 |
| Titel |
Increasing parameter certainty and data utility through multi-objective calibration of a spatially distributed temperature and solute model |
| VerfasserIn |
C. Bandaragoda, B. T. Neilson |
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| ISSN |
1027-5606
|
| Digitales Dokument |
URL |
| Erschienen |
In: Hydrology and Earth System Sciences ; 15, no. 5 ; Nr. 15, no. 5 (2011-05-20), S.1547-1561 |
| Datensatznummer |
250012788
|
| Publikation (Nr.) |
 copernicus.org/hess-15-1547-2011.pdf |
|
|
|
|
|
| Zusammenfassung |
| To support the goal of distributed hydrologic and instream model predictions
based on physical processes, we explore multi-dimensional parameterization
determined by a broad set of observations. We present a systematic approach
to using various data types at spatially distributed locations to decrease
parameter bounds sampled within calibration algorithms that ultimately
provide information regarding the extent of individual processes represented
within the model structure. Through the use of a simulation matrix,
parameter sets are first locally optimized by fitting the respective data at
one or two locations and then the best results are selected to resolve which
parameter sets perform best at all locations, or globally. This approach is
illustrated using the Two-Zone Temperature and Solute (TZTS) model for a
case study in the Virgin River, Utah, USA, where temperature and solute
tracer data were collected at multiple locations and zones within the river
that represent the fate and transport of both heat and solute through the
study reach. The result was a narrowed parameter space and increased
parameter certainty which, based on our results, would not have been as
successful if only single objective algorithms were used. We also found that
the global optimum is best defined by multiple spatially distributed local
optima, which supports the hypothesis that there is a discrete and narrowly
bounded parameter range that represents the processes controlling the dominant
hydrologic responses. Further, we illustrate that the optimization process
itself can be used to determine which observed responses and locations are
most useful for estimating the parameters that result in a global fit to
guide future data collection efforts. |
| |
|
| Teil von |
|
|
|
|
|
|
|
|