High-resolution UAV-based thermal imagery offers a non-invasive approach to assess canopy thermal dynamics by capturing canopy surface temperature, which integrates radiative loading, aerodynamic coupling, and transpiration-driven evaporative cooling. In this study, repeated UAV-based thermal surveys were conducted throughout the 2023 growing season in a boreal jack pine stand to quantify canopy-air temperature differences (Tc - Ta) across two contrasting microsites. Data were collected between July and September 2023 over four sampling days, comprising a total of 14 flights. Using this dataset, we examine microsite influence on canopy surface temperature, compare canopy and proximal canopy environment temperatures, and evaluate relationships between Tc - Ta, sap flux density, and meteorological drivers. Our findings reveal that microsite differences were most pronounced during mid-summer as Tc - Ta was significantly higher at the drier site during late morning and midday. Elevated midday Tc - Ta values were consistent with reduced evaporative cooling under increased vapor pressure deficit during early summer. A marginally significant positive relationship was also observed between Tc - Ta and solar radiation. Canopy surface temperatures remained closely coupled to proximal canopy surface temperature, with differences predominantly within ±1 °C across the observed range of air temperatures. These results demonstrate that UAV thermal datasets, as the one presented here, may be effective in capturing spatial and diurnal variability in canopy temperature and in detecting signatures of radiative heating. However, their effectiveness in isolating specific physiological processes must be enhanced through refined acquisition strategies recommended in this study.

High-resolution climate reconstructions are needed to understand past and future climate-vegetation dynamics. However, climate reconstructions using ring width are scarce in temperate regions where climate has limited impacts on tree growth. In Temiscamingue, white pine sunken logs represent valuable archives for long-term climate reconstructions, but white pine's response to climate variations in this temperate region is unknown. Moreover, the most suitable proxies for climate reconstruction in white pine-ring widths or stable isotopes-remain unclear. This study evaluates which proxy, or combination of proxies, best captures climate signals in white pine radial growth across contrasting soil drainage conditions. Ring widths were measured in 60 trees from Opemican National Park (30 on each of fast and moderate drainage conditions). Carbon and oxygen isotope ratios were analyzed from five trees per site over the period 1950-2020. Climatic data from ERA5 were used to calibrate response functions. June-August drought index (SPEI) as well as June-July precipitation and temperature were the dominant climatic drivers on all proxies, regardless of drainage type. Bayesian linear models revealed that combining carbon isotopes with ring-width data provided the most robust reconstruction of June-July precipitation (r = 0.43), that combining oxygen isotopes with ring widths yielded the best reconstruction of June-July temperatures (r = 0.49), and that combining carbon and oxygen isotopes produced the best reconstruction of June-August SPEI (r = 0.45). This study is among the first to demonstrate the potential of a multi-proxy approach for climate reconstruction using white pine. Our results provide insights that may guide future climate reconstructions in temperate regions, where tree species with lower climate sensitivity could also be considered.

The collapse of the Dzungar Khanate in the 1750s marked the end of the world's last nomadic empire and reshaped Eurasian geopolitics. However, whether environmental and societal factors interplayed and contributed to its collapse remains unclear. Here, we reconstructed steppe productivity variability across the Dzungar heartland since 1625 using 441 tree-ring series based on an extreme growth ratio. Our record shows that the Khanate's collapse coincided with the lowest steppe productivity in the past four centuries during 1751-1761. The weakening societal resilience approaching a tipping point intensified the effects of the rapid steppe productivity decline, which cascaded into famine, conflict, and trade breakdown. Smallpox then reemerged and escalated into a catastrophic epidemic, transforming the environmental crisis into the collapse of the Dzungar Khanate. Our findings underscore the urgent need to strengthen pastoral resilience and epidemic surveillance in steppe regions as climate stress intensifies and cross-border epidemic threats escalate.

Process-based ecophysiological forest models are essential for understanding the mechanisms underlying forest responses to climate change. The MAIDENiso ecophysiological model provides an excellent framework for systematically analyzing photosynthetic production, carbon allocation strategies, and stable isotope signals in forests. However, the model involves numerous parameters that are challenging to derive from direct observations, which makes calibration especially difficult. In this study, we developed an improved genetic algorithm (GA) and combined it with the Markov chain Monte Carlo (MCMC) method, proposing a novel GA-based parameter calibration approach (MCMC-GA). Results from algorithm evaluations at two sites with distinct climatic conditions in China (PMC and HLS) demonstrated that MCMC-GA substantially improved calibration accuracy and efficiency-the weighted RMSE at convergence decreased by an average of 23% across the two sites compared with the classical GA, while calibration efficiency (the number of algorithm iterations to reach a target RMSE) increased by 65%. Based on the application of the novel GA-based approach to parameterize MAIDENiso, tree-ring widths and stable carbon isotopes (delta 13CTRC) at both sites were successfully simulated. Overall, the improvements of the GA and its integration with MCMC significantly enhanced the calibration efficiency and accuracy of process-based forest models, providing a valuable tool for parameterizing other Earth system models and offering promising opportunities for advancing forest ecological research.

In northeastern Canada, lakeshore trees are of particular interest because they eventually become lake subfossils, which serve as main archives of past climate fluctuations over the last millennium. However, the extent to which these lakeshore trees, growing at the aquatic-terrestrial interface, carry stable isotopic series representative of regional mesic forests remains unclear. The main objective of this study is to determine whether carbon (C) and oxygen (O) stable isotopes (δ13C and δ18O) and ring width indexes (RWI) series from black spruce trees [Picea mariana, (Mill.) BSP.] growing along boreal lakeshores exhibit similar variations to those from trees growing farther away from the shores (upland trees). To verify this similarity, we compared inter-annual δ13C and δ18O variations from both environments, in boreal Quebec (eastern Canada), over the 1940–2015 period. Chronologies of intracellular CO2 concentration (ci) and intrinsic water use efficiency (iWUE) were derived from the δ13C series. Our results suggest that fractionation of C and O isotopes, in both sites, is affected jointly by rising maximum temperatures (Tmax) and Vapor Pressure Deficit (VPD) during summer, controlling gas exchanges at the leaf level. By contrast to average δ18O whose average values remained stationary through time, the δ13C records presented evident changes in acclimation strategies to rising CO2, with an early phase of strong stomatal regulation followed by a more moderate response after about 1975. Overall, the conformity of isotopic series suggests that black spruce, regardless of topographic position, can act as a reliable recorder of regional scale, past temperatures, atmospheric dryness and ecophysiological adaptation of boreal forest to climate variability.

Summary Process-based tree growth models provide a mechanistic framework for investigating how climate conditions regulate tree growth across daily to annual time scales. Yet, their broader application across species and environments is constrained by the limited accessibility in open-source environments and the difficulty of estimating physiological parameters that are rarely measured directly. Here, we present virtualRings, a new R package integrating the Vaganov-Shashkin model (VSM) and the RINGS3 models, and focus on the implementation and calibration of VSM. Using tree-ring width observations from seven Northern Hemisphere sites across various environmental conditions, we compared the traditional bootstrap-based calibration approach with the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). CMA-ES improved agreement between simulated and observed radial tree growth and provided an efficient approach for model parameter estimation. We further evaluated practical CMA-ES settings to balance computational cost and performance and discussed its potential limitations. The virtualRings package provides an open and reproducible platform for tree growth simulation, facilitating the application of important process-based models across species and environments and the investigation of how temperature and moisture constraints regulate daily tree-ring formation across spatial and temporal scales.

Modeling river ice jam risks under present and future climate is challenging task due to the sudden nature of ice jams. In this study, we develop a hybrid machine-learning model to predict the daily probability of ice jam occurrence under current climate and use it to project ice jam risks under future climates. A random forest ensemble simulation with two components (temporal and spatial) was calibrated and validated using 108 historical (2005-2022) ice jam events recorded within 882 river reaches along the Chaudière and Beaurivage rivers in Eastern Canada, where ice jams cause frequent damages to nearby populations. The model integrates both hydrometeorological predictors known to influence ice resistance (e.g., freezing degree days as a proxy for ice thickness) and dislocation forces (melting degree days, rainfall, changes in streamflow), as well as channel morphological conditions prone to ice congestion (e.g., sinuosity, channel width, slope breaks). Predictor importance analysis shows that ice jam occurrence is preconditioned by cold, snowy winter conditions and rapidly rising streamflow in winter and/or spring. An increase in sinuosity and a reduction in the width of the river channels are the main causes of ice congestion. Daily times series from a bias-adjusted ensemble of dynamically downscaled climate simulations were then used to force the model into future climate under two representative concentration pathways (RCP4.5 and RCP8.5). Ice jam occurrence is projected to decrease overall in the future compared with the historical reference period, both for the RCP4.5 and RCP8.5 trajectories. A seasonal shift is also projected for the future, with ice jam events becoming more concentrated during February- March, rarer in April, and practically absent in December-January, especially for the 2080-2100 horizon. This research represents a significant advance in ice jam prediction, combining spatial and temporal approaches.

Exhaustive long-term and large-scale ice jam records are scarce in most cold river environments. Many discrete events occur in small, sparsely populated river systems and are poorly represented in open-source databases. These observation biases are transferred to predictive models of ice jams and the collective understanding of their formation mechanisms. This study addresses these observation biases by using land use as a proxy for ice jam observation probability and by combining direct human observations with dendrochronological records of ice jam activity. The probability of observing an ice jam directly by a witness or indirectly by a tree-ring dated tree scar increases with the density of urban and forest cover, respectively. The annual probability of occurrence for ice jams calculated from direct observational or dendrochronological records alone correlated poorly with geomorphological factors known to cause ice jams. Correcting the observational biases in individual records with their respective land use densities improved the correlation with the Ice jam Predisposition Index (IJPI), a spatial predictor of the probability of occurrence for ice jams. Correcting the observation bias with land use and combining multisource data (direct and dendrochronological observations) further improved the correlation between ice jams and the IJPI. A multi-source approach thus partly overcomes the observation bias of individual records. This work highlights the potential impacts of observation biases in direct and indirect (dendrochronological) ice jam records and shows that bias-corrected, multisource ice jam records could benefit the calibration and validation of ice jam prediction model.

Xylogenesis, the process through which wood cells are formed, results in the long-term storage of carbon in woody biomass, making it a key component of the global carbon cycle. Understanding how environmental drivers influence xylogenesis during the growing season is therefore of great interest. However, studying short-term drivers of wood production using xylogenetic data is complicated by the usual sampling scheme and the influence of eccentric growth, i.e., heterogeneous growth around the stem. In this study, we improve xylogenesis research by introducing a statistical approach that explicitly considers seasonal phenology, short-term growth rates, and growth eccentricity. To this end, we developed Bayesian models of xylogenesis and compared them with a conventional method based on the use of Gompertz functions. Our results show that eccentricity generated high temporal autocorrelation between successive samples, and that explicitly taking it into account improved both the representativeness of phenology and intra-ring variability. We observed consistent short-term patterns in the model residuals, suggesting the influence of an unaccounted-for environmental variable on cell production. The proposed models offer several advantages over traditional methods, including robust confidence intervals around predictions, consistency with phenology, and reduced sensitivity to extreme observations at the end of the growing season, often linked to eccentric growth. These models also provide a benchmark for mechanistic testing of short-term drivers of wood formation.

Peatlands play a crucial role in carbon storage and climate regulation. Traditional gravity-based and loss-on-ignition methods have been widely used to acquire bulk density and thus organic carbon estimates in peat sequences. However, these methods are time-consuming, and the measurement resolution frequently ranges from half to a few centimetres, hampering the understanding of carbon accumulation history at finer temporal resolution. Here, we explore the potential of non-destructive X-ray computed tomography (XCT), a method for analyzing 3D material structure and mass density, for obtaining proxy measurements for bulk density parameters using peat cores collected in eastern boreal Quebec, Canada. We find that the Hounsfield Unit (HU) of medical XCT scans is a robust surrogate for the bulk density of wet peat (BDwet). A universal linear model can be applied to calibrate HU values for a wide range of peat stratigraphy from different microforms: Sphagnum hummock, lichen hummock, lawn, and hollow. Moreover, HU of dry peat is indicative of both dry and organic matter bulk density (BDdry and BDom). It is possible to develop case-specific logarithmic models to calibrate HU with BDdry. In addition, the precise measurement of the peat sample volumes using XCT suggests that traditional methods can be subject to substantial uncertainties when estimating bulk density and carbon content. Medical XCT can be applied to quantify bulk density in peat soils in a more time-efficient manner, with a resolution up to 0.6 mm, approximately equivalent to the yearly accumulation rate.

Climate change impacts river ice phenomena, influencing ice jam regimes. The complex formation of ice jams, which depends on a combination of morphological and hydrometeorological conditions, varies interannually and between watersheds. This research aims to identify the hydroclimatic conditions, associated with the balance between resisting and dislocating forces exerted on ice, that influence the spatiotemporal variability of seasonal ice jams in ten watersheds in southern Quebec, Canada. Generalized additive mixed-effects models were used to quantify the effect of potential hydroclimatic predictors on winter and spring number of ice jams at the watershed scale during the 2005–2018 period. Winter ice jams were not frequent (n = 114) and occurred only in a few locations. A higher number of winter ice jams was found to be associated with extreme seasonal hydroclimatic conditions, including higher peak rainfall or streamflow that dislocated the ice, while the role of snowfall remains uncertain in winter. The spring breakup favors a higher number of ice jams (n = 219), particularly in conjunction with high peak flows and sustained rainfall. The number of spring ice jams also increased in response to higher antecedent winter snowfall, hinting at the importance of white ice as a mechanism for the growth of river ice cover. In spring, hydroclimatic predictors alone offered greater explanatory potential (deviance explained = 34.5

Ongoing climate change is increasing vegetation flammability, intensifying fire activity in the boreal forests of eastern North America. This situation suggests a potential tipping point in fire regimes, raising critical questions about their impact on the biodiversity and structure of these ecosystems. To gain a deeper understanding of landscape dynamics and ongoing environmental changes, it is essential to understand how this climate, vegetation and fire linkages operate across various temporal and spatial scales. By integrating paleo-datasets (charcoal, pollen, chironomids, and testate amoebae) with model simulations of vapor pressure deficit (VPD) and plant-available soil water (ASW) over the past 8,000 years, we show that drier spring conditions over the last 3,000 years led to fewer but larger and more severe fire episodes, peaking within the last 250 years. This shift in fire regimes promoted an increase in fire-adapted conifer species, particularly Pinus banksiana, across the landscape. These findings challenge previous projections of increased dominance by thermophilous species under climate change scenarios and instead suggest an expansion of pyrophilous vegetation. Such ecological transitions are set to drive significant environmental and socio-economic consequences.

Pinus banksiana Lamb. exhibits remarkable ecological adaptability, thriving across diverse environments in the Canadian boreal zone, including clay deposits, fast-draining glacial tills and rocky outcrops. However, projected rising temperature and increasing vapor pressure deficit (VPD) could increase the species' vulnerability, particularly in dry regions. In this study, we measured basal area increment (BAI) and physiological responses from isotopic fractionation across a soil gradient including three sites in the boreal mixed wood of western Quebec, Canada. The sites were a clay-rich soil (CLY, a humid site), an esker base (ESB, an intermediate site) and an esker top (EST, a sandy, well drained, dry site). Using tree-ring analysis and dual stable isotopes (δ13C and δ18O), we evaluated intrinsic water-use efficiency (iWUE) and leaf water enrichment (Δ18Olw). Our results revealed a significant correlation between Δ18Olw and VPD, indicating that stomatal regulation is the crucial physiological mechanism controlling P. banksiana's response to environmental stress across the sites. This effect was most pronounced at the dry EST site, where higher iWUE and less negative δ13C values suggest greater stomatal limitation of CO2 uptake. Increased iWUE was associated with enhanced BAI in the humid CLY site and a negative iWUE-BAI relationship emerged at EST, suggesting carbon assimilation constraints under drier conditions. Our results reveal a physiological trade-off in P. banksiana across a soil moisture gradient, demonstrating that rising atmospheric demand may decouple water-use efficiency from growth in drier environments like the EST site. By integrating isotopic signatures with growth dynamics, our study identifies a potential ecological tipping point beyond which increased iWUE may no longer sustain carbon gain under intensifying climate stress.

The stable isotopic composition of carbon (S13C) and oxygen (S18O) in tree rings is widely used to explore tree eco-physiological dynamics across various time scales. However, interpreting these isotopic signals is challenging due to multiple interacting factors, including gas exchange at the leaf level, stored carbohydrate reserves, and xylem water, whose timing and interactions during the growing season remain poorly understood. In this study, weekly S13C and S18O signals were tracked within the cambial region and forming xylem of black spruce (Picea mariana (Mill.) BSP.) in boreal forests of Quebec, Canada. The study covered three consecutive growing seasons (2019-2021) at two forest sites with differing temperature and soil water content. Weekly isotopic profiles were developed for the cambial region (S13Ccam and S18Ocam) and developing xylem cellulose (S13Cxc and S18Oxc). Strong positive correlations were observed between S13Ccam and S18Ocam, with an increasing trend along the growing season. Conversely, negative relationships were observed between S13Cxc and S18Oxc, characterized by an increasing trend in S13Cxc and a decreasing trend in S18Oxc. The results illustrated that stomatal conductance is the dominant physiological factor controlling seasonal fractionation of S13Ccam and S18Ocam. Increasing proportional exchanges between xylem water and sugars at the sites of cellulose synthesis (i.e., Pex effect) are thought to be strong enough to completely blur the observed trends in S18Ocam during the growing season. This suggests that S18Oxc signals differ from those originating in the earlier cambium sink. These findings highlight the need to carefully consider the processes influencing isotopic signals to avoid misinterpretations in dendroclimatological studies.

The combined contribution of CO2 fertilization and climate variability to arid and semi-arid forest growth remains unclear. To disentangle these multiple influences, we used a preexisting process-based ecophysiological model (MAIDENiso) to simulate tree growth changes during 1956-2010 in arid and semi-arid regions of China. Results revealed that simulated tree growth was more dependent on climate trends than on atmospheric CO2 concentration. Mechanistic analysis showed that the regulation of stomatal conductance under water stress positively affected tree growth in the arid region, but had an opposite pattern in the semi-arid region. Intrinsic water-use efficiency (iWUE, measured from tree-ring delta C-13) has increased by 29% and 44% since 1900 CE in the arid and semi-arid regions, respectively, but did not stimulate radial tree growth. This suggests that there will possibly be a continued increase (decrease) of radial forest growth in arid (semi-arid) areas of northern China if the current climate trends remain in the next decades.

The Asian Summer Monsoon (ASM) is a crucial driver of precipitation, sustaining ecological balance and socioeconomic development in North China. However, the extent to which climate change has influenced this monsoonal system, leading to detectable and attributable modifications in precipitation regimes, remains unclear. Here, we present a robust annual precipitation reconstruction spanning 1770-2020, using delta 18O from tree ring cellulose and a simple linear regression model in the North China Monsoon Marginal Region (NCMMR). Reconstructed precipitation and independent hydroclimatic records reveal a pronounced drying trend across the NCMMR since the 1950s. Multiple linear regression modelling, water vapor transport analyses using ensemble means from the Community Earth System Model-Last Millennium Ensemble, and correlation analysis indicate that precipitation variability in the NCMMR is modulated by the Indian Ocean Dipole, El Nino-Southern Oscillation, Atlantic Multidecadal Oscillation, and Interdecadal Pacific Oscillation. Nevertheless, fingerprint analysis suggests that the observed precipitation decline since the 1950s is strongly associated with greenhouse gas concentrations, albeit partially offset by the effects of anthropogenic aerosol emissions and internal variability. The impact of greenhouse gas forcing on precipitation variability is expected to intensify in the coming decades.

In eastern Canada, Black spruce ( Picea mariana Mill. B.S.P.) grows in a wide variety of climates, from maritime-oceanic conditions near the Labrador Sea, to more continental climates, inland. Along this gradient, timing and provenance of heat and moisture that support growth are uncertain, weakening our capacity to predict the response of boreal ecosystems to climate variability. Here, we measured the stable oxygen isotopic composition of black spruce tree-ring cellulose at three sites in eastern Canada and provide evidence of a rapid decrease of Labrador Sea’s influence on adjacent ecosystems. Our results report a landwards decrease in the oxygen isotope composition of both tree-ring cellulose ( δ 18 O T R C ) and precipitation water ( δ 18 O p ). We also reveal a rapid landwards decoupling between δ 18 O T R C variability (1950-2013), maximum temperature and Sea Surface Temperature variations over the Northwest Atlantic. Thus, despite their apparent ecological homogeneity, eastern Canada’s black spruce ecosystems rely on heterogeneous sources of heat and moisture.

Intra-annual variations of carbon stable isotope ratios (delta C-13) in different tree compartments could represent valuable indicators of plant carbon source-sink dynamics, at weekly time scale. Despite this significance, the absence of a methodological framework for tracking delta C-13 values in tree rings persists due to the complexity of tree ring development. To fill this knowledge gap, we developed a method to monitor weekly variability of delta C-13 in the cambium-xylem continuum of black spruce species [Picea mariana (Mill.) BSP.] during the growing season. We collected and isolated the weekly incremental growth of the cambial region and the developing tree ring from five mature spruce trees over three consecutive growing seasons (2019-21) in Simoncouche and two growing seasons (2020-21) in Bernatchez, both located in the boreal forest of Quebec, Canada. Our method allowed for the creation of intra-annual delta C-13 series for both the growing cambium (delta C-13(cam)) and developing xylem cellulose (delta C-13(xc)) in these two sites. Strong positive correlations were observed between delta C-13(cam) and delta C-13(xc) series in almost all study years. These findings suggest that a constant supply of fresh assimilates to the cambium-xylem continuum may be the dominant process feeding secondary growth in the two study sites. On the other hand, rates of carbon isotopic fractionation appeared to be poorly affected by climate variability, at an inter-weekly time scale. Hence, increasing delta C-13(cam) and delta C-13(xc) trends highlighted here possibly indicate shifts in carbon allocation strategies, likely fostering frost resistance and reducing water uptake in the late growth season. Additionally, these trends may be related to the black spruce trees' responses to the seasonal decrease in photosynthetically active radiation. Our findings provide new insights into the seasonal carbon dynamics and growth constraints of black spruce in boreal forest ecosystems, offering a novel methodological approach for studying carbon allocation at fine temporal scales.

Summer cooling is one of the most direct consequences of explosive volcanic eruptions that can affect ecosystems and human societies. Recent studies revealed a multiyear cooling impact on hemispheric and global summer temperatures after tropical eruptions, yet the volcanic responses appear to vary on regional scales. Here, we revisit volcano-induced summer cooling in eastern Canada and northern and central Europe by applying superposed epoch analy-sis on CMIP6-PMIP4 simulations and millennial temperature reconstructions based on tree-ring density. We then examine potential causes modulating region-specific volcanic impact. While confirming that, on average, tropical eruptions over the last millennium have induced a longer cooling (.4 yr) than eruptions from extratropical Northern Hemisphere in all three North Atlantic regions, we show that the peak magnitude of cooling is stronger in eastern Canada. We also find that the de-tected volcanic temperature anomalies can be strongly affected by the selection and number of volcanic events and nonvol-canic signals embedded in the climate time series. This study highlights the risks of using highly noisy proxy records to investigate volcanic impacts, especially in regions with strong unforced climate variability. The CMIP6-PMIP4 simulations generally agree with the three reconstructions on the average response to tropical eruptions, but their performance is poorer regarding the production of significant cooling after extratropical eruptions. Our results further suggest that the particular sensitivity to tropical eruptions in eastern Canada is likely related to increased sea ice surrounding Quebec- Labrador associated with the positive Arctic Oscillation and North Atlantic Oscillation formed during the first post -eruption winter.

3Department of Geography, Geomatics and Environment, University of Toronto Mississauga, Canada 4Université du Québec à Montréal, GEOTOP, Canada 5Université du Québec à Montréal, Centre d’études sur la forêt, Canada 6University of British Columbia, Department of Forest and Conservation Sciences, Canada 7Indiana State University, Department of Earth and Environmental Systems, Canada 8University of Alberta Augustana, Camrose, Alberta, Canada