• Rural Digital Europe
• US close
• Journal of Geophysical Research Atm... close

We present the results of a detailed study of the high‐pressure behavior of anorthite (CaAl2Si2O8) between ambient pressure and 30 GPa under static conditions, using in situ Raman spectroscopy and energy‐dispersive X ray diffraction. On increasing pressure under hydrostatic conditions, the transition occurs first at 2.6 GPa. At 10 GPa, a reversible polymorphic transition is observed which transforms the polymorph into a phase of higher symmetry. With further pressure increase, large changes occur in the Raman and X ray spectra between 14 and 16 GPa, premonitory to the onset of pressure‐induced amorphization. Above 16 GPa, anorthite is fully amorphous. However, only samples pressurized to above 22 GPa remain amorphous on recovery to ambient conditions. The high‐pressure behavior of anorthite is highly sensitive to deviatoric stresses. Under less hydrostatic conditions, the 10‐GPa transition described above occurs below 9 GPa, amorphization begins below 11.2 GPa, and samples decompressed from peak pressures as low as 17 GPa are fully amorphous. We have also carried out a thermodynamic analysis of the pressure‐induced amorphization of anorthite, using a two state model for the equation of state (EOS) of CaAl2Si2O8 glass in order to account for the increase in Al coordination. With this EOS, we calculate that the free energies of the crystalline and amorphous materials are equal at about 30 GPa, or slightly lower when effects of structural relaxation in the glass are taken into account, well above the observed crystalline‐amorphous transition at 16 GPa.

Recent studies have noted an asymmetrical climate change across Antarctica, with significant warming in West Antarctica and the Antarctica Peninsula, and primarily insignificant trends in East Antarctica. Due to its proximity, variations in the position and intensity of the Amundsen‐Bellingshausen Seas Low (ABSL) are a suspected atmospheric mechanism. Here, we investigate the ABSL to understand its characteristic variability and underlying synoptic‐scale influences, based on three reanalysis data sets. The ABSL is defined as the minimum monthly pressure in the 45°–75°S, 180°–60°W domain. Using this criterion, a significant north‐south and east‐west progression is noted in the climatological (1979–2001 average) ABSL, which is strongly tied to the location of the maximum cyclone system density and minimum cyclone central pressures. More than 550 cyclones a year were identified in the vicinity of the ABSL; during spring, significant trends in their central pressures are noted in the Ross Sea. The implied changes in temperature advection by these stronger systems are consistent with the warming in West Antarctica. The strongest cyclone events (i.e., the ten with the deepest central pressures) also demonstrate a connection to the climatological ABSL, albeit weaker. Moreover, these strong cyclone events are significantly linked to the Southern Annular Mode (SAM), particularly in their annual frequency and location/steering in the summer. This shows that large‐scale forcing, such as from the SAM, may influence the strongest cyclones in the region and could allow for the prediction of such events.

Recent measurements of the charge states of low-energy (about 100 keV/nucleon) solar cosmic rays at 1 AU are discussed. The measurements are consistent with models involving charge equilibrium with neutral matter at the sun only if the particles lose about 90% of their energy owing to adiabatic deceleration in the solar wind. Such an energy loss is shown to be possible only if the diffusion coefficient for 1-MeV/nucleon particles is smaller than 100 quintillion sq cm per sec. The implications of these results for models of solar-cosmic-ray acceleration are discussed.

While halogens from chlorofluorocarbons (CFCs) dominated past ozone changes, our simulations show climate change playing an increasingly important role over coming decades. Including potential climate‐induced stratospheric water vapor increases, the ozone change relative to 1980 attributable to climate change surpasses that due to halogens in the 2030s for both the upper stratosphere and the total column. Its overall impact may delay recovery of total ozone to 1980 values. These results emphasize that ozone recovery is not only a detection issue but also a detection and attribution issue, as trends must be ascribed to climate and halogens. Exploiting their different spatial patterns, we demonstrate a technique to isolate the halogen signature in the upper stratosphere by contrasting equatorial and midlatitude ozone trends at 40‐km altitude. This may allow early detection of the effectiveness of halogen emission reductions, which would show that a delayed recovery does not represent a CFC treaty failure but rather highlights the growing influence of greenhouse gases on the Earth's atmosphere.

In recent years it has been recognized that the level of shear and normal stress along a fault can vary; thus the stress is spatially and temporally inhomogeneous. Moreover, it has also been suspected that faults might interact in some way, with the result that a variety of earthquake magnitudes might be produced along a given length of fault at varying times. In order to explore these ideas we have developed a quantitative formalism, which we call the interaction matrix method, to express the influence of one fault upon another. This matrix is calculated by use of the energy change for a system of interacting cracks or faults and therefore gives energy‐consistent results. Specifically, the interaction matrix relates the area‐averaged stress on the fault segment to the area‐averaged slip state on all the other fault segments in the system. Since any fault can be subdivided into an arbitrary number of fault segments, the interaction matrix can have arbitrary dimension; in fact, the continuum limit is recovered as the dimension of the matrix approaches infinity. We combine this matrix method with a segmentation, or “patch,” model for earthquakes, in which each discrete segment of a fault has the same coseismic stress change (defined as the difference between the driving stress at which healing occurs minus the driving stress at which sliding starts) each time it slips. We show that slip on a patch during an earthquake can vary substantially, depending on how it interacts with other nearby patches. In this model it is quite possible for the spatial distribution of stress on the fault following an event to be again in a spatially inhomogeneous state, rather than in a uniform state, as is often assumed. Hence the seismic moment produced by an earthquake on a given set of patches can vary substantially, depending on the sequence of sliding and healing on the different patches. To apply these ideas, we devised a means to calculate the interaction matrix elements and used them to quantitatively examine earthquake sequences off the Colombia‐Ecuador coast and in the Nankai Trough near Japan.

In order to test Community Multiscale Air Quality (CMAQ) model performance for ultrafine particle concentrations in the Pacific Northwest, CMAQ v4.4 was modified for ternary NH3‐H2SO4‐H2O nucleation and for atmospheric processing of ultrafine particles. Sulfuric acid from sulfur dioxide oxidation is iteratively partitioned into gaseous sulfuric acid, newly condensed aerosol sulfate, and aerosol sulfuric acid contained in new 1 nm particles. Freshly nucleated particles are either coagulated to larger particles or grown by sulfuric acid condensation to 10 nm at which point they are included in CMAQ's existing Aitken mode. Multiple nucleation parameterizations were implemented into CMAQ, and one other was investigated in a sensitivity analysis. For a case study in the Pacific Northwest where aerosol number concentration and size distributions were measured, standard binary nucleation in CMAQ produces nearly no particles for this case study. Ternary nucleation can produce millions of 1 nm particles per cm3, but few of these particles survive coagulation loss and grow to 10 nm and into the Aitken mode. There are occasions when the additions to CMAQ increase the number of particles to within an order of magnitude of observations, but it is more common for number concentrations to remain underpredicted by, on average, one order of magnitude. Significant particle nucleation in CMAQ successfully produces a distinct Aitken and accumulation mode and an Aitken mode that is more prominent than the accumulation mode, although errors in the size distribution remain. A more recent ternary nucleation scheme including ammonium bisulfate clusters does not nucleate an appreciable number of particles.

Time‐mean abyssal current observations over the flanks of the East Pacific Rise (EPR) at 9–10°N and at the Juan de Fuca Ridge at 45°N document the occurrence of paired along‐ridge current jets that are trapped to the ridge flanks and sheared across the ridge in an anticyclonic sense. A coincident feature, where local hydrothermal discharge effects are not in play, is the upward bowing of isopycnals over ridge crests and isopycnals plunging into ridge flanks. It would be tempting to explain the jets primarily as geostrophic responses to the doming/plunging isopycnal distribution, though that should lead to the question as to how the isopycnal perturbations originate. A numerical model of time‐dependent flow on a cross‐ridge (x–z) transect, forced in a way to be consistent with a yearlong, hourly sampled record of currents measured at the EPR ridge crest, is used to investigate some of the underlying physics. It will be shown that the jets can arise from oscillatory flows via eddy‐momentum, eddy‐heat, and eddy‐salt fluxes that ultimately cause the isopycnals to dome over the ridge. As the probable offspring of velocity‐velocity and velocity‐density correlations that depend upon oscillatory motion, the jets are likely examples of stratified topographic flow rectification. An ancillary feature is a slight yearlong‐averaged downward current O (0.1–0.5 mm/s) over the EPR ridge crest that crosses the time‐mean, upward bowing isopycnals in a counter‐intuitive vertical direction.

Over the period July 4 to October 10, 1978, the SEASAT synthetic aperture radar (SAR) gathered 23 cm wavelength radar images of some 108 km2 of the earth's surface, mainly of ocean areas, at 25–40 m resolution. Our assessment is in terms of oceanographic and ocean monitoring objectives and is directed toward discovering the proper role of SAR imagery in these areas of interest. In general, SAR appears to have two major and somewhat overlapping roles: first, quantitative measurement of ocean phenomena, like long gravity waves and wind fields, as well as measurement of ships; second, exploratory observations of large‐scale ocean phenomena, such as the Gulf Stream and its eddies, internal waves, and ocean fronts. These roles are greatly enhanced by the ability of 23 cm SAR to operate day or night and through clouds. To begin we review some basics of synthetic aperture radar and its implementation on the SEASAT spacecraft. SEASAT SAR imagery of the ocean is fundamentally a map of the radar scattering characteristics of ∼30 cm wavelength ocean waves, distorted in some cases by ocean surface motion. We discuss how wind stress, surface currents, long gravity waves, and surface films modulate the scattering properties of these resonant waves with particular emphasis on the mechanisms that could produce images of long gravity waves. Doppler effects by ocean motion are also briefly described. Measurements of long (wavelength ≳100 m) gravity waves, using SEASAT SAR imagery, are compared with surface measurements during several experiments. Combining these results we find that dominant wavelength and direction are measured by SEASAT SAR within ±12% and ±15°, respectively. However, we note that ocean waves are not always visible in SAR images and discuss detection criteria in terms of wave height, length, and direction. SAR estimates of omnidirectional wave height spectra made by assuming that SAR image intensity is proportional to surface height fluctuations are more similar to corresponding surface measurements of wave height spectra than to wave slope spectra. Because SEASAT SAR images show the radar cross section σ° of ∼30 cm waves (neglecting doppler effects), and because these waves are raised by wind stress on the ocean surface, wind measurements are possible. Comparison between wind speeds estimated from SEASAT SAR imagery and from the SEASAT satellite scatterometer (SASS) agreed to within ±0.7 m s− over a 350‐km comparison track and for wind speeds from 2 to 15 m s−. The great potential of SAR wind measurements lies in studying the spatial structure of the wind field over a range of spatial scales of from ≲1 km to ≳100 km. At present, the spatial and temporal structure of ocean wind fields is largely unknown. Because SAR responds to short waves whose energy density is a function of wind stress at the surface rather than wind speed at some distance above the surface, variations in image intensity may also reflect changes in air‐sea temperature difference (thus complicating wind measurements by SAR). Because SAR images show the effects of surface current shear, air‐sea temperature difference, and surface films through their modulation of the ∼30 cm waves, SEASAT images can be used to locate and study the Gulf Stream and related warm water rings, tidal flows at inlets, internal waves, and slicks resulting from surface films. In many of these applications, SAR provides a remote sensing capability that is complementary to infrared imagery because the two techniques sense largely different properties, namely, surface roughness and temperature. Both stationary ships and moving ships with their attendant wakes are often seen in SAR images. Ship images can be used to estimate ship size, heading, and speed. However, ships known to be in areas imaged by SAR are not always detectable. Clearly, a variety of factors, such as image resolution, ship size, sea state, and winds could affect ship detection. Overall, the role of SAR imagery in oceanography is definitely evolving at this time, but its ultimate role is unclear. We have assessed the ability of SEASAT SAR to measure a variety of ocean phenomena and have commented briefly on applications. In the end, oceanographers and others will have to judge from these capabilities the proper place for SAR in oceanography and remote sensing of the ocean.

Throughout February and March of 1997 Okmok Volcano, in the eastern Aleutian Islands of Alaska, erupted a 6 km long lava flow of basaltic 'a'a within its caldera. A numerical model for lava flow cooling was developed and applied to the flow to better understand the nature of its cooling. Radiation and convection from the surface, as well as conduction to the ground, were used to transport the flow's heat to its surroundings in the model. Internally, a conduction‐only approach moved heat from the interior outward. Vesiculation, latent heat generation, and thermal conductivity changes with temperature are among the other factors that were dynamically accounted for. Results indicate that ambient temperature fluctuations, on the scale of days to weeks, must be taken into account to create an accurate short‐term prediction of lava surface temperature. Daily data of rainfall and ambient temperature, as opposed to yearly averages, greatly increased the accuracy of the model. Furthermore, convective cooling of the lava surface was observed to be a dominant heat loss process during the first 200 days, indicating the convective heat transfer coefficient is a prime determinant of the accuracy of the model for predicting surface temperatures. Over a longer cooling period (2 years), thermal conductivity and porosity proved to be among the dominating factors for heat loss because of the limiting role of conductive heat flow in the interior. The model's flexibility allows application to flows other than the 1997 Okmok eruption.