) CH4_per_pixel <- biomass_per_pixel$ multiply(EF_CH4_img)$ rename("CH4_kg")$ clip(aoi_country_ee$geometry()) # N₂O emissions (kg) N2O_per_pixel <- biomass_per_pixel$ multiply(EF_N2O_img)$ rename("N2O_kg")$ clip(aoi_country_ee$geometry()) # CO₂ cross-check (forest only, kg) CO2_crosscheck_per_pixel <- biomass_per_pixel$ updateMask(forest_mask)$ multiply(EF_CO2_img)$ rename("CO2_crosscheck_kg")$ clip(aoi_country_ee$geometry()) # 3. Visualize ch4_vis <- list( min = 0L, max = 50L, palette = c("yellow", "orange", "red", "darkred") ) n2o_vis <- list( min = 0L, max = 5L, palette = c("lightblue", "blue", "darkblue", "purple") ) tmap::tmap_mode("view") tmap::tm_shape(aoi_country_sf) + tmap::tm_borders(col = "white", lwd = 1.5) + tmap::tm_tiles( ee_tile_url(CH4_per_pixel$selfMask(), ch4_vis), group = "CH₄ Emissions (kg per pixel)" ) + tmap::tm_tiles( ee_tile_url(N2O_per_pixel$selfMask(), n2o_vis), group = "N₂O Emissions (kg per pixel)" ) + tmap::tm_add_legend( type = "polygons", fill = c("orange", "red", "blue", "purple"), labels = c("CH₄ Low", "CH₄ High", "N₂O Low", "N₂O High"), title = "GHG Emissions (kg/pixel)" ) + tmap::tm_scalebar(position = c("RIGHT", "BOTTOM"), text.size = 0.5) + tmap::tm_basemap("Esri.WorldImagery") ``` ```{r} #| label: emissions-gee-aggregate #| code-summary: "Aggregate pixel-level emissions — Honduras 2010" # Total CH₄ emissions CH4_total <- CH4_per_pixel$ reduceRegion( reducer = ee$Reducer$sum(), geometry = aoi_country_ee$geometry(), scale = 500L, maxPixels = 1e9 )$getInfo() # Total N₂O emissions N2O_total <- N2O_per_pixel$ reduceRegion( reducer = ee$Reducer$sum(), geometry = aoi_country_ee$geometry(), scale = 500L, maxPixels = 1e9 )$getInfo() # CO₂ cross-check (forest only) CO2_crosscheck_total <- CO2_crosscheck_per_pixel$ reduceRegion( reducer = ee$Reducer$sum(), geometry = aoi_country_ee$geometry(), scale = 500L, maxPixels = 1e9 )$getInfo() CH4_Gg <- (CH4_total[["CH4_kg"]] %||% 0) / 1e6 N2O_Gg <- (N2O_total[["N2O_kg"]] %||% 0) / 1e6 CO2_xcheck_Gg <- (CO2_crosscheck_total[["CO2_crosscheck_kg"]] %||% 0) / 1e6 cat("Honduras 2010 — GEE Pixel-Level Emissions\n") cat(" Total CH₄:", round(CH4_Gg, 4), "Gg\n") cat(" Total N₂O:", round(N2O_Gg, 5), "Gg\n") cat(" CH₄ CO₂-eq:", round(CH4_Gg * 28, 2), "Gg\n") cat(" N₂O CO₂-eq:", round(N2O_Gg * 265, 2), "Gg\n") cat(" Total CO₂-eq (non-CO₂):", round(CH4_Gg * 28 + N2O_Gg * 265, 2), "Gg\n") cat("\n Forest CO₂ cross-check:", round(CO2_xcheck_Gg, 2), "Gg\n") cat(" (Reported via stock changes, not fire emissions table)\n") ``` ------------------------------------------------------------ ## NGHGI Reporting Format {#sec-emissions-reporting} ### IPCC Common Reporting Format (CRF) {#sec-emissions-crf} Required tables for fire emissions: 1. Table 4.C - Fires on Forest Land - Area burned (ha) - Biomass burned (tonnes DM) - CH~4~ emissions (Gg) - N~2~O emissions (Gg) - Note: CO~2~ from forest fires is reported in the carbon stock change rows of Table 4.A, not here 2. Table 4.F - Fires on Grassland - Area burned (ha) - Biomass burned (tonnes DM) - CH~4~ emissions (Gg) - N~2~O emissions (Gg) - CO~2~: Not reported (Tier 1 synchrony assumption) 3. Table 4(V) - Fires in Organic Soils (Wetlands Supplement) - Area burned (ha) - CO~2~ emissions (Gg) - CH~4~ emissions (Gg) ```{r} #| label: emissions-crf-format #| code-summary: "Format emissions for UNFCCC CRF reporting — Honduras 2010" # Format for CRF Table 4.C (Forest Land) crf_4c <- emissions %>% filter(fire_category == "Forest") %>% group_by(country, year) %>% summarise( biomass_burned_t = sum(biomass_burned_Gg) * 1000, CH4_Gg = sum(emissions_Gg[gas == "CH4"]), N2O_Gg = sum(emissions_Gg[gas == "N2O"]), .groups = "drop" ) %>% mutate( IPCC_Category = "4.C.1", Subcategory = "Fires - Forest Land", Tier = "Tier 1", CO2_note = "CO₂ reported via stock changes (Table 4.A)" ) # Format for CRF Table 4.F (Grassland) crf_4f <- emissions %>% filter(fire_category == "Savanna") %>% group_by(country, year) %>% summarise( biomass_burned_t = sum(biomass_burned_Gg) * 1000, CH4_Gg = sum(emissions_Gg[gas == "CH4"]), N2O_Gg = sum(emissions_Gg[gas == "N2O"]), .groups = "drop" ) %>% mutate( IPCC_Category = "4.F.1", Subcategory = "Fires - Grassland", Tier = "Tier 1", CO2_note = "CO₂ excluded (Tier 1 synchrony assumption)" ) # Combine for full CRF submission crf_fires <- bind_rows(crf_4c, crf_4f) # Export write.csv( crf_fires, "Honduras_2010_CRF_Fires.csv", row.names = FALSE ) library(knitr) kable( crf_fires %>% select(IPCC_Category, Subcategory, biomass_burned_t, CH4_Gg, N2O_Gg, Tier, CO2_note), caption = "UNFCCC CRF Tables 4.C and 4.F — Honduras 2010 Fire Emissions", digits = c(0, 0, 0, 4, 5, 0, 0) ) ``` ### Documentation Requirements {#sec-emissions-transparency} IPCC Good Practice Guidance requires documentation of: 1. Data sources: - MODIS MCD64A1 Collection 6.1 (burned area) - MODIS MCD12Q1 Collection 6.1 (land cover) - IPCC 2019 Refinement Tables 2.4 and 2.5 2. Methods: - Tier 1 approach - Equation 2.27 for non-CO~2~ gases - Carbon stock changes (Eq. 2.7--2.14) for forest CO~2~ - Quality filters applied (uncertainty \< 20%) 3. CO~2~ accounting: - Forest: CO~2~ reported through stock changes (2006 GL §4.2.4) - Savanna: CO~2~ excluded per Tier 1 synchrony assumption (2019 Ref §2.4) - Organic soils: CO~2~ reported separately (2013 WS Ch. 2) 4. Uncertainty: - Combined uncertainty: ±65-100% - Major contributors: Fuel loads, combustion factors - Monte Carlo analysis performed 5. QA/QC procedures: - Multi-source validation (GLAD, ESRI) - Time series consistency checks - Comparison with independent estimates 6. Recalculations: - If methodology changes: Recalculate entire time series - Document changes in National Inventory Report (NIR) ------------------------------------------------------------ ## Notes and Next Steps {#sec-emissions-summary} ### Key Takeaways {#sec-emissions-takeaways} 1. Emission factors convert biomass burned to gas emissions: - CH~4~: 2.3-6.8 g kg^-1^ (savanna \< forest) - N~2~O: 0.20-0.26 g kg^-1^ (similar across types) - CO~2~: 1580-1613 g kg^-1^ (forest/savanna, Table 2.5) - CO~2~: 1703 g kg^-1^ (organic soils, Wetlands Supp.) 2. CO~2~ reporting: - Forest: CO~2~ is reported via carbon stock changes (not the fire emissions table); synchrony cannot be assumed for woody biomass (2006 GL §4.2.4) - Savanna: CO~2~ excluded at Tier 1 (synchrony assumed for non-woody grassland), with caveats for woody systems (2019 Ref §2.4) - Organic soils: CO~2~ reported directly (permanent carbon loss) 3. CO~2~-equivalents: - GWP~CH4~ = 28, GWP~N2O~ = 265 (IPCC AR5) - Total warming impact dominated by CH~4~ (65-70%) 4. Uncertainty: - Forest fires: ±68% - Savanna fires: ±85% - Driven by fuel consumption more than emission factors 5. Reporting: - UNFCCC CRF Tables 4.C (Forest) and 4.F (Grassland) - Forest CO~2~ in stock change tables (4.A) - Transparency documentation required ### Chapter 5 {#sec-emissions-to-chapter5} Chapter 5: Organic Soils and Peatland Fires will expand on: 1. Detailed peat fire methodology (IPCC Wetlands Supplement) 2. Southeast Asia case studies (Indonesia, Malaysia) 3. Drainage impacts on fire risk and emissions 4. Depth of burn estimation techniques 5. Smoldering combustion characteristics Alternative pathway: If organic soils are not relevant to your region, Chapter 5 could cover: - Quality control and validation procedures - Time series analysis and trend detection - Inter-annual variability and climate drivers ------------------------------------------------------------ [^index-1]: To confirm values, paged links to GWP values from AR5 [here](https://www.ipcc.ch/site/assets/uploads/2018/05/SYR_AR5_FINAL_full_wcover.pdf#page=104), AR2 [here](https://www.ipcc.ch/site/assets/uploads/2018/02/ipcc_sar_wg_I_full_report.pdf#page=36), and ~~AR6 here~~. [AR6](https://www.ipcc.ch/report/ar6/syr/downloads/report/IPCC_AR6_SYR_AnnexesIndex.pdf#page=6) states: "Under the Paris Rulebook \[Decision 18/CMA.1, annex, paragraph 37\], parties have agreed to use GWP100 values from the IPCC AR5 or GWP100 values from a subsequent IPCC Assessment Report to reportaggregate emissions and removals of GHGs. In addition, parties may use other metrics to report supplemental information on aggregate emissions and removals of GHGs."