Spatio-temporal analysis of melting onset dates of sea-ice in the Arctic (original) (raw)
Related papers
Large-Scale Patterns of Snow Melt on Arctic Sea Ice Mapped from Meteorological Satellite Imagery
Annals of Glaciology
The seasonal progression of snow melt on the Arctic pack ice is mapped from satellite shortwave imagery (0.4–1.1 micrometers) for four spring/summer seasons (1977, 1979, 1984 and 1985). This provides the first detailed information on the temporal change of the ice surface albedo in summer and of its year-to-year variability. The average surface albedo of the Arctic Basin for the years investigated falls from between 0.75 and 0.80 in early May to between 0.35 and 0.45 in late July and early August. In the central Arctic, where ice concentration remains high and ponding on the ice is limited, the July albedo ranges from 0.50 to 0.60. Overall, melt progresses poleward from the Kara and Barents Seas and from the southern Beaufort and Chukchi Seas, with the melt fronts meeting on the American side of the Pole. There are substantial year-to-year differences in the timing, duration and extent of the melt interval. The progression of melt in May and June of the earliest melt year (1977) was...
Satellite-Derived Evolution of Arctic Sea Ice Age: October 1978 to March 2003
IEEE Geoscience and Remote Sensing Letters, 2004
Combining gridded ice motions with daily ice extent maps, it is possible to "track" the evolution of sea ice in the Arctic region. Classifying the ice by ice age, this evolution reveals that the area of the oldest ( 4 years) ice is decreasing in the Arctic Basin and is being replaced by younger, first-year ice. As a result, the extent of the oldest ice retreats to a relatively small area north of the Canadian Archipelago, with narrow bands that spread out across the central Arctic. This new approach reinforces the work done by others showing the changes in the Arctic sea ice cover over the past two decades.
1. Abstract The recognized importance of the annual cycle of sea ice in the Arctic to heat budgets, human behavior, and ecosystem functions, requires consistent definitions of such key events in the ice cycle as break-up and freeze-up. An internally consistent and reproducible approach to characterize the timing of these events in the annual sea-ice cycle is described. An algorithm was developed to calculate the start and end dates of freeze-up and break-up and applied to time series of satellite-derived sea-ice concentration from 1979 to 2013. Our approach builds from discussions with sea-ice experts having experience observing and working on the sea ice in the Bering, Chukchi and Beaufort Seas. Applying the algorithm to the 1979–2013 satellite data reveals that freeze-up is delayed by two weeks per decade for the Chukchi coast and one week per decade for the Beaufort coast. For both regions, break-up start is arriving earlier by 5–7 days per decade and break-up end is arriving earlier by 10–12 days per decade. In the Chukchi Sea, " early " break-up is arriving earlier by one month over the 34-year period and alternates with a " late " break-up. The calculated freeze-up and break-up dates provide information helpful to understanding the dynamics of the annual sea-ice cycle and identifying the drivers that modify this cycle. The algorithm presented here, and potential refinements, can help guide future work on changes in the seasonal cycle of sea ice. The sea-ice phenology of freeze-up and break-up that results from our approach is consistent with observations of sea-ice use. It may be applied to advancing our understanding and prediction of the timing of seasonal navigation, availability of ice as a biological habitat, and assessment of numerical models.
Spatial and temporal variations in the age structure of Arctic sea ice
Geophysical Research Letters, 2005
1] Spatial and temporal variations in the age structure of Arctic sea ice are investigated using a new reversechronology algorithm that tracks ice-covered pixels to their location and date of origin based on ice motion and concentration data. The Beaufort Gyre tends to harbor the oldest (>10 years old) sea ice in the western Arctic while direct ice advection pathways toward the Transpolar Drift Stream maintain relatively young ( 5 years) ice in the eastern Arctic. Persistent net losses (À4.2% yr À1 ) in extent of ice >10 years old (10+ year age class) were observed during 1989 -2003. Since the mid-1990s, losses to the 10+ year age class lacked compensation by recruitment due to a prior depletion of all mature (6 -10 year) age classes.
Arctic sea-ice variability revisited
Annals of Glaciology, 2008
This paper explores spatial and temporal relationships between variations in Arctic sea-ice concentration (summer and winter) and near-surface atmospheric temperature and atmospheric pressure using multivariate statistical techniques. Trend, empirical orthogonal function (EOF) and singular value decomposition (SVD) analyses are used to identify spatial patterns associated with covariances and correlations between these fields. Results show that (1) in winter, the Arctic Oscillation still explains most of the variability in sea-ice concentration from 1979 to 2006; and (2) in summer, a decreasing sea-ice trend centered in the Pacific sector of the Arctic basin is clearly correlated to an Arctic-wide air temperature warming trend. These results demonstrate the applicability of multivariate methods, and in particular SVD analysis, which has not been used in earlier studies for assessment of changes in the Arctic sea-ice cover. Results are consistent with the interpretation that a warmin...
Sea ice breakup and freeze-up indicators for users of the Arctic coastal environment
The Cryosphere, 2022
The timing of sea ice retreat and advance in Arctic coastal waters varies substantially from year to year. Various activities, ranging from marine transport to the use of sea ice as a platform for industrial activity or winter travel, are affected by variations in the timing of breakup and freeze-up, resulting in a need for indicators to document the regional and temporal variations in coastal areas. The primary objective of this study is to use locally based metrics to construct indicators of breakup and freeze-up in the Arctic and subarctic coastal environment. The indicators developed here are based on daily sea ice concentrations derived from satellite passive-microwave measurements. The "day of year" indicators are designed to optimize value for users while building on past studies characterizing breakup and freeze-up dates in the open pack ice. Relative to indicators for broader adjacent seas, the coastal indicators generally show later breakup at sites known to have landfast ice. The coastal indicators also show earlier freeze-up at some sites in comparison with freeze-up for broader offshore regions, likely tied to earlier freezing of shallow-water regions and areas affected by freshwater input from nearby streams and rivers. A factor analysis performed to synthesize the local indicator variations shows that the local breakup and freeze-up indicators have greater spatial variability than corresponding metrics based on regional ice coverage. However, the trends towards earlier breakup and later freeze-up are unmistakable over the post-1979 period in the synthesized metrics of coastal breakup and freeze-up and the corresponding regional ice coverage. The findings imply that locally defined indicators can serve as key links between pan-Arctic or global indicators such as sea ice extent or volume and local uses of sea ice, with the potential to inform community-scale adaptation and response.
Spatial hierarchy in Arctic sea ice dynamics
Tellus A, 2003
We define a new classification for Arctic sea ice dynamics based on a spatial and temporal scale: floe, multifloe, aggregate, coherent, sub-basin and seasonal. The classification is supported by remote sensing and in situ observations of ice motions at scales of 1-700 km, as found in the existing scientific literature. The first significant change in sea ice behavior appears as an "emergent" property of the sea ice at the transition from the multifloe scale (2-10 km/<1 d) to the aggregate scale (10-75 km/1-3 d). This emergent behavior establishes a statistical mechanical length where sea ice can be considered a plastic continuum. A second important, or coherent scale occurs at 75-300 km and 3-7 d, where the spatial/temporal processes of sea ice dynamics best match the scales of the wind forcing, i.e., winds of this duration and fetch are necessary to fully load the internal stress field. At scales smaller than the coherent scale, the spatial dimension is important because the sea ice motions on the coherent scale provide non-local forcing to the aggregate scale. At dimensions larger than the coherent scale, including the sub-basin and seasonal scales, spatial and temporal averaging occurs, which smooths discontinuities. To understand and model sea ice dynamics at each of these scales requires an understanding of the detail at the next smallest level. Proper understanding and representation of sea ice dynamics at all scales is critical to devising a sound strategy for data assimilation into sea ice models.
A 5-year satellite climatology of winter sea ice leads in the western Arctic
Journal of Geophysical Research, 1998
The distribution of openings (leads and polynyas) in polar sea ice is not well known. This study estimates the large-scale distribution and variability of leads in the Arctic of the western hemisphere in winter, using a 5-year record of Defense Meteorological Satellite Program thermal-and visible-band imagery. The occurrence (density) and orientation of leads are derived from gridded maps made at 1 O-day intervals. Their mean value and interannual, seasonal, and monthly variabilities are estimated. Lead densities are observed to be highest in early winter, decreasing 20% from November through April. The highest densities are observed in the central Canada Basin, and the lowest are in the East Siberian Sea. There is limited interannual variability in the positions of maximum and minimum densities. Preferred lead orientations are identified as generally north-south in the Beaufort Sea sector and east-west in the East Siberian Sea sector, with transitional orientations in the intermediate area. The mean distributions of lead density and orientation are observed to be associated with large-scale mean fields of ice divergence and shear, respectively. 1. Introduction Sea ice is an important, interactive component of the Earth's climate system, both affecting and reflecting climate variability through a number of feedback mechanisms. The most important climatological aspects of sea ice arc, in probable order of significance: (1) its spatial extent (the total area within the ice margins), (2) its thickness distribution, and (3) the open water area within the ice cover. Variations in the ice extent affect and reflect climate variability and arc considered important factors in global climate change [c.g., Johannessen et al., 1995]
An iterative approach to multisensor sea ice classification
IEEE Transactions on Geoscience and Remote Sensing, 2000
Characterizing the variability in sea ice in the polar regions is fundamental to an understanding of global climate and the geophysical processes which govern climate changes. Sea ice can be grouped into a number of general classes with different characteristics. Multisensor data from NSCAT, ERS-2, and SSM/I are reconstructed into enhanced resolution imagery for use in ice type classification. The resulting 12-dimensional data set is linearly transformed through principal component analysis to reduce data dimensionality and noise levels. An iterative statistical data segmentation algorithm is developed using maximum likelihood and maximum a posteriori techniques. For a given ice type, the conditional probability distributions of observed vectors are assumed to be Gaussian. The cluster centroids, covariance matrices, and a priori distributions are estimated from the classification of a previous temporal image set. An initial classification is produced using centroid training data and a weighted nearest-neighbor classifier. Though validation is limited, the algorithm results in an ice classification which is judged to be superior to a conventional k-means approach.