A magneto-mineralogical study of basalt at DSDP Leg 73 Holes

The basalts in Holes 519A, 522B, and 524 were studied for intensity of natural remanent magnetization, magnetic hysteresis, magnetic susceptibility, stability of isothermal remanence, and thermomagnetic behavior. Some of these properties are sensitive to both the composition and the microstructure of the magnetic minerals, others to composition only. Thus it is possible to separate the two effects and to trace the variation of effective magnetic grain size and degree of alteration within a lithologic unit or over a yet larger distance or time interval. The flow in Hole 519A is highly maghemitized at the top, the degree of maghemitization decreasing with depth in the flow. Effective grain size increases with increasing depth. Electron microprobe analysis of the titanomaghemite grains in these samples provides no support for the leaching out of iron during alteration. The pillows and flows in Hole 522B are distributed among a number of cooling units, and no systematic downhole variations are apparent. The inferred magneto-petrology is consistent with the cooling and alteration history that might be expected within the units. The upper and lower sills in Hole 524 are more uniform and have a larger concentration of well-developed magnetic mineral grains than the pillows and flows in Holes 519A and 522B. Maghemitization appears to have developed from the boundaries of the sills that are in contact with the sediments between the sills.

(Table 1) Apparent ages from 40Ar/39Ar heating steps for DSDP Hole 72-516F basalts

Submarine basalts are difficult to date accurately by the potassium-argon method. Dalrymple and Moore (1968) and Dymond (1970), for example, showed that, when the conventional K-Ar method is used, pillow lavas may contain excess 40Ar. Use of the 40Ar/39Ar step-heating method has not overcome the problem, as had been hoped, and has produced some conflicting results. Ozima and Saito (1973) concluded that the excess 40Ar is retained only in high temperature sites, but Seidemann (1978) found that it could be released at all temperatures. Furthermore, addition of potassium, from seawater, to the rock after it has solidified can result in low ages (Seidemann, 1977), the opposite effect to that of excess 40Ar. Thus, apparent ages may be either greater or less than the age of extrusion. Because of this discouraging record, the present study was approached pragmatically, to investigate whether self-consistent results can be obtained by the 40Ar/39Ar step-heating method.

Model parameter

Interannual environmental variability in Peru is dominated by the El Niño Southern Oscillation (ENSO). The most dramatic changes are associated with the warm El Niño (EN) phase (opposite the cold La Niña phase), which disrupts the normal coastal upwelling and affects the dynamics of many coastal marine and terrestrial resources. This study presents a trophic model for Sechura Bay, located at the northern extension of the Peruvian upwelling system, where ENSO-induced environmental variability is most extreme. Using an initial steady-state model for the year 1996, we explore the dynamics of the ecosystem through the year 2003 (including the strong EN of 1997/98 and the weaker EN of 2002/03). Based on support from literature, we force biomass of several non-trophically-mediated 'drivers' (e.g. Scallops, Benthic detritivores, Octopus, and Littoral fish) to observe whether the fit between historical and simulated changes (by the trophic model) is improved. The results indicate that the Sechura Bay Ecosystem is a relatively inefficient system from a community energetics point of view, likely due to the periodic perturbations of ENSO. A combination of high system productivity and low trophic level target species of invertebrates (i.e. scallops) and fish (i.e. anchoveta) results in high catches and an efficient fishery. The importance of environmental drivers is suggested, given the relatively small improvements in the fit of the simulation with the addition of trophic drivers on remaining functional groups' dynamics. An additional multivariate regression model is presented for the scallop Argopecten purpuratus, which demonstrates a significant correlation between both spawning stock size and riverine discharge-mediated mortality on catch levels. These results are discussed in the context of the appropriateness of trophodynamic modeling in relatively open systems, and how management strategies may be focused given the highly environmentally influenced marine resources of the region.