Identifying degrees of freedom in pushbroom bundle adjustment

An analytical method of detecting and enumerating degrees of freedom (DOF) in bundle adjustment (BA) is developed, and used to analyze several scene geometries for pushbroom cameras. While it is well recognized in the photogrammetric community that BA can be poorly constrained, especially with pushbroom imagery, the literature is absent of techniques to study the DOF in BA except through numerical analysis, which does not delineate them except through the suggestion of their existence in degraded results. The analytical methods presented here provide insight into the DOF. The method is based on finding the singular values, and the associated singular directions, of a Hessian matrix from the minimization process in BA. These directions are scrutinized as possible DOF along which the minimal cost of BA will not change from the true solution. To demonstrate the use of the methods six different scene geometries are analyzed, some of which are commonly employed by current remote sensing satellites. Several of the scene geometries are shown to have multiple DOF beyond the well known scaling and absolute position and orientation that are not recoverable from stereo imagery without exterior orientation control of the cameras or ground control points. As a simple example, the effects of these DOF images from the HiRISE camera on the Mars Reconnaissance Orbiter are used along with the associated ephemeris data to demonstrate possible distortions in the results of a terrain reconstruction problem.     

A photometric and spectroscopic study of WW And – an Algol-type,long period binary system with an accretion disc

We have analysed the available spectra of WW And and for the first time obtained a reasonably well defined radial velocity curve of the primary star. Combined with the available radial velocity curve of the secondary component, these data led to the first determination of the spectroscopic mass ratio of the system at q_spec = 0.16 ± 0.03. We also determined the radius of the accretion disc from analysis of the double-peaked H_α emission lines. Our new, high-precision, Johnson VRI and the previously-available Strömgren vby light curves were modelled with stellar and accretion disc models. A consistent model for WW And – a semidetached system harbouring an accretion disc which is optically thick in its inner region, but optically thin in the outer parts – agrees well with both spectroscopic and photometric data.     

Local quasigeoid model creation from astrogeodetic measurements

This article describes the process of creation and testing of a local quasigeoid model using the astrogeodetic method known as astronomical levelling. The process used here was based on astronomical levelling principles combined with the least square adjustment in a triangular network and a common method of surface generation. Using this method, the authors have created a quasigeoid model of a small portion of the city of Brno. This model covers an area of approximately 1 × 2 km and is based on the astronomically determined vertical deflection at 11 stations. The average distance between the astrogeodetic stations was 500 m, which is an unusually high density (5.5 stations per km²). This high geo-spatial data density input made it possible to generate a quasigeoid model of height difference precision at mm-level over few km. Tests described in this article document the suitability of our methods for creating local quasigeoid models of high precision and resolution. Employing the least square adjustment in a planar network offers the possibility to easily compute standard deviations of both input and result values. This is a great advantage in comparison with more common astrogeodetic quasigeoid profiles, which are not suitable for simple adjustment and require more complex methods to be used for evaluation of their precision. The model described here serves the authors as a technological example, from which they learn more about the potential of the astrogeodetic method. Astrogeodetic models of a much greater extent are planned to be used for validation purposes of models generated by other independent methods (gravimetric, satellite, combined, etc.).     

Holocene marine terraces on two salt diapirs in the Persian Gulf,Iran: age,depositional history and uplift rates

Slightly inclined Holocene marine terraces cover parts of two circular salt diapirs (Hormoz and Namakdan) in the Persian Gulf. Their relative altitude above present sea level results from a combination of general marine transgression/regression affecting the whole area, and of local uplift related to salt diapirism. Differential uplift rate of the studied diapirs in centre‐to‐rim profiles was calculated from results based on: (i) radiocarbon ages of skeletal remains of benthic faunas (19 samples), which originally grew close to sea level; (ii) original altitude of samples, estimated from general sea‐level oscillation curves for the last 10 kyr, and (iii) present sample altitude measured in the field. Calculated uplift rates increase from rim to centre on both diapirs in the range from: 2 mm yr^?1 at the rim to 5–6 mm yr^?1 at the interior of Hormoz, and 1–3 mm yr^?1 at the rim to 3–5 mm yr^?1 at the interior of Namakdan. Such uplift rate distributions fit into the parabolic profile of Newtonian fluid rather than to profiles typical for pseudoplastic fluids. The increase in uplift rate with distance from rim to centre of diapirs is gradual as demonstrated also by generally smooth surface of marine terraces. No tectonic dissections were found. The depositional history on both salt diapirs is similar although they are situated more than 100 km apart. Marine sedimentation started at about 9.6k cal. yr BP on Hormoz and at 8.6k cal. yr BP on Namakdan. Owing to rapid transgression, the sea partially truncated both salt diapirs and rapidly deepened, and carbonate mud was deposited on the peripheries of both salt diapirs. Between 7 and 5k cal. yr BP beach deposition replaced carbonate mud. Soon after 5k cal. yr BP, the sea retreated from most of the marine terraces on both salt diapirs. Copyright © 2006 John Wiley & Sons, Ltd.     

A study of erosion rates on salt diapir surfaces in the Zagros Mountains, SE Iran

Salt exposures and weathering residuum on several salt diapirs in different geographic/climatic settings were studied. Anhydrite, gypsum, hematite, calcite, dolomite, quartz, and clay minerals are the main constituents of the weathering residuum covering the salt diapirs in various thicknesses. Erosion rates of residuum as well as of rock salt exposures were measured at selected sites for a period of 5 years by plastic pegs as benchmarks. Recorded data were standardized to a horizontal surface and to long-term mean precipitation. For the rock salt exposures the following long-term denudation rates were determined of 30–40 mm a⁻¹ for coastal diapirs and up to 120 mm a⁻¹ for mountain salt diapirs. Long-term mean superficial denudation rate measured on weathering residuum of low thickness reached 3.5 mm a⁻¹ on coastal diapirs. The total denudation rate estimated for the thin residuum is close to 4–7 mm a⁻¹ based on apparent correlation with the uplift rate on Hormoz and Namakdan diapirs. Denudation of rock salt exposures is much faster compared to parts of diapirs covered by weathering residuum. The extent of salt exposures is an important factor in the morphological evolution of salt diapirs as it can inhibit further expansion of the diapir. Salt exposures produce huge amounts of dissolved and clastic load, thus affecting the surrounding of the diapir.     

Efficient road geometry identification from digital vector data

A new method for the automatic identification of road geometry from digital vector data is presented. The method is capable of efficiently identifying circular curves with their radii and tangents (straight sections). The average error of identification ranged from 0.01 to 1.30 % for precisely drawn data and 4.81 % in the case of actual road data with noise in the location of vertices. The results demonstrate that the proposed method is faster and more precise than commonly used techniques. This approach can be used by road administrators to complete their databases with information concerning the geometry of roads. It can also be utilized by transport engineers or traffic safety analysts to investigate the possible dependence of traffic accidents on road geometries. The method presented is applicable as well to railroads and rivers or other line features.     

Vertical and horizontal spheroidal boundary-value problems

Vertical and horizontal spheroidal boundary-value problems (BVPs), i.e., determination of the external gravitational potential from the components of the gravitational gradient on the spheroid, are discussed in this article. The gravitational gradient is decomposed into the series of the vertical and horizontal vector spheroidal harmonics, before being orthogonalized in a weighted sense by two different approaches. The vertical and horizontal spheroidal BVPs are then formulated and solved in the spectral and spatial domains. Both orthogonalization methods provide the same analytical solutions for the vertical spheroidal BVP, and give distinct, but equivalent, analytical solutions for the horizontal spheroidal BVP. A closed-loop simulation is performed to test the correctness of the analytical solutions, and we investigate analytical properties of the sub-integral kernels. The systematic treatment of the spheroidal BVPs and the resulting mathematical equations extend the theoretical apparatus of geodesy and of the potential theory.     

Early and middle Holocene in the Aegean Sea: interplay between high and low latitude climate variability

Changes in the orbital parameters, solar output, and ocean circulation are widely considered as main drivers of the Holocene climate. Yet, the interaction between these forcings and the role that they play to produce the pattern of changes observed in different domains of the climate system remain debated. Here, we present new early to middle Holocene season-specific sea surface temperature (SST) and δ¹⁸O_seawater results, based on organic-walled dinoflagellate cyst and planktonic foraminiferal data from two sediment cores located in the central (SL21) and south-eastern (LC21) Aegean Sea (eastern Mediterranean). Today, this region is affected by high to mid latitude climate in winter and tropical/subtropical climate in summer. The reconstructed δ¹⁸O_seawater from LC21 displays a marked (～1.3%) negative shift between 10.7 and 9.7 ka BP, which represents the regional expression of the orbitally driven African monsoon intensification and attendant freshwater flooding into the eastern Mediterranean. A virtually contemporaneous shift, of the same sign and magnitude, is apparent in the δ¹⁸O_speleothem record from Soreq Cave (Northern Israel), an important part of which may therefore reflect a change in the isotopic composition of the moisture source region (Aegean and Levantine Seas). Our SST reconstructions show that Aegean winter SSTs decreased in concert with intensifications of the Siberian High, as reflected in the GISP2 nss [K⁺] record. Specifically, three distinct sea surface cooling events at 10.5, 9.5–9.03 and 8.8–7.8 ka BP in the central Aegean Sea match increases in GISP2 nss [K⁺]. These events also coincide with dry interludes in Indian monsoon, hinting at large (hemispheric) scale teleconnections during the early Holocene on centennial timescales. A prominent short-lived (～150 years) cooling event in core SL21 – centred on 8.2 ka BP – is coeval to the ‘8.2 ka BP event’ in the Greenland δ¹⁸O_ice, which is commonly linked to a melt-water related perturbation of the Atlantic Meridional Overturning Circulation and associated ocean heat transport. By deciphering the phasing between a recently published record of reduced overflow from the Nordic Seas into the northern North Atlantic, the Greenland δ¹⁸O_ice ‘8.2 ka BP event’ anomaly, and the short-lived cooling in SL21, we demonstrate severe far-field impacts of this North Atlantic event in the Aegean Sea. The Aegean is isolated from the North Atlantic oceanic circulation, so that signal transmission must have been of an atmospheric nature.     

KLENOT Project 2002–2008 contribution to NEO astrometric follow‐up

Abstract— Near‐Earth object (NEO) research plays an increasingly important role not only in solar system science but also in protecting our planetary environment as well as human society from the asteroid and comet hazard. Consequently, interest in detecting, tracking, cataloguing, and the physical characterizing of these bodies has steadily grown. The discovery rate of current NEO surveys reflects progressive improvement in a number of technical areas. An integral part of NEO discovery is astrometric follow‐up crucial for precise orbit computation and for the reasonable judging of future close encounters with the Earth, including possible impact solutions. The KLENOT Project of the Klet Observatory (South Bohemia, Czech Republic) is aimed especially at the confirmation, early follow‐up, long‐arc follow‐up, and recovery of near‐Earth objects. It ranks among the world's most prolific professional NEO follow‐up programs. The 1.06 m KLENOT telescope, put into regular operation in 2002, is the largest telescope in Europe used exclusively for observations of minor planets and comets, and full observing time is dedicated to the KLENOT team. In this paper, we present the equipment, technology, software, observing strategy, and results of the KLENOT Project obtained during its first phase from March 2002 to September 2008. The results consist of thousands of precise astrometric measurements of NEOs and also three newly discovered near‐Earth asteroids. Finally, we also discuss future plans reflecting also the role of astrometric follow‐up in connection with the modus operandi of the next generation surveys.