Because of precipitation serves as a major vehicle of nutrient input into the forest ecosystem, the accurate measurement of its volume and ion concentration is of prime importance in an evaluation of any bio-geochemical cycle. Therefore, chemistry of the precipitation and throughfall of forest ecosystem was investigated in the Norikura Highlands. The investigation period was from January, 2003 to October, 2006. The throughfall volume in growing season and dormant season were 86 % and 93 % of the precipitation volume. Throughfall pH increased with increasing K+ concentration showed that H+ was held within the canopy by cation exchange reaction. And the concentration level of K+, Mg2+ and Ca2+ in the throughfall was much higher than that in the precipitation. It was the cause of canopy leaching. In growing season, proportions of canopy leaching of K+, Mg2+ and Ca2+ were 95 %, 70 % and 43 % of the throughfall deposition respectively. At Coniferous site, the flux of dry deposition was higher in dormant season than growing season. It is suggested that aerosol of the atmosphere and leaf area might be influenced.
Hydro-chemical cycle of forest ecosystem was investigated in the Norikura Highlands. The investigation period was from June to October, 2003 and 2004. The precipitated water that reached to the forest floor contained vastly different chemical constituents than those of the precipitation. The throughfall ion concentration was higher than precipitation. Generally it was in the following order: Ao-perocolation > 20 cm soil water ≒ 50 cm soil water. The concentration levels of NH4+ and NO3- in the soil water are lower than those in the throughfall, and HCO3- was higher in soil water. It is suggested that the transformation might be caused by microbes, such as nitrogen fixation, nitrification, and uptake by vegetation. The output flux of H+ from mineral horizon was lower than input flux by precipitation in both the coniferous site and the deciduous site. It showed the effect of acid neutralization. However, the internal cycling of base cations (K++Mg2++Ca2+) were more active in the deciduous site. Therefore, the activity of acid-neutralizing reaction was higher in the deciduous site than the coniferous site.
The observations in the pre-monsoon seasons of 1996 and 1997 indicated that two moraine-dammed glacial lakes, Tsho Rolpa and Imja, in the Nepal Himalayas have the different thermal structure in spite of the similar solar radiation, air temperature and wind system at the weather stations. The difference of thermal conditions in the lakes possibly results from the depletion in wind velocity over Imja Lake, which is caused by the screening of the upwind end-moraine to dead-ice zone 20-25 m higher than the lake surface. In order to clarify the topographic screening effect, three-dimensional numerical simulations of airflow around Tsho Rolpa and Imja Lakes were performed by building up a topographic model of actual size in the calculation domain. Simulated results indicated that the wind velocity at 2 m above the Imja Lake surface is 33-42 % smaller than that above the Tsho Rolpa Lake surface for the constant wind velocites of 1-5 m/s at 2 m above the points corresponding to the weather stations. This evidences the significant screening effect on the thermal structure of Imja Lake, since the wind force per unit lake area is proportional to the square of wind velocity at a certain height above the lake surface. The airflow simulations indicate that, as a barrier, the glacier terminus (ice cliff) leeward of Imja Lake decreases the wind velocity near the lake surface as much as does the upwind end-moraine to dead-ice zone.
The subarctic Tanana River basin, Alaska, is occupied by ca. 5.6% glacierized region and the other discontinuous permafrost region. Time series of discharge, suspended sediment concentration (SSC) and water chemistry in the Tanana River were obtained in the glacier-melt seasons of 2000 to 2005. In order to estimate each contribution of glacier-melt to discharge, SSC and water chemistry, all the time series were reproduced by the tank model. As a result, the reasonable simulations (correlation coefficient R = 0.68-0.97) for the discharge time series revealed that glacial-melt discharge occupies 33-58% of the Tanana river discharge. The SSC and SiO2 time series were also reasonably simulated (R = 0.70-0.87). These calculations led to the conclusion that the major sediment and SiO2 sources are located in the glacierized and permafrost regions, respectively.
The relationship between the stream water quality and mean basin slope was examined in the mountainous region of the Azusa River and Tama River basin. The ionic compositions of stream water are of Ca-HCO3 type and of Ca · Na-HCO3 type. The HCO3-, Ca2+ and Na+ concentrations in stream water exhibit a negative correlation with the mean slope angle in the mountainous watersheds of the Tama River. In contrast, the HCO3- and Ca2+ concentrations in stream water are correlated to the mean slope angle in the mountainous watersheds of the Azusa River, and the Na+ concentrations are not correlated with the mean slope angle. The Na+/Ca2+ ratio in stream water has a negative correlation with the mean slope in the mountainous watersheds of both the Azusa River and Tama River. It is proposed that the depth of groundwater flow coupled with the basin slope is one of the factors influencing the stream water quality.
In a snowy watershed, the response of streamwater chemistry to snowmelt depends not only on meltwater chemistry but on the hydrochemical processes of the subsurface water. In this study, the δ18O values and solute concentrations of the precipitation, meltwater, subsurface water and surface water at the Moshiri experimental watershed located in the northern part of Hokkaido, Japan were obtained, and were discussed here in order to clarify the subsurface runoff processes and the formational mechanisms of the subsurface and surface water chemistry during the snowmelt season. A groundwater with a constant δ18O value (-12.3‰) and constant Cl- concentration (200 μeq/L) was widely found in the watershed and in the shallow saturated water zone. Hydrograph separations with high precision were conducted using δ18O early in the snowmelt season (Mar. 23-Apr. 14), and using a Cl- concentration during the active snowmelt season (Apr. 21-May 1). The contributions of the old water to the streamwater discharge were 82% (Mar. 23-Apr. 14) and 76% (Apr. 21-May 1). The contributions of old water were very high throughout the snowmelt season, which means that the groundwater reservoir must be significantly large. Subsequently, the temporal and spatial variations of the SiO2 concentrations of the subsurface water and surface water were observed to clarify the chemistry of the subsurface water, and revealed that there were two geographic sources (shallow groundwater and deep groundwater) in the ground. The SiO2 concentration of the deep groundwater was higher than that of the shallow groundwater. The sources of two spring waters changed from the deep groundwater to the shallow groundwater, and the SiO2 concentrations of these waters decreased while the discharges of these waters increased. The SiO2 concentration of old water was estimated using the results of isotopic and chloride hydrograph separation. Then, the result of the relationship between the amount of old water discharge and the SiO2 concentration of old water showed that the source of old water during the low outflow period was deep groundwater, and the source during the low outflow period was shallow groundwater. As the discharge of the old water reached over 0.08m3/s, the estimated SiO2 concentration of the old water became constant. These results indicated that a large amount of the old water flowing out to the river in the active snowmelt season was discharged from shallow groundwater. By observing the snowmelt runoff processes in a nival watershed, it was clarified that the old water is composed of a combination of two sources of groundwater and that the shallow groundwater chemistry greatly affects the streamwater chemistry during the large stormflow period.