Description: Northern lakes are considered a major source of atmospheric methane (CH4), a potent GHG1,2. However, large uncertainties in their emissions (7–26 Tg CH4 yr–1; ref. 2) arise from challenges in upscaling field data, including fluxes by ebullition (bubbling), the dominant emission pathway2. Remote sensing of ebullition would allow detailed mapping of regional emissions but has hitherto not been developed. Here, we show that lake ebullition can be imaged using synthetic aperture radar remote sensing during ice-cover periods by exploiting the effect of ebullition on the texture of the ice–water interface. Applying this method to five Alaska regions and combining spatial remote sensing information with year-round bubble-trap flux measurements, we create ebullition-flux maps for 5,143 Alaskan lakes. Regional lake CH4 emissions, based on satellite remote sensing analyses, were lower compared to previous estimates based on upscaling from individual lakes2,3 and were consistent with independent airborne CH4 observations. Thermokarst lakes formed by thaw of organic-rich permafrost had the highest fluxes, although lake density and lake size distributions also controlled regional emissions. This new remote sensing approach offers an opportunity to improve knowledge about Arctic CH4 fluxes and helps to explain long-standing discrepancies between estimates of CH4 emissions from atmospheric measurements and data upscaled from individual lakes.

Description: Waterbodies in the arctic permafrost zone are considered a major source of the greenhouse gas methane (CH4). However, the spatio-temporal variability of CH4 fluxes from waterbodies complicates spatial extrapolation of CH4 measurements at individual waterbodies. Therefore, the contribution of CH4 emissions from different waterbody types to the CH4 budget of the arctic permafrost zone has not yet been well constrained. To approach this problem, our study aimed i) at understanding if there are correlations between waterbodies and CH4 fluxes on a larger spatial extent containing several waterbodies and ii) at quantifying the influence of the spatial resolution of CH4 flux data on potential relations. Our two study areas of 1000 km² each are located in the northern and central part of the Mackenzie Delta, arctic Canada. We classified the waterbodies using maps from the circum-arctic Permafrost Region Pond and Lake Database (PeRL) based on TerraSAR-X data with a spatial resolution of 2.5 m x 2.5 m. We used the backscatter signals of Sentinel-1 data to determine whether or not waterbodies were freezing to the bottom to divide them into the two classes “deep” (〉 2 m depth) and “shallow” (〈 2 m depth). The CH4 flux map with a spatial resolution of 100 m x 100 m was calculated from data derived via the eddy-covariance technique from two aircraft campaigns in July 2012 and 2013. We coarsened the resolution of the CH4 flux map manually, to analyze if different spatial resolutions of CH4 flux data have an effect on the relation between waterbody characteristics (coverage, number, depth, size) and CH4 flux. We found that in both study areas, there was no correlation at any spatial resolution between the area fraction covered with water and the CH4 flux at a significance level of α = 0.05. We did not find consistent correlations or patterns between the number, size or depth of waterbodies and the CH4 flux in the two study areas. While there was no significant correlation in the central study area, in the northern study area a higher number of small or shallow waterbodies slightly increased the CH4 flux, whereas deep waterbodies decreased the CH4 flux. Our results indicate that waterbodies in permafrost landscapes do not necessarily act as significant CH4 emission hotspots on a regional scale containing both waterbodies and wetlands.

Description: Large uncertainties still exist in the global methane budget with clear disagreements between bottom-up and top-down estimates, limiting confidence in climate projections. This is particularly true in the Arctic, which is warming rapidly while storing vast amounts of organic carbon that could potentially be released as carbon dioxide and methane, adding a new greenhouse gas source of unknown magnitude. Regional scale methane emission estimates and functional relationships between potential drivers and methane fluxes are currently unavailable. The Airborne Measurements of Methane Fluxes (AIRMETH) campaigns are designed to quantitatively and spatially explicitly address this question. While ground-based eddy covariance (EC) measurements provide continuous in-situ observations of the surface-atmosphere exchange of energy and matter, they are rare in the Arctic permafrost zone and site selection is bound by logistical constraints among others. Consequently, these observations cover only small areas that are not necessarily representative of the region of interest. Airborne measurements can overcome this limitation by covering distances of hundreds of kilometers over time periods of a few hours. During the AIRMETH-2012 campaign aboard the research aircraft POLAR 5 we measured turbulent exchange fluxes of energy and methane along thousands of kilometers covering the North Slope of Alaska. Time-frequency (wavelet) analysis, footprint modeling, and machine learning techniques are used to extract spatially resolved turbulence statistics and fluxes, spatially resolved contributions of land cover and biophysical surface properties to each flux observation, as well as regionally valid functional relationships between environmental drivers and observed fluxes that can explain spatial flux patterns and – if available in temporal resolution – allow for spatio-temporal scaling of the observations. Here we present a 100 m resolution gridded methane flux map for the North Slope of Alaska, covering about 90.000 km2. We show that surface properties like elevation, temperature, and NDVI along with meteorological drivers such as shortwave radiation, water vapor mixing ratio, and horizontal wind speed are sufficient to explain and project the measured fluxes. The median methane flux for the campaign period (end of June/beginning of July) was 19.4 mg m−2 d−1 after excluding all values with 30 % standard error. The largest fluxes were observed along the coast and in the Arctic coastal plain, decreasing towards the Brooks Range.

Description: Among other regions of the world, the Arctic is strongly affected by climate change. Globally, it is the region with the most pronounced warming, leading to permafrost warming and thawing. Part of the 1,300 Pg soil organic carbon currently stored in the frozen ground is already and might be further released as carbon dioxide (CO2) and methane (CH4). CO2 is released through aerobic soil respiration and from plant roots, but also sequestered through photosynthesis. CH4 emission can be attributed to either recent microbial activity or to past microbial or thermal decomposition and is spatially heterogeneous. To our knowledge, regional assessments of the total carbon flux (CO2 and CH4) based on high frequency airborne measurements do not exist. Here we determine the regional pattern of CO2 and total carbon emissions (CO2 + CH4) of the Mackenzie Delta region, Canada, based on the Airborne Measurements of Methane Fluxes Campaign (AIRMETH) in July 2013 [Kohnert et al., 2014]. The Mackenzie Delta is the second largest arctic delta (13,000 km2). Our measurements covered an area extending 320 km from west to east (140°58’ W to 133°22’W) and of 240 km from north to south (69°33’N to 67°26’N). The study area is heterogeneous and comprises the delta itself, the adjacent Yukon coastal plain, and Richards Island north east of the delta. Part of the delta is located north of the treeline. The area surrounding the delta is described as continuous permafrost zone where the permafrost reaches a thickness of 300 m along the coastal plain and 500 m on Richards Island. In the delta itself the discontinuous permafrost reaches a maximum thickness of 100 m. For the AIRMETH campaign we used the research aircraft Polar 5. Equipped with a 5-hole probe, the usual meteorological sensors, and a fast greenhouse gas analyser (GGA 24EP, Los Gatos Research Inc.) we flew at 30 - 60 m above ground at a true airspeed of 60 m s−1. CO2 and CH4 fluxes were calculated with a timefrequency resolved version of the eddy-covariance technique [Metzger et al., 2013]. We calculated flux topographies [Mauder et al., 2008] to resolve the fluxes along a linear flight track to the area within the footprint of the measurements. The result is a 100 m resolved gridded carbon flux map within the footprints of the flight tracks. Based on the flux topographies we produce a map of the regional pattern of peak growing season carbon fluxes.

Description: Arctic wetlands associated with permafrost as well as thawing permafrost emit the greenhouse gas methane (CH4). Two important contributors are recent microbial activity in the active layer or taliks (biogenic CH4), and deeper fossil sources where pathways through the permafrost exist (geologic CH4). Current emission estimates vary strongly between different models. Moreover, there is still disagreement between bottom-up estimates from local field studies, and topdown estimates from atmospheric measurements. Here, we quantify permafrost CH4 emissions directly on the regional scale, based on the Airborne Measurements of Methane Fluxes Campaigns (AIRMETH) in the Mackenzie River Delta region, Canada, in July 2012 and 2013 [Kohnert et al., 2014]. The Mackenzie Delta is the second largest Arctic delta (13,000 km2). Our measurements covered an area extending 320 km from west to east (140°58’W to 133°22’W) and of 240 km from north to south (69°33’N to 67°26’N). The study area comprises the delta itself, the adjacent Yukon coastal plain, and Richards Island north east of the delta. The area surrounding the delta is described as continuous permafrost zone where the permafrost reaches a thickness of 300 m along the coastal plain and 500 m on Richards Island. In the delta itself the discontinuous permafrost reaches a maximum thickness of 100 m. The northern part of the study area is crossed by geological faults and underlain by oil and natural gas deposits. We analyse the regional pattern of CH4 fluxes and estimate the contribution of geologic emissions to the total CH4 budget of the delta. CH4 fluxes were calculated with a time-frequency resolved version of the eddy-covariance technique [Metzger et al., 2013], followed by the calculation of flux topographies [Mauder et al., 2008]. The result is a 100 m resolved gridded flux map within the footprints of the flight tracks. The results provide the first regional estimate of CH4 release from the Mackenzie Delta and the adjacent coastal plain. We distinguish geological gas seeps from biogenic sources by their strength, and show that geologic sources contribute strongly to the annual CH4 budget of the study area: One percent of the covered area contains the strongest geological seeps which contribute disproportionately to an annual emission estimate. The contribution of geological sources to CH4 emission warrants further attention, in particular in areas where permafrost is vulnerable to increased geologic gas migration due to thawing and opening of new pathways. The presented map can be used as a baseline for future CH4 flux studies in the Mackenzie Delta. References Kohnert, K.; Serafimovich, A.; Hartmann, J. and Sachs, T. [2014]: Airborne measurements of methane fluxes in alaskan and canadian tundra with the research aircraft “polar 5”. In Reports on Polar and Marine Research, volume 673. Alfred Wegener Institue Bremerhaven, pp. 81. Mauder, M.; Desjardins, R.L. and MacPherson, I. [2008]: Creating surface flux maps from airborne measurements: Application to the Mackenzie area GEWEX study MAGS 1999. Boundary-Layer Meteorology, 129:431–450, 2008. Metzger, S.; Junkermann, W.; Mauder, M.; Butterbach-Bahl, K.; Trancón y Widemann, B.; Neidl, F.; Schäfer, K.; Wieneke, S.; Zheng, X. H.; Schmid, H. P. and Foken, T. [2013]: Spatially explicit regionalization of airborne flux measurements using environmental response functions. Biogeosciences, 10(4):2193–2217, 2013. doi:10.5194/bg-10-2193-2013.

Description: In this study, gradual elevation change due to summer thawing of active layer in tundra permafrost landscape of Barrow, Alaska is investigated using SAR interferometry (InSAR) technique. We used a variety of SAR sensors including TerraSAR-X, ALOS, and Sentinel-1 images to assess elevation changes in summer season. Preliminary result, obtained by TerraSAR-X InSAR analysis, clearly delineates subsidence during the summer by identifying thousands of coherent pixels on the ground. InSAR time-series analysis from June to September 2013 show the progressive thaw-season subsidence on the Barrow coastal plain with maximum magnitude of 4 cm in satellite line of sight. The spatial pattern of InSAR elevation change reflects different thaw-related landscape features of permafrost. As a general pattern, detected elevation change is higher in wet thermokarst basins than tundra uplands. Barrow is one of the Global Terrestrial Network for Permafrost (GTN-P) sites. A lot of ancillary data that have been acquired in the last decades will help us in interpretation of InSAR results. We also use airborne hyperspectral data to derive characteristics like wetness and vegetation cover of the landscape to better understand the process and the link between estimated subsidence and landscape features.

Description: Recently, the United States Geological Service (USGS) released a new provisional product which estimates aquatic reflectance from Landsat 8 Operational Land Imager (OLI), called Landsat 8 Provisional Aquatic Reflectance (L8PAR). However, as indicated in the product guide, the use of this product for inland waters needs further verification and improvements. The goal of this study was to determine how the novel L8PAR product performs for different small turbid and eutrophic lakes in Northern Germany compared to in situ measurements of above water remote sensing reflectance (Rrs). For several recent scenes during our monitoring, the L8PAR product failed to produce full data for the lakes of our interest. For the best scene with in situ spectra, L8PAR was not able to retrieve any information for band 1 and not all information for bands 2, 3 and 4. The pixels with valid values for reflectance showed a weak relationship for band 2 (R2 of 0.24) and a medium relationship for bands 3 and 4 (R2 of 0.68 and 0.72, respectively). Compared to other atmospheric correction routines (ACOLITE, C2RCC, C2X, iCOR and L8SR), L8PAR was the only product which was not able to retrieve Rrs for all match up samples. This work provides an evaluation of the L8PAR product for inland waterbodies. Although more analysis and validation need to be conducted, our study suggests that the L8PAR product cannot be used for small inland lakes in its current state and has to be used with care for inland waters in general.

Description: Eutrophication of inland waters is an environmental issue that is becoming more common with climatic variability. Monitoring of this aquatic problem is commonly based on the chlorophyll-a concentration monitored by routine sampling with limited temporal and spatial coverage. Remote sensing data can be used to improve monitoring, especially after the launch of the MultiSpectral Instrument (MSI) on Sentinel-2. In this study, we compared the estimation of chlorophyll-a (chl-a) from different bio-optical algorithms using hyperspectral proximal remote sensing measurements, from simulated MSI responses and from an MSI image. For the satellite image, we also compare different atmospheric corrections routines before the comparison of different bio-optical algorithms. We used in situ data collected in 2019 from 97 sampling points across 19 different lakes. The atmospheric correction assessment showed that the performances of the routines varied for each spectral band. Therefore, we selected C2X, which performed best for bands 4 (root mean square error—RMSE = 0.003), 5 (RMSE = 0.004) and 6 (RMSE = 0.002), which are usually used for the estimation of chl-a. Considering all samples from the 19 lakes, the best performing chl-a algorithm and calibration achieved a RMSE of 16.97 mg/m3. When we consider only one lake chain composed of meso-to-eutrophic lakes, the performance improved (RMSE: 10.97 mg/m3). This shows that for the studied meso-to-eutrophic waters, we can reliably estimate chl-a concentration, whereas for oligotrophic waters, further research is needed. The assessment of chl-a from space allows us to assess spatial dynamics of the environment, which can be important for the management of water resources. However, to have an accurate product, similar optical water types are important for the overall performance of the bio-optical algorithm.