Survey estimation and marginal effects in Stata

Stata is a general purpose software package that has become popular among various disciplines such as epidemiology, economics, or social sciences. Users like Stata for its scientific approach, its robustness and reliability, and the ease with which its functionality can be extended by user written programs. In this talk I will first give a brief overview of the functionality of Stata and then discuss two specific features: survey estimation and predictive margins/marginal effects. Most surveys are based on complex samples that contain multiple sampling stages, are stratified or clustered, and feature unequal selection probabilities. Standard estimators can produce misleading results in such samples unless the peculiarities of the sampling plan are taken into account. Stata offers survey statistics for complex samples for a wide variety of estimators and supports several variance estimation procedures such as linearization, jackknife, and balanced repeated replication (see Kreuter and Valliant, 2007, Stata Journal 7: 1-21). In the talk I will illustrate these features using applied examples and I will also show how user written commands can be adapted to support complex samples. Complex can also be the models we fit to our data, making it difficult to interpret them, especially in case of nonlinear or non-additive models (Mood, 2010, European Sociological Review 26: 67-82). Stata provides a number of highly useful commands to make results of such models accessible by computing and displaying predictive margins and marginal effects. In my talk I will discuss these commands provide various examples demonstrating their use.

West Antarctic Ice Streams

Solar heat is the acknowledged driving force for climatic change. However, ice sheets are also capable of causing climatic change. This property of ice sheets derives from the facts that ice and rock are crystalline whereas the oceans and atmosphere are fluids and that ice sheets are massive enough to depress the earth's crust well below sea level. These features allow time constants for glacial flow and isostatic compensation to be much larger than those for ocean and atmospheric circulation and therefore somewhat independent of the solar variations that control this circulation. This review examines the nature of dynamic processes in ice streams that give ice sheets their degree of independent behavior and emphasizes the consequences of viscoplastic instability inherent in anisotropic polycrystalline solids such as glacial ice. Viscoplastic instability and subglacial topography are responsible for the formation of ice streams near ice sheet margins grounded below sea level. As a result the West Antarctic marine ice sheet is inherently unstable and can be rapidly carved away by calving bays which migrate up surging ice streams. Analyses of tidal flexure along floating ice stream margins, stress and velocity fields in ice streams, and ice stream boundary conditions are presented and used to interpret ERTS 1 photomosaics for West Antarctica in terms of characteristic ice sheet crevasse patterns that can be used to monitor ice stream surges and to study calving bay dynamics.

Stratigraphic Variation Within Polar Firn Caused by Differential Accumulation and Ice Flow: Interpretation of a 400 Mhz Short-Pulse Radar Profile from West Antarctica

We investigate causes of the stratigraphic variation revealed in a 177 km, 400 MHz short-pulse radar profile of firn from West Antarctica. The profile covers 56 m depth, and its direction was close to those of the ice flow and mean wind. The average, near-surface accumulation rates calculated from the time delays of one radar horizon consistently show minima on leeward slopes and maxima on windward slopes, confirming an earlier study based on stake observations. The stratigraphic variation includes up to 30 m depth variation in individual horizons over tens of km, fold limbs that become progressively steeper with depth, and fold-hinge loci that change direction or propagate down-ice with depth over distances far less than predicted by the ice speeds. We use an accumulation rate model to show how local rate anomalies and the effect of ice speed upon a periodic variation in accumulation rate cause these phenomena, and we reproduce two key features seen in the stratigraphic variations. We conclude that the model provides an explanation of changes in spatial stratigraphy and local measures of accumulation history given the constraints of surface topography, ice and wind velocities, and a general accumulation rate for an area.

Chronology of Lateglacial ice flow reorganization and deglaciation in the Gotthard Pass area, Central Swiss Alps, based on cosmogenic 10Be and in situ 14C

We reconstruct the timing of ice flow reconfiguration and deglaciation of the Central Alpine Gotthard Pass, Switzerland, using cosmogenic 10Be and in situ14C surface exposure dating. Combined with mapping of glacial erosional markers, exposure ages of bedrock surfaces reveal progressive glacier downwasting from the maximum LGM ice volume and a gradual reorganization of the paleoflow pattern with a southward migration of the ice divide. Exposure ages of ∼16–14 ka (snow corrected) give evidence for continuous early Lateglacial ice cover and indicate that the first deglaciation was contemporaneous with the decay of the large Gschnitz glacier system. In agreement with published ages from other Alpine passes, these data support the concept of large transection glaciers that persisted in the high Alps after the breakdown of the LGM ice masses in the foreland and possibly decayed as late as the onset of the Bølling warming. A younger group of ages around ∼12–13 ka records the timing of deglaciation following local glacier readvance during the Egesen stadial. Glacial erosional features and the distribution of exposure ages consistently imply that Egesen glaciers were of comparatively small volume and were following a topographically controlled paleoflow pattern. Dating of a boulder close to the pass elevation gives a minimum age of 11.1 ± 0.4 ka for final deglaciation by the end of the Younger Dryas. In situ14C data are overall in good agreement with the 10Be ages and confirm continuous exposure throughout the Holocene. However, in situ14C demonstrates that partial surface shielding, e.g. by snow, has to be incorporated in the exposure age calculations and the model of deglaciation.

Water stable oxygen isotope records from six Antarctic ice core sites for the present and last interglacial periods

We compare the present and last interglacial periods as recorded in Antarctic water stable isotope records now available at various temporal resolutions from six East Antarctic ice cores: Vostok, Taylor Dome, EPICA Dome C (EDC), EPICA Dronning Maud Land (EDML), Dome Fuji and the recent TALDICE ice core from Talos Dome. We first review the different modern site characteristics in terms of ice flow, meteorological conditions, precipitation intermittency and moisture origin, as depicted by meteorological data, atmospheric reanalyses and Lagrangian moisture source diagnostics. These different factors can indeed alter the relationships between temperature and water stable isotopes. Using five records with sufficient resolution on the EDC3 age scale, common features are quantified through principal component analyses. Consistent with instrumental records and atmospheric model results, the ice core data depict rather coherent and homogenous patterns in East Antarctica during the last two interglacials. Across the East Antarctic plateau, regional differences, with respect to the common East Antarctic signal, appear to have similar patterns during the current and last interglacials. We identify two abrupt shifts in isotopic records during the glacial inception at TALDICE and EDML, likely caused by regional sea ice expansion. These regional differences are discussed in terms of moisture origin and in terms of past changes in local elevation histories, which are compared to ice sheet model results. Our results suggest that elevation changes may contribute significantly to inter-site differences. These elevation changes may be underestimated by current ice sheet models

Grain size distribution in surface sediments samples of the continental shelf and slope of southeast Greenland

During "Meteor" Cruise 6/1966 in the northwest Atlantic a systematic survey of the bottom topography of the southeast Greenland continental margin was undertaken. Eighty-seven profiles transverse to the shelf edge at distances of 3-4 nautical miles and two longitudinal profiles parallel to the coast were carried out with the ELAC Narrow Beam Echo-Sounder giving a reliable record of even steep slopes. On the basis of the echo soundings the topography and morphology of the continental shelf and slope are evaluated. A detailed bathymetric chart and a serial profile chart were designed as working material for the morphological research. These maps along with the original echograms are morphometrically evaluated. The analysis of the sea bottom features is the basis of a subsequent morphogenetical interpretation, verified and extended by means of interpretation of magnetic data and sediment analysis (grain size, roundness, lithology). The results of the research are expressed in a geomorphological map. The primary findings can be summarized as follows: 1) The southeast Greenland shelf by its bottom topography can be clearly designated as a glacially formed area. The glacial features of the shelf can be classified into two zones nearly parallel to the coast: glacial erosion forms on the inner shelf and glacial accumulation forms on the outer shelf. The inner shelf is characterized by the rugged and hummocky topography of ice scoured plains with clear west/east slope asymmetry. On the outer shelf three types of glacial accumulation forms can be recognized: ice margin deposits with clearly expressed terminal moraines, glacial till plains and glaciomarine outwash fans. Both zones of the shelf can be subdivided into two levels of relief. The ice scoured plains, with average depths of 240 meters (m), are dissected to a maximum depth of 1060 m (Gyldenloves Trough) by trough valleys, which are the prolongations of the Greenland fjords. The banks of the outer shelf, with an average depth of 180 m, surround glacial basins with a maximum depth of 670 meters. 2) The sediments of the continental shelf can be classified as glacial due to their grain size distribution and the degree of roundness of the gravel particles. The ice margin deposits on the outer shelf can be recognized by their high percentage of gravels. On the inner shelf a rock surface is suggested, intermittently covered by glacial deposits. In the shelf troughs fine-grained sediments occur mixed with gravels. 3) Topography and sediments show that the southeast Greenland shelf was covered by an ice sheet resting on the sea floor during the Pleistocene ice-age. The large end moraines along the shelf edge probably indicate the maximum extent of the Wurm shelf ice resting on the sea floor. The breakthroughs of the end moraines in front of the glacial basins suggest that the shelf ice has floated further seaward over the increasing depths. 4) Petrographically the shelf sediments consist of gneisses, granites and basalts. While gneisses and granites occire on the nearby coast, basalt is not known to exist here. Either this material has been drifted by icebergs from the basalt province to the north or exists on the southeast Greenland shelf itself. The last interpretation is supported bythe high portion of basalt contained in the sediment samples taken and the strong magnetic anomalies probably caused by basaltic intrusions. 5) A magnetic profile allows the recognition of two magnetically differing areas which approximately coincide with the glacial erosion and accumulation zones. The inner shelf shows a strong and variable magnetic field because the glacially eroded basement forms the sea floor. The outer shelf is characterized by a weak and homogenous magnetic field, as the magnetized basement lies at greater depthy, buried by a thick cover of glacial sediments. The strong magnetic anomalies of the inner shelf are probably caused by dike swarms, similar to those observed further to the north in the Kangerdlugssuaq Fjord region. This interpretation is supported by the high basalt content of the sediment samples and the rough topography of the ice scoured plains which correlates in general with the magnetic fluctuations. The dike structures of the basement have been differentially eroded by the shelf ice. 6) The continental slope, extending from the shelf break at 313 m to a depth of 1270 m with an average slope of 11°, is characterized by delta-shaped projections in front of the shelf basins, by marginal plateaus, ridges and hills, by canyons and slumping features. The projections could be identified as glaciomarine sediment fans. This conclusion is supported by the strong decrease of magnetic field intensity. The deep sea hills and ridges with their greater magnetic intensities have to be regarded as basement outcrops projecting through the glaciomarine sediment cover. The upper continental rise, sloping seaward at about 2°, is composed of wide sediment fans and slump material. A marginal depression on the continental rise running parallel to the shelf edge has been identified. In this depression bottom currents capable of erosion have been recorded. South of Cape Farvel the depression extends to the accumulation zone of the "Eirik" sedimentary ridge. 7) By means of a study of the recent marine processes, postglacial modification of the ice-formed relief can be postulated. The retention effect of the fjord troughs and the high velocity of the East Greenland stream prevents the glacial features from being buried by sediments. Bottom currents capable of active erosion have only been found in the marginal depression on the continental rise. In addition, at the time of the lowest glacio-eustatic sea level, the shelf bottom was not situated in the zone of wave erosion. Only on the continental slope and rise bottom currents, sediment slumps and turbidity currents have led to significant recent modifications. Considering these results, the geomorphological development of the southeast Greenland continental terrace can be suggested as follows: 1. initial formation of a "peneplain", 2. fluvial incision, 3. submergence, and finally 4. glacial modification.

(Table 3) Sedimentation rates of stage 1 and last glacial maximum from different Holes during cruises of RRS Discovery in 1984-1985 and 1987-1988, and RRS James Clark Ross in 1992-1993

Within the Scotia Sea, the axis of the Antarctic Circumpolar Current (ACC) is geographically confined, and sediments therefore contain a record of palaeo-flow speed uncomplicated by ACC axis migration. We outline Holocene and Last Glacial Maximum (LGM) current-controlled sedimentation using data from 3.5-kHz profiles, cores and current meter moorings. Geophysical surveys show areas of erosion and deposition controlled by Neogene basement topography. Deposition occurs in mounded sediment drifts or flatter areas, where 500-1000 m of sediment overlies acoustic basement. 3.5-kHz profiles show parallel, continuous sub-bottom reflectors with highest sedimentation rates in the centre of the drifts, and reflectors converging towards marginal zones of non-deposition. Locally, on the flanks of continental blocks (e.g. South Georgia), downslope processes are dominant. The absence of mudwaves on the sediment drifts may result from the unsteadiness of ACC flow. A core transect from the ACC axis south to the boundary with the Weddell Gyre shows a southward decrease in biogenic content, controlled by the Polar Front and the spring sea-ice edge. Both these features lay farther north at LGM. The cores have been dated by relative abundance of the radiolarian Cycladophora davisiana, and by changes in the biogenic Ba content, a palaeoproductivity indicator. Sedimentation rates range from 3 to 17 cm/ka. The grain size of Holocene sediments shows a coarsening trend from south to north, consistent with strongest bottom-current flow near the ACC axis, though interpretation is complicated by the presence of biogenic grains. Year-long current meter records indicate mean speeds from 7 cm/s in the south to 12 cm/s in the north, with benthic storm frequency increasing northwards. LGM sediments are predominantly terrigenous and show a clearer northward-coarsening trend, with well-sorted silts in the northern Scotia Sea. Assuming a constant terrigenous source, this implies stronger ACC flow at the LGM, contrasting with weaker Weddell Gyre flow deduced from earlier work.

EPICA Dome C Ice Core 800KYr deuterium data and temperature estimates

A high-resolution deuterium profile is now available along the entire European Project for Ice Coring in Antarctica Dome C ice core, extending this climate record back to marine isotope stage 20.2, ~800,000 years ago. Experiments performed with an atmospheric general circulation model including water isotopes support its temperature interpretation. We assessed the general correspondence between Dansgaard-Oeschger events and their smoothed Antarctic counterparts for this Dome C record, which reveals the presence of such features with similar amplitudes during previous glacial periods. We suggest that the interplay between obliquity and precession accounts for the variable intensity of interglacial periods in ice core records. Temperature was estimated after correction for sea-water isotopic composition (Bintanja et al, 2005) and for ice sheet elevation (Parrenin et al, 2007) on EDC3 age scale (Parrenin et al, 2007).

Hydrate, ice and quartz content of eight holes of Leg 204

The state of preservation of natural gas hydrate samples, recovered from 6 sites drilled during ODP Leg 204 at southern summit of Hydrate Ridge, Oregon Margin, has been investigated by X-ray diffraction (XRD) and cryo-scanning-electron-microscopy (cryo-SEM) techniques. A detailed characterization of the state of decomposition of gas hydrates is necessary since no pressurized autoclave tools were used for sampling and partial dissociation must have occurred during recovery prior to the quench and storage in liquid nitrogen. Samples from 16 distinct horizons have been investigated by synchrotron X-ray diffraction measurements at HASYLAB/ Hamburg. A full profile fitting analysis ("Rietveld method") of synchrotron XRD data provides quantitative phase determinations of the major sample constituents such as gas hydrate structure I (sI), hexagonal ice (Ih) and quartz. The ice content (Ih) in each sample is related to frozen water composed of both original existing pore water and the water from decomposed hydrates. Hydrate contents as measured by diffraction vary between 0 and 68 wt.% in the samples we measured. Samples with low hydrate content usually show micro-structural features in cryo-SEM ascribed to extensive decomposition. Comparing the appearance of hydrates at different scales, the grade of preservation seems to be primarily correlated with the contiguous volume of the original existing hydrate; the dissociation front appears to be indicated by micrometer-sized pores in a dense ice matrix.

Probabilistic models for the prediction of a ship performance in dynamic ice

We introduce two probabilistic, data-driven models that predict a ship's speed and the situations where a ship is probable to get stuck in ice based on the joint effect of ice features such as the thickness and concentration of level ice, ice ridges, rafted ice, moreover ice compression is considered. To develop the models to datasets were utilized. First, the data from the Automatic Identification System about the performance of a selected ship was used. Second, a numerical ice model HELMI, developed in the Finnish Meteorological Institute, provided information about the ice field. The relations between the ice conditions and ship movements were established using Bayesian learning algorithms. The case study presented in this paper considers a single and unassisted trip of an ice-strengthened bulk carrier between two Finnish ports in the presence of challenging ice conditions, which varied in time and space. The obtained results show good prediction power of the models. This means, on average 80% for predicting the ship's speed within specified bins, and above 90% for predicting cases where a ship may get stuck in ice. We expect this new approach to facilitate the safe and effective route selection problem for ice-covered waters where the ship performance is reflected in the objective function.

Prominent deformation features and interpretation of sediment core CRP-2/2A (Table 1)

Sediment deformation features in CRP-2/2A were described during normal logging procedures and from core-scan images. In this paper the origin of soft-sediment folding, contorted bedding, microfaulting, clastic dykes, shear zones and intraformational breccias is discussed. The features have a stratigraphic distribution related to major unconformities and sequence boundaries. Hypotheses for the origins of sediment deformation include hydrofracturing, subglacial shearing, slumping, and gas hydrate formation. Shear zones, microfaults, clastic dykes and contorted bedding within rapidly deposited sediments, suggest that slumping in an ice-distal environment occurred in the early Oligocene. A till wedge beneath a diamictite at 364 mbsf the mid-Oligocene section represents the oldest evidence of grounded ice in CRP-2/2A. Shear zones with a subglacial origin in the early late Oligocene and early Miocene sections of the core are evidence of further grounding events. The interpretation of sediment deformation in CRP-2/2A is compared to other Antarctic stratigraphic records and global eustatic change between the late Eocenel/early Oligocene and the middle Miocene.