This session explores climate-ecosystem interactions across space and time, integrating research on vegetation and fauna (including humans), and the broader aspects of ecosystem feedbacks and societal adaptations. We welcome interdisciplinary contributions integrating climate science, paleoecology, evolutionary biology, archaeology, history, genetics, geography, conservation science, and proxy-based and modeling approaches. Topics of interest include,
- Species extinctions, adaptations and the ecological impacts
- Vegetation and biome dynamics
- Biodiversity changes
- Fire and disturbance regimes
- Habitat degradation
- Hominin dispersal and habitat dynamics
- Climate-human-landscape interactions
- Agricultural adaptations
- Environmental influences on trade and exchange systems, including Silk Road contexts.
By bridging past, present, and future perspective, this session aims to foster cross-disciplinary dialogue and collaboration on the intertwined histories and futures of climate, life, and society.
The session is organised in connection with the ERC Synergy Grant EUROpest (Grant Agreement No. 101166700).
Orals: Wed, 6 May, 14:00–17:55 | Room 0.14
During glacial cycles of the late Quaternary, terrestrial vegetation changed globally in response to orbitally controlled insolation changes, variable levels of carbon dioxide, and availability of ice-free land. As they developed, the emerging vegetation patterns also in turn influenced climatic and carbon cycle trends via biogeophysical and biogeochemical feedbacks. Although such trends are clearly seen in proxy data covering glacial cycles, the vegetation patterns still remain poorly constrained due to short, scarce and discontinuous plant fossil and pollen records. For understanding the varying vegetation patterns, and for assessing terrestrial sources and causes of rapid atmospheric greenhouse gas changes, we have simulated the entire last glacial cycle, covering over 130,000 years, using an Earth system model with dynamic vegetation, carbon pools and methane emissions. In line with proxy records, our simulation shows an Eemian interglacial globally warmer than the preindustrial era, with slightly more boreal forest cover and a greener Sahara, while the simulated much colder Last Glacial Maximum, has larger subtropical deserts, more boreal tundra and fragmented tropical forests. We separate regions where vegetation change is mainly bound to forcing from ice-sheet extent (and sea level) or carbon dioxide fertilization, or the result of a feedback response to climate change. Regions where the feedbacks with climate are strong, like in northern Africa where there is hydroclimate-driven vegetation growth, and eastern Siberia where there is thermally-driven taiga-tundra turnover, the vegetation responses are highly dynamic, including a clear precessional signal that propagates to land carbon allocation and greenhouse gas emissions. Such regions have the largest potential to contribute to rapid changes in atmospheric greenhouse gases, besides any fast changes that depend directly on cryosphere or sea-level dynamics. In contrast, large parts of the tropics have vegetation with a muted response to climate change, and rapid coverage changes within this region may only occur when there are sudden changes in carbon dioxide fertilization.
How to cite: Duque-Villegas, M., Kleinen, T., Brovkin, V., and Claussen, M.: Rapid vegetation changes of the late Quaternary, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15702, https://doi.org/10.5194/egusphere-egu26-15702, 2026.
Plant responses to glacial–interglacial climate change are frequently delayed by migration lags and shaped by landscape connectivity and changing biotic interactions. Yet most spatio‑temporal species distribution models (SDMs) still assume near‑equilibrium with climate, treat dispersal only implicitly, and rarely confront their hindcasts with independent, process‑relevant validation data. This limits confidence in both late‑Quaternary reconstructions and future projections, especially in regions with complex topography and strong post‑glacial ecological reorganization.
Here we present a model–proxy framework that links occurrence‑based niche modelling with dynamic, taxa‑specific dispersal and connectivity and evaluates predicted trajectories using sedimentary ancient DNA (sedaDNA). We initially parameterize SDMs for Arctic and Tibetan Plateau taxa using modern occurrences and climate (first implementations with MaxEnt), hindcast climatic suitability through late‑Quaternary paleoclimate reconstructions, and translate suitability into time‑varying accessibility using spatially explicit dispersal models and landscape‑configured networks. This enables hypothesis testing on how connectivity, terrain, and interactions modulate community change beyond shared climatic forcing.
Broader high‑latitude analyses further indicate recurrent glacial legacy effects on interglacial assemblages, identify persistent hotspots and migration corridors. It also show that future Arctic vegetation may occupy only a small fraction of emerging climate niches due to limited dispersal, leading to extirpation from declining suitability often exceeding new colonizations in driving compositional change. We also evaluate how community assembly shifts from predominantly facilitative interactions during glacial conditions to more negative interactions in the Holocene.This is coincident with post‑glacial woody encroachment and trait shifts toward taller, deeper‑rooted communities—mechanisms relevant to contemporary “arctic greening”. On the eastern Tibetan Plateau, proxy–model agreement demonstrates that complex terrain and connectivity to refugia are first‑order controls on post‑glacial vegetation trajectories: steep valley configurations enhance connectivity and reduce migration lags, whereas long gentle terrain can impose pronounced lags despite similar climate.
Finally, we outline how these proxy‑validated developments motivate a forthcoming multimodal deep‑learning foundation model (FOUNA) integrating global occurrences, paleo‑occurrences (including sedaDNA), remote sensing, and (paleo)climate to deliver transferable, decision‑relevant biodiversity predictions from decades to millennia.
How to cite: Herzschuh, U., Jia, W., Liu, S., Schild, L., Liu, Y., and Schwenkler, R.: Plant dispersal and biotic interactions across glacial–interglacial timescales: evidences from combining spatio‑temporal niche modelling with sedimentary ancient DNA proxy data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17762, https://doi.org/10.5194/egusphere-egu26-17762, 2026.
How global vegetation responded to climate change since the Last Glacial Maximum (LGM) remains incompletely understood due to the lack of continuous, global-scale reconstructions. Here we present a millennial-resolution reconstruction of global vegetation patterns since the LGM based on a synthesis of 3,286 pollen records using a biomization framework. We show that tundra dominated the mid- to high-latitudes of the Northern Hemisphere during the LGM, while steppe and open coniferous forests characterized western North America, taiga prevailed in eastern North America, and extensive steppe covered much of Eurasia. In the tropics, rainforest extent was markedly reduced, accompanied by widespread expansion of arid shrublands across Africa.
We find that forest expansion following deglaciation was spatially asynchronous across latitudes and hemispheres. Global forest cover increased by ~31% from the LGM to the mid-Holocene, before declining by ~5% during the late Holocene. In the Northern Hemisphere mid- to high-latitudes, forest cover rose rapidly after the LGM, peaked between ~7 and 5 ka BP, and subsequently declined, whereas Southern Hemisphere mid-latitudes experienced a more gradual increase, reaching maximum forest extent earlier (~12–8 ka BP) and remaining relatively stable thereafter. Tropical regions exhibited the most heterogeneous trajectories, with early deglacial fluctuations, sustained expansion to a mid-Holocene maximum (~6–4 ka BP), and enhanced variability in the late Holocene.
Our results reveal pronounced asynchrony in global vegetation evolution and provide a biome-scale perspective that refines previous global reconstructions. This dataset establishes a benchmark for evaluating palaeovegetation simulations and offers new constraints on vegetation–climate feedbacks relevant to future ecosystem change.
How to cite: Wu, H., Geng, J., Zhang, W., Li, Q., and Yu, Y.: Spatiotemporal Evolution of Global Vegetation Since the Last Glacial Maximum: Insights from Quantitative Pollen Reconstructions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9191, https://doi.org/10.5194/egusphere-egu26-9191, 2026.
Vegetation actively regulates climate through biophysical processes such as altering surface albedo, evapotranspiration, and roughness. Despite its recognized importance in modern and Quaternary systems, the role of terrestrial vegetation in shaping Earth’s climate on geological timescales remains poorly quantified. Understanding how vegetation–climate interactions vary across different background states—from icehouse to greenhouse worlds—is critical for interpreting paleoclimate proxies and for constraining future biosphere–climate feedbacks.
Here, we present a series of coupled climate–vegetation experiments using the Community Earth System Model (CESM1.2.2) with BIOME4 to systematically isolate the biophysical effects of land plants across 42 time slices from 410 Ma to the pre-industrial era. Through paired “Vegetated” and “Bare-ground” simulations, we assess the global and regional climatic impacts of vegetation across a wide range of paleogeographic and climatic conditions.
Our results show that vegetation consistently exerts a warming influence of 2–6 °C, primarily via albedo reduction, and increases precipitation by 30–105 mm yr⁻¹. This forcing is strongly state-dependent, being most pronounced during cold, high-ice climates where vegetation activates potent snow/ice-albedo feedbacks. Moreover, vegetation systematically reorganizes large-scale atmospheric circulation: it intensifies the Walker circulation, redistributing tropical rainfall, and under certain configurations can reverse the global meridional overturning circulation, thereby altering oceanic heat transport.
These findings establish terrestrial vegetation as a persistent, state-dependent climate modulator throughout the Phanerozoic, offering a unifying framework for understanding its role in past and future vegetation–climate interactions.
How to cite: Guo, J., Hu, Y., Liu, Y., and Liu, Y.: The Biophysical Forcing of Terrestrial Vegetation: A Persistent Climate Modulator with State-Dependent Efficacy Through the Phanerozoic, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17530, https://doi.org/10.5194/egusphere-egu26-17530, 2026.
The lengthening of the growing season in the Northern Hemisphere is a key response of terrestrial ecosystems to climate warming, yet long-term perspectives on Holocene growing season dynamics remain limited. Growing Degree Days (GDD), a widely used metric for assessing growing season thermal conditions, can be reconstructed using the micro-phenological method, a relatively recent proxy based on changes in leaf epidermal cell morphology. When combined with pollen-based reconstructions, this integrated approach provides robust estimates of past growing season thermal conditions.
In this study, we explore Holocene growing season variability at Linje peatland in northern Poland by combining Betula nana leaf micro-phenology with pollen-based GDD reconstructions derived from the same sediment sequence. Linje peatland represents a mid-latitude microrefugium where B. nana has persisted throughout the Holocene and where long-term peat accumulation, together with modern hydrometeorological monitoring, provides a unique opportunity for local proxy calibration. Building on an existing micro-phenological model developed for northern Finland, a site-specific inference model was established using annually collected modern leaves and applied to subfossil B. nana remains spanning approximately the last 11,350 years.
Preliminary results suggest broadly coherent long-term patterns in growing season thermal variability during the Late Holocene, while intervals of divergence between the two proxies are more pronounced during the Early Holocene. Interestingly, these differences may reflect contrasting proxy sensitivities or ecological response times. Overall, this study illustrates how combining micro-phenological and pollen proxies can be used to investigate past vegetation-climate interactions, growing season dynamics, and their relationship to prehistoric and historic human societies.
This research was supported by ESF project PRG1993, the Doctoral School of Tallinn University of Technology, the (Estonian) Ministry of Education and Research Centre of Excellence grant TK215, and the National Science Centre, Poland (grant nos. 2021/41/B/ST10/00060 and 2022/45/B/ST10/03423).
How to cite: Poolma, E., Wagner-Cremer, F., Kołaczek, P., Słowińska, S., Poska, A., Ercan, F. E. Z., Lamentowicz, M., Leszczyńska, K., Marcisz, K., Niebieszczański, J., Słowiński, M., Szambelan, W., Veski, S., and Amon, L.: Exploring Holocene growing season variability in Central Europe: Evidence from Vegetation Proxies, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18520, https://doi.org/10.5194/egusphere-egu26-18520, 2026.
Understanding how past societies adapted to climatic and environmental changes offers valuable perspectives for contemporary sustainability challenges. This study investigates the long-term interplay between climate, vegetation, and human settlement patterns in the Liaoxi Corridor—a sensitive forest-steppe transition zone at the northern margin of the East Asian Summer Monsoon. We provide a quantitative assessment of how Holocene climate variability influenced human habitat preferences through mediating changes in vegetation cover.
We integrated multi-proxy datasets to address this question. Thirty high-resolution pollen records were analyzed using the REVEALS model to reconstruct vegetation dynamics across the Holocene. Archaeological site distributions were examined through kernel density estimation and spatial clustering techniques to identify settlement aggregation patterns. For each period, we calculated site elevation, slope, and distance to rivers to assess topographic and hydrological preferences. To synthesize these variables, we applied a Human Ecological Niche Model (HENM), which allowed us to evaluate the relative importance of environmental factors in driving settlement location and to detect shifts in human habitat selection over time.
The results highlight several key findings. First, the forest-steppe boundary shifted markedly in response to Holocene monsoon variability, with forest expansion during humid phases and steppe dominance during arid intervals. Second, human settlements consistently clustered in environmentally favorable niches, but these niches changed over time. During warm-wet periods associated with forest expansion, populations dispersed into upland areas. In contrast, cooler and drier conditions led to settlement contraction into lowland river valleys, reflecting a strategic shift toward resource-security under climatic stress. Third, the HENM identified vegetation type, water accessibility, and gentle terrain as the primary factors influencing site location, with their relative weights varying across climatic phases.
This study underscores the role of vegetation as a critical intermediary between climate and human behavior. By quantifying past human-environment linkages in a climatically sensitive region, we offer a refined framework for understanding adaptive responses to environmental change. These insights not only deepen our knowledge of East Asian prehistory but also inform current models of landscape resilience and sustainable habitat planning under future climate scenarios. In an era of rapid global change, such long-term perspectives are essential for anticipating human-environment feedbacks and fostering resilient socio-ecological systems.
How to cite: Liang, C., Qin, F., Huang, B., and Li, J.: How Vegetation Mediated Human Settlement Responses to Holocene Climate Change: A Quantitative Spatiotemporal Analysis from the East Asian Transitional Zone, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18492, https://doi.org/10.5194/egusphere-egu26-18492, 2026.
Baja California, Mexico occupies a climatically sensitive peninsular setting between the cool Pacific Ocean and the comparatively warmer Gulf of California. This Mexican state is home to a large spectrum of environmental conditions and diverse ecology, due in part to the compounding effects of variable precipitation from El Niño Southern Oscillation (ENSO) cycling and the North American Monsoon (NAM) across a topographical gradient. Near the center of the state resides Sierra de San Pedro Mártir, a high elevation mountain range at the tip of the California Floristic Region, forming a California Mountains ecoregion that is drastically different in biodiversity than the area that surrounds it. Sierra de San Pedro Mártir is a pine-dominated forest that receives ~75% of its annual precipitation during winter months, making it particularly sensitive to ENSO-driven hydroclimatic variability. Notably, this forest has only recently seen the emergence of fire management strategies.
In a palaeoecological reconstruction from this region, a high-resolution fossil pollen record, coupled with macro-charcoal analysis, highlights shifting dominance between precipitation sources through the middle to late Holocene. More contemporarily, however, the impacts of fire suppression can already be seen in the palynological record. Methods of inferential statistics are employed alongside a traditional time series, and cohesion between these two methods of data analysis provides additional confidence in a compelling and robust precipitation-fire-ecology relationship detected through generalized linear regression. This finding has significant implications for the future of fire management in this unique environment, representing the integrative potential for high-resolution palaeoecological research. As this environment represents a natural laboratory for studying ENSO and NAM, this finding additionally has implications for how these two hydrological systems contribute to the future of more regional conservation and restoration.
How to cite: Enke, S., Watt, J., Codding, B., Layon, E., and Brunelle, A.: Hydroclimate-driven ecological and fire regime shifts in a unique forest biome of Baja California since the mid-Holocene, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20151, https://doi.org/10.5194/egusphere-egu26-20151, 2026.
Terrestrial vegetation models are typically based on quasi-empirical or dynamical relationships that link climatic, soil, nutrient, and land-use conditions to the presence of different plant taxa, biomes, or plant functional types. The majority of these models neither include plant seed dispersal nor herbivore grazing effects, potentially leading to a misrepresentation of climate-biosphere feedbacks. Both of these factors, however, are known to play a critical role in the global distribution of plant types.
In this presentation, we will introduce a new coupled global model with a 1x1 degree horizontal resolution that integrates vegetation dynamics with mammal herbivory. The vegetation model (ICCP global vegetation model, IGVM) simulates the fractional cover of grass, shrubs, trees, and desert using coupled reaction-diffusion equations. These quantities are translated into net primary productivity (NPP) through a non-parametric empirical method. The mammal model (ICCP global mammal model, IGMM) - also based on coupled reaction diffusion dynamics - realistically simulates the biomass of over 2,100 mammal species worldwide and has been extensively validated against observational datasets. The NPP from the vegetation model determines the biomass of individual herbivore species through their carrying capacities. In turn, herbivore grazing acts as a sink for vegetation carbon, and this effect is mapped back onto grass, shrub, and tree fractions in the vegetation model. Furthermore, the impact of mammal-mediated seed dispersal can be estimated.
By running the fully coupled model with and without mammal grazing, we determine the impact of mammal distributions on pre-Anthropocene global vegetation biogeography. This allows us to directly test the “Zimov Vegetation Hypothesis” and document the effects of trophic coupling on ecosystem functionality and stability at global-to-regional scales. We will further discuss how this new coupled modeling framework can be implemented into Earth System models.
How to cite: Mohanty, S., Timmermann, A., Venugopal, T., and Kim, I.-W.: A coupled vegetation–mammal modeling framework for Earth system models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8838, https://doi.org/10.5194/egusphere-egu26-8838, 2026.
The Mammal fossil record is long recognized as an excellent source for testing the causality of evolutionary change. The large mammal evolution shows a wide diversity of patterns including so called turnover pulses (high magnitude random impulses of extinctions and originations), periods of taxonomic stasis, while also there is plenty of evidence for long-term trends in morphological traits and taxonomic diversity. The major question is: how can we reconcile such a diversity of dynamical regimes given that we are having a single history of life?In this contribution we approach this question of dynamical regimes from the scaling perspective. The shapes of species diversity curves were reconstructed using spatio-temporal occurences of Perissodactyla, Artiodactyla (excluding Cetacea) and Carnivora (excluding Pinnipedia) from the NOW (New and Old Worlds) database by applying the Bayesian PyRate approach. The derived diversity curves were analyzed applying Haar fluctuations which were used in constructing structure functions, which reveal typical fluctuation magnitudes as a function of time scales.The results reveal that there are three separate time scales characterized by contrasting regimes. At shorter macroevolutionary time scales the episodic events produce turnover pulse patterns with the magnitudes peaking at time scales around 2 Ma. At time scales ranging from 3 to 5 Ma the stabilizing processes dominate. And the longer time scale the positive scaling trends in diversity dominate the biodiversity change. These longest time scale changes in biodiversity are directly coupled with the long-term megaclimate drift. The large mammal evolution is shaped by the superposition of qualitatively different processes operating on three time scales. The separation of time scales is shaped by the climate-macroclimate-megaclimate transitions and internal biotic feedbacks.
The study was supported by the project S-MIP-24-62 BretEvoGeneralized.
How to cite: Spiridonov, A., Lovejoy, S., Bekeraitė, S., and Stankevič, R.: Turnover pulses, intermittent stability and trends — on the time scales of large mammal evolution, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15242, https://doi.org/10.5194/egusphere-egu26-15242, 2026.
Total genomic diversity in humans increases when populations are isolated from each other for extended periods. However, the role of past astronomically forced climate conditions in the generation of human diversity remains unresolved. Here, we employ an agent-based model with genetic inheritance, a representation of culture, and realistic climate conditions to simulate the genetic history of African hominin populations throughout the Pleistocene. Our simulations of human population density and DNA changes show the dominant effect of Milanković cycles on dispersal, population structure, and genomic diversity. Warm early Pleistocene climates supported a heterogeneous patchwork of genetically diverse subpopulations across Africa. However, with the onset of colder conditions and reduced food resources at ~900 thousand years ago, human populations disappeared everywhere except in southern and eastern Africa. The corresponding simulated rapid decline in nucleotide diversity during this time is consistent with archaeological and genomic evidence. Following this regime change, humans began adapting to harsher climatic conditions, leading to rapid population expansions across the continent during interglacials. Boosted further by the spread of cultural traits and facilitated by warm, wet climate corridors over the last 400 thousand years, eastern African hominin populations eventually dispersed into Eurasia, contributing to the emergence of new geographically isolated populations and distinct genomic lineages.
How to cite: Fang, S.-W., Raia, P., Sundaresan, A., Barbieri, C., Ruan, J., Vahdati, A. R., Zeller, E., Zollikofer, C., and Timmermann, A.: Past Climate and Cultural Impacts on African Human Genetic Diversity , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9467, https://doi.org/10.5194/egusphere-egu26-9467, 2026.
Past climate change and cultural evolution played significant roles in the migration of archaic humans into new geographic areas, contributing to the diversification of the genus Homo. The Mid-Pleistocene period was a critical time when Homo heidelbergensis evolved in Africa and migrated to Eurasia, likely leading to the emergence of new human species, including Homo neanderthalensis and Denisovans. The present study investigates how past climate, rivers, and cultural changes affected possible hominin migration routes to northwest Africa and Eurasia, as well as the timing of their arrival. To identify migration pathways across Africa and Eurasia, we conducted an ensemble of sensitivity experiments using a realistic climate-forced agent-based model (ABM) with varying cultural levels and with coastal and riverine routes enabled or disabled. In the absence of coastal and riverine activation and low cultural levels, the hominin population remains confined to southern and eastern Africa. However, with higher cultural evolution, they could reach north-west Africa via the western Saharan route.
The ABM simulations with river and coastal amplification show that the hominin populations migrated to north-eastern Africa via the Nile route at low cultural levels. However, with increased levels of culture, they could reach north-western Africa through the Nile-Mediterranean coastal route, and the north-western African population shows intermittent interaction with the west-central population. Also, over a short period, they dispersed into Eurasia via the Levant and migrated into Europe. Thus, coastal and riverine amplification helped the hominin reach north-western Africa and Europe with relatively low cultural values at the beginning of the middle Pleistocene period, which closely matches the Homo heidelbergensis archaeological record. Additional analysis of our simulations reveals that the precessional cycle played a dominant role in controlling the hominin migration through the corridors. At the same time, population density in African population hotspot regions was controlled by changes in atmospheric CO2 concentration. The phylogenetic analysis of the individual virtual agents' DNA shows distinct branches for the north-west, central, and south-east African populations. Since low-frequency climate cycles isolate and reconnect north-west and central African populations with east and south African populations, they contribute to the dynamics of genetic diversity.
How to cite: Sundaresan, A., Timmermann, A., and Fang, S.-W.: Role of Climate, Culture, and Waterways in Shaping the Hominin Population Dynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2478, https://doi.org/10.5194/egusphere-egu26-2478, 2026.
Due to processes such as climate change, construction, and agricultural expansion, natural habitats have become fragmented and functionally degraded. As a result, numerous wildlife species are forced to migrate to new suitable habitats. During these migrations, their ranges increasingly overlap with human activities, creating potential risks for human-wildlife conflicts. Unlike previous studies relying merely on species distribution models, this study innovatively predicts future human-wildlife conflict risks during wildlife migration driven by habitat degradation in Southwest China in 2030 and 2050. By integrating habitat degradation assessments under multiple climate change and socio-development scenarios with species migration path simulations employing landscape ecology methods, alongside land-use modeling and human footprint data, this study quantifies conflict risks between humans and wildlife species such as takin and wild boar while classifying conflict types. Building upon historical and current conditions, the findings demonstrate that future climate change and human activities will trigger large-scale habitat degradation and significant spatial shifts in suitable habitats. Consequently, a chain reaction—involving increased conflicts, wildlife capture, and questioning of conservation actions—threatens the harmonious relationship between humans and wildlife. This research offers a complementary perspective on understanding climate change impacts on terrestrial life and holds significant value for guiding the optimization of biodiversity conservation planning and policy development.
How to cite: Li, Q. and Jin, X.: Mapping future human-wildlife conflict risks during habitat degradation and species migration driven by climate and human factors in Southwest China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15372, https://doi.org/10.5194/egusphere-egu26-15372, 2026.
Among wild pollinators, diurnal butterflies are important in natural ecosystems and contribute significantly to agricultural productivity. Worryingly, a growing body of literature suggests that Climate Change (CC) may result in the extinction and decline of many butterfly species. Understanding which species and areas are most vulnerable to CC is essential for planning conservation and mitigation efforts. In this work we present the main results obtained during LIFE project BEEadapt (LIFE21-CCA-IT-LIFE BEEadapt/101074591) which aims to improve wild pollinator climate resilience in four areas in Central Italy, including protected areas, natural and agro-ecosystems.
Results show first that CC signals are evident in all the studied areas in terms of increased temperatures, and increased extreme events, both in intensity and frequency. Furthermore, they show that butterflies have a consistent vulnerability pattern at both the species and multispecies level. In the study areas, CC appears to favor lowland and generalist species, which increase their climatic suitability under both scenarios, particularly in mountains. Mountain and specialist species are expected to have reduced climatic suitability, especially under the SSP5-8.5.
Findings are comparable with recent studies on the effects of CC on pollinators, which revealed similar sensitivity patterns based on species ecology, and provided new insights into species potential local responses to CC, allowing to set conservation priorities and direct LIFE BEEadapt mitigation actions which need to be combined with the definition of governance strategies and the involvement of key actors at different spatial levels.
How to cite: Baldi, M. and Biancolini, D.: How climate change impacts on wild pollinators: the case of butterflies in Central Italy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8952, https://doi.org/10.5194/egusphere-egu26-8952, 2026.
During the Belgian Bronze Age two major Rapid Climate Change (RCC) events occured around 4.2 and 3.2 ka. Yet, the impacts of these episodes on environments and human communities in northwestern Europe remain insufficiently understood. This contribution presents results from the Learning from the Past (LEAP) project, which examines how abrupt climate shifts influenced ecosystems, mobility patterns, and population dynamics in pre- and early-complex societies during the Middle to Late Holocene in the Meuse basin of Belgium.
Using a high-resolution, multiproxy approach, LEAP integrates palaeoclimate data (C and O isotopes, and trace elements from speleothems), palaeoenvironmental evidence (pollen and microcharcoal from raised peat bogs), and archaeological datasets including palaeomobility indicators (O and Sr isotopes from human remains) and palaeodemographic proxies (SPDs and kernel density estimates).
By statistically modelling and correlating these high-resolution archaeological, environmental, and climatic records, as well as comparative data from neighbouring regions, the project evaluates the synchronicity of environmental stressors and societal responses. Leads- or lag-responses are explored, as well as handling of unequal sampling intervals, and determining the significance of signals and potential causality.
Preliminary results point to shifts in settlement density, funerary practices, population size, and mobility that coincide with periods of climatic fluctuation and environmental change. These patterns shed light on the resilience and adaptive capacities of Belgian Bronze Age communities facing short-lived environmental changes.
How to cite: Van Maldegem, E., Pincé, P., Capuzzo, G., Boudin, M., Burlet, C., Crombé, P., De Groote, I., De Mulder, G., Leonard, H., Snoeck, C., Verheyden, S., Wojcieszak, M., Fagel, N., and Deforce, K.: Modelling Resilience and Adaptation to Abrupt Climate Changes in the Belgian Bronze Age: Insights from a High-Resolution Multiproxy Study, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2000, https://doi.org/10.5194/egusphere-egu26-2000, 2026.
In 1863, the Bear River Massacre took place in Southeastern Idaho, USA, where about 400 members of the Northwestern Band of the Shoshone Nation were murdered by the United States Government. The massacre caused significant ecological land changes from the massacre itself but also from the colonization of the land, which presented land use changes and the introduction of invasive species. The tribe has since received 350 acres of their traditional land back from the government, but the land has been vastly altered since their ancestors lived on it. The Bear River restoration project, led by the Shoshone tribe, was created with the aim to bring the native vegetation back to the land and to allow the tribe to learn about the relationship between their ancestors and the resources they used.
This research is contributing to the tribe’s restoration goals by reconstructing the past vegetation and environmental history of the Bear River Massacre Site using a quantitative, multiproxy paleoecological approach. The primary questions of concern that we aim to answer for the tribe are what was the native vegetation like when their ancestors lived on the land, and what is the environmental history of the site being a spring. The methodological approach to answer these questions will utilize pollen counts and loss on ignition from a wetland sediment core collected from a spring along the Bear River to reconstruct the paleoenvironment and identify past changes and disturbances in the environment.
How to cite: Layon, E., Watt, J., Aichele, E., Brunelle, A., and Codding, B.: Reconstructing the Environmental History of the Bear River Massacre Site, Idaho, USA, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13603, https://doi.org/10.5194/egusphere-egu26-13603, 2026.
Archaic hominin fossils from East Asia dating to the late Middle Pleistocene and Late Pleistocene display substantial morphological diversity, and their systematic classification has long remained controversial. In this Perspective, we integrate morphological evidence with recent advances in molecular research to re-evaluate the evolutionary landscape of archaic hominins in East Asia prior to the emergence of Homo sapiens, with particular focus on the Harbin cranium and its implications for the taxon Homo longi. Recent paleoproteomic and ancient DNA studies indicate that the Harbin cranium carries Denisovan-related genetic and proteomic signatures and is closely affiliated with Denisovan lineages identified at Xiahe, Penghu, and in the Altai Mountains. When viewed in a broader morphological context, the Harbin cranium and the Xiahe mandible form a sister grouping, together with fossils from Dali, Jinniushan, and Hualongdong, suggesting a coherent East Asian archaic hominin “Homo longi” clade. We propose a unifying taxonomic framework in which East Asian Denisovan populations are provisionally referred to as Homo longi, and discuss the possibility that this lineage comprised multiple deeply divergent populations with the capacity to occupy diverse ecological niches, including high-altitude environments. Future research integrating additional ancient genomes, proteomic data, and high-precision chronologies will be essential to further elucidate the origins, dispersal, environmental adaptations, and contributions of Homo longi populations to the formation of modern humans in East Asia.
How to cite: Chen, F., Ran, J., Xia, H., Xing, S., and Li, H.: Homo longi, Denisovans, Neanderthals, and other archaic hominins in eastern Asia prior to the Rise of Homo sapiens, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6137, https://doi.org/10.5194/egusphere-egu26-6137, 2026.
The past decades have seen an upsurge in the paleoenvironmental studies of Chinese archaeological sites. However, systematic investigations on human-environment interactions in river valleys are still rare in Central China and thus require further study. Here, we reconstruct the landscape evolution of the Shuangji River valley in the eastern foothills of Songshan Mountain and its relationship with climate change and human settlement patterns since the terminal Pleistocene. From 50 ka BP to the terminal Paleolithic, under cold climate conditions, approximately 20 m of fluvial-lacustrine sediments and loess-derived alluvium were deposited in the middle reaches of the river valley. A transition from fluvial-lacustrine to aeolian deposits occurred around 28 ka BP, accompanied by a decrease in deposition rate. Three stages of fluvial terraces were formed since the terminal Pleistocene. The formation of the third terrace (T3) was dated between 20~10 ka BP. It provided the ideal habitat for the last hunter-gatherers and early farmers through the terminal Paleolithic to early Neolithic. From 8 to 4 ka BP, the river valley aggraded under a warm and humid climate, while the second terrace (T2) formed slowly. Due to its suitability for human habitation, settlements gradually moved downstream and clustered on the alluvial valleys, associated with the change of subsistence strategy. After 4 ka BP, the climate aridity coincided with large-scale river downcutting, which led to the disappearance of lakes and swamps. This paralleled the emergence of urban settlements. The late Holocene valley incision and smaller-scale first terrace (T1) during the historical period shaped the present landscape. Our results contribute to a better understanding of the relationships between climate change, landscape evolution, and human settlement patterns in the cradle of Chinese civilization.
How to cite: Ren, X. and Mo, D.: Climate-human-landscape interaction in the eastern foothills of Songshan Mountain, Central China since the terminal Pleistocene, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21407, https://doi.org/10.5194/egusphere-egu26-21407, 2026.
Reconstructing the spatiotemporal features of past climate variability and assessing its influence on societal change are essential for understanding long-term human–environment co-evolution and for informing contemporary climate adaptation. The “4.2 ka event” (around 4.2 ka BP) has long been regarded as a major hydroclimatic anomaly marking the onset of the Late Holocene and has frequently been invoked to explain major societal disruptions across multiple regions. However, expanding and increasingly detailed proxy records have challenged both the presumed global uniformity of this event and the magnitude of its societal impacts.
To address these debates, this study conducts a global transient simulation for 4.5–3.5 ka BP using the Community Earth System Model (CESM) to obtain continuous climate fields across Eurasia and to evaluate whether the 4.2 ka anomaly represents coherent regional change or spatially heterogeneous variability. In parallel, we compile archaeological cultural sequences and key regional syntheses across Eurasia, and delineate seven sub-regions based on subsistence strategies and environmental settings. By comparing the spatiotemporal pattern of climate anomalies with trajectories of social change within each sub-region, we examine plausible pathways through which climate perturbations may have shaped early societal dynamics.
Our results indicate that the 4.2 ka signal is widespread across Eurasia but is far from uniform: anomaly intensity, persistence, and hydroclimatic expression exhibit pronounced spatial heterogeneity. Such heterogeneity implies region-specific societal consequences, ranging from amplified stress and risk accumulation in some socio-ecological settings to the reorganization of resources and interregional connectivity in others. These differential impacts may have contributed to divergent developmental pathways among early societies, including those in early China, the Harappan world, Mesopotamia, the Mediterranean, and the Eurasian steppe. Overall, our findings underscore the need to move beyond deterministic “collapse vs. flourishing” narratives and toward process-based, regionally explicit mechanisms linking climate variability and social change.
How to cite: Han, L., Yang, H., and Liu, M.: Spatiotemporal Characteristics of the 4.2 ka Climate Event Across Eurasia and Its Implications for Early Societal Trajectories, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9369, https://doi.org/10.5194/egusphere-egu26-9369, 2026.
Throughout the first millennium of East Asian imperial history, Chinese empires periodically extended military control over the semi-arid to arid southern Central and Inner Asia. During warmer, wetter climatic phases, states undertook efforts to establish or restore agricultural settlements in the region. Crop production was primarily intended to sustain frontier garrisons and their livestock through systems of state-organized military colonies (tuntian), in which soldier-farmers cultivated the land. The military colonies addressed growing demand for land and agricultural output to support both people and livestock. Recent scholarship on Han military farming (Trombert 2020) argues that agricultural colonies in the Hexi Corridor and Tarim Basin were largely unproductive and placed heavy demands on soldier-farmers, who had to balance cultivation with military service; the main significance of military colonies was likely not in their productivity, but rather in their role as strategic footholds that facilitated subsequent civilian settlement. Building on this argument, I examine the Tang military colonies along the Yellow River in the northwestern fringes of the empire, from the Qinghai Plateau to the Ordos Loop. This semi-arid region, situated at the edge of the monsoon zone, was traditionally more conducive to an agropastoral economy. It comprised the southern segment of the trade routes connecting Central Asia, extending through the Hexi Corridor. The Tang established large, permanent armies in the region and expanded agricultural settlements to sustain them. Scholars argue that the expansion of cropland into typically unsuitable areas was likely enabled by a particularly favorable climatic period in the 7th century, characterized by warm and humid conditions. Rising temperatures could have enabled earlier planting dates and extended growing seasons, while also expanding arable land into higher-altitude regions. Increased precipitation would have further supported crop growth by boosting water availability. Unlike earlier periods, Tang administrative records detail the civil and military populations, livestock numbers, farm counts, crop types, and the man-days of labor needed to cultivate each crop. This extensive data can serve as a proxy for productivity and sustainability. By combining historical and administrative data with climatic data, the paper emphasizes the importance of studying how state institutions addressed environmental challenges and climate variability in the empire's semi-arid peripheries. It shows how military farms relied on continuous state intervention, particularly in land distribution, irrigation system maintenance, and labor enforcement.
How to cite: Barenghi, M.: State Agriculture on the Ecological Margins of Empire: Military Colonies and Environmental Adaptation in the Sui–Tang Period (6th–8th Centuries CE), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18040, https://doi.org/10.5194/egusphere-egu26-18040, 2026.
The Greater Ordos Region (GOR), located at the interface between the East Asian Summer Monsoon and mid-latitude westerly circulation systems, is highly sensitive to both oceanic forcing and regional land–atmosphere interactions. This study synthesises annually resolved tree-ring and documentary records with lower-resolution evidence from lake sediments, aeolian archives, and pollen data to reconstruct hydroclimatic and temperature variability over the past ~3500 years. The multi-proxy evidence reveals pronounced alternations between wetter and drier conditions across successive dynastic periods. High-resolution records resolve the timing, duration, and severity of extreme events, including multi-decadal droughts during the late Han and Tang periods and a widespread megadrought in the early seventeenth century CE associated with crop failures and societal stress. Lower-resolution archives provide longer-term context, documenting progressive shifts towards increased aridity, steppe expansion, and desertification, particularly following major drought episodes. The combined proxy approach demonstrates how recurrent hydroclimatic extremes, interspersed with phases of recovery, have exerted a persistent influence on agricultural systems, land-use dynamics, and societal stability. Integrating high- and low-resolution climate records allows assessment of both abrupt climate shocks and longer-term environmental trends that have shaped regional vulnerability through time.
How to cite: Xoplaki, E., Luterbacher, J., Chen, F., Liu, B., Qin, C., Yang, B., Su, R., Kempf, M., Ma, S., Hao, Z., Haupt, M., Barenghi, M., Bello, D., and Di Cosmo, N.: Paleoclimatic Evidence across the Ordos Region and Yellow River Loop, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19430, https://doi.org/10.5194/egusphere-egu26-19430, 2026.
Hydroclimatic variability has significantly influenced societal dynamics in arid Central Asia (ACA), by triggering periods of unrest and migration. The western part of ACA, a key node of the ancient Silk Roads, remains poorly investigated due to limited climatic and environmental records, hindering our understanding of how environmental changes shaped the evolution of civilization in this region. This study uses records of high-resolution scanning XRF (X-Ray Fluorescence), n-fatty acids, and grain size from the sediments of Green Spring Lake, in northeastern Iran, within the historically prominent region of Greater Khorasan. This record is then used to reconstruct the regional hydroclimatic variability over the past 1,350 years. It reveals a generally dry and stable climate during the Medieval Climate Anomaly (950–1250 CE), a pronounced drought during 1250–1350 CE, and a transition to wetter conditions accompanied by increased hydroclimatic variability during the Little Ice Age (1400–1850 CE). During the LIA, increased moisture supply to northeastern Iran was caused by a negative NAO phase in spring, coupled with anomalous ascending atmospheric motion caused by the weakening of the Siberian High, which jointly led to increased spring precipitation in this region. Additionally, drought events during 970–1040 CE, 1220–1250 CE, 1480–1520 CE, 1600–1650 CE, and 1850–1910 CE align with documented events in eastern Iran, including severe famines and population declines, as recorded in historical Iranian sources.
How to cite: Xie, H.: Decadal hydroclimate fluctuations recorded in lake sediments from northeastern Iran over the past 1350 years, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4595, https://doi.org/10.5194/egusphere-egu26-4595, 2026.
Investigating the intricate connection among the combustion types (smoldering v.s. flaming), human activity and climatic patterns is is crucial for understanding the mechanisms of paleo-fire occurrences. This interaction requires further clarification, especcially in the NE Tibetan Plateau, due to the dramatic climatic and human activity shifts during the mid-late Holocene. Here, we examined the black carbon (BC, comprising char and soot) content within sediments from Caodalian (CDL) lake spanning the past 7600 years. The findings indicate paleo-fire intensity was consistent with variations in combustion types revealing by the ratio of char/soot on different timescales during the mid-late Holocene. An integrtated paleo-fire index shows that the fire activity underscores a pattern of low middle Holocene and rapidly increasing late Holocene fires, with two distinct peaks occurring during the Bronze Age and historical period. This variation aligns with the shifts observed in the char/soot ratio, indicating that enhanced flaming fires were more prevalent during the late Holocene. A comparative analysis of regional paleo-climate data has revealed that the progressive aridification of the climate, along with rising spring temperatures, contributed to the increase in paleo-fires. And the expansion of grasslands likely fueled the rise in flaming fires, thereby intensifying paleo-fire activity. Notably, we contend that human fire practices heightened the incidence of late Holoceng regional flaming fires, which in turn contributed to the intensification of paleo-fire regimes. High-intensity human activities (e.g. land reclamation, pottery production, bronze crafting) that have been prevalent since 4000 BP, along with the increased warfare since 1200 BP, were significant factors behind this outcome.
How to cite: Zhang, S.: Enhanced late Holocene flaming fires in the NE Tibetan Plateau: coupled impacts of climate and human activities, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16560, https://doi.org/10.5194/egusphere-egu26-16560, 2026.
Global vegetation patterns are not only determined by climate, water availability, and soil conditions, but also by the dynamics of seed/plant dispersal, competition, herbivory, and fire. To account for these processes, we developed a new dynamical vegetation model (ICCP Global Vegetation Model) based on coupled 2D Lotka-Volterra equations that also includes plant and fire diffusion. The model simulates the area fraction of three plant functional types (grass, shrubs, and trees) and fire. Fire is introduced as a stochastic predator that "feeds" on available burnable carbon and emerges when climate conditions are suitable. The climate dependence of the competing plant functional types is calculated from a species distribution model that calculates habitat suitability from key climatic parameters. The model can also account for herbivore grazing, which is estimated from the ICCP Global Mammal model. In this presentation, we will compare transient Pleistocene simulations conducted with the ICCP Global Vegetation Model with Biome4 model simulations and time-slice vegetation reconstructions for the mid-Holocene (6 ka) and the Last Glacial Maximum (21 ka).
How to cite: Kim, I.-W., Timmermann, A., and Mohanty, S.: A predator-prey model for Pleistocene global vegetation and wildfire dynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8730, https://doi.org/10.5194/egusphere-egu26-8730, 2026.
The Holocene in the Korean Peninsula offers an ideal natural laboratory for evaluating long-term interactions among climate, vegetation, fire, and human activity. We present high-resolution pollen and sedimentary charcoal records from paleo-lake Gaho in southern Korea covering the last ~7,000 years, and derive quantitative reconstructions of mean annual temperature and annual precipitation using the modern analogue technique and an extensive modern pollen dataset. The temperature estimates (approximately 9–12°C) capture millennial-scale variability linked to changes in the East Asian winter monsoon and Bond-scale events, whereas reconstructed precipitation (around 1,250–1,540 mm) follows shifts in the Intertropical Convergence Zone and the strength of the East Asian summer monsoon. Hydroclimate signals inferred from pollen are consistent with lake-level changes, geochemical indicators, and multivariate statistical analyses. Charcoal influx records indicate persistent fire occurrence throughout the Holocene, with a marked rise in large-scale burning around 5.0–4.0 ka BP, likely associated with progressive drying and increased fuel availability. After ~3.0 ka BP, the appearance of abundant large Poaceae pollen (>40 μm) suggests expansion of agriculture, and pronounced fluctuations in Pinus and Quercus after ~2.0 ka BP indicate intensifying human disturbance. We infer that late Holocene fires were increasingly anthropogenic, associated with land clearance, warfare, and metallurgical activities rather than purely climatic forcing. Overall, our results demonstrate the coupled evolution of climate, ecosystem dynamics, and human impact in southern Korea during the Holocene, providing important context for anticipating ecosystem responses under ongoing climate change.
How to cite: Lee, J. and Yi, S.: Tracking Holocene climate, fire, and human activities in southern Korea using pollen and charcoal records., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8569, https://doi.org/10.5194/egusphere-egu26-8569, 2026.
Climate and vegetation are inherently intertwined through feedbacks that are still not fully understood. Reconstructing past climate-vegetation interactions is challenging because plants have undergone evolutionary, physiological, and ecological changes that cannot be inferred from the present day vegetation. Here we investigate a range of potential vegetation/climate states that could have existed 3 million years ago with the use of the BIOME4 vegetation model, pollen data, and iCESM1.3 mid-Pleistocene model simulations,. The BIOME4 model is widely used to reconstruct paleo vegetation although the plant phenology parameterization is based on modern day vegetation types. We explore the sensitivity of the model to the plant phenology parameterization and the space of possible vegetation distributions to find best fits to pollen data. With the use of iCESM1.3 we estimate the range of vegetation related climate uncertainties showing that this can cause local to global changes impacting e.g. arctic amplification and global circulations.
How to cite: Zeller, E.: Vegetation related climate uncertainty during the mid-Pleistocene warm period, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13794, https://doi.org/10.5194/egusphere-egu26-13794, 2026.
The transition between the Early and Middle Pleistocene (1400-400 ka) represents a key phase of climatic system reorganization, marked by the shift in dominant orbital periodicity from 41 kyr to 100 kyr cycles. In the Mediterranean region, this transition is associated with a trend toward increasing aridity. A multi-proxy analysis was conducted on the continuous marine sediment sequence from ODP Leg 161- Site 976, in the Alboran Sea. The study integrates pollen, isotopic, and clay mineral data together with a multi-method pollen-based climate-reconstruction approach (CAM, MAT, WA-PLS, RF and BRT) and reveals contrasting dynamics among the different indicators. The isotopic records, pollen assemblages, and quantitative climate reconstructions all display variability structured by glacial–interglacial alternation. These proxies show progressive yet well-defined transitions between open cold and dry phases and warmer and more humid conditions. Between 1100 and 870 ka, the record shows a decrease in the amplitude of temperate deciduous forest development. In contrast, the clay mineral composition exhibits a distinct cyclicity that diverges from the classical glacial–interglacial rhythm, and is characterized by a series of abrupt changes, particularly pronounced between 1235 and 870 ka. These mineralogical disruptions suggest rapid reorganizations of hydro-sedimentary conditions, potentially linked to regional modifications in ocean and/or sea circulation and aeolian inputs. Altogether, the results highlight the complexity of the Early–Middle Pleistocene transition in the western Mediterranean, shaped by the superposition of global climatic forcing and regionally specific responses.
How to cite: Catrain, M., Combourieu-Nebout, N., Lebreton, V., Fauquette, S., Peyron, O., Joannin, S., Bout-Roumazeille, V., Fischer-Fries, M., Richard, P., Dubost, L., Delattre, M., Bertini, A., Toti, F., and Moncel, M.-H.: Climatic and environmental variability of the Early–Middle Pleistocene transition in the western Mediterranean, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-501, https://doi.org/10.5194/egusphere-egu26-501, 2026.
Marine Isotope Stages (MIS) 14 to 12 (~563-424 ka) precede the Mid-Brunhes Event (~424 ka, MIS 12/11 transition), which marks a significant shift in the amplitude of glacial-interglacial cycles, and an increase in interglacial temperatures. This interval would also encompass the end of the Early-Middle Pleistocene transition (EMPT; 1.4-0.4 Ma) as defined by Head & Gibbard (2015). Despite ongoing debate regarding the precise EMPT boundaries, the MIS 14-12 interval remains crucial for understanding Pleistocene climate dynamics; particularly MIS 12 (~478–424 ka), one of the Pleistocene’s most intense Northern Hemisphere glaciations. The climatic oscillations during MIS14-12 profoundly influenced environmental conditions, potentially creating areas that were more or less favourable for human settlements in southern Spain. Archaeological data indeed show an absence of sites south of Spain during MIS 14-12 period, a pattern that could be interpreted as a possible response to environmental and climatic constraints. However, vegetation dynamics in the region during this key interval are still poorly understood due to the scarcity of available records.
Here, we present a new, continuous and regional pollen record from marine ODP Site 976 in the Alboran Sea, located south of the Iberian Peninsula. This record covers MIS 14 to MIS 12. to document the vegetation response in a climatically sensitive region. A multi-method approach, combining modern analogues, regression models and machine-learning techniques (e.g., Modern Analogue Technique (MAT), Weighted Average Partial Least Squares (WAPLS), Boosted Regression Trees (BRT), Random Forest (RF) & Climatic Amplitude Method (CAM)), are applied to the pollen data to reconstruct annual and seasonal temperature and precipitation. These results are compared with those from other western Mediterranean pollen records, as well as with the timing of human occupations recorded in an archaeological database to strengthen our understanding of settlement dynamics and their relationship with environmental changes.
The pollen data and climate reconstructions reveal significant shifts in vegetation and climate: a steppe-dominated landscape under less severe conditions during MIS 14, two phases of temperate forest expansion during MIS 13; and an enhanced steppe development under the very cold MIS 12 climate. Additionally, this study reveals three distinct climatic phases in southern Spain during MIS 13 and MIS 12, which are also recorded in several marine, pollen and Chinese loess archives. Although these archives were able to distinguish several phases within MIS 14, the resolution of our pollen record is insufficient to detect them.
How to cite: Fourcade, T., Combourieu-Nebout, N., Lebreton, V., Fauquette, S., Peyron, O., Sassoon, D., Robles, M., Dubost, L., Cucart-Mora, C., and Moncel, M.-H.: Western Mediterranean vegetation and climate responses to MIS 14-12 glacial-interglacial variability: new insights from ODP Site 976 (Alboran Sea), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-500, https://doi.org/10.5194/egusphere-egu26-500, 2026.
The onset of the Eemian interglacial (MIS 5e) represents a critical interval for investigating vegetation responses to rapid climate warming and changes in hydroclimatic seasonality in the Mediterranean region. While forest expansion during early interglacial phases is rather well-documented, the degree to which reforestation pathways differed within the same geographic region across contrasting physiographic settings, remains insufficiently explored. Here we present a palynological comparison of early Eemian vegetation development at two central Italian sites characterised by different topographic and climatic contexts: Valle di Castiglione, a lowland crater lake of Alban Hills, located in the alluvial plain close to Rome, and the Fucino Basin, a wide intramontane dried out lake situated in the central Apennines. The explicit aim of this comparison is to assess whether early Eemian reforestation followed synchronous or differentiated trajectories in lowland versus intramontane settings, and to evaluate the role of altitude, basin morphology and continentality on modulating forest establishment and stability.
Preliminary pollen data from Valle di Castiglione carried out in the frame of AMUSED project (https://progetti.ingv.it/it/amused), indicate a rapid expansion of arboreal taxa at the MIS 6–MIS 5 transition, associated with increasing temperature and precipitation and the rapid establishment of predominantly mesophilous and Mediterranean vegetation during MIS 5e. These lowland dynamics are compared with evidence from the Fucino Basin, one of the most complete and sensitive terrestrial archives in the Mediterranean and a key reference for vegetation–climate relationships in central Italy. The comparison is explicitly designed to explore whether early Eemian reforestation followed synchronous or differentiated pathways in lowland coastal versus intramontane environments. This lowland record is compared with the high-resolution, radiometrically constrained pollen sequence from the Fucino Basin recently presented by Roberts et al. (2025), which provides a detailed reconstruction of vegetation dynamics between ~139 and 107 ka based on a robust tephrochronological framework. The Fucino record shows that early Eemian forest expansion was not monotonic but involved a rapid increase in many temperate deciduous taxa interspersed with centennial-scale fluctuations and transient reductions in forest cover.
This contribution is conceived as a pilot and contextual study in support of the ICDP proposal for the drilling of the Fucino paleolake (MEME project – the longest continuous terrestrial archive in the Mediterranean recording the last five million years of Earth system history), which aims to recover a continuous, high-resolution and chronologically robust record of environmental change across the entire basin infill. By placing early Eemian vegetation dynamics from Valle di Castiglione into a regional comparative framework with Fucino, this study provides a first step towards disentangling the role of altitude, basin setting and climatic gradients in shaping interglacial reforestation patterns in central Italy.
Roberts, C. A., Zanchetta, G., Giaccio, B., Nomade, S., Mannella, G., Sadori, L., ... & Tzedakis, P. C. (2025). A radiometrically-constrained reference record of Last Interglacial climate and vegetation changes from the Fucino Basin, Central Italy. Quaternary Science Reviews, 363, 109377.
How to cite: Borgognone, C., Giaccio, B., Macrì, P., Roberts, C. A., Sadori, L., Tzedakis, C., Zanchetta, G., and Masi, A.: Eemian reforestation patterns in central Italy, a palynological comparison between a lowland basin (Valle di Castiglione) and an intramontane basin (Piana del Fucino)., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17644, https://doi.org/10.5194/egusphere-egu26-17644, 2026.
The Late Palaeozoic Ice Age (LPIA), was one of Earth's most extensive and long-lasting glacial episodes, spanning roughly from 350 to 260 Ma. The Arabian Peninsula is long known to have experienced the LPIA at its position at the northern edge of Gondwana throughout the Late Carboniferous (Pennsylvanian) to the Early Permian (Cisuralian) and particularly the deglaciation that occurred from the latest Gzhelian, through late Sakmarian/early Artinskian to mid-Kungurian. Modelling of Mean Annual Surface Temperature (MAST; Li et al. 2022) for this period superimposed on palaeogeographic maps based on PaleoDEM and points/polyline/polygon (rotation and geometry files) of Scotese and Wright (2018) allows temperature-calibration of the succession of palynological assemblages. A number of trends and generalisations are possible related to MAST change between -0.2°C to -3.4°C (latest Gzhelian) and 9.3°C to 11.1°C (mid-Kungurian). As a group, the plants that produced monosaccate pollen (now extinct) appear amongst the most tolerant of MAST increase, with certain genera, for example Plicatipollenites and Cannanoropollis, being common throughout. Punctatisporites probably produced by the simplest most cold-adapted plants such as mosses were the most sensitive to climate warming. Cingulicamerate spores and fern spores of Microbaculispora and Horriditriletes are similarly sensitive to warming conditions particularly as MAST reaches above 0°C. MAST above 0°C appears to have stimulated a surge of caytonialean-type, probably upland, trees or shrubs that produced Pteruchipollenites indarraensis, although continued warming seems to have been at least partly responsible for restricting their distribution because such plants are almost absent at 208 Ma where MAST is ~10°C. Kingiacolpites subcircularis, probably produced by a cycad, may, also have been stimulated by MAST reaching above 0°C. Some of these trends in palaeovegetation in response to climate warming may have relevance in studies of modern environmental change.
How to cite: Stephenson, M. H., Shen, S., Fan, J., Hu, L., and Qi, J.: Palaeotemperature calibration of palynological assemblages and palaeovegetation through the last stage of the Late Palaeozoic Ice Age , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7510, https://doi.org/10.5194/egusphere-egu26-7510, 2026.
Understanding the seasonal timing of vegetation growth ("Green Wave") is crucial for modeling prehistoric human mobility and settlement patterns. However, high-resolution vegetation data is only available for the modern satellite era. To reconstruct these dynamics in the deep past, we present a hybrid modeling approach that combines domain-specific knowledge of seasonality with the flexibility of supervised machine learning.
Our core premise is that while we cannot observe the past directly, we can learn the rules of phenology from the present. We utilize global modern datasets to learn a mapping between climatic conditions and vegetation greenness, which can then be applied to paleoclimate simulations.
Our method decomposes the problem. First, we compress modern satellite observations into compact, interpretable parameters using a harmonic seasonal model. Second, we train a machine learning regressor to learn the complex, non-linear mapping between bioclimatic drivers and these phenological parameters. By treating the seasonal shape as a prediction target, we ensure that our reconstructions maintain structural integrity. We validate the model using spatially-disjoint cross-validation to account for spatial autocorrelation, ensuring robust generalization. The resulting framework allows us to translate paleoclimate simulations into high-resolution maps of ancient vegetation seasons, providing new quantitative inputs for archaeological hypotheses.
How to cite: Schlüter, P. and Shao, Y.: Learning the Green Wave: A Hybrid Machine Learning Framework for Reconstructing Past Vegetation Dynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7184, https://doi.org/10.5194/egusphere-egu26-7184, 2026.
Climate strongly constrains vegetation structure and the availability of surface water, thereby shaping the distribution, quality, and accessibility of resources for mammals, including hominins, over long timescales. Unfortunately, fossil and archaeological archives are discontinuous and biased by preservation and sampling, which limits the use of deterministic or fully probabilistic approaches that typically rely on continuous time series, homogeneous observations, and well-constrained parameters.
To address these constraints, we formalized current knowledge and competing hypotheses on climate–ecosystem linkages into a qualitative, possibilistic dynamic model. This approach was designed to (i) accommodate incomplete and heterogeneous evidence, (ii) make causal assumptions explicit, and (iii) systematically explore the range of environmental trajectories compatible with those assumptions. We implemented this knowledge base within the EDEN (Ecological Discrete Event Network) framework as a set of formal rules « if-then » describing interactions among climate-related variables, vegetation states, and surface-water availability.
It then reconstructed the resulting transition graph linking successive states through admissible event sequences. The resulting scenario ensemble provided a structured view of which combinations of climate, vegetation, and surface-water availability corresponded to feasible system states, which transitions were enabled or disabled by the rule base and its causal constraints, and where key uncertainties in rule definition and variable discretization most strongly affected the inferred habitat dynamics.
How to cite: Angeli, A., Ramstein, G., Fluteau, F., Tan, N., Barboni, D., and Gaucherel, C.: Environmental scenarios for hominin habitats in East Africa during the Plio-Pleistocene (4-1 Ma), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19420, https://doi.org/10.5194/egusphere-egu26-19420, 2026.
The dispersal of Homo sapiens out of Africa represents a key transition in human prehistory, yet the timing, routes, and environmental mechanisms underlying the expansion into East Asia remain debated. Fossil and archaeological evidence suggests the presence of anatomically modern humans in southern China by at least ~80 ka, but the relative importance of different migration corridors is still unresolved.
Three potential dispersal pathways into East Asia have been proposed: a southern coastal route through South and Southeast Asia, a northern inland route via Central Asia and southern Siberia, and a more speculative interior route through the Tarim Basin. While these routes have been widely discussed, most previous studies remain qualitative, and quantitative assessments of how Late Pleistocene climate variability shaped human existence potential and migration pathways are limited.
Here, we apply a Human Dispersal Model (HDM) based on the Human Existence Potential (HEP) framework to explore climate-based constraints on human migration into East Asia between 80 and 30 ka. Palaeoclimate simulations and archaeological site data are combined using a logistic regression approach to estimate HEP through time. The resulting HEP fields are then used to drive dispersal simulations, allowing us to explore potential migration pathways and corridors under different climatic conditions. This modelling framework provides a quantitative perspective on how Late Pleistocene climate variability may have influenced human dispersal into East Asia.
How to cite: Liang, G. and Shao, Y.: Modelling climate-based human existence potential and dispersal of Homo sapiens into East Asia (80–30 ka), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3069, https://doi.org/10.5194/egusphere-egu26-3069, 2026.
The rising number of climate reconstructions of the past utilizing both proxy data and climate models enable further numerical model based research of human occupation and settlement patterns. In conjunction with statistical, numerical or machine learning powered spatial downscaling methods, spatial resolutions can be achieved that better fit with the scale of human-landscape interactions. This study gives an overview of combinations of archaeological site distribution data with paleoclimate reconstructions and additional environmental data by using the Human Existence Potential (HEP) model. The focus lies on European societies starting from the first hunter-gatherer occupation of modern humans in Europe roughly 50k years before present to the earliest farming societies that settled down around 7.5k years ago. The resulting potential fields allow occupation and settlement pattern analysis as well as further use for other applications within the “Our Way” Framework of population dynamics modeling. This framework includes an agent-based model for small scale dynamics and a density based model for continental scale dispersal.
How to cite: Wegener, C. and Shao, Y.: Occupation and settlement through time: Applying the human existence potential model to European societies, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6843, https://doi.org/10.5194/egusphere-egu26-6843, 2026.
Natural disturbances shape ecosystems by redistributing biomass, resources, and mortality across space and time. While the ecological effects of individual disturbance types (e.g. fire, floods, storms) are well studied, a globally consistent assessment of how multiple disturbance types combine into long-term disturbance regimes, and how these regimes relate to biodiversity patterns, is still lacking at the macroecological scale.
Previous work (Kropf et al. in prep) presented 'hazomes,' a novel classification system of the earth based on hazard profiles, which is distinct from existing frameworks such as climate zones that categorize earth according to average conditions. Building up on this, we utilize a disturbance score based on eight different natural hazards (including heavy precipitation, earth quakes, tropical cyclones, cold spells, heat stress, coastal and river floods, water deficit, and wildfires), and their frequency of occurrence at different intensities. Unlike other commonly used climate descriptors such as mean temperature and precipitation, this approach captures the historical disturbance regimes ecosystems have been exposed to, providing a complementary perspective on the environmental drivers of biodiversity.
By correlating the disturbance index with biodiversity indicators, such as species richness across taxa, we find biome-specific disturbance-biodiversity relationships. While climate is understood to be a key driver of global biodiversity patterns, our research implies disturbance regimes may be key to understanding biodiversity patterns within areas of similar climatic conditions. These findings highlight disturbance regimes as an underexplored dimension of biogeography and suggest that biodiversity patterns reflect long-term exposure to disturbance, not only to climate. As climate change increasingly alters the frequency and intensity of natural hazards, understanding how ecosystems have been shaped by historical disturbance regimes is critical for anticipating future biodiversity responses.
Kropf, C. M., Hülsen, S., Stalhandske, Z., Hantson, S., Ward, P. J., Wens, M., Peleg, N., Bresch, D. N., & Steinmann, C. B. (in prep). Hazomes: Earth’s natural multi-hazard terrestrial disturbance regimes. EarthArXiv. https://eartharxiv.org/repository/view/10580/
How to cite: Hülsen, S., Runge, K., Kropf, C., Bresch, D., and Dee, L.: Understanding multi-hazard disturbance regimes as macro-ecological drivers of biodiversity, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2986, https://doi.org/10.5194/egusphere-egu26-2986, 2026.
The impacts of both natural and human-induced climate change are evident across the pine (Pinus) dominated forests of the Northern Rocky Mountains, USA. Paleoecological records have been used to investigate climate driven vegetation change and the complexities of fire disturbance in these forests over the Holocene, providing important information to the development of forest management plans and fire suppression protocols.
Mountain pine beetle (Dendroctonus ponderosae) (MPB) outbreaks have also influenced ecosystem change in the Northern Rocky Mountains. However, little is known about the occurrence of MPB outbreaks beyond the historic time period (past 200 years). Without direct evidence (fossil beetle remains) to identify MPB outbreaks in paleoecological records, it has been challenging to identify the timing and frequency of outbreaks over a longer time period (Holocene) and to demonstrate how unusual the patterns of the past 200 years are. The increase in MPB outbreaks in the historic time period has been attributed to increasing temperatures affecting the reproductive cycle of the beetles and the weakening of the defense mechanisms in pine species. Understanding the frequency of MPB outbreaks and forest response over the Holocene is helping land managers better plan for the future management strategies of these iconic landscapes.
This project used both traditional paleoecological time series analysis and quantitative analysis (regression analysis and machine learning) to investigate the frequency and timing of MPB outbreaks over the Holocene and identify patterns related to climate change. The data presented are from a series of sites across an elevational and latitudinal gradient in the Northern Rocky Mountains, USA and includes periods of high resolution (every cm) paleoecological proxy data. Initial findings indicate a positive relationship between MPB outbreaks and pine (Pinus) dominance on the landscape and more frequent MPB outbreaks during the historic time period than during any other time throughout the Holocene.
How to cite: Watt, J., Codding, B., and Brunelle, A.: Using Regression Analysis and Machine Learning to Investigate the Connection Between Mountain Pine Beetle (Dendroctonus ponderosae) Outbreaks and Human-Induced Climate Change, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15812, https://doi.org/10.5194/egusphere-egu26-15812, 2026.
Global megafaunal populations experienced widespread decline and extinctions during the Late Quaternary period. Adverse climatic conditions during the Last Glacial Maximum (LGM), and the emergence and global spread of modern humans are widely considered as the primary drivers of megafaunal loss and the associated decline in global biodiversity. However, the relative contributions of climate change and human influence on the unprecedented Late Quaternary megafaunal extinctions remain unresolved, largely due to the scarcity of palaeoecological evidence. Here, we employ a new spatially explicit dynamical model (ICCP Global Mammal Model, IGMM) to simulate climate-driven changes in the distribution of about 2000 terrestrial mammal species (including humans), incorporating biotic interactions through predation and competition, across space and through time on a global scale. While adverse climatic conditions during the LGM, marked by dramatic changes in habitat suitability, created a favorable background for the megafaunal decline, our model reveals that the global spread of culturally advanced modern humans played a crucial role in exacerbating the population loss of iconic species including mammoths, mastodons, stegodons, and giant sloths, ultimately leading to their extinction during the Late Quaternary period.
How to cite: Venugopal, T., Timmermann, A., Ruan, J., Raia, P., Yun, K.-S., Zeller, E., Mohanty, S., Castiglione, S., and Girardi, G.: Coupled climate-human drivers of Late Quaternary megafaunal decline, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19819, https://doi.org/10.5194/egusphere-egu26-19819, 2026.
