Thomas Janssen

Before human settlement, lightning was the predominant ignition source shaping the expansion of fire-adapted savannas and grasslands such as the Cerrado. However, humans have since altered ignition frequency, timing and intensity of fires, further influencing these ecosystems. Even though humans are considered the main cause of fires in the tropics, no study has quantitatively disentangled the relative contributions of lightning and human ignitions across the entire Cerrado. Here, we present a region-wide assessment of lightning and human-caused fire regimes, linking individual fire events and lightning occurrence through an attribution algorithm. We show that humans account for nearly 90% of ignitions, effectively redefining the natural fire regime of the Cerrado. Lightning ignitions peak during the dry-to-wet transition, whereas humans ignite fires throughout the dry season extending its duration threefold. Further, humans disproportionately drive fires with extreme behaviour, which have outsized ecological impacts, by igniting fires about ten times larger or two times more intense than lightning. These findings highlight the dominant role of humans in shaping Cerrado fire regimes, with profound implications for biodiversity conservation, ecosystem resilience and carbon cycling in tropical savannas, stressing the need for fire management policies that differentiate between traditional, prescribed and agricultural ignitions.

Lightning plays a critical role in the Earth system by shaping biogeochemical cycles, while also posing significant natural hazards and serving as a key geophysical indicator for storm monitoring and wildfire early warning. However, existing publicly available global lightning datasets are often limited in either spatial or temporal resolution and do not distinguish between intra-cloud (IC) and cloud-to-ground (CG) lightning, restricting their applicability for many scientific studies. Here, we present a newly developed global gridded lightning dataset, the Flash Location Aggregation from Strokes into a High-resolution Multi-scale Analysis Product (FLASHMAP). FLASHMAP is derived from lightning observations provided by Vaisala’s Global Lightning Detection Network (GLD360) and currently covers the period from 2019 to 2024. A gridding framework is applied to convert point-based lightning stroke detections into multi-scale products at 0.1° hourly, 0.25° daily, and 0.5° monthly resolutions. FLASHMAP provides comprehensive lightning characteristics, including counts of IC and CG strokes and flashes, stroke location uncertainty and peak current, and flash multiplicity. FLASHMAP can report more total lightning strokes than existing global lightning products in most of the land regions. Comparisons with regional lightning detection networks in Alaska (USA), Spain, and New South Wales and the Australian Capital Territory (Australia) indicate that FLASHMAP reports comparable CG stroke counts while detecting fewer IC strokes. FLASHMAP is expected to advance interdisciplinary research on global and regional lightning climatology, lightning-ignited wildfires, thunderstorm identification, and ecosystem impacts.

Major advances have been made in understanding carbon emissions from boreal forest fires, yet substantial knowledge gaps remain for tundra fires. Tundra ecosystems are increasingly susceptible to fire activity due to climate change, posing a threat to their large organic soil carbon stocks and the permafrost they insulate. Additionally, studies assessing remote sensing fire severity metrics in tundra landscapes are limited. Here, we report carbon (C) combustion estimates from two large tussock tundra fires (1048 km²) that occurred in Southwest Alaska, USA, in 2022. We quantified above-and belowground carbon stocks and combustion (kg C m⁻²) at 45 field plots (36 burned, 9 unburned) one year post-fire. Soil burn depth was determined using paired moss patches, and aboveground carbon stocks were calculated using allometric equations for shrubs and tussocks. Fire severity was estimated using the field-based geometrically structured Composite Burn Index (GeoCBI) at these 45 plots and at 41 additional plots where only GeoCBI was measured. We upscaled our field-based findings using the differenced normalized burn ratio (dNBR) derived from Sentinel-2 imagery. Tussock tundra landscapes emitted an average of (± standard deviation) 1.59 ± 0.55 kg C m⁻², consuming 11% of the total pre-fire carbon stock. The majority of the loss (75%) came from belowground stocks, with an average burn depth of 6.9 ± 2.1 cm. By scaling carbon combustion with the dNBR ( R² = 0.42, p < 0.001), we estimated total carbon emissions from tussock tundra across both fires at 0.81 ± 0.22 Tg. GeoCBI estimates were moderately correlated with the dNBR ( R² = 0.63, p < 0.001) across vegetation types. This work provides important information on the impacts of fires on the carbon balance of tundra ecosystems and demonstrates that dNBR, a well-established remote sensing proxy for forest fires, is also effective for mapping fire severity in tundra landscapes.

The Brazilian Amazon contains approximately 40% of the world's tropical rainforest and plays a critical role in preserving biodiversity and regulating water, energy and carbon cycles. However, deforestation and increasingly frequent droughts, heatwaves and wildfires threaten these rainforests. Amazonian fires are generally assumed to be entirely anthropogenic, which has led to lightning-ignited fires being underexplored. Here, we present the first detailed assessment of the spatiotemporal patterns of lightning-ignited fires in the Amazon rainforest to elucidate the role of lightning and human ignitions in shaping Amazon fire dynamics. To do this, we matched cloud-to-ground lightning strokes from the Global Lightning Dataset (GLD360) with individual fire events between 2019 and 2024 to obtain a probability of lightning ignition for each fire. We also calculated a human-ignition probability index using proximity to roads, waterways, and human land cover as proxies for human activity. By combining both probabilistic indices with ground-observed lightning ignitions from eight protected areas, we could optimize the threshold that determines if an ignition is more likely to be caused by lightning or human activities. We estimate that in the Brazilian Amazon, lightning caused on average 0.2–0.4% of all fires each year (234–407 ignitions per year) and 1.1–1.2% of the annually burned area (1,226–1,358 km2 per year) between 2019 and 2024. More than 89% of these fires occurred in the late dry season between August and November, peaking in September and October. Despite lightning-ignited fires contributing a small proportion of all Amazonian fires, they constitute over 25% of the fires in identified grid clusters in parts of the states of Pará (particularly in the Breves region), Amazonas, and Rondônia. This study provides the first estimation of the role of natural ignitions in Amazon fire dynamics and a scientific basis for understanding their contribution within the region.

Lightning is a fundamental Earth system process that influences the world's major forest biomes and their carbon storage through two primary pathways. Lightning is the major cause of boreal forest fires, while lightning strikes kill patches of trees in tropical forests. We summarized the current understanding of these processes and identified knowledge gaps.

Boreal forest regions, including East Siberia, have experienced elevated fire activity in recent years, leading to record‐breaking greenhouse gas emissions and severe air pollution. However, our understanding of the factors that eventually halt fire spread and thus limit fire growth remains incomplete, hindering our ability to model their dynamics and predict their impacts. We investigated the locations and timing of 2.2 million fire stops—defined as 300 m unburned pixels along fire perimeters—across the vast East Siberian taiga. Fire stops were retrieved from remote sensing data covering over 27,000 individual fires that collectively burned 80 Mha between 2012 and 2022. Several geospatial datasets, including hourly fire weather data and landscape variables, were used to identify the factors contributing to individual fire stops. Our analysis attributed 87% of all fire stops to a statistically significant (p < 0.01) change in one or more of these drivers, with fire‐weather drivers limiting fire growth over time and landscape drivers constraining it across space. We found clear regional and temporal variations in the importance of these drivers. For instance, landscape drivers—such as less flammable land cover and the presence of roads—were key constraints on fire growth in southeastern Siberia, where the landscape is more populated and fragmented. In contrast, fire weather was the primary constraint on fire growth in the remote northern taiga. Additionally, in central Yakutia, a major fire hotspot in recent years, fuel limitations from previous fires increasingly restricted fire spread. The methodology we present is adaptable to other biomes and can be applied globally, providing a framework for future attribution studies on global fire growth limitations. In northeast Siberia, we found that with increasing droughts and heatwaves, remote northern fires could potentially grow even larger in the future, with major implications for the global carbon cycle and climate.

Lightning is a key atmospheric phenomenon that modulates atmospheric chemistry and impacts terrestrial carbon dynamics through the ignition of wildfires and direct tree mortality. Despite its importance, there is a data gap in publicly available global lightning datasets with high spatial and temporal resolution for scientific use. In this study, we present our progress towards creating a global gridded lightning dataset derived from Vaisala’s Global Lightning Detection Network (GLD360), covering the period from 2019 to 2024, with potential annual updates thereafter. This dataset is produced through a systematic gridding procedure that converts raw GLD360 lightning event data into 0.1º hourly, 0.25º daily, and 0.5º monthly gridded values. It includes key variables such as positive and negative cloud-to-ground and intra-cloud stroke count/density, stroke peak current, stroke location uncertainty, and flash count/density, making it valuable for a wide range of scientific applications. We are evaluating the gridded dataset using local lightning detection networks in Alaska (USA), Spain, South Africa, and the New South Wales and Australian Capital Territory (Australia). Meanwhile, we are comparing stroke density with the Global Lightning Climatology (WGLC) dataset derived from the World Wide Lightning Location Network (WWLLN) and flash density with the Lightning Imaging Sensor/Optical Transient Detector (LIS/OTD). The dataset could be particularly useful for advancing studies on lightning climatology, the role of lightning in wildfire ignition, thunderstorm identification, and other related topics. Its high spatial and temporal resolution also supports regional studies of lightning-related hazards and ecosystem impacts. Our goal is to make this dataset publicly available to the scientific community to facilitate new insights into the role of lightning in the Earth system.