• Energy Research
• Restricted close
• GB close
• CA close
• Global Change Biology close

AbstractThe recent rise in atmospheric methane (CH4) concentrations accelerates climate change and offsets mitigation efforts. Although wetlands are the largest natural CH4 source, estimates of global wetland CH4 emissions vary widely among approaches taken by bottom‐up (BU) process‐based biogeochemical models and top‐down (TD) atmospheric inversion methods. Here, we integrate in situ measurements, multi‐model ensembles, and a machine learning upscaling product into the International Land Model Benchmarking system to examine the relationship between wetland CH4 emission estimates and model performance. We find that using better‐performing models identified by observational constraints reduces the spread of wetland CH4 emission estimates by 62% and 39% for BU‐ and TD‐based approaches, respectively. However, global BU and TD CH4 emission estimate discrepancies increased by about 15% (from 31 to 36 TgCH4 year−1) when the top 20% models were used, although we consider this result moderately uncertain given the unevenly distributed global observations. Our analyses demonstrate that model performance ranking is subject to benchmark selection due to large inter‐site variability, highlighting the importance of expanding coverage of benchmark sites to diverse environmental conditions. We encourage future development of wetland CH4 models to move beyond static benchmarking and focus on evaluating site‐specific and ecosystem‐specific variabilities inferred from observations.

Global climate models suggest enhanced warming of the tropical mid and upper troposphere, with larger temperature rise rates at higher elevations. Changes in fire activity are amongst the most significant ecological consequences of rising temperatures and changing hydrological properties in mountainous ecosystems, and there is a global evidence of increased fire activity with elevation. Whilst fire research has become popular in the tropical lowlands, much less is known of the tropical high Andean region (>2000 masl, from Colombia to Bolivia). This study examines fire trends in the high Andes for three ecosystems, the Puna, the Paramo and the Yungas, for the period 1982-2006. We pose three questions: (i) is there an increased fire response with elevation? (ii) does the El Niño- Southern Oscillation control fire activity in this region? (iii) are the observed fire trends human driven (e.g., human practices and their effects on fuel build-up) or climate driven? We did not find evidence of increased fire activity with elevation but, instead, a quasicyclic and synchronous fire response in Ecuador, Peru and Bolivia, suggesting the influence of high-frequency climate forcing on fire responses on a subcontinental scale, in the high Andes. ENSO variability did not show a significant relation to fire activity for these three countries, partly because ENSO variability did not significantly relate to precipitation extremes, although it strongly did to temperature extremes. Whilst ENSO did not individually lead the observed regional fire trends, our results suggest a climate influence on fire activity, mainly through a sawtooth pattern of precipitation (increased rainfall before fire-peak seasons (t-1) followed by drought spells and unusual low temperatures (t0), which is particularly common where fire is carried by low fuel loads (e.g., grasslands and fine fuel). This climatic sawtooth appeared as the main driver of fire trends, above local human influences and fuel build-up cyclicity.

AbstractBoreal wildfires are increasing in intensity, extent, and frequency, potentially intensifying carbon emissions and transitioning the region from a globally significant carbon sink to a source. The productive southern boreal forests of central Canada already experience relatively high frequencies of fire, and as such may serve as an analog of future carbon dynamics for more northern forests. Fire–carbon dynamics in southern boreal systems are relatively understudied, with limited investigation into the drivers of pre‐fire carbon stocks or subsequent combustion. As part of NASA's Arctic‐Boreal Vulnerability Experiment, we sampled 79 stands (47 burned, 32 unburned) throughout central Saskatchewan to characterize above‐ and belowground carbon stocks and combustion rates in relation to historical land use, vegetation characteristics, and geophysical attributes. We found southern boreal forests emitted an average of 3.3 ± 1.1 kg C/m2from field sites. The emissions from southern boreal stands varied as a function of stand age, fire weather conditions, ecozone, and soil moisture class. Sites affected by historical timber harvesting had greater combustion rates due to faster carbon stock recovery rates than sites recovering from wildfire events, indicating that different boreal forest land use practices can generate divergent carbon legacy effects. We estimate the 2015 fire season in Saskatchewan emitted a total of 36.3 ± 15.0 Tg C, emphasizing the importance of southern boreal fires for regional carbon budgets. Using the southern boreal as an analog, the northern boreal may undergo fundamental shifts in forest structure and carbon dynamics, becoming dominated by stands <70 years old that hold 2–7 kg C/m2less than current mature northern boreal stands. Our latitudinal approach reinforces previous studies showing that northern boreal stands are at a high risk of holding less carbon under changing disturbance conditions.

AbstractIt is well known that increased spring temperatures cause earlier onset dates of leaf unfolding and flowering. However, a temperature increase in winter may be associated with delayed development when species' chilling requirements are not fulfilled. Furthermore, photosensitivity is supposed to interfere with temperature triggers. To date, neither the relative importance nor possible interactions of these three factors have been elucidated. In this study, we present a multispecies climate chamber experiment to test the effects of chilling and photoperiod on the spring phenology of 36 woody species. Several hypotheses regarding their variation with species traits (successional strategy, floristic status, climate of their native range) were tested. Long photoperiods advanced budburst for one‐third of the studied species, but magnitudes of these effects were generally minor. In contrast to prior hypotheses, photosensitive responses were not restricted to climax or oceanic species. Increased chilling length advanced budburst for almost all species; its effect greatly exceeding that of photoperiod. Moreover, we suggest that photosensitivity and chilling effects have to be rigorously disentangled, as the response to photoperiod was restricted to individuals that had not been fully chilled. The results indicate that temperature requirements and successional strategy are linked, with climax species having higher chilling and forcing requirements than pioneer species. Temperature requirements of invasive species closely matched those of native species, suggesting that high phenological concordance is a prerequisite for successful establishment. Lack of chilling not only led to a considerable delay in budburst but also caused substantial changes in the chronological order of species' budburst. The results reveal that increased winter temperatures might impact forest ecosystems more than formerly assumed. Species with lower chilling requirements, such as pioneer or invasive species, might profit from warming winters, if late spring frost events would in parallel occur earlier.

AbstractExtreme weather events can have strong negative impacts on species survival and community structure when surpassing lethal thresholds. Extreme, short‐lived, winter warming events in the Arctic rapidly melt snow and expose ecosystems to unseasonably warm air (for instance, 2–10 °C for 2–14 days) but upon return to normal winter climate exposes the ecosystem to much colder temperatures due to the loss of insulating snow. Single events have been shown to reduce plant reproduction and increase shoot mortality, but impacts of multiple events are little understood as are the broader impacts on community structure, growth, carbon balance, and nutrient cycling. To address these issues, we simulated week‐long extreme winter warming events – using infrared heating lamps and soil warming cables – for 3 consecutive years in a sub‐Arctic heathland dominated by the dwarf shrubsEmpetrum hermaphroditum, Vaccinium vitis‐idaea(both evergreen) andVaccinium myrtillus(deciduous). During the growing seasons after the second and third winter event, spring bud burst was delayed by up to a week forE. hermaphroditumandV. myrtillus, and berry production reduced by 11–75% and 52–95% forE. hermaphroditumandV. myrtillus, respectively. Greater shoot mortality occurred inE. hermaphroditum(up to 52%),V. vitis‐idaea(51%), andV. myrtillus(80%). Root growth was reduced by more than 25% but soil nutrient availability remained unaffected. Gross primary productivity was reduced by more than 50% in the summer following the third simulation. Overall, the extent of damage was considerable, and critically plant responses were opposite in direction to the increased growth seen in long‐term summer warming simulations and the ‘greening’ seen for some arctic regions. Given the Arctic is warming more in winter than summer, and extreme events are predicted to become more frequent, this generates large uncertainty in our current understanding of arctic ecosystem responses to climate change.

AbstractThe 20th century was a pivotal period at high northern latitudes as it marked the onset of rapid climatic warming brought on by major anthropogenic changes in global atmospheric composition. In parallel, Arctic sea ice extent has been decreasing over the period of available satellite data records. Here, we document how these changes influenced vegetation productivity in adjacent eastern boreal North America. To do this, we used normalized difference vegetation index (NDVI) data, model simulations of net primary productivity (NPP) and tree‐ring width measurements covering the last 300 years. Climatic and proxy‐climatic data sets were used to explore the relationships between vegetation productivity and Arctic sea ice concentration and extent, and temperatures. Results indicate that an unusually large number of black spruce (Picea mariana) trees entered into a period of growth decline during the late‐20th century (62% of sampled trees; n = 724 cross sections of age >70 years). This finding is coherent with evidence encoded in NDVI and simulated NPP data. Analyses of climatic and vegetation productivity relationships indicate that the influence of recent climatic changes in the studied forests has been via the enhanced moisture stress (i.e. greater water demands) and autotrophic respiration amplified by the declining sea ice concentration in Hudson Bay and Hudson Strait. The recent decline strongly contrasts with other growth reduction events that occurred during the 19th century, which were associated with cooling and high sea ice severity. The recent decline of vegetation productivity is the first one to occur under circumstances related to excess heat in a 300‐year period, and further culminates with an intensifying wildfire regime in the region. Our results concur with observations from other forest ecosystems about intensifying temperature‐driven drought stress and tree mortality with ongoing climatic changes.