A Satellite View of Wildfires Since the 50’s

In this map, wildfires are shown in orange. Private lands are shown in purple while public lands are clear (no color). From Earth Matters.

I ran across this interesting approach to western fires from NASA. Followers of this blog have read the fire observations of folks like Leiburg in the teens, in which the Native American burning patterns were still visible. Then came big fires and fire suppression efforts, and since 1950 many changes in forest and fire management, invasive species, people living in the forest, people igniting the forest and so on. So to me, anyway, picking the dates 50 til present is within that larger dynamic and understanding the larger dynamic could explain some of the observations.
Data sources for the study are given at the end of the article.

1. There are more fires.
Over the past six decades, there has been a steady increase in the number of fires in the western U.S. In fact, the majority of western fires—61 percent—have occurred since 2000 (shown in the graph below).

(I think we’ve talked about the historic fire databases before.)

2. And those fires are larger.
Those fires are also burning more acres of land. The average annual amount of acres burned has been steadily increasing since 1950. The number of megafires—fires that burn more than 100,000 acres (156 square miles)—has increased in the past two decades. In fact, no documented megafires occurred before 1970.

(This is interesting, even if only from 1950 to 1970. It’s not clear to me what the frequency is on the y axis of this graph).

3. A small percentage of the West has burned.

Even though fire frequency and size has increased, only a small percentage of western lands— 11 percent—has burned since 1950. In this map, wildfires are shown in orange. Private lands are shown in purple while public lands are clear (no color). The location of wildfires was random; that is, there was no bias toward fires affecting private or public land.

Keith Weber, a professor at Idaho State University who led the analysis, was surprised at the 11 percent figure. There’s no clear reason yet for why more of the region hasn’t burned. “Some of the 89% may not burn because it has low susceptibility—not dry enough or it has low fuel (vegetation),” said Weber. “Some areas may be really ripe for a fire, but they have not had an ignition source yet.”

(I’m not sure about the public vs private land maps here.. check out the San Joaquin Valley of California. Also not sure if Tribal lands are considered public or private).

4. The same areas keep burning.
How has only 11 percent of the west burned, yet the annual number of acres burned and the frequency of fire increased? It turns out that many fires are occurring in areas that have already experienced fires, known as burn-on-burn effects. About 3 percent—almost a third of the burned land—has seen repeated fire activity.

The map here shows the locations of repeated fire activity. While you can’t see it at this map’s resolution, some areas have experienced as many as 11 fires since 1950. In those areas, fires occurred about every seven years, said Weber, which is about the amount of time it takes for an ecosystem to build up enough vegetation to burn again.

(It would be interesting to know exactly where those are).

5. Recent fires are burning more coniferous forests than other types of landscape.
Since 2000, wildfires have shifted from burning shrub-lands to burning conifers. The Southern Rocky Mountains Ponderosa Pine Woodland landscape has experienced the most acres burned—more than 3 million.

(Again, this is very interesting but not clear on what it means. Certainly plenty of shrublands are still burning.)

6.Wildfires are going to have a big impact on our future.

Research suggests that global warming is predicted to increase the number of very large fires (more than 50,000 acres) in the western United States by the middle of the century (2041-2070).

The map below shows the projected increase in the number of “very large fire weeks”—periods where conditions will be conducive to very large fires—by mid-century (2041-2070) compared to the recent past (1971-2000). The projections are based on scenarios where carbon dioxide emissions continue to increase.

(This assumes that more bad fire weather=more bad fires. That’s not a bad assumption (projections of future weather changes based on unknown levels of carbon) but reasonable people could also include assumptions about improvements in firefighting, community resilience, warming causing plant life to grow more slowly and lead to lower fuel loadings at less density, and so on.). It’s a bit like predicting deaths from a disease without considering health care efforts.

15 thoughts on “A Satellite View of Wildfires Since the 50’s”

  1. While seeing the fire perimeters (at least I assume that is what we are seeing?) is instructive, but the impact or effect of the fire varies considerably within that perimeter. In some cases, a fair amount of the burn is from backfires that were lit to create a containment line. In other cases, as noted, there wasn’t enough fuel or the sites weren’t that conducive to burning; in others there is 100% mortality. I think we need to pay attention to the proportion of fires that is high severity (high levels of tree mortality) and if that is changing. Generally that is a fairly small proportion – much smaller than areas that burn with lower levels of mortality. There are some years however (2015 comes to mind) where the high levels of mortality were much greater.

    • I think it is also quite important to measure mortality 5 years after the fire. Many trees will survive the fire but, not the bloom of bark beetles and cambium kill. I worked on the Biscuit Fire salvage efforts, and was surprised by how much extra mortality there was after we went through and marked the snags.

  2. With so much scientific knowledge in today’s world, I wonder how and why some studies seem to just look at wildfires in terms of the past 50 to 75 years. Is that really a long enough time frame to draw many definitive conclusions?

    See also:

    • Thanks for those graphs, Matthew! That’s what I was trying to get at verbally. I liked the one that showed the apparent influence of the PDO (I believe that’s the Pacific Decadal Oscillation). I emailed Keith Weber, the lead author, as to why he started in the 50’s.

      “Our study focuses on wildfires in the western US from 1950-present. This was simply done out of necessity as the availability of consistent and complete GIS fire perimeters reduced drastically prior to 1950. To remove the uncertainty related to incompleteness of available records we chose 1950 as a start point.”
      This caused me to think, hey, they didn’t have GIS back in those days..so I asked Keith that question and he replied

      Keith Weber
      8:32 AM (9 hours ago)
      to me

      “In the 90’s there was an effort to digitize historic fire records into GIS. Here at ISU, we did a lot of that work for the Federal agencies back into the 1930’s. Other states seem to have done similar projects with most going back only to 1950. No one was doing GIS in 1950, but the records were retroactively digitized into GIS… but it becomes spotty very quickly.”

      • Maybe he could also explain the map better. Does the orange (and 11%) include the brown (which is not explained)? The white/no color areas (like the Central Valley in CA) are NOT public lands, but they are not purple (like eastern CO.). It would be interesting to see how this lines up with the Forest Service maps delineating areas at risk due to fire regime condition class or areas “in need of restoration.”

        • Here’s the answer from Keith.
          “These maps were derived from the BLM’s Surface Management Agency (SMA) GIS layer. When zoomed out to the scale of the western US, some small patches of lands may not appear due to scale effects.”
          I wish I had some GIS skills…

  3. Basically, we have more fuels in our forests than ever in recorded history. Here in California, those fuels get bone-dry, every single summer. Plus we have those millions and millions of dead trees, too. Going back to a pre-human landscape is impossible. Then, there are also so many re-burns, which burn at high intensity, especially if they have not been thinned of snags. Paradise had no dedicated fuelbreaks or fuels modifications, and we’ve seen how tragic that can be. Pretending that “doing nothing” would have had any reduced effect on the Camp Fire is just ridiculous and dangerous idiocy.

    • For whatever it’s worth…None of the four wildfire charts I provided show a “pre-human landscape.” One chart does go back to a “pre-industrial” period, which they identified as about 1500-1800 AD. The other charts go back to 1910, 1919 and 1930. Also, not sure anyone is advocating “doing nothing”…that’s just another strawman you’re so fond of Larry. Also, pretty sure that Paradise did have many examples of “fuels modifications” in and around the community, and even in the forest lands to the east. Also, pretty sure not all re-burns burn at high-intensity, just like I’m pretty sure not all wildfires burn at high-intensity.

      • Well, we cannot go back to those times, either. You’re pushing the idea that larger and more frequent fires would be wonderful and oh, so ‘natural’ (while ignoring the ignition sources and extreme damages that TODAY’S wildfires do).

        Indeed, there are legions of idiots who insist on letting all wildfires burn, wishing that Smokey Bear was dead (along with his message).

        • More strawman arguments from Larry…including “there are legions of idiots who insist on letting all wildfires burn,” which was a nice touch.

  4. Apparently, as well, the ‘Fire Funding Fix’ has not been made into law, and time is running out for the Republican Congress to act. We’re still stuck in an undesirable situation, treating the symptoms, instead of providing solutions. We’ll see if the new Congress will be any different.

      • Matt, thanks for sharing the HCN article. According to the article, the two “environmental rollbacks” are a CatEx for fuel-reduction logging projects less than 3,000 acres and removing the requirement for the FS to reconsult on Forest Plans whenever a new species is listed or new habitat designated. (Consultation would still be required for individual project implementation.) Both of these seem like common-sense ways to expedite the needed work of managing National Forests.

        However, if you don’t like FS’s management direction, I understand how any expediting would be percieved as bad, but IMO it is misleading to call these “environmental rollbacks”.

  5. Because CE’s are fairly complicated to understand, and so is anything to do with the ESA and consultation, it’s easier for reporters to quote interest group representatives who don’t like something and accept that what they say as true. I don’t know of any case in which it’s been shown that using a CE led to less protective project decisions than an EA. We have had the 2014 CE’s in place for four years now so you’d think there would be some evidence if that were the case.

  6. I don’t know how you would analyze whether a CE led to less, but I suspect there is plenty of evidence of EAs that led to more because of additional public involvement and/or more attention to alternatives.


Leave a Reply to Anonymous Cancel reply