A State of Emergency has been declared in British Columbia because of all the forest fires, and the usual suspects have started their usual game of blaming the forest fires on climate change. The problem is, as I will demonstrate in this blog post, the science is pretty clear that climate change cannot be blamed for this two-year uptick in fire activity. This is not to belittle climate change as a long-term threat to our forests. I cannot make this clear enough, climate change will eventually increase our fire danger. But we can’t blame climate change for everything, all the time. There are lots of moving parts in nature and this fire season is likely due to some other natural or anthropogenic feature, one we have to correctly identify if we are to avoid a repeat next year.
Now I can already hear my detractors saying: “but Blair climate change is an existential threat to our nation. We have to take advantage of the public interest to help us fight climate change”. My response is simple:
The argument being made is inconsistent with the state of the science. When you argue against the science you undermine your credibility.
But let’s think of it in a scientific manner: if climate change is not to blame for the two-year increase in forest fire activity then blaming it on climate change will mean we are missing some important other factor. Think of this as the wrongfully convicted murderer theory of environmental policy. When you wrongfully convict someone of murder you may feel better that someone is in jail but since it is the wrong person that means the real murderer is still out there running free. So identifying the actual cause of the forest fires becomes a really important topic. Why is this true? Because when climate change eventually increases our fire risk having it build on whatever is actually causing our current fires would make a bad situation even worse. So let’s look at what the science says about this topic.
The first misconception we have to clear up is the belief that climate change means everywhere in the world will get hotter and drier. This cannot be farther from the truth. Climate change is predicted to affect different areas differently. Most dry areas will get drier and many wet areas are going to get wetter. In B.C. the global climate models (GCMs) indicate that in the early years of climate change we will have wetter winters AND wetter summers. This will have the effect of initially reducing our risk of forest fires. Consider the most highly regarded of the papers on the topic of climate change and wildfires: Climate-induced variations in global wildfire danger from 1979 to 2013. The paper’s abstract reads like it should provide overwhelming support for the premise that climate change will increase our risk of fire. But a careful read of the paper shows that global trends will not always be seen regionally.
Since I want to make it easy just look at Figures 3 and 4 in the paper. Figure 3 displays “Global patterns of fire weather season length changes from 1979 to 2013“. Looking at B.C. on Figure 3a you see that the length of the fire season has significantly shortened between 1979 and 2013. Figure 3b “shows regions that have experienced changes in the frequency of long fire weather seasons (>1σ above historical mean) during the second half of the study period (1996–2013) compared with the number of events observed during the first half (1979–1996)”. Looking at Figure 3b we see that B.C. has experienced a shortening of the fire season between the start and end of the study period. Figure 4, meanwhile, reinforces the fact that the fire season has decreased in British Columbia over time.
The other thing I have heard is: “the paper says that globally fire incidence is expected to increase”. Well that can be true without it meaning that B.C. will get more fires. Rather the GCMs indicate that B.C. will initially have shorter fire seasons. Why is this the case? Well the simple answer is that as we have warmed B.C. has become wetter. Let’s look at a graph of long-term change in precipitation in BC. Virtually every part of B.C. has seen significantly increasing amounts of precipitation in the last 113 years. Moreover, that precipitation has increased across every season. Winter, spring, summer and fall have all been getting wetter. Now I have already had several people on Twitter argue that the time-window ends in 2013? Considering it is a 113 year time-frame this is an incredible minor quibble, but just to satisfy the pedants let’s look at what has happened in the last 5 years. The data shows that the last five years were even wetter than the 5 years before. The data makes it clear, the wet trend continues. So that argument is simply wrong.
The next argument I have repeatedly heard is that because it is warmer (even though it is wetter) we will still get more fires. Well here is a paper that considers that issue: Fuel moisture sensitivity to temperature and precipitation: climate change implications. What does the paper conclude? That the argument will eventually be true but that it is not true right now. Given the current increase in precipitation compared to the current amount of heating, the precipitation exceeds the values necessary to avoid the increase in fire risk associated with warming. This is why Natural Resources Canada (NRC) has concluded:
This complex combination of influences makes it difficult to identify clearly whether any measurable changes in the patterns of wildland fire over the last few decades can be linked directly to climate change. Nevertheless, pattern changes do appear to be underway.
In Canada’s northwestern boreal regions, for example, the annual amount of forest area burned by wildland fires rose steadily over the second half of the 20th century. Some of this increase has been attributed to climate change.
By contrast, in Canada’s southern boreal forest, the annual amount of area burned seems to have decreased during the 20th century. This trend might be the result of climate change causing greater amounts of precipitation over time in these regions.
However, analyses of fire history suggest that it is the effect of climate variability on precipitation regimes that is the primary reason for the decreasing fire activity in southern regions.
In reading the NRC conclusion recognize that none of B.C. is in the “northwestern boreal region”. We are in the part of the country where we have seen decreasing fire activity. What the NRC and the Flannigan et al. paper make clear: in the long-term (by the 2091-2100 fire regimes) climate change, if it continues unabated, should result in increased number and severity of fires. However, what the data says is that right now this signal is not yet evident.
So to conclude this post lets sum the story up: the science is pretty clear, climate change will increase our forest fire danger, but it is NOT responsible for this year’s forest fires. Why is it so important to make this clear? Because in the last two years we have had a significant jump in forest fires and figuring out what is actually causing the increase in area burned is pretty darned important. We need to determine what changes in forest condition, forest management or whatever has resulted in the last two years of fire behaviour. Understanding that the last two years’ fire seasons are due to something other than global warming frees us to figure out what actually has caused the increase. Most importantly, we need to figure this out before the effects of global warming actually start influencing our fire seasons.
A Chemist in Langley 
Interesting article Blair. I don’t believe we are going to find any single cause of these increased fires, as I am sure it will be a complex combination.
However, in reading through your article, the section about precipitation stuck out to me. My personal impression was that it seems to be getting drier, not wetter. I looked at the site you referenced about it being wetter the past 5 years. I decided to look at the historical data available on that site for annual precipitation. I went through and averaged it out by decade for both Kamloops and Vancouver (which you pointed to). The data only goes back to the late 30s for Vancouver and starts in 1950 for Kamloops. What both locations show is that it is getting very slightly wetter for the first several decades, then a big jump for the 80s and 90s, then dropping back down for the 00’s and the first 7 years of this decade. So I would be curious to know where that level of precipitation mitigates the temperature rise.
Another thought on the precipitation is that with the warmer temperatures, less of it is falling as snow. With that precip not sticking around to slowly water the environment, but instead running off , it would seem to me the forests will dry out faster, making them more susceptible to fire. I hear reports in the media during the winter about the % of normal the snow pack is, but I didn’t find a historical record of that tonight after a quick search. If the late spring snow pack is trending downwards, I think that would be one of the biggest culprits in the increased fire activity we are seeing. If that is the case, then could we not argue that climate change (in the form of higher temperatures we are seeing) is indeed a cause of the increased fire activity the past couple of years?
Regards,
Brad
LikeLike
Pine beetle damage?
LikeLike
I concur. Pine beetle damage has left combustible material on the forest floor from Waterton Lakes to Nelson, a gigantic swathe of really hot burning material. I am told it’s burning too hot for the normal recoveries and the end result is going to be desertification.
LikeLike
After the Fort Mac fire in Alberta, I found this comment (referred to in your article in HuffPost Alberta, May 10, 2016): “Before major wildfire suppression programs, boreal forests historically burned on an average cycle ranging from 50 to 200 years as a result of lightning and human-caused wildfires. Wildfire suppression has significantly reduced the area burned in Alberta’s boreal forests. However, due to reduced wildfire activity, forests of Alberta are aging, which ultimately changes ecosystems and is beginning to increase the risk of large and potentially costly catastrophic wildfires. – Flat Top Complex Wildfire Review Committee Report, May 2012 –”
Would this observation be applicable to BC too?
LikeLike
A nice analysis Blair. I looked at a science study of wildfires versus rise in annual temperature in the Northern Rockies in 2006. See LINK below. A warming of less than 0.9°C caused the wildfires in western USA observed in spring and summer. Bigger fires were caused by higher temperatures in the area: higher temperatures in spring and summer cause earlier snowmelt, longer summers, drier forests in summer, etc.
LINK = http://science.sciencemag.org/content/313/5789/940.full
Whether the changes observed are the result of global warming or only a natural fluctuation, such as El Nino, was beyond the scope of the study. BUT virtually all climate-model projections indicate that warmer springs and summers will occur over the western USA in coming decades. The Intergovernmental Panel on Climate Change predicted that global surface temperature will rise by 1.5° — 5.8°C by the end of this century. This is considerably larger than the recent warming — of less than 0.9°C — that caused the wildfires in the northern Rockies. The above discussion comes from my blog on the subject at https://www.iandexterpalmer.com/climate-change-part-4-wildfires-in-california/
Conclusion: The referenced article and the wildfires of the western USA raging right now do seem to point to global warming. I’d be interested in your comments.
LikeLike
I agree absolutely. The Nature article makes it clear that the US is responding much worse than Canada and that virtually all the increase in fire response could be attributable to climate change.
LikeLike
“Fuel moisture sensitivity to temperature and precipitation: climate change implications. What does the paper conclude? That the argument will eventually be true but that it is not true right now.” – I read the article and it does not state this. Could you clarify? South BC is projected at -24%-0% change in precipitation, a large amount is 1%-10%, and most of the change is 2-3C warmer. The last sentence in the article states: “The bottom line is that a warmer world will have more fire.” This paper suggests that BC will be warmer. There are also other peer-reviewed papers that are linking current forest fires to climate change (e.g., https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2004GL020876#.W3b0nVFJgwg.linkedin). The other issue with this opinion blog is that you don’t correctly address the issue of cause (see http://www.aaai.org/ojs/index.php/aimagazine/article/download/1612/1511 – for a more in depth look into the identification of cause by one of the worlds leading experts on that topic). I think this topic is more complicated than your article suggests. Browning of forests is occurring, there are beetle infestations, human dynamics, and most studies acknowledge that the fire season has been increasing. I disagree with your conclusion and I don’t think that this was well thought out.
LikeLike
Look at the actual precipitation changes in the last 100 years. We have real data for that stat so we don’t have to rely on modelling results.
LikeLiked by 1 person
So you are making a philosophical argument that statistics on historical data is to be trusted and more scientifically reliable than models of prediction? That is a separate argument from the published science. Further, there is still the problem that information within the article that you cite (Flannigan et al. 2016) does not accord with your summary of it. The authors do not state nor does it claim that “That the argument will eventually be true but that it is not true right now”.
If you want me to go and look at data and analyze it, then that is the type of argument that climate deniers use. I’m not a climate scientist. I’m a practicing ecologist-herpetologist. I leave the climate science, modelling, data analysis, and interpretation to the experts in that field and their peer-reviewed literature that is available for me to read and consider in relation to my knowledge. However, the subject clause in the title of your blog is “Looking at the science”, which I would assume to mean that you are not offering a new interpretation of the science but giving an objective summary of what scientists are concluding; I take objective journalism to mean something different than the post-truth era of pitting 1:1 opposing views in an arena, but rather would expect the journalist to have done some research with background knowledge to fact-check, reason, and accurately represent the information to be conveyed.
This is a very important topic that you are covering and it is a complex one. Many of us are being affected personally by what is going on. Our air quality has become hazardous, many of us have had to leave our homes, I put my kids to bed worried and stressed about a warming climate, and I go to work to do scientific work to the best of my abilities because I think it might help in some way. I am very careful in the conclusions I reach and I try to be very skeptical of my own beliefs as I listen to and interpret the evidence within my limited capacity for reason. I would hope that you would take another look at your article and give it an update with these considerations in mind.
I think you need to study and incorporate the philosophy of cause into your argument. Judea Pearl is a good start, but Bill Shipley also put out a great book on “Cause and correlation in biology: A user’s guide to path analysis, structural equations and causal inference”. If you want to make an argument that models should be weighed with less authority than statistics on data, then I think your argument rests on unstable footing. There is much to consider in the philosophy at work here in terms of evidence and method of inference (abduction, induction, deduction) – where statistical reasoning is just one form of inductive inference. Your literature review also misses some key references (e.g., http://www.pnas.org/content/pnas/113/42/11770.full.pdf) that conclude:
“Anthropogenic increases in temperature and vapor pressure deficit significantly enhanced fuel aridity across western US forests over the past several decades and, during 2000–2015, contributed to 75% more forested area experiencing high (>1 σ) fire-season fuel aridity and an average of nine additional days per year of high fire potential. Anthropogenic climate change accounted for ∼55% of observed increases in fuel aridity from 1979 to 2015 across western US forests, highlighting both anthropogenic climate change and natural climate variability as important contributors to increased wildfire potential in recent decades. ”
Granted that is western-US but note that those authors also cite the Flannigan et al. (2016) article that you cite, they have a very different take on it, and other researchers into fire activity in Alaska attribute recent fire activity to climate change (e.g., https://journals.ametsoc.org/doi/pdf/10.1175/BAMS-D-16-0149.1, http://iopscience.iop.org/article/10.1088/1748-9326/aaa136/meta). See also Dr. Stephen Dery’s research (that you’ve missed in your narrative here), which gives another published overview that differs considerably from your opinion on this subject matter (http://web.unbc.ca/~sdery/).
Thank you!
LikeLike
Your argument is that models are more reliable than the actual data? That is a pretty brave suggestion. Brave because no sane scientist would suggest we ignore real data and instead rely on models to look at past conditions. Model are use to help predict future but we have real data to examine historical conditions. As for the Flannigan article it specifically says that the modelling will be true in the 2050 era not now.
LikeLiked by 1 person
Look at Allchin and Dery (2017) (https://www.cambridge.org/core/services/aop-cambridge-core/content/view/7FB6363E70CB38090F18373B5FC5CFD0/S0260305517000477a.pdf/div-class-title-a-spatio-temporal-analysis-of-trends-in-northern-hemisphere-snow-dominated-area-and-duration-1971-2014-div.pdf). Annual snow cover duration is significantly lower across BC except along one section of the coast – that’s current state. The authors also discuss some of the limitations in simplistic analyses of the data. Why would you direct a herpetologist to “Look at the actual precipitation changes in the last 100 years”, when my capability on the matter would be about as good as a “Chemist in Langley”? I’m not attempting to offend you here, but I would expect that people of science would known and respect their limits. Are you a climate-science chemist? This does not mean that you cannot move in that direction, but unless you have some peer-reviewed papers under your belt in the topic of climate science prediction and statistical analysis, then I would expect only a summary of the peer-reviewed data and not some ad-hoc unprofessional summary of “precipitation changes in the last 100 years”. I wouldn’t do this – it’s unreasonable and I wouldn’t trust my own interpretation of it. Could you point me to a summary article on the precipitation changes in BC over the past 100 years? This article (http://people.forestry.oregonstate.edu/richard-waring/sites/people.forestry.oregonstate.edu.richard-waring/files/publications/Mathys%20et%20al.%202018.pdf) points to “increasing air temperatures and reduced snowpack” (in agreement with Dery’s et al’s investigations) and by “incorporating a large regeneration dataset together with climate data in physiological modelling…this study indicated that warmer temperatures, reduced frost, and increased drought occurred in some places in recent years, causing shifts in forest composition”. This does not sound like a wetter province.
The more I look into this, the more I’m finding disagreement in the peer-reviewed science. Where’s the disconnect? I’m really hoping that you can point me to the science that supports your opinion.
Thanks.
LikeLike
You present a paper that deals with the Arctic circle to describe conditions outside the arctic circle and ignore the actual data presented by the specialists at BC Env. As for your suggestion that I am not a climate change scientist. I do data analysis for a living. This is a data analysis problem period. Anyone who thinks that the rules underlying statistics somehow go away in climate science really needs to get out more.
LikeLiked by 1 person
“Your argument is that models are more reliable than the actual data?” – No. My post does not say this at all. I made a comment on your reasoning that you would like to weight the relative inferential trustworthiness of a) analysis of historical data v. b) input of data into a model. Both approaches have their roles in scientific method and I would be hesitant to wade into the debate on one being more reliable than the other. There could be concerns with the historical data and the way it was collected (e.g., how frequent, what was the spacing, who was collecting the data, what instruments were being used?) and how it is analyzed statistically (are we using null-hypothesis testing, Bayesian approach, likelihood, mixed modelling approach?). Statisticians have their own issues (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0202121). Models have an important role in science as well and have their issues as well, but I wouldn’t be so glib in my view to reject one approach outright rather than weight the reasoning that is involved in any particular analysis (good scientific practice). Models are used to make predictions about the world – we couldn’t make predictions without them. They are theories purposefully constructed to not match reality, such as a model airplane that we can use to study airfoil drag predictions, for example. Model and theory grade into each other – depending on what type of model you are using, but they are definitely universal in science and have an important role. You are diving into a classic debate in the philosophy of science on synthetic versus analytical reasoning, but it should not be used as a justification to discredit climate models outright. That’s poor scientific workmanship.
LikeLike
“As for the Flannigan article it specifically says that the modelling will be true in the 2050 era not now.” -> clarify this to the fact that they “For precipitation, the decadal future monthly GCM averages were divided by the 30-year GCM monthly baselines to get a ratio of future precipitation over baseline precipitation. We prepared 9 maps (3 GCMs × 3 scenarios) for 2041–2050 and 2091–2100”. However, they give other information that is not consistent with your summary of it. For example: “Not even a 40 % increase in precipitation amount could compensate for an increase in temperature of a degree or more”. The authors of that article to not make the claim that you make – you are extrapolating beyond their words in support your position. The paper does not conclude: “That the argument will eventually be true but that it is not true right now.” – That is false.
LikeLike
So you take one item (Fine Fuel Moisture Code greater than 91) and pretend it counts for all the conclusions. That is about as intellectually dishonest as you can get. The claim I make is specifically made in the paper. You on the other hand take a single quotation out of context and pretend it counts for the entire paper. I’m not sure what you think you are doing here, but anyone who reads the paper can tell how completely that one quotation is taken out of context.
LikeLiked by 1 person
The forests and grasslands are combustible. Period. The flammability may vary with sustained heat and dryness due to weather and or climate, but we should not neglect the causes of the fires, mainly “Person” caused and “Lightning” caused. Environment Canada detects every lightning strike. Is there a trend in lightning strikes?
https://www.canada.ca/en/environment-climate-change/services/lightning/canadian-detection-network.html
There are currently lots of lightning caused fires in the northern and coastal parts of Vancouver Island. The locals say there was an incredible lightning show, with some hail but without the usual downpours.
LikeLike
The BC Wildfire Service has changed their ways of fighting fires. They no longer direct attack the fires heavily like they used to. Instead they draw boxes around the fires with retardant and call it contained (Until it jumps). Secondly since their report last year, the BCWS has decided to go after fires with retardant etc, close to cities/towns. Fires in remote areas with no threat to civilization, let it burn. This “Let it burn” policy is the #1 reason our province is filled with smoke and more uncontrollable fires than ever, leaving our entire province a complete mushroom cloud. Sure they say that the forests with lots of underbrush needs to burn, BUT under a controlled burn, not by letting mother nature zapping lightning fires all over and not attacking them. That is entirely irresponsible of them they know better. Now they have so many fires burning they don’t have any resources to stop them. So many 2Ha fires have grown beyond 100Ha in days or less, some are in the 1000s of Hectares. This is something that needs to be changed and senior leaders of BCWS need to be held accountable for irresponsible management.
LikeLike
Cliff Mass states that wildfire data from California shows a decrease in size and intensity over the last century and that fire activity is not well corellated with global warming. He suggests that increased forest fuel loads resulting from long term fire suppression may be one cause of firestorms.
http://cliffmass.blogspot.com/2018/08/california-wildfires-is-global-warming.html
LikeLike
Cliff Mass says this at the end of today’s post:
Cliff is worth paying attention to because he is very knowledgable and rational.
He has had posts in previous years about the role of wind in wildfires, both for the safety of the firefighters and the predictability of danger conditions.
LikeLike
http://cliffmass.blogspot.com/2018/08/northwest-wildfires-are-we-seeing-new.html
LikeLike
Jeepers, this is an absolutely unpopular stance to take these days. It’s virtually impossible to argue against the climate change/forest fire correlation without getting insulted for being an “ignorant idiot,” etc.
LikeLike
This post concludes with the unequivocal statement that climate change “is NOT responsible for this year’s forest fires.” I have two problems with this statement:
1. It incorrectly represents the references that you use in your post.
From the NRC web page that you referenced: “This complex combination of influences makes it difficult to identify clearly whether any measurable changes in the patterns of wildland fire over the last few decades can be linked directly to climate change.” This is hardly an unequivocal assertion of the converse.
You refer to “the most highly regarded of the papers on the topic of climate change and wildfires” to make the point that “global trends will not always be seen regionally”. This paper looks at historical trends but does not attempt to make any predictions beyond its data (1979 to 2013). You base your statement on a couple of figures from the paper and the fact that the last few years have been quite wet. However, the same paper, when discussing why “the US has witnessed a marked increase in large wildfire frequency and duration” says “These trends are widely attributed to shifts towards earlier snowmelt timing … and variability in the timing of spring precipitation”. Look at the monthly rainfall in BC (https://vancouver.weatherstats.ca/charts/precipitation-monthly.html) and you will see that the distribution of precipitation paints a different picture than your oversimplified look at the yearly numbers. Your references do not support your conclusion.
Your references also DO NOT support the conclusion that: “climate change IS responsible for this year’s forest fires”. And that brings me to my second problem with your statement:
2. An unequivocal statement like yours that climate change “is NOT responsible for this year’s forest fires.” is unscientific. In your own words: “The argument being made is inconsistent with the state of the science. When you argue against the science you undermine your credibility.” The state of the science is not yet at the point where we can unequivocally assert that a regional weather event is a consequence of (or NOT a consequence of) climate change. The state of the science is that we are just beginning to be able to say with reasonable confidence how much more or less likely a particular regional weather event was as a consequence of climate change (http://assets.climatecentral.org/pdfs/WWA_NRC_Attribution_Report_March2016.pdf).
Your argument is inconsistent with the state of the science and you have undermined your credibility.
Aside from that, your essential message is a good one. Climate change is very likely not the only reason why we have had two devastating fire seasons in a row. A better understanding of the vulnerabilities of our forests and a plan to better manage them is called for.
I just wish that you hadn’t made your point with bad science.
LikeLike
Point 1) you are demonstrably wrong and miss the geographic details associated with the modeling and presented in the Nature paper (which distinguishes the US results from elsewhere) I actually had a back-and-forth with a US expert on this topic who agreed with me on this point. As for the EC results they show it has become wetter in summer so your use of one year’s data is a blip. My results are consistent with the research. You just need to read deeper into the topic.
Point 2) and once again you look at the American data and conflate it with the Canadian results. If you can’t tell the difference between BC and California then I can’t help you. The GMs say BC will get wetter (and warmer) and to date the wet is still outweighing the warm. The US has seen a very different result because dry areas have got drier. Thus the Northwest and mid-west fires in the US.
LikeLike
Thanks for responding to my post.
Point 1) Fair enough Blair. Its your blog, so I won’t get into my read of the data vs. yours. I’ll just say that it doesn’t look to me that your read is the slam dunk that you represent it as. You wouldn’t be the first to have underestimated “environmental complexity”.
Point 2) I did not conflate American data and Canadian results. My link to the NRC attribution report was to draw attention to the “state of the science” of the attribution of regional weather events to the effects of climate change. You made an unequivocal statement about BC forest fires attribution that the state of the science absolutely does not support.
You wrote that climate change “is NOT responsible for this year’s forest fires.”
Your post was misleading because an unequivocal statement like that is bad science.
Had I written that climate change IS responsible for this year’s forest fires on a comment on this blog, I expect that you would have (rightly) called me out for bad science on exactly the same grounds.
I personally suspect that two seasons in a row of serious early flooding followed by greater than average heat, drought and forest fires is probably related to climate change, but I will wait for some attribution research for THIS specific regional weather event to be published before I make up my mind how likely that is. Stay tuned, the science of near-real-time attribution is advancing quickly.
LikeLike
Yes, there are many “moving parts” to wildfire dynamics, but we seem to be getting lost in the analytic details. We wonder if the record burn-up of BC’s forests is due to climate change, when we already know the answer. Physics 101: a warmer atmosphere will intensify natural weather and climate cycles including drought and rainfall. (And don’t forget that climate change is a major factor in the pine beetle epidemic.) So the extensive wildfires of 2017 and 2018 may have occurred anyway due to other factors, but a warmer climate certainly contributed to their intensity and scope. By how much? Who cares?
We know enough to realize we have to stop greenhouse gas pollution and clean up what has accumulated as quickly as possible.
LikeLiked by 1 person