The Use of Mass Flux or Concentration
for Sensitivity Analysis of the Alpine Hydrochemical Model
Thomas Meixner1, Hoshin V. Gupta1, Luis A. Bastidas1, and Roger C. Bales1
Sensitivity analysis for hydrochemical models requires consideration of the multivariate nature of
watershed response. A robust, multiobjective, generalized sensitivity analysis (MOGSA) algo-
rithm, recently developed at the University of Arizona, was used to fully investigate the parameter
sensitivity of the Alpine Hydrochemical Model (AHM). In this paper, we discuss differences in
parameter sensitivity that result from using mass flux objective functions as opposed to concentra-
tion objective functions. A total of 20,000 simulations for the 1986 and 1987 water years (water
year day 1 = October 1st) were conducted for the Emerald Lake watershed in Sequoia National
Park, California (3635′N, 11840′W). A total of 21 objective functions, discharge and both con-
centration and mass flux for 10 chemical species, were evaluated for each simulation.
The MOGSA algorithm determined that only 2000 simulations were necessary to determine para
meter sensitivity for the 24 model parameters used in these simulations. We found significant
differences in parameter sensitivity for concentration and mass flux objective functions. For exam-
ple, the snowpack elution parameter and a number of hydrologic parameters were sensitive for Cl
concentration, while only the snowpack elution parameter was sensitive for Cl mass flux. For Ca2+
and the other cations, fewer hydrologic and mineral weathering parameters were sensitive, but soil
exchange parameters were more sensitive for mass flux objective functions.
Our results indicated differences in hydrologic parameter sensitivity between the soil and talus
subunits of the AHM. Flow rate parameters were more sensitive on the soil subunit, while volume
parameters were more important for the talus subunit. These results indicate that future field efforts
need to emphasize improved measurement of the effective reaction volume of talus slopes in alpine
watersheds.
The differences in parameter sensitivity between the mass flux and concentration objective func-
tions are due to the pronounced seasonality of discharge in alpine basins and the varied influence of
different hydrologic and chemical processes across the seasons. Mass flux calculations emphasize
the spring snowmelt and peak discharge events of the early summer, while concentration objective
functions equally weight the entire year. Our results indicate that using mass instead of concentra
tion permits better identification of the model parameters that most affect stream conditions during
peak springtime flows and that some combination of mass flux and concentration objectives should
be used in evaluating model performance.
1
Department of Hydrology and Water Resources, University of Arizona, Room 122B, Building #11, Tucson,
Arizona 85712, USA
16
Previous Page
