Daiji Kobayashi

Separation of the snowmelt hydrograph by stream temperatures

Abstract The temperature of a stream under thick snow cover in northern Hokkaido Island, Japan, was found to be 3° to 4°C during the snowmelt flood. Because this was considerably higher than the temperature of snowmelt (0°C), and the stream was perfectly insulated from heat sources from above by snow cover, it was reasoned that soil heat was furnishing most of the heat in runoff. An attempt was made to “separate” the daily snowmelt hydrograph into surface and subsurface components, using the conservation low of mass and heat. Surface runoff begins when the daily snowmelt hydrograph rises, but accounts for only 15–20% of streamflow even at daily peak runoff. As the hydrograph recedes, all of the flow is subsurface in origin, according to heat-mixing analysis. The results are contrary to the assumptions underlying conventional hydrograph analysis, but are consistent with the results by environmental isotope analysis and with the variable source-area concept of runoff in headwaters.

Stream temperature, specific conductance and runoff process in mountain watersheds

Stream temperature ranged from 3 to 4°C at an experimental site during snowmelt on Hokkaido Island, Japan, which provided direct evidence of major contributions of subsurface water to stream water. In contrast, stream temperatures during rainstorms in summer decreased gradually after stream flow peaked, attaining a nearly constant temperature ranging from 9 to 11°C. During storm flow recession, stream temperatures during summer or snowmelt were similar to the soil temperature at 1·8 m below the land surface, suggesting that subsurface water contributions to stream flow are derived from this depth. The hygrographs during two rainstorms, August 1987 and September 1989, were separated using temperature. The stream temperature was assumed to depend on the mixing of surface flow, having a temperature ranging from that of rainfall to that of shallow (50 cm deep) soil water, and subsurface flow, having the temperature of the soil at 1·8 m below the land surface. Subsurface flow was estimated to contribute 85–90% of the total stream flow during each rainstorm. A two-component hydrograph separation was also evaluated using specific conductance. Runoff contributions from the two sources for the temperature and specific conductance analysis were similar. Analysis of the temperature and conductance–discharge hysteresis loop, and of individual flow components for storm hygrographs, provide a general picture of the runoff process in the experimental basin. Copyright

Snow accumulation on a narrow board

Abstract In calm weather falling snow flakes sometimes bounce even on the soft surface of freshly deposited snow. Activity of such bouncing snow flakes influences the efficiency of snow accumulation on a narrow board. Snow flakes which fall on the board near the edge frequently jump over the edge. Activity of bouncing increases and cohesion decreases as temperature falls. Consequently, the amount of snow accumulation on a unit area of a board reduces with decreasing width of the board and falling temperature. It was found impossible for snow flakes to accumulate on the board when its width was narrower than the critical value.

Separation of a snowmelt hydrograph by stream conductance

Abstract By the continuous measurement of stream conductance and discharge for a long snowmelt period, the runoff components were separated and the dominancy of subsurface flow was confirmed as found in the previous paper on stream temperature analyses. A plot of conductance versus discharge gave a shift between rising limbs and falling limbs of a diurnal hydrograph and the shift was inversed after a snow-free area emerged adjacent to stream channels; and the shift corresponded to the change of the peak lag of the runoff components. The results provided important information for discussing and reinforcing the variable source-area concept on the runoff process in headwaters.

Formation process and direction distribution of snow cornices

Abstract The formation process of a snow cornice was traced, with the finding that a cornice grew by extending leeward as a horizontal slab and by depositing windblown snow particles on a cantilever slab, notwithstanding the behavior of eddies on the leeward of a mountain ridge. The collection coefficient of windblown snow particles by a cornice was found to range from 2% to 50%. Investigation of the direction distribution of cornices on the mountain ridge revealed that the cornice arranged itself perpendicular to each ridge direction.