Snow measurement techniques for land-surface-atmosphere exchange studies in boreal landscapes

Summary  Snow has been studied widely in hydrology for many decades whereas recent meteorological interest in snow is caused by increased emphasis on high latitudes and wintertime in climate-change research as well as by the need to improve weather-forecast models during these conditions. Ground-based measurements of snow properties are needed both to improve understanding of surface-atmosphere exchange processes and to provide ground truth to new remote-sensing algorithms. This justifies a review of techniques to measure snow in combination with establishment of criteria for the suitability of the methods for process studies. This review assesses the state-of-art in ground-based snow-measurement techniques in the end of the 1990s in view of their accuracy, time resolution, possibility to automate, practicality and suitability in different terrain. Methods for snow-pack water equivalent, depth, density, growth, quality, liquid-water content and water leaving the snow pack are reviewed. Synoptic snow measurements in Fennoscandian countries are widely varying and there is no single standard on which process-related studies can build. A long-term, continuous monitoring of mass and energy properties of a snow cover requires a combination of point-measurement techniques. Areally representative values of snow properties can be achieved through a combination of automatically collected point data with repeated manual, areally covering measurements, remote-sensing data and digital elevation models, preferably in a GIS framework. Received August 27, 1999     

A Model of Magmatic Crystallization

A computer model simulating fractional crystallization at oneatmosphere pressure incorporates nine broadly-defined minerals—magnetite,olivine, hypersthene, augite, quartz, plagioclase, orthoclase,leucite, and nepheline. The crystallization temperature of eachmineral is considered to be a smooth function of the compositionof the magmatic liquid. These mineral temperature equationsare obtained by multiple linear regression analysis of informationfrom published silicate systems and rock melting experiments.The nine equations are solved for any primary liquid, withinthe broad range of common magma types, to select the crystallizingmineral or minerals. Partition ratios from published experimentsand analyses of lavas and phenocrysts permit calculation ofthe composition of the crystallizing mineral assemblage. Subtractionof a small amount of that composition from the primary liquidyields a new liquid, which may be recycled to yield a sequenceof liquids during fractional crystallization. The crystallizationmodel handles assemblages of co-precipitating minerals, andcan trace progressive saturation in new minerals, substitutionof a new mineral for an old mineral, and cessation of crystallizationof a mineral. The sequences of minerals and liquids derivedfrom a broad set of primary liquids are geologically realistic,so the model is useful in predicting phenocrysts in volcanicrocks and events during crystallization of shallow intrusions.