Caribou herds in North America seem to be declining. Is warming climate or it’s effects on habitat to blame? The relationship of caribou to lichen-rich winter ranges and fire is often oversimplified. Many factors besides habitat affect caribou numbers, which undergo large fluctuations naturally. In this new Research Brief, we highlight recent publications on caribou-fire relationships and explore some of the factors that make it complicated to predict exactly what will happen and when if old-growth caribou habitats diminish with warming climate and more frequent burning.
Although vegetation treatments can reduce fire potential, they may have unintended ecological effects, but there has been little published on possible impacts—especially for Alaska. So the recent publication (Melvin, et al. 2017) of a study on interior Alaska fuel treatments by an interdisciplinary team of researchers is an important addition to regional management resources. In fact, it probably represents the FIRST published paper specifically on how fuel-reduction affects carbon and nutrient pools, permafrost thaw, and forest successional trajectories. The analysis included 19 sites managed by numerous Alaska agencies covering a large swath from Nenana to Deltana, and were sampled at various ages from 2-12 years post-thinning or shearblading. Our third AFSC Research Brief of 2017 is a digest of the study results.
Incident fire behavior analysts predicted the 2011 Las Conchas fire would calm down at night, but instead they witnessed a night-time blow-up between 10 p.m. and 3 a.m. where 35-ft high “rolling barrels of fire” advanced rapidly downhill, quadrupling the fire’s size.
Article in July-Aug 2017 Popular Science by Kyle Dickman, sheds light on extreme fire behavior.
Rod Linn at the fire Los Alamos National Laboratory has been studying wildfires for 22 years, using computational models including weather and topography to explain unexpected behavior. In a recent Popular Science article he sheds light on some very interesting scenarios that caught the analysts off guard, including how an inversion developing in the evening spilled out of the Valles Grande basin like an overflowing bathtub and spawned the 26 ft/sec downslope night winds that blew up the Las Conchas fire. The article is very readable and sheds light on several other species of extreme fire behavior that will be of interest to anyone on the fireline. Pick up the July/August Popular Science or read it for free online here: https://www.popsci.com/las-conchas-wildfire-pillar-of-fire
As climate warming brings more wildfire to the North, scientists and citizens wonder how the landscape will be transformed. Will forests continue their 2000’s-era trend toward less spruce and more hardwoods, catalyzed by larger fires and more frequent burning? If so, that might slow down the trend for larger and more intense fires. However, will hotter summers with more effective drying lead to increased fire re-entry into the early successional hardwoods, making them less strategic barriers for fire protection? A research team modeling the former question just unveiled an interactive web tool to model forest changes under various future climate scenarios (Feb. 1 webinar recording available HERE). With the new web tool, funded by JFSP, Paul Duffy and Courtney Schultz will be working with fire managers in Alaska to look at fire occurrence and cost in the future. Try it for yourself at http://uasnap.shinyapps.io/jfsp-v10/
The subject of a new study (and a recent AFSC webinar by Sean Parks of the USFS Rocky Mountain Research Station) introduces a novel way to look at fire regime changes through time over a landscape using the idea of “climate analogs”. We’ve all seen maps showing future changes in temperature and precipitation based on climate projection models. Spatial analysis can locate a “future” climate analog for any pixel on a map using projections of variables like temperature, precipitation, or modeled evapotranspiration. Parks et al. 2016 provide a way to “look next door to see the future”, i.e. our pixel or region of interest, may be expected to show a fire return interval, burn severity, etc. similar to that now reflected in its analog which has those climate characteristics today. If the average annual temperature in Fairbanks was 30⁰F in 2015, for example, we could map the nearest points that may have similar temperatures by 2085—possibly at higher elevations around Fairbanks. If a “path of least resistance” with respect to skirting areas that may have way different temperatures due to topographic features is added, you get a figure kind of like the one below from Yellowstone park (Dobrowski and Parks 2016). The authors have used the method to look at future availability of wildlife habitats, and to hypothesize fire regime characteristics of parks and wilderness areas in the mountain west. Among their findings were thresholds for climate moisture deficit which seemed to make fire frequency jump up and other areas which seemed to indicate fuel limitations may lead to lower fire severity. So far, the approach has not been tried in Alaska, but might provide an interesting comparison to vegetation and future fire modeling being done by SNAP.
From Dobrowski and Parks, 2016, Fig. 1: Climate trajectories are defined by a source pixel (start) with a given temperature under current conditions (1981–2010) and a destination pixel (end) with a similar temperature under future conditions (2071–2100). Curved path (2) minimizes traversing pixels with large differences in temperature.
The most important ecological effects of fire may not be evident for many years after burning. Take permafrost, for example: just-published research is revealing extensive thawing and drying of soils in the aftermath of the Boundary Fire in interior Alaska. Brown et al. 2016 found almost all the severely burned plots in their study had thawed by 10 years after the 2004 fire. Without permafrost the burned areas were better drained, leading to drier soils, and influencing vegetation succession.
Typical burn appearance after 3 years (R. Jandt)
Another interesting facet of their study was the array of remotely-sensed data that Brown and colleagues employed, including optical and infrared spectra (Landsat 7 & 8), radar (L-band Synthetic Aperture Radar, or ALOS-PALSAR), and topographic (Light Detection and Ranging–LiDAR) datasets. Infrared indices used in the study were strongly correlated with soil moisture–allowing researchers to map the distribution of permafrost and compare it to burn severity maps.
A paper just published by the indefatigable Adam Young, a PhD candidate at the University of Idaho, and colleagues pulls together a lot of information about climate, forest, tundra and fire to offer a glimpse of potential future fire regimes in different parts of Alaska. By looking at fire occurrence at a multi-decadal time scale, the researchers drill down into how fire rotations are likely to respond to climate projections at a regional scale.
Exerpt from Fig. 6, Young et al. 2016. Figures in the paper not only show the observed fire rotation for 19 subregions of Alaska (Figure A2 in supplement) with 60 years of fire occurrence data, but also project future rotations under various climate scenarios (in this case a mean of of 5 global climate models).
The use of advanced statistical models to build fire-landscape response models for boreal forest and tundra reaffirms prior findings of the sensitivity of fire regime to summer temperatures and moisture deficit. However, the effect is not uniform among regions: they identify a threshold at about 56⁰ F (30-yr mean temperature of the warmest month) and another threshold for annual precipitation where fire occurrence really seems to jump. This latter finding accounts for results which project large increases in 30-year probability of burning for areas where these thresholds will be crossed in the next several decades. For example, models project the Brooks Range foothills of the North Slope, Noatak tundra and the Y-K Delta may see increases in fire 4-20x greater than historical levels. Some tundra areas are likely to experience fire frequency increase to levels not observed in the paleo record, spanning the past 6,000-35,000 years. Across most of the boreal forest, fire rotation periods are projected to be less than 100 years by end of the 21st century. This is useful information for natural resources management as well as fire protection agencies—a concise, well-researched, well-illustrated paper—put it on your summer reading list.
Young, A. M., Higuera, P. E., Duffy, P. A. and Hu, F. S. (2016), Climatic thresholds shape northern high-latitude fire regimes and imply vulnerability to future climate change. Ecography 39: 1-12. http://dx.doi.org/10.1111/ecog.02205
Estimates of carbon released from combustion of vegetation and organic soil during wildfires have improved dramatically over the past decade. Biomass inventory, fire effects and fire severity studies have contributed more accurate data to improve these models. (See Ottmar 2007, Brendan Rogers webinar 2015) However, figuring out the net effect of all the various effects of fire, the recovery phase and warming climate on the carbon stored in Alaska’s forests and tundra is a lot more challenging! You’d have to consider changes in burn extent and/or severity, increases in plant productivity in recovering burns, changes in species composition and what that means for productivity, changes in permafrost distribution and soil C decomposition, methane emissions and carbon fluxes in lake systems and wetlands–etc.! A team lead by Dr. Dave McGuire at UAF has taken on this modeling challenge by applying their Integrated Ecosystem Model (IEM) which includes modules for fire, permafrost, and carbon cycling. Dave recently presented an overview of their findings at an IARPC-WCT/AFSC joint webinar (presentation slides available HERE). In a nutshell, they found: 1) tundra holds 2x the carbon that boreal forest does in the same area 2) there has been a net C loss from boreal land area of about 8 Tg/yr over the last 60 years, primarily driven by large fires during the 2000’s 3) arctic tundra and SE Alaska still act as C sinks, compensating for these losses so that overall, Alaska sequesters about 3.7 Tg/yr, 4) increases in fire extent predicted with with warming climate will release even more C, but longer growing seasons and increased plant growth (as much as 8-19% increased productivity throughout the remainder of this century) with warmer climate and higher CO2 concentration in the atmosphere are estimated to offset these losses under most of the climate projection scenarios. Since this nutshell summary glosses over a lot, you should take a look at the slides and the SNAP projects page with information on scenarios and the individual models used.
On the surface Alaska fire management and boreal ecosystem carbon studies have little in common. But a deeper look reveals the connections between them. Carbon scientists in the last decade have become increasingly interested in fire effects on the legacy carbon locked up in permafrost and the deep, slow-to-decompose organic layer of boreal forest floor (Kasischke et al. 2013, Genet et al. 2013). Projections indicating more extensive, frequent and/or severe fires in northern latitudes with a rapidly warming climate, longer fire seasons, and more lightning (Romps, et al. 2014) lend a certain urgency to attempts to quantify the potential impacts of fire-released carbon on greenhouse warming. Fire management agencies are less interested in long-term impacts of fire-released gasses but they are more and more driven to assess impacts of smoke on communities. Work at the boundary between the two sets of interests has started to yield some interesting results. For example, Michigan Tech Research Institute has joined their consumption field data from NASA studies to the USFS Consume Model and FCCS fuels maps and LANDFIRE fire perimeters in a web-based tool that provides users a simple interface for computing wildland fire emissions (1-km spatial resolution). The Wildland Fire Emissions Information System (WFEIS) can calculate tons of CO2 or other gases from large fires across the US and Canada from 1984-2010. Although this tool is for post-facto emissions analysis it is a good example of how large spatial data sets and complex equations can be united in a simple graphical interface allowing one to–say–query the forest fire emissions from the 231,000 acres burned in Alaska in 2010 (10.9 million tons CO2, 95,000 tons PM 2.5). The hope is that weather modeling and research linkages with the common fire danger and risk rating system used in northern latitudes (CFFDRS) will soon bring this kind of application into the real-time and forecast prediction realm.