In California, wildfires feel like a looming threat that we can’t seem to get a handle on. Each fire season, voices from experts flood radio stations and newspapers, but the information is confusing and often feels conflicting. As climate change creates a nearly year-round fire season, fire agencies call for the allocation of more funding for better equipment, planes, and personnel. Worried experts agree that our landscapes are a ticking tinder-bomb, but they simultaneously argue that fire is a natural and necessary part of California’s ecosystems, and that we should urgently be burning more land. When we see wildfires decimate our forests and leave a lifeless moonscape, how can that be true?
Fire is…complicated. And the way that it is talked about it can be confusing. So we want to break it down here.
Fire is a Spectrum
Good communication about the science of fires is hard. Often we are left with graphs like this:
library(ggplot2)
library(dplyr)
library(ggdark)
library(ggh4x)
library(ggiraph)
library(ggrepel)
library(ggtext)
library(knitr)
library(markdown)
library(plotly)
library(RColorBrewer)
library(rmarkdown)
library(scales)
library(tidyverse)#Introduce dataset, which we've renamed from the original name to "FireBurnSeverityData"
df <- read.csv("FireBurnSeverityData.csv") #Summarise area burned by year in acres
YearlyTotals <- df %>%
group_by(Fire_Year) %>%
summarise(YearlyHighHec = sum(High_Severity.ha., na.rm = TRUE), YearlyModHec = sum(Moderate_Severity.ha., na.rm = TRUE), YearlyLowHec = sum(Low_Severity.ha., na.rm = TRUE), YearlyUnchangHec = sum(Unchanged.ha., na.rm = TRUE), YearlyTotal = sum(Total_area_burned.ha., na.rm = TRUE)) %>%
mutate(YearlyHighAc = YearlyHighHec * 2.47105, YearlyModAc = YearlyModHec * 2.47105, YearlyLowAc = YearlyLowHec * 2.47105, YearlyUnchangAc = YearlyUnchangHec * 2.47105, YearlyTotalAc = YearlyTotal * 2.47105) #Getting number of fires per year
FireFreq <- df %>%
count(Fire_Year)#Joining df's together & creating graph that is most commonly used in wildfire communication
YearlyTotals <- left_join(YearlyTotals, FireFreq, by = "Fire_Year")
YearlyTotals %>%
ggplot(aes(x = Fire_Year)) +
geom_col(aes(y = YearlyTotalAc), fill="darkred") +
scale_x_continuous(breaks = c(1984,1990,1995,2000,2005,2010,2015,2020)) +
geom_smooth(aes(y = n*35000), color = "goldenrod2", span = 0.15, size = 1, alpha = 0.1) +
scale_y_continuous("Acres", sec.axis = sec_axis(trans = ~./35000, name = "Number of Fires"), labels = comma) +
dark_mode(theme_minimal()) +
labs(x = NULL , y = "Acres" , title = "Total CA Acres Burned by Year" , subtitle = "" , caption = "") +
annotate("text", x=1996, y=3250000, label="Number of Fires", size = 5.5, color = "goldenrod2") +
annotate("segment", x = 1990, xend = 1988, y = 3200000, yend = 3000000, color = "goldenrod2", size = 0.4, arrow = arrow()) +
theme(axis.title.y = element_text(vjust = +2.5, size = 15), axis.title.x = element_text(vjust = -0.75), axis.title.y.right = element_text(vjust = +2.5), axis.text.x = element_text(size = 10, face = "bold"), axis.text.y = element_text(size = 10, face = "bold"))
How much land is actually burning? How long does it take for ecosystems to recover? The answer is…it depends.
Experts promoting burning more land feels illogical in part because we’re often taught that all fire is destructive. But that isn’t true! There are several metrics that firefighters and researchers use to determine how “bad” a wildfire was. One is size, of course, but the other is what we call “severity,” a measure of how much the ecosystem changed as a direct result of the fire. Severity is a spectrum. “Low-severity” fires that primarily burn ground fuels and small trees can be beneficial to forests. On the other end of the spectrum, raging fires that kill nearly all vegetation and living soil, are “high severity.”
