Search

Significant progress in permafrost carbon science made over the past decades include the identification of vast permafrost carbon stocks, the development of new pan‐Arctic permafrost maps, an increase in terrestrial measurement sites for CO2 and methane fluxes, and important factors affecting carbon cycling, including vegetation changes, periods of soil freezing and thawing, wildfire, and other disturbance events. Process‐based modeling studies now include key elements of permafrost carbon cycling and advances in statistical modeling and inverse modeling enhance understanding of permafrost region C budgets. By combining existing data syntheses and model outputs, the permafrost region is likely a wetland methane source and small terrestrial ecosystem CO2 sink with lower net CO2 uptake toward higher latitudes, excluding wildfire emissions. For 2002–2014, the strongest CO2 sink was located in western Canada (median: −52 g C m−2 y−1) and smallest sinks in Alaska, Canadian tundra, and Siberian tundra (medians: −5 to −9 g C m−2 y−1). Eurasian regions had the largest median wetland methane fluxes (16–18 g CH4 m−2 y−1). Quantifying the regional scale carbon balance remains challenging because of high spatial and temporal variability and relatively low density of observations. More accurate permafrost region carbon fluxes require: (a) the development of better maps characterizing wetlands and dynamics of vegetation and disturbances, including abrupt permafrost thaw; (b) the establishment of new year‐round CO2 and methane flux sites in underrepresented areas; and (c) improved models that better represent important permafrost carbon cycle dynamics, including non‐growing season emissions and disturbance effects.

Significant progress in permafrost carbon science made over the past decades include the identification of vast permafrost carbon stocks, the development of new pan-Arctic permafrost maps, an increase in terrestrial measurement sites for CO2 and methane fluxes, and important factors affecting carbon cycling, including vegetation changes, periods of soil freezing and thawing, wildfire, and other disturbance events. Process-based modeling studies now include key elements of permafrost carbon cycling and advances in statistical modeling and inverse modeling enhance understanding of permafrost region C budgets. By combining existing data syntheses and model outputs, the permafrost region is likely a wetland methane source and small terrestrial ecosystem CO2 sink with lower net CO2 uptake toward higher latitudes, excluding wildfire emissions. For 2002–2014, the strongest CO2 sink was located in western Canada (median: −52 g C m−2 y−1) and smallest sinks in Alaska, Canadian tundra, and Siberian tundra (medians: −5 to −9 g C m−2 y−1). Eurasian regions had the largest median wetland methane fluxes (16–18 g CH4 m−2 y−1). Quantifying the regional scale carbon balance remains challenging because of high spatial and temporal variability and relatively low density of observations. More accurate permafrost region carbon fluxes require: (a) the development of better maps characterizing wetlands and dynamics of vegetation and disturbances, including abrupt permafrost thaw; (b) the establishment of new year-round CO2 and methane flux sites in underrepresented areas; and (c) improved models that better represent important permafrost carbon cycle dynamics, including non-growing season emissions and disturbance effects.

We have observed tree die-offs in a variety of regions in the world. Understanding the diverse causes of tree mortality requires exact information about which trees are dying and where. With the increased user-friendliness of drones and the availability of airborne imagery, high-resolution imagery of forests is becoming widely available. Delineating standing deadwood in such aerial imagery has become a classic segmentation task and several models with varying accuracy have been developed. However, these machine-learning based models are not generic and limited to specific image resolutions, sensor characteristics, geographic regions, and forest ecosystems. The reason for this lack of generality is that previous models have been trained using only datasets representative of specific regions and obtained from a single source. In this study, we obtain a diverse dataset spanning more than a dozen countries across continents and implement a single convolutional neural network (CNN) model that is able to cope with most forest ecosystems, varying image quality, and spatial resolutions.

Excess tree mortality in the wake of climate extremes has been observed globally. However, we still lack precise data on mortality at global scale to understand respective drivers and spatiotemporal dynamics. The Sentinel-2 satellite fleet, equipped with the MultiSpectral Instrument (MSI), covers the entire earth on average every five days at spatial resolutions ranging from 10 m to 60 m. Mapping tree mortality from Sentinel-2 globally in diverse ecosystems requires equally diverse reference data. Using globally distributed high-resolution aerial orthoimagery reference data and artificial intelligence methods, we can translate spectral signatures of remote sensing into deadwood. Specifically, in this study we show how to predict the share of standing deadwood for a 10 m pixel in a specific year. The method takes into account temporal patterns, spatial context, as well as all Sentinel-2 spectral bands. This will enable us to map tree mortality globally at a new level of precision.

The subarctic forest tundra transition zone is one of the most vulnerable ecological regions worldwide and susceptible to climate change. Forest changes could lead to biodiversity losses when tundra areas become colonized. However, the impact of complex landscapes with barriers and channels for seed dispersal is highly understudied. Hence, we investigated potential tree aboveground biomass (AGB) change in mountainous central Chukotka (Siberia) with the individual-based spatially explicit vegetation model Larix vegetation simulator (LAVESI). In a climate sensitivity study, we simulate forest dynamics until 3000 CE for Representative Concentration Pathways (RCPs) with and without hypothetical cooling after 2300 CE to twentieth-century levels. The current state and spatiotemporal dynamics of tree AGB are validated against field and satellite-derived data. Our results suggest densification of existing tree stands and a lagged forest expansion depending on the distance to the current tree line (~39 percent of the total study area, RCP 8.5) under all considered climate scenarios. In scenarios with cooling after 2300 CE, forests stopped expanding and then gradually retreated to their pre-twenty-first-century position (~10 percent, RCP 8.5). However, forest remnants remain in the colonized area, leaving an imprint of forests in former tundra areas, which will likely have an adverse impact on tundra biodiversity.

Future precipitation levels remain uncertain because climate models have struggled to reproduce observed variations in temperature-precipitation correlations. Our analyses of Holocene proxy-based temperature-precipitation correlations and hydrological sensitivities from 2,237 Northern Hemisphere extratropical pollen records reveal a significant latitudinal dependence and temporal variations among the early, middle, and late Holocene. These proxy-based variations are largely consistent with patterns obtained from transient climate simulations (TraCE21k). While high latitudes and subtropical monsoon areas show mainly stable positive correlations throughout the Holocene, the mid-latitude pattern is temporally and spatially more variable. In particular, we identified a reversal from positive to negative temperature-precipitation correlations in the eastern North American and European mid-latitudes from the early to mid-Holocene that mainly related to slowed down westerlies and a switch to moisture-limited convection under a warm climate. Our palaeoevidence of past temperature-precipitation correlation shifts identifies those regions where simulating past and future precipitation levels might be particularly challenging.

We compile a data set of forest surveys from expeditions to the northeast of the Russian Federation, in Krasnoyarsk Krai, the Republic of Sakha (Yakutia), and the Chukotka Autonomous Okrug (59-73°N, 97-169°E), performed between the years 2011 and 2021. The region is characterized by permafrost soils and forests dominated by larch (Larix gmelinii Rupr. and Larix cajanderi Mayr). Our data set consists of a plot database describing 226 georeferenced vegetation survey plots and a tree database with information about all the trees on these plots. The tree database, consisting of two tables with the same column names, contains information on the height, species, and vitality of 40289 trees. A subset of the trees was subject to a more detailed inventory, which recorded the stem diameter at base and at breast height, crown diameter, and height of the beginning of the crown. We recorded heights up to 28.5m (median 2.5m) and stand densities up to 120000 trees per hectare (median 1197ha-1), with both values tending to be higher in the more southerly areas. Observed taxa include Larix Mill., Pinus L., Picea A. Dietr., Abies Mill., Salix L., Betula L., Populus L., Alnus Mill., and Ulmus L. In this study, we present the forest inventory data aggregated per plot. Additionally, we connect the data with different remote sensing data products to find out how accurately forest structure can be predicted from such products. Allometries were calculated to obtain the diameter from height measurements for every species group. For Larix, the most frequent of 10 species groups, allometries depended also on the stand density, as denser stands are characterized by thinner trees, relative to height. The remote sensing products used to compare against the inventory data include climate, forest biomass, canopy height, and forest loss or disturbance. We find that the forest metrics measured in the field can only be reconstructed from the remote sensing data to a limited extent, as they depend on

Boreal forests cover over half of the global permafrost area and protect underlying permafrost. Boreal forest development, therefore, has an impact on permafrost evolution, especially under a warming climate. Forest disturbances and changing climate conditions cause vegetation shifts and potentially destabilize the carbon stored within the vegetation and permafrost. Disturbed permafrost-forest ecosystems can develop into a dry or swampy bush- or grasslands, shift toward broadleaf- or evergreen needleleaf-dominated forests, or recover to the pre-disturbance state. An increase in the number and intensity of fires, as well as intensified logging activities, could lead to a partial or complete ecosystem and permafrost degradation. We study the impact of forest disturbances (logging, surface, and canopy fires) on the thermal and hydrological permafrost conditions and ecosystem resilience. We use a dynamic multilayer canopy-permafrost model to simulate different scenarios at a study site in eastern Siberia. We implement expected mortality, defoliation, and ground surface changes and analyze the interplay between forest recovery and permafrost. We find that forest loss induces soil drying of up to 44%, leading to lower active layer thicknesses and abrupt or steady decline of a larch forest, depending on disturbance intensity. Only after surface fires, the most common disturbances, inducing low mortality rates, forests can recover and overpass pre-disturbance leaf area index values. We find that the trajectory of larch forests after surface fires is dependent on the precipitation conditions in the years after the disturbance. Dryer years can drastically change the direction of the larch forest development within the studied period.

The year 2021 set new records for wildfire extent in the Republic of Sakha (Yakutia) in eastern Siberia, Russia. Wildfire seasons in this unique region, characterized by its deciduous boreal forest and permafrost landforms, are becoming more intense. Some fires are threatening local communities, while their smoke covers vast stretches of land every summer, posing health risks to people even in the distance. At the same time, the larch trees of the eastern Siberian boreal forest stabilize the permafrost soils below as guardians of a continental-scale storage of terrestrial carbon. It is still largely unknown how the current trend of wildfire intensification will develop in the future, and how it will modify the structure of the boreal forests within the next decades to centuries. However, even though needed for a well-founded evaluation of long-term impacts of changing fire regimes, data on past trends of wildfire activity still remains scarce in eastern Siberia. Here, we present a new reconstruction of boreal fire and vegetation dynamics, spanning the last ca. 10.8 ka. Continuously analyzed macroscopic charcoal particles and a REVEALS-transformed pollen record from a sediment core from Lake Satagay (N 63.078, E 117.998) give insight into long-term trends and relationships between changes in fire regime and vegetation composition and coverage. The data indicates that modern larch-dominated forests co-exist with a lower severity fire regime, whereas early Holocene open larch-birch woodlands enabled increased charcoal accumulation and thus supported a higher severity fire regime. Considering the expected increase in tree mortality caused by wildfires and insect damage, likely to thin out currently denser tree stands, this fire-vegetation relationship suggests a potential upcoming positive feedback on intensifying fire regimes.

Abstract The Lena Delta in Siberia is the largest delta in the Arctic and as a snow-dominated ecosystem particularly vulnerable to climate change. Using the two decades of MODerate resolution Imaging Spectroradiometer satellite acquisitions, this study investigates interannual and spatial variability of snow-cover duration and summer vegetation vitality in the Lena Delta. We approximated snow by the application of the normalized difference snow index and vegetation greenness by the normalized difference vegetation index (NDVI). We consolidated the analyses by integrating reanalysis products on air temperature from 2001 to 2021, and air temperature, ground temperature, and the date of snow-melt from time-lapse camera (TLC) observations from the Samoylov observatory located in the central delta. We extracted spring snow-cover duration determined by a latitudinal gradient. The ‘regular year’ snow-melt is transgressing from mid-May to late May within a time window of 10 days across the delta. We calculated yearly deviations per grid cell for two defined regions, one for the delta, and one focusing on the central delta. We identified an ensemble of early snow-melt years from 2012 to 2014, with snow-melt already starting in early May, and two late snow-melt years in 2004 and 2017, with snow-melt starting in June. In the times of TLC recording, the years of early and late snow-melt were confirmed. In the three summers after early snow-melt, summer vegetation greenness showed neither positive nor negative deviations. Whereas, vegetation greenness was reduced in 2004 after late snow-melt together with the lowest June monthly air temperature of the time series record. Since 2005, vegetation greenness is rising, with maxima in 2018 and 2021. The NDVI rise since 2018 is preceded by up to 4 °C warmer than average June air temperature. The ongoing operation of satellite missions allows to monitor a wide range of land surface properties and process

The SiDroForest (Siberian drone-mapped forest inventory) data collection is an attempt to remedy the scarcity of forest structure data in the circumboreal region by providing adjusted and labeled tree-level and vegetation plot-level data for machine learning and upscaling purposes. We present datasets of vegetation composition and tree and plot level forest structure for two important vegetation transition zones in Siberia, Russia; the summergreen–evergreen transition zone in Central Yakutia and the tundra–taiga transition zone in Chukotka (NE Siberia). The SiDroForest data collection consists of four datasets that contain different complementary data types that together support in-depth analyses from different perspectives of Siberian Forest plot data for multi-purpose applications. i. Dataset 1 provides unmanned aerial vehicle (UAV)-borne data products covering the vegetation plots surveyed during fieldwork (Kruse et al., 2021, https://doi.org/10.1594/PANGAEA.933263). The dataset includes structure-from-motion (SfM) point clouds and red–green–blue (RGB) and red–green–near-infrared (RGN) orthomosaics. From the orthomosaics, point-cloud products were created such as the digital elevation model (DEM), canopy height model (CHM), digital surface model (DSM) and the digital terrain model (DTM). The point-cloud products provide information on the three-dimensional (3D) structure of the forest at each plot. ii. Dataset 2 contains spatial data in the form of point and polygon shapefiles of 872 individually labeled trees and shrubs that were recorded during fieldwork at the same vegetation plots (van Geffen et al., 2021c, https://doi.org/10.1594/PANGAEA.932821). The dataset contains information on tree height, crown diameter, and species type. These tree and shrub individually labeled point and polygon shapefiles were generated on top of the RGB UVA orthoimages. The individual tree information collected during the expedition such as tree height, crown diameter, and vitality