Human-caused ignitions and plant invasion drive fire-regime change in a dryland biodiversity hotspot

K. C. Rodman; A. L. Mahood; A. J. Antoninka; M. A. Bowker; H. I. Rowe

doi:10.64898/2026.11.03.764038

AI summary · DOI: 10.64898/2026.11.03.764038 · bioRxiv: Ecology

Human-caused ignitions and plant invasion drive fire-regime change in a dryland biodiversity hotspot

K. C. Rodman, A. L. Mahood, A. J. Antoninka, M. A. Bowker, H. I. Rowe

Original paper Published on 05 Nov. 2026 bioRxiv preprint License: CC BY-NC-ND 4.0 Also in: fr

How to read this summary? Illustrated guide

The essentials

Across the U.S. Sonoran Desert, human activity largely determines where fires ignite, while herbaceous fuels and bare-ground gaps influence whether fires spread.

Question

Which human, environmental, and vegetation factors are associated with fire ignition, fire size, and fire spread in the Sonoran Desert? The study also asks how these relationships may contribute to fire-regime change in this historically infrequently burned ecosystem.

Method

The authors analyzed 73,207 fire records from 1992–2020 to model human- and lightning-caused ignition likelihood and fire size using Random Forests, with 50,000 pseudo-absence points for the ignition models. For fire spread, they assessed vegetation cover at the edges and interiors of 346 fire perimeters from 1987–2024, using the Fire Perimeter Effect and testing herbaceous-cover thresholds from 5% to 30%.

Findings

Result Human activities accounted for 95.5% of known-cause fires in the Sonoran Desert, and human ignitions were associated with proximity to buildings, roads, and transmission lines.
Result Lightning-caused ignitions were uncommon in the Sonoran Desert and were more likely at elevations above 1,500 m, mainly during summer.
Result Fires tended to be larger when ignitions occurred at lower elevations, in areas with greater perennial herbaceous cover, under high Ignition Component conditions, and farther from infrastructure.
Result Annual and perennial herbaceous cover were associated with greater fire spread, while bare ground was associated with inhibited spread; the authors found little evidence of a herbaceous-cover threshold.
Authors’ interpretation The authors interpret the results as evidence that human ignitions and herbaceous fuels jointly support a novel fire regime in the Sonoran Desert.

Limitations

Fire occurrence records have location uncertainty, and vegetation-cover estimates have inherent error that limits precise inference about cover–spread relationships and possible flammability thresholds.
The study excluded the southern portion of the Sonoran Desert in Mexico because key spatial datasets were unavailable there.
Fire-size model performance was modest, limiting the precision of its predictions.

Why it matters

The authors argue that different drivers govern distinct parts of the fire regime: human activity shapes where fires start, herbaceous fuels shape whether they spread, and climate variability affects fuels and fire behavior. They present this framework as useful for anticipating dryland fire-regime transformation and targeting prevention, suppression, and invasive-plant management.

Written by AI from the paper. Passages are checked word for word against the text we analysed; the rest may contain errors.

AI-generated keywords: Sonoran Desert; fire regimes; human-caused ignitions; herbaceous fuels; fire spread

Read further

Comprehensive Summary
Layman's Summary
Blog article
Original abstract

The paper examines how human activity, vegetation, and environmental conditions shape fire-regime change in the Sonoran Desert, a biodiverse region where fire was historically infrequent. It asks three related questions: where human- and lightning-caused fires occur, which factors are associated with larger fires, and how pre-fire vegetation cover relates to fire spread. The authors distinguish between ignition and spread: human activity may influence where fires start, while fuel availability may affect whether they expand. For the ignition and fire-size analyses, the researchers used U.S. wildfire records from 1992 to 2020. They retained 73,207 fires with known causes across the Sonoran Desert and adjacent mountain ecoregions. Using random forest models, they related fire records to vegetation, terrain, fire-weather, and infrastructure variables. The ignition analyses included spatial cross-validation and pseudo-absence locations. To study fire spread, they analyzed 346 fire perimeters from 1987 to 2024, comparing pre-fire vegetation cover near each perimeter with cover inside the fire perimeter. This measure was used to assess which cover types were associated with increased spread or its limitation. Human activity accounted for 95.5% of recorded fires with known causes in the Sonoran Desert, while lightning accounted for 4.5%. The pattern differed in the adjacent mountain ecoregions, where 54% of fires were lightning-caused. Lightning ignitions were concentrated in uplands and peaked during the summer monsoon; human ignitions were most common in late spring. Reported human-related causes included escaped campfires, arson, debris burning, and vehicle use. The human-ignition model associated higher relative ignition likelihood with proximity to buildings, roads, and transmission lines, as well as with spring and early-summer dates. Its out-of-bag AUC was 0.85. The authors interpreted transmission lines mainly as a proxy for human activity because only a small fraction of fires were attributed directly to them. Lightning ignition likelihood was most strongly associated with elevation and season: it rose substantially above about 1,500 meters and peaked from June through September. The lightning model’s AUC was 0.91. Fire size showed a different pattern. Fires tended to be larger when they started below about 1,500 meters, in areas with more than 25% perennial herbaceous cover, and under conditions associated with high Ignition Component values. Fires that started farther from roads, transmission lines, and buildings also tended to be larger, a pattern consistent with more limited access for suppression. The fire-size model was less predictive than the ignition models, with an out-of-bag correlation of 0.34. In the perimeter analysis, annual and perennial herbaceous cover were most strongly associated with increased spread, and the relationship strengthened as cover increased. The authors found no clear threshold response; instead, the pattern was broadly linear. Bare ground was associated with reduced spread, while shrub cover showed no significant relationship. Litter and tree cover had statistically detectable but modest associations. The authors conclude that human-caused ignitions and herbaceous fuels jointly contribute to a changing fire regime: human activity shapes where fires begin, while herbaceous cover helps determine whether they spread. They suggest reducing human-caused ignitions, managing invasive herbaceous fuels, preserving bare-ground interspaces, and suppressing fires quickly. The authors caution that ignition locations may be uncertain by up to about 1.6 kilometers, vegetation-cover estimates have reported errors of roughly 6–14%, and fire records may be more complete near infrastructure. Although sensitivity analyses suggested limited effects from location uncertainty and reporting bias, these limitations constrain the precision of the findings. The study also covers only the U.S. portion of the Sonoran Desert, excluding much of the region in Mexico.

How People and Plants Are Changing Fire in the Sonoran Desert

Fire has long been uncommon across much of the Sonoran Desert. But human activity and changes in vegetation may be altering where fires begin and how far they spread. In “Human-caused ignitions and plant invasion drive fire-regime change in a dryland biodiversity hotspot,” K. C. Rodman, A. L. Mahood, A. J. Antoninka, M. A. Bowker, and H. I. Rowe examine the forces behind this changing fire regime.

The study separates two parts of the fire process: ignition and spread. Human activity can influence where a fire starts, while available vegetation can determine whether it grows. To investigate these patterns, the researchers analyzed wildfire records from 1992 to 2020 and fire perimeters from 1987 to 2024 across the U.S. Sonoran Desert and nearby mountain ecoregions.

Human activity dominates desert fire starts

Among 73,207 fires with known causes, 95.5% of recorded fires in the Sonoran Desert were human-caused, compared with 4.5% sparked by lightning. The neighboring mountain ecoregions showed a strikingly different balance: lightning caused 54% of fires there.

The timing and location of ignitions also varied. Human-caused fires were most common in late spring, while lightning ignitions were concentrated in uplands and peaked during the summer monsoon, from June through September. Reported human causes included escaped campfires, arson, debris burning, and vehicle use.

In a model of human ignitions, fires were more likely to occur near buildings, roads, and transmission lines, as well as during spring and early summer. The model’s out-of-bag AUC—a measure of how well it distinguished ignition locations from comparison locations—was 0.85. The authors interpret transmission lines mainly as a marker of human presence: only a small fraction of fires were directly attributed to them.

Lightning followed a different pattern. Elevation and season were its strongest predictors, with ignition likelihood rising substantially above about 1,500 meters and peaking during the summer monsoon. The lightning model had an AUC of 0.91.

Starting a fire is not the same as spreading one

The factors associated with ignition did not necessarily predict fire size. Fires tended to be larger when they started below about 1,500 meters, where perennial herbaceous vegetation covered more than a quarter of the area, and under conditions associated with high Ignition Component values. This index reflects weather and fuel conditions linked to the potential for fire to start.

Fires also tended to grow larger when they began farther from roads, transmission lines, and buildings. One possible explanation is that fires in more remote locations may be harder for crews to reach and suppress. Still, the fire-size model was less predictive than the ignition models, with an out-of-bag correlation of 0.34. These relationships therefore offer clues, not a precise way to forecast the size of any particular fire.

Herbaceous cover helps fire move across the landscape

To study spread, the researchers compared pre-fire vegetation cover near 346 fire perimeters with cover inside those perimeters. Annual and perennial herbaceous cover had the clearest association with increased spread. The relationship strengthened as herbaceous cover increased and was broadly linear, rather than showing a distinct cover threshold beyond which spread suddenly accelerated.

Bare ground was associated with reduced spread, suggesting that open spaces between plants can help interrupt the continuity of fuels. Shrub cover showed no significant relationship with spread. Litter and tree cover had statistically detectable but modest associations.

Together, the results point to a two-part explanation for changing fire patterns: human activity helps determine where fires start, while herbaceous fuels help determine whether they spread. In a desert where fire was historically infrequent, that combination may have important consequences for ecosystems adapted to relatively limited fire.

Implications—and important limits

The authors suggest several practical responses: reduce human-caused ignitions, manage invasive herbaceous fuels, preserve bare-ground interspaces, and suppress fires quickly. The findings do not imply that all herbaceous plants have the same effects or that vegetation alone determines outcomes. Weather, terrain, access, and other conditions also matter.

The study has limits. Reported ignition locations may be uncertain by up to about 1.6 kilometers, vegetation-cover estimates have errors of roughly 6–14%, and fire records may be more complete near infrastructure. Sensitivity analyses suggested that location uncertainty and reporting bias had limited effects on the overall patterns, but they still constrain how precisely those patterns can be mapped. The research also covers only the U.S. portion of the Sonoran Desert, leaving much of the region in Mexico outside its scope.

Even with these caveats, the central message is clear: in this dryland biodiversity hotspot, people and plant fuels play distinct but connected roles. Reducing unnecessary ignitions and understanding where herbaceous cover can carry fire may help communities and land managers respond to a fire regime that is changing.

Dryland ecosystems cover much of Earths terrestrial surface and are increasingly affected by climate warming, land-use change, and biological invasion. The Sonoran Desert of western North America, a biodiversity hotspot that historically experienced infrequent fire, exemplifies how these global-change processes are driving ecosystem transformation through the emergence of novel disturbance regimes. Using records of more than 73,000 wildland fire events, we quantified anthropogenic and environmental controls on fire ignition, size, and spread across the Sonoran Desert. Human-caused ignitions accounted for 95.5% of all recorded fires since the early 1990s and were concentrated near roads, buildings, and other areas of human activity, particularly during late spring. In contrast, lightning-caused ignitions were rare within the Sonoran Desert and occurred primarily in adjacent montane forests during late summer months. Consistent with broader trends across the western United States, fires were typically largest when they ignited far from roads, where access is limited, and fire management often relies on indirect attack strategies. Pre-fire annual and perennial herbaceous vegetation both promoted fire spread, with detectable effects at cover values as low as 5%, while bare ground reduced fuel connectivity and constrained fire growth. Rather than exhibiting a distinct threshold response, fire spread increased linearly with herbaceous cover, suggesting that even modest increases in fine fuels can alter fire behavior in this historically fuel-limited ecosystem. Together, our results illustrate that human ignitions determine where fires occur, whereas herbaceous fuels determine whether those ignitions spread, driving fire-regime transformation across an iconic North American desert. Managing and adapting to this novel fire regime will require coordinated efforts to reduce human-caused ignitions, limit invasive herbaceous fuels, maintain bare-ground interspaces, and rapidly suppress fires. More broadly, our findings identify the mechanisms controlling fire regime change in dryland ecosystems and provide a framework for anticipating ecosystem transformation under ongoing global change.

Created on 09 Nov. 2026

Was this synthesis useful?

Score: 0

Ask questions about this paper to our AI assistant

How to chat with this paper? Illustrated guide

You can also chat with multiple papers at once here.

AI assistant instructions?

Tools for this paper

Explain it differently 1 free rewrite left today

Save to a collection Group papers in collections to compare them, chat with them and export them. Continue with Google Create a free account Log in

Export BibTeX · RIS (Zotero, Mendeley, EndNote) · Obsidian (Markdown) · Notion (Markdown)

Advanced tools Premium

For researchers who follow a field closely: citation alerts, journal-club slides, version comparison and Zotero export. Discover Premium

Zotero

Citation alerts Citation alerts guide

Who works on it? The people and institutions who publish on its topic

Weekly letter Presented in "Climate and Earth this week" of 9 November 2026. Climate and Earth this week: Hidden shortages, spreading fires and signals beneath the surface

Journal club Premium+ Slides to present it Propose for the journal club

Talk about it in public Premium+ Press kit: release, questions, hype check Press kit guide

Papers around this one

See them on a map

Ranked by how close their title and abstract are to this paper’s. Select some to read them together.

Closest papers

Related Broader context
A mismatch in wildfire impact reduction and ecological restoration in western U.S. fuel treatments
T. Metz, J. N. Hendershot, C. Song · 2026 · bioRxiv preprint · AI summary available

Fuel treatments in western forests affect tree survival and seedling composition, whereas this study examines ignition and fire spread in the Sonoran Desert.

In short
Across western U.S. forests, fuel treatments improve tree survival but disproportionately benefit fire-intolerant trees and do not shift seedling composition toward fire-tolerant species.
Related Broader context
Typicality of the 2021 Western North America Summer Heatwave
Valerio Lucarini, Vera Melinda Galfi, Gabriele Messori et al. · 2022 · Atmospheric and Oceanic Physics · cited 39 times · Not summarized yet

This study evaluates the statistical typicality of a western North American heatwave, while this paper models wildfire ignition, size, and spread in the Sonoran Desert.

Abstract
Elucidating the statistical properties of extreme meteo-climatic events and capturing the physical processes responsible for their occurrence are key steps for improving our understanding of climate variability and climate change and for better evaluating the associated hazards. It has recently become apparent that large deviation theory is very useful for investigating persistent extreme events, and specifically, for flexibly estimating long return periods and for introducing a notion of dynamical typicality. Using a methodological framework based on large deviation theory and taking advantage of long simulations by a state-of-the-art Earth System Model, we investigate the 2021 North Americ…
Related Another approach to the problem
A Multi-Modal Wildfire Prediction and Personalized Early-Warning System Based on a Novel Machine Learning Framework
Rohan Tan Bhowmik · 2022 · Machine Learning · preprint, not peer reviewed · cited 2 times · AI summary available

This system uses machine learning and sensor data for wildfire prediction and alerts, while this paper models fire ignition, size, and spread from fire records and vegetation data.

Abstract
Wildfires are increasingly impacting the environment, human health and safety. Among the top 20 California wildfires, those in 2020-2021 burned more acres than the last century combined. California's 2018 wildfire season caused damages of $148.5 billion. Among millions of impacted people, those living with disabilities (around 15% of the world population) are disproportionately impacted due to inadequate means of alerts. In this project, a multi-modal wildfire prediction and personalized early warning system has been developed based on an advanced machine learning architecture. Sensor data from the Environmental Protection Agency and historical wildfire data from 2012 to 2018 have been compi…
Related Broader context
Reconfiguration of Amazon's Connectivity in the Climate System
Adam Giammarese, Jacob Brown, Nishant Malik · 2023 · Atmospheric and Oceanic Physics · cited 4 times · Not summarized yet

This study tracks Amazon climate-system connectivity amid deforestation and fires, while this paper examines human ignitions and vegetation-driven fire-regime change in the Sonoran Desert.

Abstract
With the recent increase in deforestation, forest fires, and regional temperatures, the concerns around the rapid and complete collapse of the Amazon rainforest ecosystem have heightened. The thresholds of deforestation and the temperature increase required for such a catastrophic event are still uncertain. However, our analysis presented here shows that signatures of changing Amazon are already apparent in historical climate data sets. Here, we extend the methods of climate network analysis and apply them to study the temporal evolution of the connectivity between the Amazon rainforest and the global climate system. We observe that the Amazon rainforest is losing short-range connectivity an…
Related Broader context
Variations in climate habitability parameters and their effect on Earth's biosphere during the Phanerozoic Eon
Iva Vilović, Dirk Schulze-Makuch, René Heller · 2023 · Sci Rep 13, 12663 (2023) · cited 2 times · AI summary available

This study correlates Phanerozoic environmental conditions with biosphere properties, while this paper relates vegetation and human activity to wildfire patterns in a desert ecosystem.

Abstract
Essential insights on the characterization and quality of a detectable biosphere are gained by analyzing the effects of its environmental parameters. We compiled environmental and biological properties of the Phanerozoic Eon from various published data sets and conducted a correlation analysis to assess variations in parameters relevant to the habitability of Earth's biosphere. We showed that environmental parameters such as oxygen, global average surface temperatures, runoff rates and carbon dioxide are interrelated and play a key role in the changes of biomass and biodiversity. We showed that there were several periods with a highly thriving biosphere, with one even surpassing present day …
Related Broader context
Changing spatial distribution of water flow charts major change in Mars' greenhouse effect
Edwin S. Kite, Michael A. Mischna, Bowen Fan et al. · 2022 · Science Advances 8(21), (2022) · cited 26 times · Not summarized yet

This study links Mars river distributions to past climate cooling, while this paper examines wildfire patterns and vegetation in the Sonoran Desert.

Abstract
Early Mars had rivers, but the cause of Mars' wet-to-dry transition remains unknown. Past climate on Mars can be probed using the spatial distribution of climate-sensitive landforms. We analyzed global databases of water-worked landforms and identified changes in the spatial distribution of rivers over time. These changes are simply explained by comparison to a simplified meltwater model driven by an ensemble of global climate model simulations, as the result of $\gtrsim$10 K global cooling, from global average surface temperature (T) $\ge$ 268 K to T $\sim$ 258 K, due to a weaker greenhouse effect. In other words, river-forming climates on Early Mars were warm and wet first, and cold and we…
Related Broader context
3D Global Climate Simulations of Synthetic Mars Warming
Tom N. Landefeld, Harald Lesch · 2026 · Earth and Planetary Astrophysics · AI summary available

This study simulates aerosol-driven warming on Mars, while this paper investigates human ignitions and herbaceous fuels as drivers of desert fire spread.

In short
In a one-year Mars climate simulation, a single substitution of natural dust with aluminium nanoparticles failed to sustain warming and instead cooled and dried the atmosphere.
Related Another approach to the problem
Wildfire Autonomous Response and Prediction Using Cellular Automata (WARP-CA)
Abdelrahman Ramadan · 2024 · Artificial Intelligence · preprint, not peer reviewed · cited 1 time · Not summarized yet

WARP-CA simulates wildfire spread with cellular automata and autonomous agents, whereas this paper analyzes observed fire records and vegetation cover to explain spread.

Abstract
Wildfires pose a severe challenge to ecosystems and human settlements, exacerbated by climate change and environmental factors. Traditional wildfire modeling, while useful, often fails to adapt to the rapid dynamics of such events. This report introduces the (Wildfire Autonomous Response and Prediction Using Cellular Automata) WARP-CA model, a novel approach that integrates terrain generation using Perlin noise with the dynamism of Cellular Automata (CA) to simulate wildfire spread. We explore the potential of Multi-Agent Reinforcement Learning (MARL) to manage wildfires by simulating autonomous agents, such as UAVs and UGVs, within a collaborative framework. Our methodology combines world s…

Go further

How are these papers found?

We compare the title and abstract of this paper with those of every paper summarized on this site, and with the papers Semantic Scholar recommends, those it cites and those citing it (for a much-cited paper, its most recent citations and, from OpenAlex, the most cited ones). Among the prior work and the follow-ups, the papers that use each other in depth, a method it uses and the much-cited follow-ups come a little higher. The levels (very close, close, related) come from the similarity of the texts, not from a judgment of quality; * marks a paper without an abstract, compared by its title. The sentence under a paper is written by an AI from the two abstracts: check it in the papers. A paper marked not relevant by three readers leaves the list.

Updated on 9 November 2026.

Disclaimer: The AI-based summarization tool and virtual assistant provided on this website may not always provide accurate and complete summaries or responses. We encourage you to carefully review and evaluate the generated content to ensure its quality and relevance to your needs.

Human-caused ignitions and plant invasion drive fire-regime change in a dryland biodiversity hotspot

The essentials

Question

Method

Findings

Limitations

Why it matters

Read further

Definitions

How People and Plants Are Changing Fire in the Sonoran Desert

Human activity dominates desert fire starts

Starting a fire is not the same as spreading one

Herbaceous cover helps fire move across the landscape

Implications—and important limits

Ask questions about this paper to our AI assistant

Tools for this paper

Advanced tools Premium

Papers around this one

Closest papers

Go further

Cite this paper (BibTeX)