The application of a water balance model to assess the role of fog in water yields from catchments in the east Otago uplands, South Island, New Zealand

摘要

The degree to which fog interception can augment water yields from snow tussock grasslands in the uplands of east Otago has been debated for over 40 years. Initially, the discussion centred on conflicting results from lysimetry-based studies. More recently, results from a stable isotope analysis suggest that fog and rain may contribute to the soil and groundwater stores in approximately equal proportions, raising the possibility that fog deposition may add as much to stream discharge as rainfall. To test this notion, we compare the mean annual water yield measured from five catchments in the east Otago uplands ranging in size from 1.2 to 60 km~2 (Glendhu, Elbow Creek, Deep Creek at Muster Huts, Deep Creek, and Upper Deep Stream) with predicted water yield using the WATYIELD water balance model. Because the model makes no allowance for fog input in its water balance calculations, our hypothesis is that any excess in measured over modelled water yield may at the outset be assigned to fog. Deep Creek and Upper Deep Stream were the only catchments to show any excess. For Deep Creek the excess averaged 30 mm a year, which is less than the error with which the average annual water yield of 836 mm is measured for this catchment. At Upper Deep Stream the excess averaged 230 mm per year over a 7-year period, which is approximately 15% of the catchment area rainfall. However, not all of this can be automatically attributed to extra input from unmeasured fog. Rain gauge under-catching, and suppressed transpiration, may also explain some of this difference. Thus, while fog, under some circumstances, may be capable of contributing to annual water yield at the catchment scale, it is not doing so in approximately equal proportions with respect to rainfall in the examined catchments.