An intelligent geospatial processing unit for image classification based on geographic vector agents (GVAs)

Spatial modeling methods usually use pixels and image objects as fundamental processing units to address real‐world objects, geo‐objects, in image space. To do this, both pixel‐based and object‐based approaches typically employ a linear two‐staged workflow of segmentation and classification. Pixel‐based methods segment a classified image to address geo‐objects in image space. In contrast, object‐based approaches classify a segmented image to identify geo‐objects from raster datasets. These methods lack the ability to simultaneously integrate the geometry and theme of geo‐objects in image space. This article explores Geographical Vector Agents (GVAs) as an automated and intelligent processing unit to directly address real‐world objects in the process of remote sensing image classification. The GVA is a distinct type of geographic automata characterized by elastic geometry, dynamic internal structure, neighborhoods and their respective rules. We test this concept by modeling a set of objects on a subset IKONOS image and LiDAR DSM datasets without the setting parameters (e.g. scale, shape information), usually applied in conventional Geographic Object‐Based Image Analysis (GEOBIA) approaches. The results show that the GVA approach achieves more than 3.5% improvement for correctness, 2% improvement for quality, although no significant improvement for completeness to GEOBIA, thus demonstrating the competitive performance of GVAs classification.     

Remote flare brightenings and type III reverse slope bursts

We present two large flares which were exceptional in that each produced an extensive chain of H emission patches in remote quiet regions more than 10⁵ km away from the main flare site. They were also unusual in that a large group of the rare type III reverse slope bursts accompanied each flare.The observations suggest that this is no coincidence, but that the two phenomena are directly connected. The onset of about half of the remote H emission patches were found to be nearly simultaneous with RS bursts. One of the flares (August 26, 1979) was also observed in hard X-rays; the RS bursts occurred during hard X-ray spikes. For the other flare (June 16, 1973), soft X-ray filtergrams show coronal loops connecting from the main flare site to the remote H brightenings. There were no other flares in progress during either flare; this, along with the X-ray observations, indicates that the RS burst electrons were generated in these flares and not elsewhere on the Sun. The remote H brightenings were apparently not produced by a blast wave from the main flare; no Moreton waves were observed, and the spatially disordered development of the remote H chains is further evidence against a blast wave. From geometry, time and energy considerations we propose: (1) That the remote H brightenings were initiated by direct heating of the chromosphere by RS burst electrons traveling in closed magnetic loops connecting the flare site to the remote patches; and (2) that after onset, the brightenings were heated by thermal conduction by slower thermal electrons (kT1 keV) which immediately follow the RS burst electrons along the same loops.     

Study of the post-flare loops on 29 July 1973

We present revised values of temperature and density for the flare loops of 29 July 1973 and compare the revised parameters with those obtained aboard the SMM for the two-ribbon flare of 21 May 1980. The 21 May flare occurred in a developed sunspot group; the 29 July event was a spotless two-ribbon flare. We find that the loops in the spotless flare extended higher (by a factor of 1.4–2.2), were less dense (by a factor of 5 or more in the first hour of development), were generally hotter, and the whole loop system decayed much slower than in the spotted flare (i.e. staying at higher temperature for a longer time). We also align the hot X-ray loops of the 29 July flare with the bright H ribbons and show that the H emission is brightest at the places where the spatial density of the hot elementary loops is enhanced.presently at Big Bear Solar Observatory, Caltech, California, U.S.A.presently at The Marshall Space Flight Center, Huntsville, Alabama, U.S.A.     

The effect of dissolved water on the oxidation state of iron in natural silicate liquids

Ferric-ferrous ratios have been measured in 22 experiments on three natural compositions equilibrated at known temperature (950°–1100° C) and oxygen fugacity, and at water-saturated conditions over a pressure range from 0.05 to 0.2 GPa. There does not appear to be any reaction between the melt and the capsule material that affects the redox state of the iron in the melt. An empirical expression for the anhydrous behavior of the redox state of iron in each of these compositions has also been determined at 1 bar as a function of temperature and oxygen fugacity. A direct comparison of the hydrous ferric-ferrous values with the calculated anhydrous values shows that the dissolution of water in a per-alkaline rhyolite, andesite, and an augite minette has no effect on the redox state of the iron in these melts. This result parallels the effect of water on sulfide speciation in basaltic melts, and confirms published results on experimental hydrous basalts.     

On the calibration and temperature behaviour of single-beam infrared hygrometers

The single-beam infrared hygrometer (Hyson and Hicks, 1975) is increasingly popular as a device capable of measuring atmospheric humidity fluctuation for use in eddy correlation studies. However, it is generally applied without recognition of the effect of temperature on its performance. Calibration data from a number of such instruments indicate that the resulting calibration coefficients vary with mean temperature as a direct consequence of temperature dependence in photodetector response. Fluctuations in atmospheric and instrumental temperatures can also introduce significant errors in the measurement of humidity fluctuations.This paper draws attention to the problem of temperature dependence in the calibration of these devices and provides a framework for investigating the magnitude of other errors involved in their use.     

Petrogenesis of the Elzevir batholith and related trondhjemitic intrusions in the grenville province of eastern Ontario,Canada

The Elzevir batholith belongs to a suite of trondhjemitic intrusions emplaced at ca. 1,240 Ma in the Grenville Province of eastern Ontario. New major and trace element data, including REE, combined with isotopic and petrographic data indicate that:

1. the batholith has calc-alkalic affinities;
2. the Elzevir parental magma is very similar to that of dacites in the nearby, coeval metavolcanic rocks; the magma formed by partial melting of crustal material at granulite grade;
3. chemical differences between the plutonic and volcanic rocks can be best explained by accumulation of plagioclase in the plutonic environment;
4. fractionation was dominated by plagioclase and quartz, with lesser biotite and epidote, and minor zircon and apatite.

It is suggested that melting of sialic crust took place during the ‘docking’ of a partly-evolved, originally ensimatic arc system against the main cratonic mass to the northwest.     

Estimates of integral time scales from a 100-M meteorological tower at a plains site

Four distinct types of autocorrelograms were observed using high-frequency vertical velocity data measured at 100 m above a flat terrain. Several types of nonstationary atmospheric motions due to low frequency fluctuations were examined. Under nighttime stable conditions, these phenomena were found to lead to abnormally slow exponential decay of the autocorrelation function. Several different techniques for estimating Eulerian integral time scales were compared in order to select an appropriate method of estimation. When grouped by stability classes, the Eulerian integral time scales decrease slightly with increasing stability, but generally exhibit no significant correlation with other meteorological parameters. Using a postulated relation, estimates of the Lagrangian to Eulerian integral time scale ratio range from 3 to 5 under unstable conditions and 15 to 25 under stable conditions. Under unstable conditions the average Lagrangian integral time scale is on the order of 40 s and exhibits no significant correlation with several pertinent meteorological parameters. Under stable conditions, the Lagrangian integral time scale correlates well with the Monin-Obukhov length and temperature lapse rate.Work conducted while a visiting scholar at Systems Applications, Inc.     

Ground water geochemistry of 228Ra, 226Ra and 222Rn

²²⁸Ra, ²²⁶Ra, and ²²²Rn activities were determined on over 150 ground water samples collected from drilled, public water supply wells throughout South Carolina. A wide range of aquifer lithologies were sampled including the crystalline rocks of the Piedmont and sedimentary deposits of the Coastal Plain. A significant linear relationship between log ²²⁸Ra and log ²²⁶Ra (n = 182, r = 0.83) was indistinguishable between Piedmont and Coastal Plain ground water. Median ${}^{228}\text{Ra}{}^{226}\text{Ra}$ activity ratios for the Piedmont, 1.2, and Coastal Plain, 1.3, ground water are close to estimated average crustal ${}^{232}\text{Th}{}^{238}\text{U}$ activity ratios of 1.2 to 1.5 corresponding to Th/U weight ratios of 3.5 to 4.5. A linear correlation was also found between log ²²²Rn and log ²²⁶Ra for Piedmont (n = 68, r = 0.62) and Coastal Plain (n = 89, r = 0.64) ground water. However, the median ${}^{222}\text{Rn}{}^{226}\text{Ra}$ activity ratio for Piedmont ground water, 6100, was much higher than for Coastal Plain ground water, 230. Higher excess ²²²Rn activities may be due to greater retention of ²²⁶Ra by the chemically active Piedmont aquifers compared to the more inert sand aquifers sampled in the Coastal Plain. The relationship between log ²²⁸Ra and log ²²⁶Ra was used to predict total Ra (${}^{228}\text{Ra +}{}^{226}\text{Ra}$) distributions in Appalachian and Atlantic and Gulf Coastal Plain ground water. Predictions estimate that 2.4% of Appalachian and 5.3% of Atlantic and Gulf Coastal Plain ground water supplies contain total Ra activities in excess of the 5 pCi/l limit established by the U.S. Environmental Protection Agency. These predictions also indicate that 40–50% of these ground water wells may be overlooked using the presently suggested screening activity of 3.0 pCi/l of ²²⁶Ra for ²²⁸Ra analysis.     

Chemistry and isotope ratios of sulfur in basalts and volcanic gases at Kilauea volcano,Hawaii

Eighteen basalts and some volcanic gases from the submarine and subaerial parts of Kilauea volcano were analyzed for the concentration and isotope ratios of sulfur. By means of a newly developed technique, sulfide and sulfate sulfur in the basalts were separately but simultaneously determined. The submarine basalt has 700 ± 100 ppm total sulfur with δ³⁴S_Σs of $\text{0.7 ± 0.1 ‰}$. The sulfate/sulfide molar ratio ranges from 0.15 to 0.56 and the fractionation factor between sulfate and sulfide is $\text{+7.5 ± 1.5‰}$. On the other hand, the concentration and δ³⁴S_Σs values of the total sulfur in the subaerial basalt are reduced to 150 ± 50 ppm and $\text{?0.8 ± 0.2‰}$, respectively. The sulfate to sulfide ratio and the fractionation factor between them are also smaller, 0.01 to 0.25 and +3.0‰, respectively. Chemical and isotopic evidence strongly suggests that sulfate and sulfide in the submarine basalt are in chemical and isotopic equilibria with each other at magmatic conditions. Their relative abundance and the isotope fractionation factors may be used to estimate the $\text{?o}{}_2$ and temperature of these basalts at the time of their extrusion onto the sea floor. The observed change in sulfur chemistry and isotopic ratios from the submarine to subaerial basalts can be interpreted as degassing of the SO₂ from basalt thereby depleting sulfate and ³⁴S in basalt.The volcanic sulfur gases, predominantly SO₂, from the 1971 and 1974 fissures in Kilauea Crater have δ³⁴S values of 0.8 to 0.9%., slightly heavier than the total sulfur in the submarine basalts and definitely heavier than the subaerial basalts, in accord with the above model. However, the δ³⁴S value of sulfur gases (largely SO₂) from Sulfur Bank is 8.0%., implying a secondary origin of the sulfur. The δ³⁴S values of native sulfur deposits at various sites of Kilauea and Mauna Loa volcanos, sulfate ions of four deep wells and hydrogen sulfide from a geothermal well along the east rift zone are also reported. The high δ³⁴S values (+5 to +6%._o) found for the hydrogen sulfide might be an indication of hot basaltseawater reaction beneath the east rift zone.