No abstract

The coast of Ceará State in NE Brazil is covered by vast fields of active and stabilized coastal sand dunes. Its tropical climate is characterized by two seasons, wet and dry, with wind intensity determined by the meridional shift of the Intertropical Convergence Zone. The wind power is negatively correlated with precipitation, and precipitation is negatively correlated with the difference between sea surface temperatures of the tropical Atlantic north and south of the equator. We present a model suggesting that during the Late Pleistocene wind power determined the mobility and stability of the dunes. Sand dunes accumulated during periods of high wind power (as it is today) and stabilized when wind power was low. Once the dunes were stabilized by vegetation they could not be activated even by increased wind power. Samples that were taken for luminescence dating from 25 stabilized dunes along the coasts of Ceará gave ages ranging from135 ka to < 100 yr. We postulate that these luminescence ages fall at the beginning of wet periods in NE Brazil characterized by low wind power. These paleoclimatic wet periods correlate well with the cold periods of stades in Greenland ice-core records.

Residual dune ridges are often formed by vegetation growing along a line some distance upwind of the lower stoss slope of migrating dunes. This process is common in areas where vegetation germinates along the edge of the water during the rainy period when the water level is higher and interdune areas are flooded. The phenomenon occurs on a large scale in North-east Brazil, because of the rise and fall in groundwater level at the end of the rainy season. Each residual dune ridge corresponds to the position of the dune during the wet period in each year. Therefore, variations in the distance between these residual dune ridges could be used potentially to monitor climatic fluctuations in rainfall and wind. To examine the potential use of these residual dune ridges for the reconstruction of past climatic fluctuations, a model that simulates them under varying conditions of wind, rainfall and evaporation rates was formulated. The model was tested for sensitivity to climatic variability in North-east Brazil and validated against residual dune ridge displacements as measured in the field and from high spatial resolution satellite images. Based on the results, it is concluded that residual dune ridges may not form in North-east Brazil in years which are exceptionally dry, as may happen during El-Niñ o events. When this type of event happens, the distance between adjacent residual dune ridges corresponds to more than one year and, therefore, the correlation between dune displacements and wind power becomes weak or even disappears. Additionally, because of biotic, aeolian and hydrological processes, these arcuate residual dune ridges may not preserve their initial shape for long periods. The presence of residual dune ridges testifies to the temporary flooding which may or may not be seasonal. However, the potential for using residual dune ridges to reconstruct the palaeo-climate of wind regime on a yearly basis or to identify past El-Niñ o events seems to be limited.

No abstract

Transforming surveyed elevations and water depths to desired vertical datums is an essential step in building a regional coastal management plan. Regional coastal management plans are based on sediment volume changes and numerical simulations of regional coastal change. Computation of sediment volume changes are possible only if the survey data sets compared share the same vertical datum. Some numerical simulations of regional coastal change require a baseline data set that is referenced to a particular stage of the tide.Until recently, hydrographic and topographic surveys covered areas that were sufficiently small to require only a simple vertical shift to convert the survey data to the desired vertical datum based on local established benchmarks. Data sets that cover large areas are now available through rapid survey techniques like airborne lidar, and through digital publishing of data, like that found on nautical charts. These data sets are not easily converted to a common datum. The magnitude of this problem for regional applications is being recognized only now. The vertical location of tidal, geodetic, and ellipsoidal datums can vary widely over the large areas that these data sets cover. The datums are derived at discrete points distributed sparsely through an area. This paper outlines methodologies for developing and applying regional datum conversions. The methods presented are designed both to realistically represent vertical datums as surfaces instead of discrete points within a region and to minimize error in volume computations and numerical simulations for regional coastal management.