Premium
Systematic Comparison of ILWAS, MAGIC, and ETD Watershed Acidification Models: 2. Monte Carlo Analysis Under Regional Variability
Author(s) -
Rose K. A.,
Brenkert A. L.,
Cook R. B.,
Gardner R. H.,
Hettelingh J. P.
Publication year - 1991
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/91wr01717
Subject(s) - monte carlo method , watershed , environmental science , deposition (geology) , scale (ratio) , hydrology (agriculture) , statistics , soil science , statistical physics , mathematics , geology , computer science , geography , cartography , physics , paleontology , geotechnical engineering , machine learning , sediment
A combination of input mapping (Rose et al., this issue) and Monte Carlo simulation was used to quantitatively compare the predictions of the Integrated Lake Watershed Acidification Study (ILWAS), Model of Acidification of Groundwater in Catchments (MAGIC), and Enhanced Trickle Down (ETD) watershed acidification models. Monte Carlo simulation was used to impose regional variability on selected model inputs of two dissimilar watersheds resembling those found in the northeastern United States for four different SO 4 2− deposition scenarios. Model predictions were viewed in aggregate (distributions, medians) corresponding to population‐level predictions, and on an iteration‐by‐iteration basis corresponding to watershed‐specific predictions. For each of these, predicted values of acid‐neutralizing capacity (ANC) were treated as relative scale (changes in ANC in response to changes in deposition) and absolute scale (ANC concentrations, number of acidic lakes) predictions. Aggregate and iteration‐specific predictions viewed on a relative scale were similar for three models under small to moderate changes in SO 4 2− deposition. Aggregate and iteration‐specific predictions differed among models when viewed on an absolute scale. Absolute scale predictions were most similar under conditions of high variability in deposition. Proper interpretation of multiple model forecasts requires objective quantification of the effects of model differences on predictions.