1,722,678 research outputs found
Lead classification maps from helicopter-borne surface temperatures during the MOSAiC expedition
No full textLeads (open water and thin ice) were classified in helicopter-borne thermal infrared observations. Lead classification maps, gridded in 1 m resolution, are provided for 35 flights between 02.10.2019 and 23.04.2020 during the MOSAiC expedition in the Arctic ocean. There is one file for every flight, either on a local (MOSAiC central observatory) or regional scale (MOSAiC distributed network). The flights can be identified by two campaign specific IDs (the event-related Device Operation label or Flight ID). The lead classification maps are derived from the surface temperature maps (doi:10.1594/PANGAEA.941017) as described in Thielke et al (in preparation). The 5 m resolution data (based on block averaged surface temperature) are included to provide data with a smaller file size so they are easier accessible and available for the comparison of the effect of different spatial resolutions. The binary lead classification is performed with a temperature threshold. In this data set, in addition to the lead classification maps, also the surface temperature and time-fixed surface temperature maps are included (the same data as included in the temperature maps: doi:10.1594/PANGAEA.941017). This time-fixed surface temperature is necessary to perform the classification and included for direct comparison to the lead classification result. All data are georeferenced, also as relative coordinates to the position of RV Polarstern which allows a Polarstern centered, Lagrangian view on the lead development
Reducing Weather Influences on Sea Ice Concentration Retrieval using Spaceborne 89 GHz Passive Microwave Observations
No full textThe major challenge of deriving sea ice concentration from the high resolution 89 GHz passive microwave observation is the strong atmospheric attenuation caused by water vapor and liquid water path, and surface variability induced by wind and temperature. In this study, we improve an 89 GHz sea ice concentration retrieval algorithm called the Arctic Radiation and Turbulence Interaction STudy (ARTIST) Sea Ice (ASI) algorithm, by correcting the observed brightness temperatures for these weather influences before they enter the algorithm. The instrument used is the Advanced Microwave Scanning Radiometer - Earth Observing System (EOS) (AMSR-E) on board NASA’s Aqua satellite. The weather correction is realized by simulating changes induced by weather influences in the top of atmosphere brightness temperatures through a radiative transfer
model. Two correction schemes are tested, one utilizing the numerical weather prediction data as input, and the other the retrievals of an optimal estimation method. Two improved versions of the ASI algorithm, ASI2 and ASI3, are developed respectively based on the corrected brightness
temperatures and new tie points. For both the influence of the atmosphere on the 89 GHz brightness temperature is successfully reduced. Main results are a better representation of low ice concentration in the marginal ice zone, and a reduction in RMS of 2.3% over high ice compared to Landsat data
Melt Ponds on Arctic Summer Sea Ice from Optical Satellite Data
No full textThe presence of melt ponds on Arctic summer sea ice strongly alters the absorption of solar radiation by the sea ice-ocean system and thereby the Arctic energy budget. Therefore, melt ponds are key to the positive sea ice-albedo feedback, which is one of the main drivers of the amplified Arctic warming observed in recent decades, and even affects the global climate. To analyze the mechanisms of melt pond evolution and their implications on the sea ice state, and to improve their representation in climate models, comprehensive observational data are needed.
This dissertation presents a new approach to retrieve melt pond, sea ice and open ocean fractions at pan-Arctic scales from Sentinel-3 optical satellite data. The newly developed Melt Pond Detection 2 (MPD2) algorithm is the first fully physical retrieval that can distinguish these three surface types at the spatial resolution of 1.2 km. Because multiple combinations of surface type fractions result in similar observations at this coarse resolution, prior information are required for retrieval.
As part of the development process, a reference data set of 33 local melt pond fraction maps with a spatial resolution of 10 m has been created from Sentinel-2 satellite data. Parts of these data were then used to calibrate an empirical pre-retrieval to provide preliminary estimates of surface type fractions. In addition, the correlation between sea ice optical properties and air temperature history has been investigated using measurement data from field campaigns. This correlation and the results of the pre-retrieval are used to initialize and constrain the physical retrieval. The results are validated against the full extent of the reference data set, leading to an uncertainty estimate of 7.8 % and 9 % for the melt pond and open ocean fractions, respectively. The MPD2 algorithm has been applied to seven years of Sentinel-3 observations from 2017 to 2023. This data set can be continued for future years and expanded by the application to previous satellite sensors.
Finally, the newly produced data set has been used to study regional differences in melt pond evolution: the lowest melt pond fractions are found in the Central Arctic with low seasonal variability, and the highest fractions are observed in the landfast ice-dominated Canadian Archipelago; the highest seasonal and interannual variability are observed in the Beaufort Sea. Additionally, a pan-Arctic analysis correlating the melt pond fraction product with sea ice surface roughness data has been carried out: this showed that flat sea ice features higher melt pond fractions at the beginning of the melt season, while later in the season melt pond fractions tend to be higher on deformed sea ice
Pathways and variability of the circulation in the subpolar eastern North Atlantic studied with inverted echo sounders and model data
No full textThe North Atlantic Current (NAC) as part of the Atlantic Meridional Overturning Circulation (AMOC) is the major pathway for warm and saline water from the subtropics into the subpolar North Atlantic. Due to buoyancy loss along its flow path and subsequent deep water formation, it connects the upper warm limb of the AMOC with the deeper cold limb. Associated volume fluxes and their variability are thus of great interest, especially in the context of climate change.
The main branch of the NAC and related transports are widely studied. The NAC crosses 47°/48°N in the western North Atlantic and further north the Mid-Atlantic Ridge (MAR) before entering the eastern subpolar basin where it partly feeds the Subpolar Gyre or flows into the Nordic Seas. To quantify the meridional exchange of water between the subtropical and subpolar regime in the interior eastern North Atlantic where studies are scarce, in this work, long-term (1993 to 2017) transport time series were calculated by combining data from inverted echo sounders taken in 2016 and 2017 with satellite altimetry. The results obtained from observational data are complemented with transport time series calculated from high resolution model output of the ANHA12 configuration of the NEMO model and with the analysis of particle trajectories calculated from the Lagrangian model ARIANE.
The observational data reveal an additional more direct pathway from the south across 47°/48°N into the subpolar eastern North Atlantic with a mean northward transport of +9.1 Sv ± 0.8 Sv contributing about 22% to the total inflow of +41.4 Sv into the eastern basin. The meridional transport of this pathway is significantly anticorrelated to the transport across the MAR (R = −0.7), damping the interannual variability of the total inflow into the subpolar eastern North Atlantic. Moreover, for the meridional transport in the interior eastern basin, a positive trend of +2.0 Sv ± 1.5 Sv per decade is found, partly balancing the negative decadal trend of −6.0 Sv ± 5.7 Sv observed for the interior western basin. The mean transport imbalance at the 47°/48°N transect between Newfoundland and 15°W was found to be −2.2 Sv which is likely to be compensated by the flow east of 15°W.
In the model, the overall circulation pattern in the subpolar North Atlantic as well as the main regions for water mass transformation are very similar to what is found from observations. However, also substantial differences between the model and observations were found such as a surplus northward flow across 47°/48°N in the western basin, a weaker coupling between the western and eastern basin, and a smaller total inflow into the eastern subpolar North Atlantic of +24.2 Sv. Moreover, the analysis of particle trajectories reveals that about 60% of the water at 47°/48°N and the MAR originates in the subtropics and about 11% flows into the Nordic Seas
Understanding polar atmosphere-ocean-sea ice momentum transfer using remote sensing and modeling techniques
No full textOver the last half a century, the Arctic sea ice extent and volume have been decreasing as a result of the amplified warming taking place in the Arctic. Similarly, the Antarctic summertime sea ice extent maximum has been the lowest in the satellite record for the last three years. As sea ice at both poles is changing in a warming climate, it is necessary to better understand the fundamental processes that determine sea ice properties such as extent, thickness, volume and drift. These processes, namely dynamic and thermodynamic ones, are triggered by the surrounding atmosphere and ocean. The overarching goal of this dissertation is to study dynamic processes while also considering thermodynamic aspects. Chapter 3 delves into the abovementioned dynamic and thermodynamic processes at mesoscale in the study of polynya events and thin sea ice anomalies above Maud Rise in the Antarctic. Chapter 4 looks at parameters that quantify dynamics, specifically at drag coefficients (Cd) that determine the momentum transfer between the atmosphere and sea ice, on a pan-Arctic scale. Finally, Chapter 5 implements the derived estimates of drag from observations into a coupled regional atmosphere-ocean-sea ice model in order to investigate the impact of variable drag on sea ice properties Arctic-wide.
The Weddell Sea Polynya (occurring in 1974-1976 and 2016-2017) is an excellent case study in the impact of mesoscale as well as synoptic scale processes on sea ice. My analysis of the events corroborates past studies that identify the Weddell Sea polynya as one that is driven by dynamic as well as thermodynamic processes. In addition, using satellite-borne microwave imaging radiometers, large thin sea ice anomalies have been identified in polynya-free years (2010-2020). Given the reported links between the polynya and different dynamic and thermodynamic ocean and atmosphere processes, our results suggest that when an insufficient amount of these processes are active, a thin sea ice anomaly may emerge instead.
The neutral sea ice-atmosphere Cd data-set is the first-ever assessment of drag on both pan- Arctic spatial and sub-yearly temporal scales. Leveraging the high resolution of Ice, Cloud and land Elevation Satellite 2 (IS2), as well as near-coincident Operation IceBridge (OIB) airborne surveys of sea ice topography, it was possible to observe the spatiotemporal evolution of drag from November 2018 to May 2022. My results showed the ice area directly north of the Canadian Archipelago and Greenland to have a Cd consistently above 2.0 × 10−3, while for most of the multiyear ice portion of the Arctic it is typically around ∼1.5 × 10−3. The first-year and young ice portion of the Arctic has a comparatively lower Cd (∼9 × 10−4) with an increase along the marginal ice zone that exceeds 1.5 × 10−3. This dataset was then used to derive a parameterization linking Cd to coincident IS2 sea ice thickness measurements, which was implemented into the regional atmosphere-ocean-sea ice model HIRHAM-NAOSIM. By running the model with and without the implementation, my results showed reasonable albeit small differences between the sea ice properties modelled by the two runs. Using sensitivity studies that varied the coefficients and integration of the Cd parameterization, I was then able to explain the differences observed. The main findings from the model study are that atmospheric and oceanic drag have the opposite effect on both sea ice drift and thickness on a pan-Arctic scale, and that over a period of three years, regardless of the range in drag variability, the impact of drag on sea ice in a coupled model is typically small in magnitude (<5% differences in both sea ice drift and thickness)
A study of Arctic sea-ice surface albedo and its uncertainty: impact of varying insolation and instrument characteristics
No full textThe Arctic sea-ice surface albedo is observed by ground-based, airborne, and satellite instruments. However, the observed albedo is influenced by varying insolation, instrument characteristics, and associated retrieval processes. This thesis quantifies these influences, develops and improves corrective schemes, and gives recommendations to increase the accuracy of the surface broadband albedo products. The analysis is based on simulations with the radiative transfer model SCIATRAN. A single-scattering property database of nine ice crystal habits is implemented into SCIATRAN to enable a realistic snow surface reflection. The associated far-field assumption of snow grains is justified. A comparison of simulated and measured reflectance factors and albedo spectra reveals discrepancies of usually less than 0.05, legitimating the utilized SCIATRAN set-up. The atmosphere decreases the black-sky surface broadband albedo by a RMSD of less than 0.04. An optically thin (thick) cloud additionally influences the albedo of up to a RMSD of 0.02 (0.06). The albedo from irradiance measuring devices suffers from the instrument’s cosine error (RMSD<0.15) and has to be corrected for by calibration factors (RMSD<0.03). The total uncertainty of the satellite broadband albedo (RMSD<0.26) is mainly controlled by the anisotropic correction (RMSD<0.25) and the narrow-to-broadband conversion (NTBC) (RMSD<0.09). New, most accurate NTBCs are developed (RMSD≤0.02). Applying those and the RossThick-LiSparseReciprocal or the modified Walthall model as angular correction almost halve the total satellite albedo uncertainty. The new NTBC improves the MERIS derived sea-ice surface broadband albedo by up to one third. Surface inhomogeneities also significantly affect the observed surface albedo. A seasonal comparison of the Arctic sea-ice surface broadband albedo from MERIS, CLARA-A2, and ERA5 reanalysis reveals RMSDs exceeding 0.10 which is partly due to the above-mentioned uncertainties
Unraveling atmosphere and sea ice in the Arctic : advancements in a multi-parameter retrieval using satellite microwave radiometer data
No full textThe Arctic undergoes accelerated warming compared to Global Warming, known as Arctic amplification. To understand this phenomenon, studying key variables at various scales is crucial. Every day, satellite radiometers measure Arctic-wide emissions of microwave radiation in terms of brightness temperatures. We use the distinct sensitivities of observations at different frequencies to atmospheric and surface parameters, aiming to disentangle the satellite signal and improve a multi-parameter retrieval. The retrieval involves inverting a forward model with an optimal estimation method to attribute satellite measurements (6.9 to 89 GHz) to a specific geophysical state.
For that, surface emissions need to be well represented in the forward model.
We study surface emissions by considering the theoretical concept of emissivity followed by an analysis of brightness temperature measurements and derived emissivities that we obtained during a summer ship campaign.
In a case study, we examine the impact of changing surface emissions caused by warm air intrusions on sea ice concentration (SIC) satellite retrievals.
We improve the multi-parameter retrieval by a better representation of the sea ice and snow emissions in the forward model for non-melting conditions. Both forward model and retrieval are evaluated against ground truth, including MOSAiC expedition data. The forward model succeeds in simulating realistic brightness temperatures. The retrieval output agrees well with reference data with regard to SIC. While the retrieval of cloud liquid water path and snow depth are promising, some disagreements are identified. Focusing on atmospheric total water vapor, our analysis demonstrates a good agreement with numerous reference datasets. We find a substantial improvement over the previous version of the method.
After applying the retrieval to satellite data from the last two decades we further analyze the spatio-temporal variability of atmospheric water vapor in winter
Remote sensing of sea ice leads with Sentinel-1 C-band synthetic aperture radar
No full textThe presence of leads with open water or thin ice is an important feature of the Arctic sea ice cover. Leads regulate the heat, gas, and moisture fluxes between the ocean and atmosphere and are areas of high ice growth rates during periods of freezing conditions. In the present study an algorithm providing an automatic lead detection based on Synthetic Aperture Radar (SAR) images is developed using traditional machine learning techniques and deep learning methods. The algorithm is applied to a wide range of Sentinel-1 scenes taken over the Arctic Ocean. Distribution of the detected leads in the Arctic during winter seasons 2016--2021 is then analyzed.
An important part of the algorithm development is the data preprocessing as the classification quality depends on the quality of the input images.
An advanced data preparation technique improves consistency of the cross-polarization channel and enables the use of dual-polarization SAR images. By using both the HH and the HV channels instead of single co-polarized observations the algorithm is able to detect more leads compared to the use of the HH polarization only.
First, a traditional machine learning approach is described. It is based on polarimetric features and texture features derived from the grey level co-occurrence matrix. The Random Forest classifier is used to investigate the individual feature importance on the lead detection. The precision-recall curve representing the quality of the classification is assessed to define a threshold for the binary lead/sea ice classification. The algorithm produces a lead classification with more than 90% precision with 60% of all leads classified, as evaluated on the test data. The precision can be increased by the cost of the amount of leads detected. Classification quality is improved by introducing an advanced binarization method based on watershed segmentation. Further improvements include object shape analysis resulting in a shape-based filter, which efficiently removes objects appearing due to noise patterns over young ice.
Second, an algorithm based on a convolutional neural network is developed. It shows more robust results compared to the algorithm based on the gray level co-occurrence matrix with Random Forest classification
and is applicable to the entire Arctic Ocean. Classification results are evaluated against the dataset which does not include training or testing data, and are also compared to Sentinel-2 optical satellite images.
Finally, the lead detection algorithm is applied to all Sentinel-1 EW GRDM scenes taken in five winter seasons, 1 November - 30 April of 2016-2021 years. 3-day composite pan-Arctic lead maps with the native Sentinel-1 40~meters pixel spacing are produces. The frequency of lead occurrence derived from these maps is compared with MODIS thermal infrared lead detection results.
The lead area fraction is compared with the AMSR2 passive microwave observations. The lead area distribution, lead length, and lead width distributions, as well as the lead orientation distributions, are analyzed
in the following regions of the Arctic Ocean: Fram Strait, Barents Sea, Kara Sea, Laptev Sea, East Siberian Sea, Chukchi Sea, Beaufort Sea, Central Arctic. Each region shows the presence of regularity in lead orientation, the preferred orientation has little variation from year to year and during season. The lead width distribution is found to follow the power low with the exponent of 1.86 with 0.16 standard deviation. The yearly mean lead area fraction derived from Sentinel-1 images varies from 2.5% to 3.7% during winter seasons 2016-2021
Snow Depth on Arctic Sea Ice from Microwave Radiometers
No full textSchnee auf Meereis gehört zu den wichtigsten Größen im arktischen Klimasystem. Trotzdem gibt es keine verlässliche Methode, die Schneedicke arktisweit zu bestimmen. In dieser Arbeit wurde ein neuer Algorithmus zur Bestimmung der Schneedicke auf arktischem Meereis aus Satellitenbeobachtungen entwickelt. Die Unsicherheiten des Algorithmus wurden mittels Monte-Carlo Modellierung abgeschätzt. Über saisonalem Eis liegt die abgeleitete Schneedicke im Mittel sehr nah an in-situ Beobachtungen, die Standartabweichung beträgt weniger als 5 cm und ist in der Größenordnung der ermittelten Unsicherheit. Über mehrjährigem Eis sind die Unsicherheiten der gewonnen Schneedicke deutlich größer und die Standartabweichung zu in-situ Messungen ist 8 cm. Es wurden zwei Schneedicke auf arktischem Meereis Datensätze veröffentlicht, welche genutzt werden können um Beispielsweise die Schneedicke in Klimamodellen zu evaluieren
A 1 km sea-ice concentration dataset from merged thermal infrared and microwave radiometer satellite observations: more than the sum of its parts
No full textEs wird ein neuartiger Meereiskonzentrations-Datensatz mit 1 km räumlicher Auflösung präsentiert, der die Vorteile von Radiometermessungen im thermalen Infrarot- und im Mikrowellen-Spektralbereich kombiniert. Der Vorteil gegenüber reinen Infrarot-Daten besteht in der räumlichen Kontinuität und der statistischen Konsistenz mit den etablierten Mikrowellen-Radiometermessungen. Der Vorteil gegenüber Mikrowellen-Daten mit einer räumlichen Auflösung von 5 km besteht in der feineren räumlichen Auflösung von 1 km und einer besseren Erkennung von Rinnen im Eis. Wir erweitern einen existierenden Meereiskonzentrations-Algorithmus für Infrarot-Messungen, indem wir die räumliche Abdeckung verbessern und ein glatteres Referenz-Temperaturfeld erhalten. Die dadurch berechnete Meereiskonzentration wird mit Mikrowellen-Messungen kombiniert, indem sowohl der Mittelwert der Mikrowellen-Daten auf einer Skala von 5 km erhalten bleibt als auch die Variabilität der Infrarot-Daten auf einer Skala von 1 km. Der Vorteil der feineren Auflösung wird aufgezeigt, indem die Fläche der Pixel mit einer Meereiskonzentration von höchstens 85 % berechnet wird. In einer Fallstudie liegt dieser Wert bei 771 km² und damit deutlich näher an einem Referenz-Datensatz (878 km²) als die gröber aufgelösten Mikrowellendaten (182 km²). In einem zweiten Fall, in dem eine unentdeckte Wolke und hohe Eisoberflächen-Temperaturen die Qualität der Infrarot-Meereiskonzentration beeinträchtigen, wird die Wurzel der mittleren quadratischen Abweichung (engl. RMSD) gegenüber den Referenzdaten von 23.5 % (Infrarot-Daten) auf 9.3 % (kombinierter Datensatz) reduziert. Der Vorteil des kombinierten Datensatzes wird darüber hinaus anhand einer Polynya demonstriert, die im Februar 2018 nördlich von Grönland entstanden ist. Die feinere Auflösung ermöglicht eine genauere Beobachtung der Entstehungsphase der Polynya als mit Mikrowellen-Daten, während die räumliche Kontinuität eine lückenlose Beobachtung erlaubt, die mit Infrarot-Daten nicht möglich ist
- …
