Cheatgrass (Bromus tectorum) is an invasive grass pervasive across the Intermountain Western US and linked to major increases in fire frequency. Despite widespread ecological impacts associated with cheatgrass, we lack a spatially extensive model of cheatgrass invasion in the Intermountain West. Here, we leverage satellite phenology predictors and thousands of field surveys of cheatgrass abundance to create regional models of cheatgrass distribution and percent cover. We compare cheatgrass presence to fire probability, fire seasonality and ignition source. Regional models of percent cover had low predictive power (34% of variance explained), but distribution models based on a threshold of 15% cover to differentiate high abundance from low abundance had an overall accuracy of 74%. Cheatgrass achieves ≥15% cover over 210,000 km 2 (31%) of the Intermountain West. These lands were twice as likely to burn as those with low abundance, and four times more likely to burn multiple times between 2000-2015. Fire probability increased rapidly at low cheatgrass cover (1-5%) but remained similar at higher cover, suggesting that even small amounts of cheatgrass in an ecosystem can increase fire risk. Abundant cheatgrass was also associated with a 10 day earlier fire seasonality and interacted strongly with anthropogenic ignitions. Fire in cheatgrass was particularly associated with human activity, suggesting that increased awareness of fire danger in invaded areas could reduce risk. This study suggests that cheatgrass is much more spatially extensive and abundant than previously documented and that invasion greatly increases fire frequency, even at low percent cover.

No abstract

Article de revue (Article scientifique dans une revue à comité de lecture)International audience

According to new ecological theories, many savannas are inherently in disequilibrium and can flip from tree-dominated to grass-dominated landscapes depending upon the disturbance regime. In particular, a shift in a fire regime to a more frequent and intensive one can radically alter the tree-to-grass ratio in a given savanna. Drawing upon the ecological buffering model we argue that savanna persistence requires a relatively stable fire regime. We hypothesize that anthropogenic burning practices perform this function by producing a regular annual spatiotemporal pattern of fire that is linked to vegetation type. We test this hypothesis using a study of two areas, one in Mali and the other Burkina Faso. We use two sources of satellite data to produce an 11-year time series of the spatiotemporal pattern of fires and an example of the annual burned area pattern these fires produce. We combine the analysis of satellite imagery with interviews of rural inhabitants who set fires to understand the logic underlying the patterns of fire. Analysis of a time series of imagery reveals a strikingly regular annual spatiotemporal pattern of burning for both study areas, which cannot be explained by the regional climatic pattern alone. We conclude that the regularity of the annual fire regime in West Africa is a human-ecological phenomenon closely linked to vegetation type and controlled by people's burning practices. We argue that the anthropogenic burning regime serves to buffer the savanna and maintain its ecological stability.

According to contemporary ecological theory, the mechanisms governing tree cover in savannas vary by precipitation level. In tropical areas with mesic rainfall levels, savannas are unstable systems in which disturbances, such as fire, determine the ratio of trees to grasses. Precipitation in these so-called ''disturbance-driven savannas'' is sufficient to support forest but frequent disturbances prevent transition to a closed canopy state. Building on a savanna buffering model we argue that a consistent fire regime is required to maintain savannas in mesic areas. We hypothesize that the spatiotemporal pattern of fires is highly regular and stable in these areas. Furthermore, because tree growth rates in savannas are a function of precipitation, we hypothesize that savannas with the highest rainfall levels will have the most consistent fire pattern and the most intense fires-thus the strongest buffering mechanisms. We analyzed the spatiotemporal pattern of burning over 11 years for a large subset of the West African savanna using a moderate resolution imaging spectroradiometer active fire product to document the fire regime for three savanna belts with different precipitation levels. We used LISA analysis to quantify the spatiotemporal patterns of fires, coefficient of variance to quantify differences in peak fire dates, and center or gravity pathways to characterize the spatiotemporal patterns of the fires for each area. Our analysis confirms that spatiotemporal regularity of the fire regime is greater for mesic areas that for areas where precipitation is lower and that areas with more precipitation have more regular fire regimes.

A fundamental principal of savanna fire ecology is that the fire regime determines vegetation cover, especially as it pertains to trees. A corollary is that late fires are more damaging to trees than early fires. Much evidence in support of this principle has been derived from a series of long-term burning experiments based on the pioneering work of André Aubréville. Eighty years ago, Aubréville devised an experiment to study the impacts of fire on savanna trees in Africa. The design conventions of this study remain highly influential. It is now clear, however, that the dates chosen by Aubréville and his followers do not reflect the burning practices of West African people. Dates that were chosen for “early” and “late” are not representative of actual fire timing; they represent extremes. This study has four goals: (i) to critically review the results of the burning experiments; (ii) to examine them in the context of results from recent savanna fire studies; (iii) to evaluate their limitations based on data for actual burning practices and fires from West Africa; and (iv) to critically evaluate the use of the early/late terminology in contemporary fire research. We find the majority of West African fires occur during the “middle” of the fire season. Our field studies find that fire temperature and burn completeness are highest in the middle-season. We conclude that the early/late fire dichotomy is not sufficient for understanding the impacts of anthropogenic fires in the region and we make suggestions for rethinking its use more broadly.

No abstract

Managed wildfires, i.e., naturally ignited wildfires that are managed for resource benefits, have the potential to reduce fuel loads, minimize the effects of future wildfires, and restore critical natural processes across many forest landscapes. In the United States, the 2009 federal wildland fire policy guidance was designed to provide greater flexibility in the use of managed wildfires, but the effects of this policy on wildfires in the western US are not yet fully understood. Our goal was to compare managed and full suppression wildfires and to also analyze the differences between managed wildfires across space (Arizona/New Mexico and California) and time (before and after 2009) using four metrics for each wildfire: (1) distance to wilderness, (2) distance to the wildland–urban interface (WUI), (3) the percentage of area burned with high severity, and (4) the number of land management agencies. Across the study area, we found that managed wildfires were significantly closer to wilderness areas, were farther from the WUI, had a lower percentage of area that was burned at high severity, and had fewer agencies involved in managing the fire compared to full suppression wildfires. In California, managed wildfires occurred closer to wilderness and had a larger percentage of high-severity burn area compared to those in the southwest US (Arizona and New Mexico). Within each region, however, there were no significant geographic differences between managed wildfires before and after the implementation of the 2009 policy guidance. Despite the greater flexibility of the 2009 policy guidance, the basic geographic properties of managed wildfires in these two regions have not changed. As the climate warms and droughts intensify, the use of managed wildfires will need to expand during favorable weather conditions in order to address the threat of large and uncharacteristic wildfires to people and ecosystems.

No abstract