The following is the established format for referencing this article:Long, J. W., S. A. Drury, S. G. Evans, C. J. Maxwell, and R. M. Scheller. 2022. Comparing smoke emissions and impacts under alternative forest management regimes. Ecology and Society 27(4):26.
ABSTRACTSmoke from wildfires has become a growing public health issue around the world but especially in western North America and California. At the same time, managers and scientists recommend thinning and intentional use of wildland fires to restore forest health and reduce smoke from poorly controlled wildfires. Because of the changing climate and management paradigms, the evaluation of smoke impacts needs to shift evaluations from the scale of individual fire events to long-term fire regimes and regional impacts under different management strategies. To confront this challenge, we integrated three widely used modeling tools to analyze smoke impacts across different management scenarios within a future of changing climate. We applied this multi-stage framework to a case study analysis in the Lake Tahoe basin, in which managers proposed scenarios that involved varying levels of hand- and mechanical-thinning treatments and prescribed fires. We began by using the LANDIS-II model to project daily emissions of fine particulate matter from wildland fires under various climate and management scenarios over a century. We also modeled dispersion and health impacts based upon individual wildfire events selected to be representative of different management scenarios. For those events, we modeled smoke conveyance to downwind communities from representative future fires using the BlueSky smoke dispersion model. Lastly, we estimated human health impacts resulting from the modeled smoke using the U.S. Environmental Protection Agency's BenMAP model. Our results suggest that emissions from wildfires will substantially increase in future decades; however, increased levels of forest thinning could substantially reduce those emissions and harmful health impacts from large wildfires. We also found that increased use of prescribed burning could reduce the health impacts associated with large wildfires but would also increase the frequency of low levels of emissions. Furthermore, the modeling results suggested that individual prescribed fires could have substantial health impacts if dispersion conditions are unfavorable. Our results suggest that increased management is likely to yield important benefits given expected increases in wildfire activity associated with climate change. However, there remain many challenges to projecting the effects of alternative management regimes, especially ones that involve substantial increases in intentional burning.
Smoke exposure from wildfires is a growing problem in California and North America because of impacts to public health and associated social and economic values including outdoor recreation (Fann et al. 2018, Chen et al. 2021, Gellman et al. 2022). This health risk is expected to grow substantially over the coming decades due to climate change (Li et al. 2020, Burke et al. 2021). Wildfires are a significant concern within the Lake Tahoe basin because of their potential to damage forest and aquatic ecosystems, life and property, and air quality (Stevens et al. 2016, Abelson et al. 2021). The basin is prone to inversions that trap smoke from both local fires and fires in other parts of California (Gertler et al. 2010). Wildfire emissions add to other air pollutants originating within and being transported from long distances into the basin (Gertler et al. 2010). Consequently, smoke impacts have become a major concern for both land and fire managers and residents in the basin and wider region (Cisneros et al. 2018) (Fig. 1).
Managers and scientists have recommended treatments such as forest thinning and the use of wildland fire to restore forests, which in turn would reduce the health impacts of smoke from poorly controlled wildfires (Schweizer and Cisneros 2017, Schweizer et al. 2019, D’Evelyn et al. 2022). Thinning for forest restoration under such recommendations, and as typically applied in the basin, refers to the harvest of smaller trees along with reduction of the resulting “activity fuels” through pile burning, mastication, or removal off-site (Safford et al. 2009, Harrison et al. 2016, Low et al. 2021). Use of wildland fire can include prescribed burning, management of naturally-ignited wildland fires to achieve resource objectives (North et al. 2021), and cultural burning by Indigenous fire practitioners (Long et al. 2021) (Fig. 2). Previous studies have compared smoke impacts from individual wildfires under alternative treatment scenarios (Navarro et al. 2016, Stevens et al. 2016, Schweizer et al. 2020). Calls to ramp up use of wildland fire, as well as expected increases in unplanned wildfires, require that evaluations shift from the scale of individual fire events to decades or centuries of landscape fire dynamics under different management strategies.
Our research was part of a broader evaluation of different management approaches for the Lake Tahoe West Restoration Partnership (https://www.nationalforests.org/who-we-are/regional-offices/california-program/laketahoewest). We sought to answer two questions that were important to managers and the broader public. First, how would different management scenarios compare in terms of smoke emissions over the long term? Second, under different management scenarios, how might emissions from extreme wildfire events (those producing the most intense emissions), as well as typical prescribed burn events, impact downwind human communities?
We analyzed several metrics of smoke emissions that had been suggested as effective representations of air quality and potential health impacts in previous studies (Long et al. 2018, Striplin et al. 2020). Alternative management scenarios should be evaluated over long periods to account for the potential of treatments to mitigate future wildfire impacts. However, modeling the thousands of fires that are expected to burn over many decades is extremely computationally intensive; even modeling a single year of smoke dispersion poses a substantial computing challenge. To make an analysis of trade-offs more tractable, we sought metrics that could be computed relatively easily to approximate smoke impacts.
Projections of health impacts entail wide uncertainty due to variable dispersion and resulting population exposure (Long et al. 2018, Mueller et al. 2020). For example, smoke from the early summer Angora wildfire of 2007 was trapped in the basin and caused extreme particulate levels at locations directly within the plume (Cahill 2009). However, due to the ventilation dynamics in the Lake Tahoe basin, smoke emissions in the summer commonly disperse more than during the late fall, which has traditionally been the preferred period for prescribed burning (Cahill 2009). Low-intensity fires may not generate sufficient energy to loft smoke high into the air column where it can be more widely dispersed. Physiochemical processes cause the smoke emissions of wild and prescribed fires to differ in ways that are complex to model (Williamson et al. 2016). Accounting for dispersion is important, although it adds greater complexity to modeling effects of smoke. To demonstrate how different levels of PM2.5 emissions may translate into health impacts, we modeled dispersion and health impacts of events that we selected to be representative of different management scenarios. As a final metric to help managers evaluate different management approaches, we considered the number of days of prescribed burning, both understory burning and pile burning, under each scenario.
The Lake Tahoe basin lies in the Sierra Nevada mountains along the boundary between California and Nevada, USA (Fig. 3). The basin includes approximately 134,000 ha ranging in elevation from 1900 to 3050 meters, with 36% of the area occupied by Lake Tahoe itself and 7% of the area occupied by urban development. The remaining surrounding lands are dominated by conifer forests with smaller areas of shrublands, meadows, aspen groves, and riparian areas. The United States Forest Service manages about 75% of the land area in the basin, with various state, local, private, and tribal landowners managing the rest. The basin lies in the heart of the ancestral territory of the Washoe Tribe of California and Nevada, which owns several parcels in the basin and also has been actively involved in cooperative restoration of meadows in the Lake Tahoe basin, including preparation to reestablish cultural burning (Davenport 2019). Forests at lower elevations in the basin burned frequently due to both lightning ignitions and burning by Washoe people to cultivate desired resource conditions, however, fire suppression since Euro-American colonization displaced both types of ignitions (Lindstrom 2000).
The Lake Tahoe West (LTW) study area includes over 23,600 ha on the western part of the basin. Like the rest of the basin, LTW has a high proportion of Wildland-Urban Interface (WUI) zones, which delineates developed areas amidst forests. Draining from the eastern slope of the Sierra Nevada, LTW has a higher proportion of wilderness (about one-quarter of its area) than the basin as a whole (Fig. 3), and it tends to have more mesic site moisture conditions (Beaty and Taylor 2008) than sites on the east side of the basin (Taylor 2004).
Modeling fire and management regimes, emissions, and smoke impacts
We constructed our analysis based upon a previously proposed three-step framework (Long et al. 2018): 1) project daily smoke emissions, 2) project daily smoke dispersal based upon weather conditions, and 3) project smoke impacts on air quality conditions in health or economic terms (Fig. 4). This framework is comparable to one proposed by Williamson et al. (2016). Each step was associated with a different model, and we connected each step to evaluate overall outcomes. As a foundational step, we first needed to simulate the forest dynamics, including burning, that drive smoke emissions and impacts.
Projecting landscape dynamics based upon climate and management scenarios
We began by using the LANDIS-II integrated landscape change model (https://www.landis-ii.org/home) to forecast forest vegetation change and fire dynamics over a century (2010–2110) across different climate and management scenarios. LANDIS-II is a process-based simulation model that integrates forest growth and succession, climate change, and disturbances (Scheller et al. 2007). The model has previously been used to model landscape dynamics in the basin. As part of the Lake Tahoe West collaborative modeling effort, we ran the LANDIS-II model with the Net Ecosystem Carbon and Nitrogen (NECN) succession extension (v. 6.1)(Scheller et al. 2011), the Social-Climate Related Pyrogenic Processes and their Landscape Effects (SCRPPLE) fire extension (v. 2.1)(Scheller et al. 2019), and biomass harvest extension (v. 2.0) to simulate management activities and natural processes, and their interaction, over the century across the entire Lake Tahoe basin landscape. We ran the model with each pixel representing 100 m by 100 m (1 ha). Wildfires in the model are stochastic with ignitions varying spatially and temporally across the landscape. Reflecting complex terrain, shifting wind directions, and variable fuels, the model can simulate complex burns with patchy fire intensity including unburned patches (Scheller et al. 2019). The model generates outputs of burned areas with variable fire intensity classes that are considered functionally equivalent to severity (Scheller et al. 2019). Multiple instances of the model are run to account for the effects of stochasticity on fire location and severity. This replication is important for better understanding of the impacts of management and climate scenarios on landscape fire behavior, as well as smoke emissions and air quality impacts.
The five management scenarios
The five management scenarios were developed through a collaborative process by an interagency team of resource managers that developed integrated goals and targets with input from agency decision-makers, technical experts, and stakeholder committees representing conservation, fire protection, recreation, homeowners and businesses, and local governments (Lake Tahoe West Restoration Partnership 2019). The management scenarios were intended to represent contrasting management approaches, rather than to precisely emulate specific sequences of treatments. Nevertheless, we obtained and used extensive input from managers to ensure that assumptions were generally consistent with likely practices. Scenario 1 involved no treatment other than suppression; scenario 2 emulated the recent history of thinning in WUI areas; scenario 3 represented more intensive and extensive thinning in all zones, while scenarios 4 and 5 included prescribed understory burning at modest and high levels, along with thinning at a rate like that in scenario 2. Key differences among the scenarios in terms of thinning, prescribed burning, and wildfire suppression within WUI and non-WUI zones are shown in Table 1, and additional details about the modeling are included in the supplemental materials.
Representing thinning and residual fuel treatments
The LANDIS-II model operates using species–age cohorts for computational tractability, so the size limits associated with treatments were translated into age classes. Analyses of recent fuel treatments within the basin were used to estimate the volume of biomass removed by species and age cohorts. Mechanical treatment was applied to represent whole removal of larger trees (generally up to 76 cm DBH, although some larger trees up to 97 cm DBH were removed in the most intensive thinning in scenario 3) in areas that were generally accessible based upon slope and distance to roads, while hand thinning was limited to smaller trees (less than 36 cm DBH). Managers suggested that residuals from hand-thinning treatments would be burned in piles but that mechanical treatments could have biomass removed off-site since those areas would be closer to roads and generate economic returns that could offset hauling costs. We assumed that such biomass removals would result in no local emissions of PM2.5. These assumptions reflect expected practices in the basin, although pile burning has often predominated due to the costs of removing the thinned biomass, even including boles of medium and large trees. Managers were hopeful that there would be opportunities to remove forest slash for bioenergy production, which could reduce emissions of PM2.5 and other pollutants (Springsteen et al. 2011).
Use of wildland fire
Two “fire-focused management scenarios” were developed to represent use of prescribed understory burning. These burns were fixed to burn at a low intensity within single days with various assumptions regarding timing, size, and fire weather based upon input from managers (Table 2). The simulated understory burns result in compact square shapes since they spread in a less sophisticated way than wildfires in the model to ensure that they attain a desired target size. Scenario 5 was added after the first four to compare effects of using more prescribed burning to achieve the high treatment levels that scenario 3 achieved using thinning. To facilitate the increase in prescribed burning compared to scenario 4, prescribed burns in scenario 5 were expanded to 73 ha/day and they were allowed to burn whenever fire weather conditions were suitable. These burns were concentrated in the spring and fall (consistent with typical practices), but some occurred throughout the year. A notable difference between the understory burns and thinning treatments is that prescribed fires could occur in vegetation types that were not dominated by trees, i.e., in shrub-dominated areas. Wildfire suppression was also modified under scenarios 4 and 5 to represent management of wildfires to achieve resource objectives outside of WUI zones, although manager guidance suggested that such practices would be highly constrained (in part because only 42% of Lake Tahoe West is not designated as WUI, and that includes large rocky areas at high elevations where large fires are unlikely to start and spread). As a result, only relatively small areas (about 40 ha/yr) experienced such burns.
The LANDIS-II model incorporates climate projections to inform future landscape disturbances. The main landscape modeling effort used four general circulation models (GCMs) that were recommended in California’s Fourth Climate Change Assessment (Pierce et al. 2018) including Hadley Center Global Environment Model (HadGEM2), Canadian Earth System Model (CanESM2), Centre National de Recherches Meteorologiques (CNRM5), and Model for Interdisciplinary Research on Climate (MIROC5). These climate projections included those GCMs run under two representative concentration pathways (RCP) 4.5 and 8.5, which respectively represent projections for medium emissions, representing substantial emissions mitigation policies, and “higher emissions, representing a continuation of current emissions policies (Pierce et al. 2018, Schwalm et al. 2020). LANDIS-II results were obtained for three replicates under each of the eight combinations of GCM and RCP, for a total of 24 runs of outputs for each management scenario. We present emissions results based upon the average of those runs. However, smoke dispersion and impacts are inherently spatial, so it was not possible to average the many LANDIS-II outputs from the dozens of replicates based upon different climate projections. Instead, we limited our analysis to a smaller set of runs from an initial round of modeling that was based upon a single climate change projection (the combination of CanESM2 GCM and RCP 4.5 emissions pathway) that was replicated 10 times for management scenarios 1 through 4. While Pierce et al. (2018) recommended working with a broad range of climate projections to address uncertainty, they regarded the CanESM2 GCM as being closest to the “average” projection.
Quantifying total and daily smoke emissions under different management scenarios using LANDIS-II
We focused on emissions of fine particulate matter (airborne particles less than 2.5 microns in diameter, or PM2.5) because it is one of the pollutants of chief concern for human health in the region (Cisneros et al. 2014), it is used as an epidemiologic proxy for health impacts of smoke (Williamson et al. 2016), and it is expected to be strongly influenced by wildfire emissions under climate change (Jacob and Winner 2009, Williamson et al. 2016). Studies have shown that fuels, burn rate, and intensity, which combine to yield daily emissions, are key factors that influence smoke impacts on populations. For example, Schweizer et al. (2019) found that “wildfires at lower-intensity and burn rates limit the smoke extent and largely capture emission on forest land while large high-intensity fires transport further and impact large high-density populations”; they added that “high burn rates” of more than 250 hectares per day were associated with a 10-fold increase in person-days of exposure per hectare compared to slower-burning wildfires and about a 73-fold increase compared to prescribed fires. Because of such relationships, daily emissions of PM2.5 could be a useful, albeit imperfect, proxy indicator for health impacts (Long et al. 2018). They are also a useful indicators because they can be generated from landscape-scale disturbance models that projects burn extent and intensity for each day.
We calculated both total and daily emissions of PM2.5 over time under the various management scenarios. We quantified smoke emissions based upon 1) the footprint and severity of wildfires and prescribed understory burns; 2) the biomass consumed within those fire footprints; and 3) emissions of PM2.5 released from the biomass consumed during the fire events. We estimated area burned and biomass available to burn for each simulated fire over the full century for each management scenario based on the LANDIS-II modeling conducted across the Lake Tahoe West landscape (Maxwell et al. 2022). For each simulated fire, we estimated fuel consumed in the flaming phase and smoldering phase for each fuel component, based upon fire severity, as shown in Table 3, with consumption values informed by previous work by Drury et al. (2014). Fire severity is determined within the SCRPPLE extension to LANDIS-II based upon fuel loads (fine and ladder) and neighborhood effects such that high-intensity fire spreads from pixel to pixel as high-intensity fire (Scheller et al. 2019). The Net Ecosystem Carbon Nitrogen (NECN) extension to LANDIS-II keeps track of biomass that moves through pools of litter, duff, dead woody biomass, and live woody biomass. When a fire occurs, LANDIS-II, the SCRPPLE generates maps of flaming and smoldering consumption based on the biomass pools from NECN and standard emissions factors for wildfires in conifer forests in the northwest (Urbanski 2014). Smoldering emissions are associated with duff, organic soils, and rotten logs (Ottmar 2014). Wildfire emissions often extend across multiple days with a gradual shift from flaming to smoldering. However, there is no established method for allocating emissions across multiple days (Long et al. 2018), so we applied professional judgment informed by Drury et al. (2014) to allocate the fine particle emissions over three-day periods with declining smoldering consumption in the duff and dead wood fuel components (Table 3). Specifically, we assigned all the flaming emissions and 50% of the smoldering emissions to the Julian date LANDIS-II reported the pixel burned (burn day 1). Thirty percent of the smoldering emissions were assigned to one day after LANDIS burn date (burn day 2) and 20% of the smoldering emissions were assigned to burn day 3. During multiday burn periods when multiple pixels were burned, the total emissions estimated to be released on that day included flaming emissions from pixels that burned on day 1 and smoldering emissions from day 2 or day 3 released from surrounding pixels in the modeling space. We calculated both total and daily emissions of PM2.5 over time under the various management scenarios.
Adding pile burning to annual emissions
LANDIS-II does not directly model emissions from pile burning or from wood that is removed from the landscape. However, accounting for those emissions can provide valuable information for evaluating management strategies. We did not include emissions removed off-site under mechanical harvest; such material could be processed into wood products or used for energy production. Previous research from the region indicated that using the waste material in a biomass power cogeneration facility would reduce PM2.5 emissions by 98% compared to pile burning (Springsteen et al. 2015). We calculated how pile burning associated with hand thinning would add to annual fine particle emissions. To convert biomass harvested by hand thinning into PM2.5 emissions, we used the pile burn emissions calculator tool (Fire and Environmental Research Applications Team 2022), which had been developed with input from experts within the Tahoe basin. We entered values for typical pile shape (half-ellipsoid) and median dimensions (3.1-m diameter and 1.2-m high) based upon recent field sampling of piles within the Tahoe basin (Hubbert et al. 2013) and the tool’s default consumption value of 90%. This yield applied an emissions factor of 6 kg PM2.5 per Mg of biomass harvested via hand thinning, which is slightly higher than the average of 5.3 kg PM2.5 per Mg of “dry forest slash” reported by (Springsteen et al. 2015). We added these emissions to the annual totals from wildfires and prescribed understory burns simulated in LANDIS-II (although in practice there would typically be a few years lag between harvest and pile burning to allow the cut biomass to dry). We did not include pile burning emissions in analyses at the daily scale, since we assumed that managers could time such burns to avoid creating daily impacts based upon manager guidance and previous work (Malamakal et al. 2013).
Evaluating daily emissions
Given a lack of established criteria for evaluating daily emissions, we thought it would be useful to categorize them into bins to highlight the levels of emissions that have been observed to more frequently cause smoke impacts. We developed those bins (Table 4) based upon guidance from Leland Tarnay, an ecologist with the U.S. Forest Service Region 5 Remote Sensing Lab, who has extensive experience monitoring and modeling fire emissions (Fusina et al. 2007, Preisler et al. 2015, Mueller et al. 2020) and studying air quality in the Lake Tahoe basin (Tarnay et al. 2005). In addition, we drew upon results from a previous study of different types of fires (Long et al. 2018), which found that daily emissions of PM2.5from prescribed burns in the Yosemite area remained below 200 Mg PM2.5/day, while resource objective wildfires sometimes had localized smoke impacts when emissions were between 200 Mg and 500 Mg PM2.5/day. Resource objective wildfires may have greater impacts than prescribed burns because they may burn faster, across larger areas, and with less emphasis on minimizing smoke impacts (Schweizer et al. 2020). The extreme Rim Fire of 2013, however, frequently exceeded 500 Mg PM2.5 and peaked at nearly 10,000 Mg PM2.5/day (Long et al. 2018), causing widespread smoke impacts across the region. The Rim Fire and other extreme smoke events, including the Caldor Fire of 2021, have affected the Tahoe basin (Fig. 1). Schweizer et al. (2019) concluded that a high wildfire burn rate of 250 ha/day would be more likely to cause smoke impacts. Multiplying that daily threshold by an emissions rate of 0.45 Mg PM2.5/ha, which is associated with relatively heavy forest fuels (Long et al. 2018), yields a threshold of 120 Mg PM2.5/day, which falls into the “high daily emissions” bin (Table 4). Specific relationships between burn rates and smoke impacts could vary substantially from airshed to airshed due to different topography, dispersion, and fuels. Consequently, we urge caution in extending these criteria to other regions.
Projecting smoke dispersion
We used the BlueSky Playground (2.0 beta, https://playground.airfire.org/) to model dispersion of fine particulate matter (PM2.5). Fusina et al. (2007) previously demonstrated the platform’s utility in predicting emissions and PM2.5 concentrations at surface stations from wildfires in Northern California. The BlueSky system requires a dispersion model; a weather forecast; and the location, timing, amount, and composition of emissions. For the smoke dispersion model, we selected the HYSPLIT option which has been shown to provide good predictions within areas of complex topography such as the Lake Tahoe basin (Malamakal et al. 2013). However, the HYSPLIT option limits users to the weather forecasts hardwired into BlueSky Playground rather than customizing meteorological conditions. To represent three smoke-dispersal forecasts for each fire event we modeled, we used dates of fire emissions discussed previously and then applied fire weather data for the same Julian dates corresponding to the three most recent years of data (2016, 2017, 2018) from the highly resolved (2-km scale) Weather Research and Forecasting (WRF) forecast for California and Nevada (https://cansac.dri.edu/). For the emissions, BlueSky Playground allows for customizing data on fuel consumption and emissions when measured data are available. For each BlueSky Playground run, we created a default emissions scenario based on modeled fire locations from LANDIS-II runs and then substituted our values for emissions and energy release from LANDIS-II. This approach provided more precise data for projecting the lofting of emissions into the atmosphere. We estimated the energy produced during the flaming and smoldering phases of combustion using the inherent heat content of forest fuels of 18608 Kj/kg of fuel consumed. The energy released simulates the lifting of emitted particles into the air where the particles enter the dispersion model.
To demonstrate how our modeled projections might translate into smoke impacts, we winnowed our LANDIS outputs down to a more manageable subset of data that represented alternative management scenarios under the CanESM2 climate projection that appeared to generate average results (Pierce et al. 2018). We extracted results from a single model year 30 (2039 on the LANDIS-II temporal scale) for our modeling exercise based upon the assumption that 30 years into the future was a reasonable period to allow for different management scenarios to influence fuels and fire regimes across the landscape. For this step, we did not model scenario 5, representing increased burning, since it had not yet been developed when we conducted the dispersion modeling. LANDIS-II modeling produced ten replicates for each management scenario under the selected climate projection. To approximate both the extreme and more typical outcomes, we ranked the replicates from 1 to 10 based on the maximum daily emissions in year 30 and then analyzed the air quality impacts associated with the worst days of emissions from the 1st, 5th, and 6th ranked replicates. The replicate with the highest maximum daily value represented the worst outcome for each scenario, while results from the 5th and 6th ranked replicates can be averaged to represent median outcomes.
As a general comparison, we also modeled a fall season prescribed burn under scenario 4. Previous research has suggested that prescribed burns, including pile burns, in the Lake Tahoe basin may cause minor smoke impacts (e.g., less than a standard of 35 µg PM2.5 m-3 over 24 h) but are unlikely to cause serious impacts because typically only relatively small areas have been burned during favorable dispersion windows (Malamakal et al. 2013, Striplin et al. 2020). Furthermore, managers tend to favor fall season burns because spring burns are considered out-of-season burns, may be less effective in consuming fuels (Striplin et al. 2020), and may conflict with regulations to protect wildlife breeding. We did not attempt to model dispersion of pile burns or spring prescribed understory burning because of the expectation that those practices would not typically cause significant smoke impacts and because of computational constraints. These assumptions and potential outcomes warrant further attention if prescribed burning is greatly expanded.
Evaluating smoke impacts
In the third step in our analysis, we followed the methodology described in Jones et al. (2016) and used the BenMAP-Community Edition tool to model health impacts and associated economic values (Sacks et al. 2018, U.S. EPA 2021). The user provides the model with gridded pollution data and a region of interest for assessing impacts. For this study, the gridded pollution data was PM2.5 data obtained from fires simulated using BlueSky in step two.
We estimated both morbidity and mortality impacts of smoke using a willingness to pay (WTP) approach. We first estimated the physical impact of the smoke exposure and then translated those impacts into a monetary value using an estimate of willingness to pay to avoid the mortality or morbidity damages of the smoke exposure. BenMAP also allows users to estimate the expected expenditures associated with specific health endpoints, such as respiratory conditions, after pollution exposure. This is often referred to as a cost-of-illness approach and generally represents a lower bound estimate of the economic damages of pollution (Jones et al. (2016) for a comparison of the two approaches in the context of wildfire smoke). When data are available, the WTP approach is preferred since it assesses the economic costs of pollution more fully (Jaafar et al. 2018).
The BenMAP model first estimates the physical impact of the pollution on user-specified non-overlapping health endpoints. For each health endpoint, the user can choose to either utilize predetermined health impact functions (also referred to as dose-response functions) or provide their own from the literature. For this study, we used minor restricted activity days (MRADs) as the health endpoint for the non-fatal health impacts of smoke exposure. As noted in Jones et al. (2016), MRADs are defined as “any day on which an individual was forced to alter his or her normal activities due to minor illnesses, including both respiratory and non-respiratory conditions.” The health impact function for estimating MRADs is taken from Ostro and Rothschild (1989) and input directly into BenMAP. For the all-cause mortality health endpoint, we used the dose-response function estimated in Johnston et al. (2011).
The final step in the BenMAP analysis is to assign an economic value to the physical health endpoints. For a non-fatal MRAD health endpoint, we used a marginal WTP estimate from Jones et al. (2016). Based on a survey of residents in Albuquerque, New Mexico, the authors found the residents would be willing to pay $130.79 to avoid adverse health effects associated with wildfire events. We used this estimate after updating it for inflation to 2018 values. This WTP estimate of general willingness to pay for avoided health damages aligns conceptually with the MRAD health endpoint. To estimate the economic cost of all-cause mortality, we used the value of a statistical life (VSL) concept. The value of a statistical life is a commonly used money metric for valuing willingness to pay to reduce mortality risk at the population level in cost-benefit analysis (Kniesner and Viscusi 2019). The BenMAP manual (U.S. EPA 2021) recommended using a VSL of $7.4 million (2006 dollars), which we adjusted for inflation to $9.3 million in 2018 equivalents.
For each simulated event in this modeling space, we estimated the health impact associated with three days of PM2.5 exposure generated in Step 2. After three days, we assumed that wildfire smoke would be fully dispersed with negligible additional health effects, recognizing that this was a simplifying assumption (and not representative of many giant wildfires that have occurred primarily outside of the Lake Tahoe basin in recent years). While some wildfires burned longer in LANDIS-II, the ones with highest emissions tended to drop rapidly after they peaked. The three replicates were ones with the first (highest), fifth, and sixth (median) peak daily emissions out of the 10 replicates generated with LANDIS-II modeling for year 2039. In total, we ran 108 BenMAP simulations of wildfire smoke days (3 replicates × 3 weather patterns × 4 management scenarios × 3 days) and 9 BenMAP runs for the single prescribed fire under scenario 4 (3 weather patterns × 3 days).
Modeling frequency of prescribed burning
As a supplementary analysis to evaluate the feasibility of prescribed burning, we compared the number of days of prescribed and pile burning compared to the number of days historically considered suitable for such burns. First, we counted days with prescribed understory burns simulated in LANDIS-II. Then, we divided the area treated by hand thinning each year, as reported in the LANDIS-II results, by 125 acres (50.6 ha) per day; managers in the advisory group suggested that daily rate would be a reasonable constraint. That value is slightly higher than the typical amount of less than 40 ha reported by Malamakal et al. (2013); however, it was twice as large as the largest prescribed burn (24 ha) reported through Prescribed Fire Information Reporting System (PFIRS, https://ssl.arb.ca.gov/pfirs/) for Lake Tahoe during the years 2014–2017. Many factors, including proximity to homes, staffing within different land management organizations, weather, and air quality, govern how such burns are accomplished or might be accomplished in the future, but we thought the results would be clearer if we applied the same rate to all management scenarios. We assumed that days allocated for pile burning and understory burning would be additive since pile burning might effectively displace prescribed understory burning (especially if relatively large areas were being burned) and because pile burning can occur during winter when understory burning is not feasible.
We then compared the resulting average number of burn days under each management scenario against the number of “available burn days.” Striplin et al. (2020) determined that there was an average of 96 days available for prescribed burning in the Lake Tahoe basin based upon three conditions: approval by air regulators for burning (“approved burn day”), suitable fire weather and fuel moisture (“within prescription”), and sufficient availability of firefighting resources. We did not attempt to analyze whether each day of burning would be feasible on a day-by-day, or season-by-season basis, although the LANDIS-II model did constrain prescribed understory burning to days of suitable fire weather (Scheller et al. 2019).
Cumulative emissions of fine particulate matter
We projected that while annual emissions of fine particulate matter would greatly increase over the next century across most management scenarios, the choice of management scenario would strongly influence those trajectories. Figure 5 compares the mean and standard deviation for cumulative emissions from wildfires and prescribed burns under each of the five scenarios (emissions from wildfires and prescribed understory burns are shown in 5A; emissions from hand piles are shown separately in 5B, and the total is shown in 5C; average values are shown in Supplemental Table S8). As an example of how wildfire emissions would accelerate, under scenario 2, such emissions over the late decade of the modeled century were nine times greater, on average, than under the first decade. Under the suppression-only scenario 1, wildfire emissions over the last decade were nearly 20 times greater than during the first decade.
For all scenarios that involved harvest (scenarios 2–5), emissions from pile burning (5B) raised total emissions while mitigating emissions from wildfires (5A). For example, the no-treatment scenario 1 resulted in 96% greater cumulative emissions from wildfires but only 48% greater cumulative emissions after accounting for pile burning emissions in scenario 2. Under the high thinning scenario 3, most emissions would result from pile burning of residual materials; it reduced the cumulative emissions from wildfires relative to scenario 2 by 61%, but only by 3% after accounting for pile burning emissions.
Adding understory prescribed burning in scenarios 4 and 5 raised total emissions compared to the other scenarios while tamping down on wildfire emissions, especially later in the century as wildfires become more widespread. Scenario 4 resulted in total cumulative emissions that were 69% greater than under scenario 2 and 14% higher than under scenario 1. Meanwhile, dramatically increasing prescribed understory burning under scenario 5 resulted in the highest total emissions compared to other scenarios; for example, total average emissions were 187% greater than under scenario 2.
However, both scenarios involving high levels of treatments (scenarios 3 and 5) reduced the variation in emissions across replicates and climate projections compared to the other three scenarios. Indeed, scenario 1 had the highest variation across replicates and climate projections, so that some replicates under no-treatment scenario resulted in higher total emissions than under scenario 4, even though scenario 4 had a higher average.
Daily emissions of fine particulate matter
We found that increased treatment reduced days of very high and extreme emissions of fine particulate matter from wildfires while increasing use of prescribed fire (scenarios 4 and 5) would increase days of more modest emissions (Fig. 6). Consequently, prescribed fires effectively shifted emissions into more frequent releases of lower daily emissions. Suppression-only management (scenario 1) averaged over 10 days of very high or extreme emissions per decade, while scenarios with the most treatment (scenarios 3 and 5) reduced the average number of such days below two per decade. Scenario 3 with the most thinning reduced the risk of very high emission days, while scenario 1 with no treatment greatly increased that risk. More modest levels of treatment (scenarios 2 and 4) were intermediate, averaging 7 and 5.5 days per decade, respectively, because each scenario resulted in comparable levels of treatment. Our results suggested that air quality would deteriorate due to fire activity over time under all scenarios. However, high emission days did become more frequent in earlier decades under scenarios 1, 2, and 4, while greatly increasing overall treatment (scenarios 3 and 5) sharply reduced the incidence of such days in early decades.
Our analysis of conditions in a single future year (Fig. 7) indicated that greatly increasing the amount of treatment under scenarios 3 and 4 resulted in lower peaks and fewer days of very high fine particulate emissions compared to scenarios 1 and 2. Prescribed understory burns resulted in more frequent, low-level emissions concentrated in late fall under scenario 4. Those patterns are reflected in increased days of “moderate” emissions in Figure 5.
Figure 8 shows the maximum daily emissions within that future model year (2039) for each of the ten replicates run for the first four management scenarios under a single climate change projection (the CanESM2 GCM × RCP4.5 emissions pathway). The results show that daily maximum emissions varied substantially across replicates, from 705 Mg down to a negligible 2 Mg. Scenario 3 consistently generated the lowest maximum emissions across the 10 replicates compared to the other scenarios, indicating that it would consistently abate high impacts. Scenario 1 resulted in higher daily maximums than any other scenario for the worst 6 replicates, while scenario 2 had higher daily maximums than other scenarios for the four best replicates. The results in Figure 8 represent only a single future model year, but they appear consistent with the full-century trend (Fig. 6), in which increased thinning treatment under scenario 3 reduced maximum daily emissions, while the lower treatment scenarios (1 and 2) risked incurring higher maximum daily emissions.
Our analysis of select future wildfire events indicated that increases in treatment would likely reduce the resulting smoke impacts from future wildfire events. Our BlueSky modeling depicted the effects of individual fire events on downwind particulate matter levels relative to standards to protect human health. It revealed how the most extreme wildfire in a future model year (2039) would have more regionally extensive and severe impacts than a modest prescribed burn, although the latter could still elevate pollution in populated areas within the Lake Tahoe basin and nearby greater Reno area (Fig. 9). Figure 10 spatially depicts results for the fifth-ranked replicates (approximating a median outcome for that future year), while highlighting the projected fine particle levels in the urban centers of Reno and Carson City (notably without factoring in background levels or other potential pollutant sources) across each of the first four management scenarios. This figure illustrates that huge areas could be affected by emissions from large wildfires especially under scenarios 1 and 2, which involved the least amount of fuel reduction treatment. Scenario 3, which involved the most extensive and intensive thinning, was projected to result in much more mild smoke impacts. This comparison of representative events reinforces the emission days analysis in illustrating how increased treatment would effectively reduce the incidence of extreme smoke events overall and on potential impact to large urban populations.
Health impacts and associated economic costs
Our results indicate that increases in treatment would reduce health impacts from future wildfire events. Higher levels of daily particulate emissions are associated with increased economic impacts, measured in terms of WTP to avoid mortality and morbidity impacts (Fig. 11). We found that extreme wildfires could have very large health impacts on downwind communities, with potential for single large wildfires in the basin to cause up to $80 million of economic losses based upon health risks alone. Those estimates are dominated by the value of a statistical life, with mortality accounting for 96% of total value if one uses WTP for morbidity and more than 99% if one uses COI estimates for morbidity. The results of the representative events analysis indicated that future wildfires under the suppression-only scenario would likely result in much greater harm than wildfires under scenarios with high levels of treatment. However, our rendition of a single prescribed burn also indicated that there could be substantial impacts especially if conducted under less favorable wind patterns.
Annual days of prescribed burning
The average number of days of prescribed burning per year associated with implementing each scenario increased substantially from management scenario 1 through to scenario 5 (Table 5). For scenarios 2 and 3, these values reflected increased levels of hand thinning compared to no thinning under scenario 1. Because the modeled size of pile burning operations was higher than the levels reported in recent PFIRS data, it may not represent “business-as-usual” but instead a more optimistic scenario of expanded burning. The levels of hand thinning, and hence pile burning under scenarios 4 and 5, were comparable to scenario 2, but those scenarios involved understory burning of 30 days per year under scenario 4 and 92 days under the more aggressive scenario 5. Indeed, the 99.1-day average estimated in scenario 5 was slightly higher than a reference value of 96 available days, which was the historical average of one-day windows available for burning reported by Striplin et al. (2020).
This study demonstrates an approach for projecting the smoke impacts of alternative management regimes. We had to adopt numerous simplifying assumptions to model a very complex and dynamic social-ecological system. Further investigations could refine key assumptions regarding consumption and emissions and expand the temporal and spatial scope. Specific results should not be regarded as forecasts, but rather they spotlight key trade-offs. Because there are no consensus yardsticks for evaluating air quality impacts of management regimes, our results suggest how emissions of fire particles, as well as smoke impacts, are likely to vary over near and long terms under different management approaches. We first discuss how our findings suggest that increased treatment could provide multiple benefits, and then we note challenges associated with increasing prescribed burning. We then consider methodological challenges in evaluating air quality effects of management regimes and the increasing opportunities to better project long-term smoke impacts.
Management implication: treatments are likely to yield benefits
Our findings indicate that increasing forest treatment, both thinning and prescribed burning, would reduce expected emissions from wildfires, days of very high emissions associated with wildfire events, and economic impacts of such events. Mitigating wildfire emissions from the study area appears likely to benefit air quality for heavily populated areas including the greater Reno–Sparks metropolitan area and the Central Valley in California. Management that expands thinning (scenario 3) appeared particularly effective in reducing wildfire emissions by reducing the amount of fuel to be consumed in such events. This finding is consistent with previous modeling of management regimes on carbon emissions (Hurteau and North 2008, Loudermilk et al. 2017) and field studies of wildfires (North and Hurteau 2011). The air quality benefits of such a strategy would be greatest when cut materials could be used in controlled biomass or energy production rather than being disposed of through pile burning, as suggested by previous field studies (Springsteen et al. 2011, Springsteen et al. 2015).
We found that strategies that rely heavily on increased prescribed burning would also be effective in reducing extreme daily emissions and associated smoke events. Ecologists have championed increasing use of wildland fire in the Sierra Nevada, particularly to maintain areas treated initially by thinning, to promote a variety of resource objectives (North et al. 2012, D’Evelyn et al. 2022). Previous research in the Lake Tahoe basin suggested that prescribed burns would be effective in mitigating smoke impacts from wildfire (Cahill 2009, Malamakal et al. 2013). As reported by managers and in research by Malamakal et al. (2013), both pile burning and understory burning are expected to remain within daily burn targets, which results in many days of low emissions but avoids high daily emissions. Schweizer et al. (2019) suggested two important mechanisms that are reflected in our findings: 1) that spreading emissions across many days keeps smoke levels below thresholds that are expected to harm both sensitive and general populations, and 2) that favoring more localized, managed burns tends to reduce the number of people exposed to smoke by reducing the potential for large wildfires that resist suppression and impact large urban populations. We found that increases in prescribed fire, in conjunction with modest thinning, would also reduce high daily emissions from wildfires, but that very frequent prescribed burning would greatly increase overall fine particulate emissions. A recent global review of studies that compared prescribed fire and wildfire regimes similarly concluded that increased use of prescribed fire can reduce total wildfire emissions by reducing the size and intensity of wildfires, but also that such regimes often increase the total amount of fire and emissions within a landscape over time (Hunter and Robles 2020). However, they found few studies that evaluated effects on economics or resilience and did not report any that specifically compared effects on air quality. This dearth of literature on this important topic may reflect the challenges in modeling such complex systems.
Climate change and opportunities for adaptation
Our analysis of emission and smoke impacts focused on differences among management scenarios rather than differences under climate change scenarios. As reported in other work in this special issue (Maxwell et al. 2022), the LANDIS-II modeling projected substantial increases in wildfire activity (in addition to drought-related tree mortality) across management and climate scenarios as conditions become warmer. However, there was also considerable variation in fire regimes across individual climate projections; specifically, some projections resulted in different combinations of drier and warmer conditions, which intensified wildfire disturbances (Maxwell et al. 2022). However, the projection based upon the CanESM GCM × RCP4.5 emissions pathway appeared to be a “middle-of-the-road” projection out of the ones examined in the LANDIS-II modeling (Maxwell et al. 2022). These findings suggest that our results are reasonable projections in light of what we might expect from climate change, and they are consistent with research regarding future impacts of climate change on air quality (Jacob and Winner 2009). Our modeling projections suggest what the future could bring, but managers can adapt their strategies as climate change unfolds, as fuel conditions change, and as weather changes. For example, they could shift their use of fire from year-to-year and decade-to-decade.
Challenges with increasing burning
Our results suggest that substantial increases in prescribed burning would also face implementation challenges. We found that the high level of prescribed burning under scenario 5 would often exceed the number of days historically likely to be available for such burning. More prescribed burning could be accomplished within those constraints by burning even larger areas than the 73 ha per day assumed in our model, but such large burns could still require multiday burn windows in the fall. Therefore, implementing scenario 5 would be difficult without policy changes to facilitate more burn days during those critical periods. In addition, warmer, drier conditions and greater wildfire activity resulting from climate change may further winnow the opportunities for burning or force relaxation of the criteria for burning, which could entail greater risks to various objectives for public health, safety, biodiversity conservation, restoration, and air quality (Kupfer et al. 2020).
Recent research on burn day windows in the Lake Tahoe basin suggested that increasing availability of specialized fire personnel in the fall and spring and adjusting prescriptions to burn more in the spring might be necessary to substantially increase burning (Striplin et al. 2020). An even more recent analysis (York et al. 2021) at the Blodgett Forest Research Station, to the west of Lake Tahoe, suggested that burning opportunities would be even narrower than Striplin et al. (2020) had indicated, particularly because larger prescribed understory burns would require longer windows of dry fuels to be successful. In contrast, pile burns can be completed even under relatively wet conditions when the residual materials are kept dry under plastic sheets (Aurell et al. 2017). Additional research may be warranted to evaluate potential effects of conducting more burning in winter and spring, since such practices have not been widely studied and air quality may also be a constraint in the winter (York et al. 2021).
While our results suggest that increasing treatments would achieve substantial benefits in terms of averting extreme wildfire emissions, we did not identify a point at which increasing prescribed burning might no longer improve net air quality and associated economic costs. Implementing the scale of prescribed burning under scenario 4, and especially scenario 5, would increase the probability that some of those burns would result in smoke impacts especially for communities near treated areas within the basin, despite efforts by managers to mitigate them. The probability of impacts likely would increase if managers used more marginal burn days to achieve those targets. Our results demonstrated that a prescribed burn during unfavorable weather conditions could result in unhealthy air quality in some downwind areas. For example, a small (8 ha) prescribed understory burn in the west side of the basin resulted in elevated levels of PM2.5 in Tahoe City in November 2019 when a nighttime inversion inhibited dispersion (Hobbs 2020).
Limitations in evaluating emission impacts of fire regimes
We highlight several sources of uncertainty in modeling smoke impacts under alternative fire regimes. The first concerns fuels consumption and emissions from fires of different types over time, including prescribed burns, wildfires managed for resources objectives, and extreme wildfires. Recent work in the 2013 Rim Fire, which burned over 1 km2 of mixed conifer forests where fire had been excluded for many decades, suggested that emissions were similar across low, moderate, and high severity burn areas (Harris et al. 2019). They noted that most of the fuel consumed was in surface fuels (litter, duff, and small wood), that consumption of those fuels was high across burn severities, and that consumption of live trees did not vary enough to produce substantial differences in direct emissions (Harris et al. 2019). However, other researchers have noted that fires are managed for resource objectives only under less extreme weather, which would moderate consumption (White 2017). Another study in a repeatedly prescribed burned forest found that fuel loads were nearly two times higher in the first burn than in the second and third, and that percent fuel consumption (compared to pre-burn levels) was highest (65%) for the first burn but dropped for the second (45%) and third (29%) burns (Levine et al. 2020). Our landscape modeling used a simplifying assumption that rates of consumption would be consistent regardless of how recently the area had burned. We also did not apply different emissions factors to wildfires versus prescribed burns or to different seasons of burning. Research has suggested that PM2.5 concentrations might be higher in smoke from prescribed fires, although that reported difference may reflect that prescribed fire smoke was measured closer to the burns (Navarro et al. 2018). Other research has noted that emissions rates vary seasonally (Hiers et al. 2020). These factors suggest opportunities to refine modeling tools to better represent the complexities of fuels, consumption, and emissions and translate those into projections of future smoke impacts.
Our analysis accounts for changes in climate on fire activity including warming and drying, which are likely to be a dominant driver of changing fire effects (Williams et al. 2019); however, we did not attempt to account for changes in wind patterns or human populations. Because the analysis only evaluated effects of fires occurring within the Lake Tahoe West landscape, it does not directly consider how fire activity in other parts of the basin, and smoke from outside of the basin, could also impact air quality and the ability of managers to use fire within the basin. These sources of uncertainty are likely to affect the absolute projections of impacts, but they are less likely to alter the relative performance of different management regimes.
Other sources of uncertainty associated with physical and social factors pose challenges to simulating intentional burning regimes. There is great uncertainty in modeling the location, timing, and smoke dispersion from such burns over decades, which is necessary to understand how a regime based upon frequent and large prescribed burns, or other applications of managed wildland fire, would affect air quality in downwind communities. Research studies are being initiated to help to better quantify the smoke effects of large burns (Prichard et al. 2019), but there are few examples to draw upon from the Lake Tahoe basin and surrounding areas. Furthermore, mitigation strategies can be used by managers to minimize smoke impacts by timing and constraining managed fires and providing advanced warning of potential smoke impacts, particularly for sensitive groups (Long et al. 2018). Such practices are increasingly being incorporated into operations and policies, although our models did not encompass such sophisticated mitigations. Ideally, projections would account for smoke transport patterns and constraints such as crew availability when modeling individual burns.
Improving understanding of health impacts
Recent extreme wildfire seasons have provided opportunities for researchers to better understand and quantify the health impacts of extreme wildfire events. Such investigations are revealing that health impacts from wildfires are likely to have unprecedented costs and to increase dramatically with a hotter climate (Johnston et al. 2021). Our results were consistent with a recent study of the California wildfires of 2018, which reported over $32.2 billion in health costs, with much of those damages occurring in densely populated locations very distant from the fires (Wang et al. 2021). That study similarly found that increased mortality was the dominant portion of those costs. Continued study of such extreme events, as well as cumulative effects across a range of events magnitudes, will help to improve estimates of health and economic impacts of long-term regimes. There could be thresholds or other non-linear relationships that are important to consider when comparing infrequent but extreme smoke events with frequent but low-level emissions. Such information would be useful in comparing trade-offs between highly contrasting management scenarios.
Our results reveal how management regimes that rely on increased thinning with fuels treatment are likely to benefit air quality and human health by reducing impacts from future wildfires. Our results suggest that increases in prescribed burning can also reduce the impacts of wildfires, but they result in much more frequent, low-level particulate emissions overall. Consequently, management regimes that would greatly increase prescribed burning could be challenging to implement. Our analysis represents an initial approximation of health impacts; a more complete analysis would require processing the full duration of multiple fire events within years and over many decades. Advances in computing power and modeling frameworks will enhance the capacity to model the air quality benefits from alternative forest management regimes. Adaptive management that includes monitoring of actual burns will help to better understand how fire can be used effectively while minimizing smoke impacts. In the meantime, averting threats to public health in fire-prone regions could depend on careful planning and execution of intentional burning and proactive engagement with affected communities.
RESPONSES TO THIS ARTICLEResponses to this article are invited. If accepted for publication, your response will be hyperlinked to the article. To submit a response, follow this link. To read responses already accepted, follow this link.
J. L. proposed the initial framework and took the lead in writing the manuscript. C. M., R. S., S. D., and J. L. established assumptions, methods, and protocols used to project emissions based upon scenarios identified by the Lake Tahoe West collaborative. J. L., C. M., S. E., and S. D. developed figures and tables. C. M., R. S., and S. E. developed code used in the analyses. C. M. led computations of emissions with review by S. D. and J. L.; S. D. led the BlueSky analysis, and S. E. led the BenMap analysis. All authors discussed the results and contributed to the final manuscript.
We thank all the contributors to the Lake Tahoe West collaborative, including members of the advisory groups, with special thanks to Brian Garrett for providing insights regarding management practices and constraints, coordinator Sarah DiVittorio, and science team leader Pat Manley. We are grateful to Lee Tarnay for technical suggestions, Ann Hobbs for sharing practical experience in addressing air quality impacts in Lake Tahoe, and Randy Striplin for sharing data on burn day windows. We thank Alec Kretchun for leading early runs of the LANDIS-II model used in the analysis. We also thank the two anonymous reviewers for their very constructive feedback on an earlier version of this manuscript. Lastly, this research was supported in general by the U.S. Department of Agriculture, U.S. Forest Service, and specifically by the project funders for the larger Lake Tahoe West modeling effort, which include the National Forest Foundation, the California Tahoe Conservancy, the U.S. Bureau of Land Management through the Southern Nevada Public Land Management Act, the U.S. Forest Service Lake Tahoe Basin Management Unit, and the U.S. Forest Service Pacific Southwest Research Station. The findings and conclusions in this article are those of the author(s) and should not be construed to represent any official USDA or U.S. Government determination or policy.
Coding and data used to generate the results are available here: https://github.com/LANDIS-II-Foundation/Project-Lake-Tahoe-2017, which has been archived here: https://doi.org/10.5281/zenodo.4644579
Abelson, E. S., K. M. Reynolds, A. M. White, J. W. Long, C. Maxwell, and P. N. Manley. 2021. Promoting long-term forest landscape resilience in the Lake Tahoe basin. bioRxiv: 2021.2006.2029.450434. https://doi.org/10.1101/2021.06.29.450434
Aurell, J., B. K. Gullett, D. Tabor, and N. Yonker. 2017. Emissions from prescribed burning of timber slash piles in Oregon. Atmospheric Environment 150:395-406. https://doi.org/10.1016/j.atmosenv.2016.11.034
Beaty, R. M., and A. H. Taylor. 2008. Fire history and the structure and dynamics of a mixed conifer forest landscape in the northern Sierra Nevada, Lake Tahoe Basin, California, USA. Forest Ecology and Management 255(3-4):707-719. https://doi.org/10.1016/j.foreco.2007.09.044
Burke, M., A. Driscoll, S. Heft-Neal, J. Xue, J. Burney, and M. Wara. 2021. The changing risk and burden of wildfire in the United States. Proceedings of the National Academy of Sciences 118(2):e2011048118. https://doi.org/10.1073/pnas.2011048118
Cahill, T. A. 2009. Effects of wild and prescribed fires on Lake Tahoe air quality. Pages 184-224 in J. W. Long, editor. Effects of fuels management in the Lake Tahoe basin: a scientific literature review. USDA Forest Service, Pacific Southwest Research Station, Albany, CA.
Chen, H., J. M. Samet, P. A. Bromberg, and H. Tong. 2021. Cardiovascular health impacts of wildfire smoke exposure. Particle and Fibre Toxicology 18(1):2. https://doi.org/10.1186/s12989-020-00394-8
Cisneros, R., D. Schweizer, H. Preisler, D. H. Bennett, G. Shaw, and A. Bytnerowicz. 2014. Spatial and seasonal patterns of particulate matter less than 2.5 microns in the Sierra Nevada Mountains, California. Atmospheric Pollution Research 5(4):581-590. https://doi.org/10.5094/APR.2014.067
Cisneros, R., D. Schweizer, L. Tarnay, K. Navarro, D. Veloz, and C. T. Procter. 2018. Climate Change, Forest Fires, and Health in California. Pages 99-130 in R. Akhtar and C. Palagiano, editors. Climate Change and Air Pollution: The Impact on Human Health in Developed and Developing Countries. Springer International Publishing, Cham, Switzerland. https://doi.org/10.1007/978-3-319-61346-8_8
D’Evelyn, S. M., J. Jung, E. Alvarado, J. Baumgartner, P. Caligiuri, R. K. Hagmann, S. B. Henderson, P. F. Hessburg, S. Hopkins, E. J. Kasner, M. A. Krawchuk, J. E. Krenz, J. M. Lydersen, M. E. Marlier, Y. J. Masuda, K. Metlen, G. Mittelstaedt, S. J. Prichard, C. L. Schollaert, E. B. Smith, J. T. Stevens, C. W. Tessum, C. Reeb-Whitaker, J. L. Wilkins, N. H. Wolff, L. M. Wood, R. D. Haugo, and J. T. Spector. 2022. Wildfire, smoke exposure, human health, and environmental justice need to be integrated into forest restoration and management. Current Environmental Health Reports. 2022 May 7:1-20. https://doi.org/10.1007/s40572-022-00355-7
Davenport, N. E. 2019. Naming, Remembering, and Experiencing We’lmelt’iʔ (northern Washoe) Cultural Spaces in Wa she shu It Deh (Washoe Land). Dissertation. University of Nevada, Reno, NV.
Drury, S. A., N. S. Larkin, T. T. Strand, S. Huang, S. J. Strenfel, E. M. Banwell, T. E. O’Brien, and S. M. Raffuse. 2014. Intercomparison of fire size, fuel loading, fuel consumption, and smoke emissions estimates on the 2006 Tripod Fire, Washington, USA. Fire Ecology 10(1):56-83. https://doi.org/10.4996/fireecology.1001056
Fann, N., B. Alman, R. A. Broome, G. G. Morgan, F. H. Johnston, G. Pouliot, and A. G. Rappold. 2018. The health impacts and economic value of wildland fire episodes in the U.S.: 2008-2012. Science of the Total Environment 610-611:802-809. https://doi.org/10.1016/j.scitotenv.2017.08.024
Fire and Environmental Research Applications Team. 2022. Piled fuels biomass and emissions calculator. USDA Forest Service Pacific Northwest Research Station, Seattle, WA.
Fusina, L., S. Zhong, J. Koracin, T. Brown, A. Esperanza, L. Tarnay, and H. Preisler. 2007. Validation of BlueSky smoke prediction system using surface and satellite observations during major wildland fire events in Northern California. Proceedings, USDA Forest Service, Rocky Mountain Research Station, Fort Collins, CO.
Gellman, J., M. Walls, and M. Wibbenmeyer. 2022. Wildfire, smoke, and outdoor recreation in the western United States. Forest Policy and Economics 134:102619. https://doi.org/10.1016/j.forpol.2021.102619
Gertler, A. W., T. A. Cahill, T. M. Cahill , S. C. Cliff, C. Emmett, J. Koracin, H. Kuhns, J. V. Molenar, J. Quashnick, and J. E. Reuter. 2010. Air quality. Pages 37-82 in Z. P. Hymanson and M. W. Collopy, editors. An integrated science plan for the Lake Tahoe basin: conceptual framework and research strategies. U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station, Albany, CA.
Harris, L. B., A. E. Scholl, A. B. Young, B. L. Estes, and A. H. Taylor. 2019. Spatial and temporal dynamics of 20th century carbon storage and emissions after wildfire in an old-growth forest landscape. Forest Ecology and Management 449:117461. https://doi.org/10.1016/j.foreco.2019.117461
Harrison, N. M., A. P. Stubblefield, J. M. Varner, and E. E. Knapp. 2016. Finding balance between fire hazard reduction and erosion control in the Lake Tahoe Basin, California-Nevada. Forest Ecology and Management 360:40-51. https://doi.org/10.1016/j.foreco.2015.10.030
Hiers, J. K., J. J. O’Brien, J. M. Varner, B. W. Butler, M. Dickinson, J. Furman, M. Gallagher, D. Godwin, S. L. Goodrick, S. M. Hood, A. Hudak, L. N. Kobziar, R. Linn, E. L. Loudermilk, S. McCaffrey, K. Robertson, E. M. Rowell, N. Skowronski, A. C. Watts, and K. M. Yedinak. 2020. Prescribed fire science: the case for a refined research agenda. Fire Ecology 16(1):11. https://doi.org/10.1186/s42408-020-0070-8
Hobbs, A. 2020. Smoke management: the risk factor. 3rd International Smoke Symposium. Raleigh, North Carolina, and University of California, Davis, CA, USA.
Hubbert, K., M. D. Busse, and S. Overby. 2013. Effects of pile burning in the LTB on soil and water quality. Final Report SNPLMA 12576, USDA Pacific Southwest Research Station, Albany, CA. https://www.fs.usda.gov/Internet/FSE_DOCUMENTS/stelprdb5441924.pdf
Hunter, M. E., and M. D. Robles. 2020. Tamm review: The effects of prescribed fire on wildfire regimes and impacts: A framework for comparison. Forest Ecology and Management 475:118435. https://doi.org/10.1016/j.foreco.2020.118435
Hurteau, M., and M. North. 2008. Fuel treatment effects on tree-based forest carbon storage and emissions under modeled wildfire scenarios. Frontiers in Ecology and the Environment 7(8):409-414. https://doi.org/10.1890/080049
Jaafar, H., N. A. Razi, A. Azzeri, M. Isahak, and M. Dahlui. 2018. A systematic review of financial implications of air pollution on health in Asia. Environmental Science and Pollution Research 25(30):30009-30020. https://doi.org/10.1007/s11356-018-3049-0
Jacob, D. J., and D. A. Winner. 2009. Effect of climate change on air quality. Atmospheric Environment 43(1):51-63. https://doi.org/10.1016/j.atmosenv.2008.09.051
Johnston, F., I. Hanigan, S. Henderson, G. Morgan, and D. Bowman. 2011. Extreme air pollution events from bushfires and dust storms and their association with mortality in Sydney, Australia 1994-2007. Environmental Research 111(6):811-816. https://doi.org/10.1016/j.envres.2011.05.007
Johnston, F. H., N. Borchers-Arriagada, G. G. Morgan, B. Jalaludin, A. J. Palmer, G. J. Williamson, and D. M. J. S. Bowman. 2021. Unprecedented health costs of smoke-related PM2.5 from the 2019–20 Australian megafires. Nature Sustainability 4(1):42-47. https://doi.org/10.1038/s41893-020-00610-5
Jones, B. A., J. A. Thacher, J. M. Chermak, and R. P. Berrens. 2016. Wildfire smoke health costs: a methods case study for a Southwestern US ‘mega-fire’. Journal of Environmental Economics and Policy 5(2):181-199. https://doi.org/10.1080/21606544.2015.1070765
Kniesner, T., and W. Viscusi. 2019. The value of a statistical life: Oxford research encyclopedia of economics and finance. Oxford University Press. https://doi.org/10.1093/acrefore/9780190625979.013.138
Kupfer, J. A., A. J. Terando, P. Gao, C. Teske, and J. K. Hiers. 2020. Climate change projected to reduce prescribed burning opportunities in the south-eastern United States. International Journal of Wildland Fire 29(9):764-778. https://doi.org/10.1071/WF19198
Lake Tahoe West Restoration Partnership. 2019. Lake Tahoe West Restoration Strategy. National Forest Foundation. https://www.nationalforests.org/assets/images/LTW-Landscape-Restoration-Strategy-02Dec2019-FINAL.pdf
Levine, J. I., B. M. Collins, R. A. York, D. E. Foster, D. L. Fry, and S. L. Stephens. 2020. Forest stand and site characteristics influence fuel consumption in repeat prescribed burns. International Journal of Wildland Fire 29(2):148-159. https://doi.org/10.1071/WF19043
Li, Y., L. J. Mickley, P. Liu, and J. O. Kaplan. 2020. Trends and spatial shifts in lightning fires and smoke concentrations in response to 21st century climate over the national forests and parks of the western United States. Atmospheric Chemistry and Physics 20(14):8827-8838. https://doi.org/10.5194/acp-20-8827-2020
Lindstrom, S. 2000. A contextual overview of human land use and environmental conditions. Gen. Tech. Rep. PSW-GTR-175, US Department of Agriculture Forest Service, Pacific Southwest Research Station, Albany, CA.
Long, J. W., F. K. Lake, and R. W. Goode. 2021. The importance of Indigenous cultural burning in forested regions of the Pacific West, USA. Forest Ecology and Management 500:119597. https://doi.org/10.1016/j.foreco.2021.119597
Long, J. W., L. W. Tarnay, and M. P. North. 2018. Aligning smoke management with ecological and public health goals. Journal of Forestry 116(1):76-86. https://doi.org/10.5849/jof.16-042
Loudermilk, E. L., R. M. Scheller, P. J. Weisberg, and A. Kretchun. 2017. Bending the carbon curve: fire management for carbon resilience under climate change. Landscape Ecology 32(7):1461-1472. https://doi.org/10.1007/s10980-016-0447-x
Low, K. E., B. M. Collins, A. Bernal, J. E. Sanders, D. Pastor, P. N. Manley, A. M. White, and S. L. Stephens. 2021. Longer-term impacts of fuel reduction treatments on forest structure, surface fuels, and drought resistance in the Lake Tahoe Basin. Forest Ecology and Management 479:118609. https://doi.org/10.1016/j.foreco.2020.118609
Malamakal, T., L. W. A. Chen, X. L. Wang, M. C. Green, S. Gronstal, J. C. Chow, and J. G. Watson. 2013. Prescribed burn smoke impact in the Lake Tahoe Basin: model simulation and field verification. International Journal of Environment and Pollution 52(3-4):225-243. https://doi.org/10.1504/IJEP.2013.058457
Maxwell, C., R. M. Scheller, J. W. Long, and P. Manley. 2022. Frequency of disturbance mitigates high-severity fire in the Lake Tahoe Basin, California and Nevada. Ecology and Society 27(1):21. https://doi.org/10.5751/ES-12954-270121
Mueller, S., L. Tarnay, S. O’Neill, and S. Raffuse. 2020. Apportioning Smoke Impacts of 2018 Wildfires on Eastern Sierra Nevada Sites. Atmosphere 11(9):970. https://doi.org/10.3390/atmos11090970
Navarro, K., D. Schweizer, J. Balmes, and R. Cisneros. 2018. A Review of Community Smoke Exposure from Wildfire Compared to Prescribed Fire in the United States. Atmosphere 9(5):185. https://doi.org/10.3390/atmos9050185
Navarro, K. M., R. Cisneros, S. M. O'Neill, D. Schweizer, N. K. Larkin, and J. R. Balmes. 2016. Air quality impacts and intake fraction of PM2.5 during the 2013 Rim mega fire. Environmental Science & Technology 50(21):11965-11973. https://doi.org/10.1021/acs.est.6b02252
North, M., B. M. Collins, and S. L. Stephens. 2012. Using fire to increase the scale, benefits and future maintenance of fuels treatments. Journal of Forestry 110(7):392-401. https://doi.org/10.5849/jof.12-021
North, M. P., and M. D. Hurteau. 2011. High-severity wildfire effects on carbon stocks and emissions in fuels treated and untreated forest. Forest Ecology and Management 261(6):1115-1120. https://doi.org/10.1016/j.foreco.2010.12.039
North, M. P., R. A. York, B. M. Collins, M. D. Hurteau, G. M. Jones, E. E. Knapp, L. N. Kobziar, H. McCann, M. D. Meyer, S. L. Stephens, R. E. Tompkins, and C. L. Tubbesing. 2021. Pyrosilviculture needed for landscape resilience of dry Western U.S. forests. Journal of Forestry 119(5):520-544. https://doi.org/10.1093/jofore/fvab026
Ostro, B. D., and S. Rothschild. 1989. Air pollution and acute respiratory morbidity: an observational study of multiple pollutants. Environmental Research 50(2):238-247. https://doi.org/10.1016/S0013-9351(89)80004-0
Ottmar, R. D. 2014. Wildland fire emissions, carbon, and climate: Modeling fuel consumption. Forest Ecology and Management 317:41-50. https://doi.org/10.1016/j.foreco.2013.06.010
Pierce, D. W., J. Kalansky, and D. Cayan. 2018. Climate, drought, and sea level rise scenarios for the fourth California climate assessment. Tech. report CNRA-CEC-2018-006, California Energy Commission, Sacramento, CA.
Preisler, H. K., D. Schweizer, R. Cisneros, T. Procter, M. Ruminski, and L. Tarnay. 2015. A statistical model for determining impact of wildland fires on Particulate Matter (PM2.5) in Central California aided by satellite imagery of smoke. Environmental Pollution 205(October):340-349. https://doi.org/10.1016/j.envpol.2015.06.018
Prichard, S., N. S. Larkin, R. Ottmar, N. H. F. French, K. Baker, T. Brown, C. Clements, M. Dickinson, A. Hudak, A. Kochanski, R. Linn, Y. Liu, B. Potter, W. Mell, D. Tanzer, S. Urbanski, and A. Watts. 2019. The fire and smoke model evaluation experiment-a plan for integrated, large fire-atmosphere field campaigns. Atmosphere 10(2):66-66. https://doi.org/10.3390/atmos10020066
Sacks, J. D., J. M. Lloyd, Y. Zhu, J. Anderton, C. J. Jang, B. Hubbell, and N. Fann. 2018. The Environmental Benefits Mapping and Analysis Program-Community Edition (BenMAP-CE): A tool to estimate the health and economic benefits of reducing air pollution. Environmental Modelling & Software 104:118-129. https://doi.org/10.1016/j.envsoft.2018.02.009
Safford, H. D., D. A. Schmidt, and C. H. Carlson. 2009. Effects of fuel treatments on fire severity in an area of wildland-urban interface, Angora Fire, Lake Tahoe Basin, California. Forest Ecology and Management 258(5):773-787. https://doi.org/10.1016/j.foreco.2009.05.024
Scheller, R. M., J. B. Domingo, B. R. Sturtevant, J. S. Williams, A. Rudy, E. J. Gustafson, and D. J. Mladenoff. 2007. Design, development, and application of LANDIS-II, a spatial landscape simulation model with flexible temporal and spatial resolution. Ecological Modelling 201(3):409-419. https://doi.org/10.1016/j.ecolmodel.2006.10.009
Scheller, R. M., D. Hua, P. V. Bolstad, R. A. Birdsey, and D. J. Mladenoff. 2011. The effects of forest harvest intensity in combination with wind disturbance on carbon dynamics in Lake States Mesic Forests. Ecological Modelling 222(1):144-153. https://doi.org/10.1016/j.ecolmodel.2010.09.009
Scheller, R., A. Kretchun, T. J. Hawbaker, and P. D. Henne. 2019. A landscape model of variable social-ecological fire regimes. Ecological Modelling 401:85-93. https://doi.org/10.1016/j.ecolmodel.2019.03.022
Schwalm, C. R., S. Glendon, and P. B. Duffy. 2020. RCP8.5 tracks cumulative CO2 emissions. Proceedings of the National Academy of Sciences 117(33):19656-19657. https://doi.org/10.1073/pnas.2007117117
Schweizer, D., R. Cisneros, and K. Navarro. 2020. The effectiveness of adding fire for air quality benefits challenged: A case study of increased fine particulate matter from wilderness fire smoke with more active fire management. Forest Ecology and Management 458:117761. https://doi.org/10.1016/j.foreco.2019.117761
Schweizer, D., H. K. Preisler, and R. Cisneros. 2019. Assessing relative differences in smoke exposure from prescribed, managed, and full suppression wildland fire. Air Quality, Atmosphere & Health 12(1):87-95. https://doi.org/10.1007/s11869-018-0633-x
Schweizer, D. W., and R. Cisneros. 2017. Forest fire policy: change conventional thinking of smoke management to prioritize long-term air quality and public health. Air Quality, Atmosphere & Health 10:33-36. https://doi.org/10.1007/s11869-016-0405-4
Springsteen, B., T. Christofk, S. Eubanks, T. Mason, C. Clavin, and B. Storey. 2011. Emission reductions from woody biomass waste for energy as an alternative to open burning. Journal of the Air and Waste Management Association 61(1):63-68. https://doi.org/10.3155/1047-32126.96.36.199
Springsteen, B., T. Christofk, R. A. York, T. Mason, S. Baker, E. Lincoln, B. R. Hartsough, and T. Yoshioka. 2015. Forest biomass diversion in the Sierra Nevada: Energy, economics and emissions. California Agriculture 69(3):142-149. https://doi.org/10.3733/ca.v069n03p142
Stevens, J. T., B. M. Collins, J. W. Long, M. P. North, S. J. Prichard, L. W. Tarnay, and A. M. White. 2016. Evaluating potential trade-offs among fuel treatment strategies in mixed-conifer forests of the Sierra Nevada. Ecosphere 7(9):e01445. https://doi.org/10.1002/ecs2.1445
Striplin, R., S. A. McAfee, H. D. Safford, and M. J. Papa. 2020. Retrospective analysis of burn windows for fire and fuels management: an example from the Lake Tahoe Basin, California, USA. Fire Ecology 16(1):1-16. https://doi.org/10.1186/s42408-020-00071-3
Tarnay, L., D. Johnson, and A. W. Gertler. 2005. Modeled inputs of atmospheric nitrogen to the Lake Tahoe Basin due to gaseous pollutant deposition. Journal Nevada Water Resources Association 2:41-57.
Taylor, A. H. 2004. Identifying forest reference conditions on early cut-over lands, Lake Tahoe basin, USA. Ecological Applications 14(6):1903-1920. https://doi.org/10.1890/02-5257
United States Environmental Protection Agency (U.S. EPA). 2021. Environmental Benefits Mapping and Analysis Program - Community Edition: User's Manual. U.S. Environmental Protection Agency, Washington, DC, USA.
Urbanski, S. 2014. Wildland fire emissions, carbon, and climate: Emission factors. Forest Ecology and Management 317:51-60. https://doi.org/10.1016/j.foreco.2013.05.045
Wang, D., D. Guan, S. Zhu, M. M. Kinnon, G. Geng, Q. Zhang, H. Zheng, T. Lei, S. Shao, P. Gong, and S. J. Davis. 2021. Economic footprint of California wildfires in 2018. Nature Sustainability 4(3):252-260. https://doi.org/10.1038/s41893-020-00646-7
White, R., P. Hessburg, S. Larkin, and M. Varner. 2017. Smoke in a new era of fire. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. Science Update 24. Portland, Oregon, USA. https://www.fs.usda.gov/pnw/publications/smoke-new-era-fire
Williams, A., J. Abatzoglou, A. Gershunov, J. Guzman‐Morales, D. Bishop, J. Balch, and D. Lettenmaier. 2019. Observed impacts of anthropogenic climate change on wildfire in California. Earth's Future 7(8):892-910. https://doi.org/10.1029/2019EF001210
Williamson, G. J., D. M. J. S. Bowman, O. F. Price, S. B. Henderson, and F. H. Johnston. 2016. A transdisciplinary approach to understanding the health effects of wildfire and prescribed fire smoke regimes. Environmental Research Letters 11(12):125009. https://doi.org/10.1088/1748-9326/11/12/125009
York, R. A., J. Levine, K. Russell, and J. Restaino. 2021. Opportunities for winter prescribed burning in mixed conifer plantations of the Sierra Nevada. Fire Ecology 17(1):33. https://doi.org/10.1186/s42408-021-00120-5
Table 1. Summary of management scenarios.
|Scenario number||Summary||Annual average treatment of vegetated landscape||Forest
|WUI Zones||Non-WUI Zones||Pile burning||Understory burning|
|1||Fire suppression only||0%||None||None||None||None||Suppression of all wildfires in all zones|
|2||Wildland-urban interface (WUI) thinning||1.8%||Yes||No||Hand thinned areas||None||Suppression of all wildfires in all zones|
|3||Extensive and intensive thinning||6.7%||Yes||Yes||Hand thinned areas||None||Suppression of all wildfires in all zones|
|4||Fire-based approach||4.0%||Yes||No||Hand thinned areas||Target of 220 ha/y across all zones||Suppression of all wildfires in WUI zones, but ~40 ha/y of “managed wildfires,” representing less active suppression of naturally-ignited fires in non-WUI zones|
|5||Extensive fire-based approach||7.2%||Yes||No||Hand-thinned areas||Target of 1050 ha per year across all zones|
|†Forest thinning involved mechanical equipment with removal of biomass in areas near roads or thinning by hand-crews followed by pile burning in less accessible areas. Thinning treatments were limited from recurring no more than every 20 years in scenarios 2, 4, and 5, while they could recur every 11 years under scenario 3.|
See the Supplemental materials for additional background on the LANDIS-II modeling of the management scenarios.
Table 2. Prescribed fire parameters used in fire-focused management scenarios 4 and 5.
|Prescribed fire parameters||Scenario 4||Scenario 5|
|Maximum wind speed||6.6 (m/s)||6.6 (m/s)|
|Maximum Fire Weather Index†||55 (unitless)||55 (unitless)|
|Minimum Fire Weather Index†||10 (unitless)||10 (unitless)|
|Maximum Fire Intensity||1 (low)||1 (low)|
|Number of fires allowable per year within the Lake Tahoe Basin area||30 (fixed)||364, subject to suitable weather|
|Allowable days||October 15–November 15||Any day, subject to suitable weather|
|Target size||40 (hectares)||72 (hectares)|
|†Canadian Fire Weather Index; see Scheller et al. (2019) for more details.|
Table 3. Assumptions regarding consumption of LANDIS-II biomass fuel components by fire intensity class and associated PM2.5 emissions factors used in modeling emissions from prescribed fire and wildfire in the Lake Tahoe basin.
|Biomass consumed categorized by fire severity||Flaming phase||Smoldering phase|
|Fuel component||Unburned||Low||Moderate||High||Proportion||Emission factor (g/kg)||Proportion||Emission factor (g/kg)|
|Duff (soil organic matter)||0%||100%||100%||100%||0%||N/A||100%||50|
Table 4. Classes used to categorize daily PM2.5 emissions.
|Class||Daily emissions of PM2.5 in Mg|
|Negligible||0 < 10|
|Low||10 < 30|
|Moderate||30 < 60|
|High||60 < 200|
|Very High||200 < 500|
|Extreme||> = 500|
Table 5. Average days of prescribed burning (pile burning and understory burning) per year by management scenario.
|Management scenario||Pile burning||Understory burning||Total|