Principal components in active galactic nuclei variability data and the estimation of the flux contributions from different components

It has been found that the near-infrared flux variations of Seyfert galaxies satisfy relations of the form   F_i ≈α _{i j} +β _{i j} F_j   , where F_i , F_j are the fluxes in filters i and j ; and  α _{i , j} , β _{i , j}   are constants. These relations have been used to estimate the constant contributions of the non-variable underlying galaxies. The paper attempts a formal treatment of the estimation procedure, allowing for the possible presence of a third component, namely non-variable hot dust. In an analysis of a sample of 38 Seyfert galaxies, inclusion of the hot dust component improves the model fit in approximately half the cases. All derived dust temperatures are below 300 K, in the range 540–860 K or above 1300 K. A noteworthy feature is the estimation of confidence intervals for the component contributions: this is achieved by bootstrapping. It is also pointed out that the model implies that such data could be fruitfully analysed in terms of principal components.     

Geochemical exploration for platinum-group elements in the Bushveld Complex, South Africa

Analyses of stream sediment and soil samples from the Bushveld Complex, South Africa have revealed enhanced precious metal concentrations, which can be related both to mining activities and the presence of hidden concentrations of platinum-group elements (PGEs) and gold. The economically important PGE deposits hosted by the Upper Critical Zone of the Rustenburg Layered Suite are revealed by a high PGE and Au content in the overlying soils. A second zone of elevated precious metal concentrations straddles the boundary between the Main and Upper Zones and has to date been traced for more than 100 km. This zone follows the igneous layering of the Rustenburg Layered Suite and is offset by the Brits Graben. It is therefore thought to be the reflection of a magmatic PGE-Au mineralisation. Received: 31 May 1996 / Accepted: 7 January 1997     

Modelling forest canopy gaps using LiDAR-derived variables

Remote sensing has revolutionized forest management and has been widely employed to model canopy gaps. In this study, a canopy height model (CHM) and an intensity raster (IR) derived from light detection and ranging (LiDAR) data were used to model canopy gaps within a four-year-old Eucalyptus grandis forest using an object-based image analysis (OBIA) approach. Model thematic accuracies using the CHM, intensity raster and combined data set (CHM and IR) were all above 90%, with KHAT values ranging from 0.88 to 0.96. Independent test thematic accuracies were also above 90%, with KHAT values ranging from 0.82 to 0.91. A comparative area-based assessment yielded accuracies ranging from 70 to 90%, with the highest accuracies achieved using the combined data set. The results of this study show that using a CHM and intensity raster, and an OBIA approach, provides a viable framework to accurately detect and delineate canopy gaps within a commercial forest environment.     

Seasonally and spatially referenced estimates of recreational shore-angling effort,catch composition,catch rates and total catch in the Goukamma Marine Protected Area,South Africa

A roving creel survey of the recreational shore fishery along the 16.4-km coastline in the Goukamma Marine Protected Area on the south coast of South Africa was conducted from 2009 to 2011. Some 838 patrols were stratified equally among months, areas and years, but intentionally biased towards weekends. Angler densities at Buffalo Bay and Groenvlei were 0.59 and 0.28 anglers km?1, respectively. Weekend densities were double to quadruple weekday densities and fishing during winter was more popular than during summer. Area, habitat and distance to access points explained variation in angler densities. Shannon–Wiener diversity in catches declined from 2.18 in an earlier (1993–2002) survey to 1.79. Although the order of species abundance in the catches remained largely unchanged, blacktail Diplodus capensis dominance increased to 57.3% by number, at the expense of galjoen Dichistius capensis. Habitat explained 27% of the variance in catch composition. The catch per unit effort (CPUE) for the top nine species ranged from 0.19 to 6.35 fish 100-h?1. The CPUE of all species, except spotted grunter Pomadasys commersonnii, declined. Blacktail and galjoen CPUE declined by 17% and 77%, respectively. The total catch estimate was 2 986 fish y?1. Transgressions of size limits were common. The results suggest that the fishery is overexploited and that catch rates are declining.     

Perspectives on present-day sea level change: a tribute to Christian le Provost

In this paper, we first discuss the controversial result of the work by Cabanes et al. (Science 294:840–842, 2001), who suggested that the rate of past century sea level rise may have been overestimated, considering the limited and heterogeneous location of historical tide gauges and the high regional variability of thermal expansion which was supposed to dominate the observed sea level. If correct, this conclusion would have solved the problem raised by the IPCC third assessment report [Church et al, Cambridge University Press, Cambridge, pp 881, 2001], namely, the factor two difference between the 20th century observed sea level rise and the computed climatic contributions. However, recent investigations based on new ocean temperature data sets indicate that thermal expansion only explains part (about 0.4 mm/year) of the 1.8 mm/year observed sea level rise of the past few decades. In fact, the Cabanes et al.’s conclusion was incorrect due to a contamination of abnormally high ocean temperature data in the Gulf Stream area that led to an overestimate of thermal expansion in this region. In this paper, we also estimate thermal expansion over the last decade (1993–2003), using a new ocean temperature and salinity database. We compare our result with three other estimates, two being based on global gridded data sets, and one based on an approach similar to that developed here. It is found that the mean rate of thermosteric sea level rise over the past decade is 1.5±0.3 mm/year, i.e. 50% of the observed 3 mm/year by satellite altimetry. For both time spans, past few decades and last decade, a contribution of 1.4 mm/year is not explained by thermal expansion, thus needs to be of water mass origin. Direct estimates of land ice melt for the recent years account for about 1 mm/year sea level rise. Thus, at least for the last decade, we have moved closer to explaining the observed rate of sea level rise than the IPCC third assessment report.     

Land water storage contribution to sea level from GRACE geoid data over 2003–2006

Seasonal and inter-annual change in land water storage (expressed in terms of water volume change) over 27 large river basins worldwide are estimated from monthly GRACE geoids solutions computed at GFZ from February 2003 to February 2006. The largest annual water volume change is found in the Amazon basin, followed by the Parana, Ob, Orinoco, Tocantins, Niger, Congo, Ganges, Mekong, and Brahmaputra. In terms of trend over the 3-year period, positive and negative values are observed but in a number of cases computed trends are at the noise level. However significant negative trends are found in the Amazon, Ganges, Mississippi, Nile, Parana, and Zambezi basins, indicating water mass loss over that period. Positive trends (water mass gain) are marginally significant. We have computed the land water contribution to sea level change. On average over the 3-year time span, we find that the net effect is positive (net loss of water in terrestrial reservoirs), on the order of 0.19 +/− 0.06 mm/yr. If sustained over a longer time span than considered here, such a value may become comparable to the ice sheets contribution to sea level rise.     

Regional patterns of observed sea level change: insights from a 1/4° global ocean/sea-ice hindcast

A global eddy-admitting ocean/sea-ice simulation driven over 1958–2004 by daily atmospheric forcing is used to evaluate spatial patterns of sea level change between 1993 and 2001. In the present study, no data assimilation is performed. The model is based on the Nucleus for European Models of the Ocean code at the 1/4° resolution, and the simulation was performed without data assimilation by the DRAKKAR project. We show that this simulation correctly reproduces the observed regional sea level trend patterns computed using satellite altimetry data over 1993–2001. Generally, we find that regional sea level change is best simulated in the tropical band and northern oceans, whereas the Southern Ocean is poorly simulated. We examine the respective contributions of steric and bottom pressure changes to the total regional sea level changes. For the steric component, we analyze separately the contributions of temperature and salinity changes as well as upper and lower ocean contributions. Generally, the model results show that most regional sea level changes arise from temperature changes in the upper 750 m of the ocean. However, contributions of salinity changes and deep steric changes can be locally important. We also propose a map of ocean bottom pressure changes. Finally, we assess the robustness of such a model by comparing this simulation with a second simulation performed by MERCATOR-Ocean based on the same core model, but differing by its short length of integration (1992–2001) and its surface forcing data set. The long simulation presents better performance over 1993–2001 than the short simulation, especially in the Southern Ocean where a long adjustment time seems to be needed. In memory of my little brother Jean-Eudes, whose thirst for science filled out the rich discussions we had about my investigations and his job as user-service provider for MERCATOR-Ocean.     

Classification of marine bioregions on the east coast of South Africa

Marine bioregional planning requires a meaningful classification and spatial delineation of the ocean environment using biological and physical characteristics. The relative inaccessibility of much of the ocean and the paucity of directly measured data spanning entire planning regions mean that surrogate data, such as satellite imagery, are frequently used to develop spatial classifications. However, due to a lack of appropriate biological data, these classifications often rely on abiotic variables, which act as surrogates for biodiversity. The aim of this study was to produce a fine-scale bioregional classification, using multivariate clustering, for the inshore and offshore marine environment off the east coast of South Africa, adjacent to the province of KwaZulu-Natal and out to the boundary of the exclusive economic zone (EEZ), 200 nautical miles offshore. We used remotely sensed data of sea surface temperature, chlorophyll a and turbidity, together with interpolated bathymetry and continental-slope data, as well as additional inshore data on sediments, seabed oxygen and bottom temperature. A multivariate k-means analysis was used to produce a fine-scale marine bioregionalisation, with three bioregions subdivided into 12 biozones. The offshore classification was primarily a pelagic bioregionalisation, whereas the inshore classification (on the continental shelf) was a coupled benthopelagic bioregionalisation, owing to the availability of benthic data for this area. The resulting classification was used as a base layer for a systematic conservation plan developed for the province, and provided the methods for subsequent planning conducted for the entire South African EEZ. Validation of the classification is currently being conducted in marine research programmes that are sampling benthic biota and habitats in a sampling design stratified according to the biozones delineated in this study.     

An alternative to O−C analysis of variable star periods

For more than a century now astronomers have used the O−C (Observed minus Calculated) method to detect the presence of systematic changes in the periods of variable stars. The method is based on an analysis of residuals from a linear fit to the observed epochs. A rather common error in applications of the method is a failure to make provision for autocorrelation which exists in the data. In this paper we consider a model that accounts for the presence of autocorrelation and develop an alternative to the O−C method of analysis. The proposed method focuses on the frequency domain characteristics of observed periods. Its use is illustrated by application to data from the variable stars X Aurigae and RY Sagittarii.