A Method for Direct Assessment of the “Non Rainfall” Atmospheric Water Cycle: Input and Evaporation From the Soil

“Non rainfall” atmospheric water (dew, fog, vapour adsorption) supplies a small amount of water to the soil surface that may be important for arid soil micro-hydrology and ecology. Research into the direct effects of this water on soil is, however, lacking due to instrument and technical constraints. We report on the design, development, construction and findings of an automated microlysimeter instrument to directly measure this soil water cycle in Stellenbosch, South Africa during winter. Performance of the microlysimeter was satisfactory and results obtained were compared to literature and fell within the expected range. “Non rainfall” atmospheric water input into bare soil (river sand) was between 0.88 and 1.10?mm per night while evaporation was between 1.39 and 2.71?mm per day. The study also attempted to differentiate the composition of “non rainfall” atmospheric water and results showed that vapour adsorption contributed the bulk of this input.     

Estimation of daytime downward longwave radiation at the surface from satellite and grid point data

Summary A hybrid method is developed for the estimation of the daytime downward longwave radiation flux (DLF) at the surface. The method makes use of the grid point thermodynamic fields at the surface and at the 1000 and 850 hPa levels. The cloud parameters are derived from the infrared and visible image data of the satellite METEOSAT-2. The calculation of the DLF is split into a clear-sky contribution, which is calculated from empirical formulae, and a cloud contribution which depends on cloud amount, cloud base height, and temperature.A sensitivity test to perturbations in the relevant parameters resembles closely the response of a multispectral radiation scheme, however the humidity dependence under clear sky has been identified as a weak point. A comparison of the method using observed input parameters with simultaneous measurements of the DLF yields an RMS error of 12.2 W/m².The method is applied for mapping the DLF over western Europe and the Mediterranean using satellite images for two days near midday and ancillary grid point data obtained during the ALPEX experiment. A comparison with ground measurements of 4 stations in West Germany shows a mean deviation of 3.9 W/m² and an RMS error of 11.6 W/m². The method is a first attempt to estimate the DLF at regional scales from satellite data on the cloud field and grid point analysis data on the thermodynamic field.

---

Zusammenfassung Zur Schätzung des abwärts gerichteten langwelligen Strahlungsflusses (DLF) während des Tages an der Erdoberfläche wird eine Hybridmethode entwickelt. Die Methode verwendet Gitterpunkt-thermodynamische Felder am Boden und auf den 1000 und 850 h Pa-Flächen. Die Wolkenparameter stammen von den Bilddaten im infraroten und sichtbaren Bereich des Satelliten METEOSAT-2. Die Berechnung des DLF teilt sich in einen, aus empirischen Formeln berechneten Beitrag des wolkenlosen Himmels und einen Beitrag durch Wolken, welcher von der Wolkenmenge, von der Höhe der Wolkenbasis und der Temperatur abhängt.Ein Test der Empfindlichkeit auf Störungen in den relevanten Parametern ist der Reaktion eines multispektraligen Strahlungsschemas sehr ähnlich, obwohl die Feuchtigkeitsabhängigkeit unter wolkenlosem Himmel als Schwachpunkt herausgefunden wurde. Ein Vergleich der Methode unter Verwendung beobachteter Inputparameter bei gleichzeitigen Messungen des DLF ergibt einen RMS-Fehler von 12.2 W/m².Diese Methode wird für die Kartierung des DLF über Westeuropa und dem Mittelmeer angewendet. Dazu werden Satellitenbilder von zwei Tagen um Mittag und zusätzliche Gitterpunktdaten, die während des ALPEX-Experiments gewonnen wurden, verwendet. Ein Vergleich mit Bodenmessungen von 4 Stationen in West-deutschland zeigt eine mittlere Abweichung von 3.9 W/m² und einen RMS-Fehler von 11.6 W/m². Diese Methode ist ein erster Versuch, den DLF im regionalen Bereich aus Satellitendaten des Wolkenfeldes und Gitterpunkt-Analysedaten des thermodynamischen Feldes zu schätzen.

---

---

With 10 Figures     

Dehydration melting of tonalites. Part I. Beginning of melting

 The beginning of dehydration melting in the tonalite system (biotite-plagioclase-quartz) is investigated in the pressure range of 2–12 kbar. A special method consisting of surrounding a crystal of natural plagioclase (An₄₅) with a biotite-quartz mixture, and observing reactions at the plagioclase margin was employed for precise determination of the solidus for dehydration melting. The beginning of dehydration melting was worked out at 5 kbar for a range of compositions of biotite varying from iron-free phlogopite to iron-rich Ann₇₀, with and without titanium, fluorine and extra aluminium in the biotite. The dehydration melting of phlogopite + plagioclase (An₄₅) + quartz begins between 750 and 770°C at pressures of 2 and 5 kbar, at approximately 740°C at 8 kbar and between 700 and 730°C at 10 kbar. At 12 kbar, the first melts are observed at temperatures as low as 700°C. The data indicate an almost vertical dehydration melting solidus curve at low pressures which bends backward to lower temperatures at higher pressures (> 5 kbar). The new phases observed at pressures ≤ 10 kbar are melt + enstatite + clinopyroxene + potassium feldspar ± amphibole. In addition to these, zoisite was also observed at 12 kbar. With increasing temperature, phlogopite becomes enriched in aluminium and deficient in potassium. Substitution of octahedral magnesium by aluminium and titanium in the phlogopite, as well as substitution of hydroxyl by fluorine, have little effect on the beginning of dehydration melting temperatures in this system. The dehydration melting of biotite (Ann₅₀) + plagioclase (An₄₅) + quartz begins 50°C below that of phlogopite bearing starting composition. Solid reaction products are orthopyroxene + clinopyroxene + potassium feldspar ± amphibole. Epidote was also observed above 8 kbar, and garnet at 12 kbar (750°C). The experiments on the iron-bearing system performed at ≤ 5 kbar were buffered with NiNiO. The f _{O 2} in high pressure runs lies close to CoCoO. With the substitution of octahedral magnesium and iron by aluminium and titanium, and replacement of hydroxyl by fluorine in biotite, the beginning of dehydration melting temperatures in this system increase up to 780°C at 5 kbar, which is 70°C above the beginning of dehydration melting of the assemblage containing biotite (Ann₅₀) of ideal composition. The dehydration melting at 5 kbar in the more iron-rich Ann₇₀-bearing starting composition begins at 730°C, and in the Ann₂₅-bearing assemblage at 710°C. This indicates that quartz-biotite-plagioclase assemblages with intermediate compositions of biotite (Ann₂₅ and Ann₅₀) melt at lower temperatures as compared to those containing Fe-richer or Mg-richer biotites. This study shows that the dehydration melting of tonalites may begin at considerably lower temperatures than previously thought, especially at high pressures (>5 kbar). Received: 27 December 1995 / Accepted: 7 May 1996