INTERPLAY BETWEEN PARAMETER UNCERTAINTY AND MODEL AGGREGATION ERROR1

ABSTRACTS: Modeling error can be divided into two basic components: use of an incorrect model and input parameter uncertainty. Incorrect model usage can be further subdivided into inappropriate model selection and inherent modeling error due to process aggregation. Total modeling error is a culmination of these various modeling error components, with overall optimization requiring reductions in all. A technique, utilizing Monte Carlo analysis, is employed to investigate the relative importance of input parameter uncertainty versus process aggregation error. An expanded form of the Streeter-Phelps dissolved oxygen equation is used to demonstrate the application of this technique. A variety of scenarios are analyzed to illustrate the relative obfuscation of each modeling error component. Under certain circumstances an aggregated model performs better than a more complex model, which perfectly simulates the real system. Alternately, process aggregation error dominates total modeling error for other situations. The ability to differentiate modeling error impact is a function of the desired or imposed model performance level (accuracy tolerance).     

Effects of physical disturbance on nematode communities in sand and mud: a microcosm experiment

A microcosm experiment was carried out to evaluate the effects of continuous and spasmodic physical disturbance of differing frequency on the structure of nematode communities of intertidal sand and mud. There was a marked, characteristic change in abundance and diversity for both sediment types. In the sand microcosms, the majority of univariate measures of community structure, including species diversity, were lowest in the sediments subjected to a high frequency of disturbance. For the mud microcosms, most univariate measures reached their highest values in the treatments with an intermediate frequency of disturbance and were lower in treatments subjected to both higher and lower frequencies. Multivariate ordinations for both nematode assemblages showed a clear separation of undisturbed controls and disturbed treatments, but only for the muddy sediment was there a graded change in community composition with increasing frequency of disturbance. These results confirmed our a priori expectation that nematode assemblages from mobile sandy sediments would be more resilient to physical disturbance than those from sheltered muds, and these observations are considered in the context of Connell's intermediate disturbance hypothesis. Received: 7 May 1997 / Accepted: 10 October 1997     

Modeling the Highly Dynamic Loading of Mercury Species in the Carson River and Lahontan Reservoir System,Nevada

The semiarid Carson River — Lahontan Reservoir system in Nevada, United States is highly contaminated with mercury (Hg) from historic mining with contamination dispersed throughout channel and floodplain deposits. Work builds on previous research using a fully dynamic numerical model to outline a complete conceptualization of the system that includes transport and fate of both sorbed and dissolved constituents. Flow regimes are defined to capture significant mechanisms of Hg loading that include diffusion, channel pore water advective flux, bank erosion, and overbank deposition. Advective flux of pore water is required to reduce dilution and likely represents colloidal‐mediated transport. Fluvial concentrations span several orders of magnitude with spatial and temporal trends simulated within 10‐24% error for all modeled species. Over the simulation period, 1991‐2008, simulated loads are 582 kg/yr (THg²⁺), 4.72 kg/yr (DHg²⁺), 0.54 kg/yr (TMeHg), and 0.07 kg/yr (DMeHg) with bank erosion processes the principal mechanism of loading for both total and dissolved species. Prediction error in the reservoir is within one‐order of magnitude and considered qualitative; however, simulated results indicate internal cycling within the receiving reservoir accounts for only 1% of the reservoir's water column contamination, with river channel sediment sources more influential in the upper reservoir and bank erosion processes having greater influence in the lower reservoir.     

Behavior of Two Phenyl Urea Herbicides in Clayey Soils and Effect of Alternating Dry-Wet Conditions on their Availability

Adsorption and mobility of linuron (3-(3,4-dichlorophenyl)-1-methoxy-1-methylurea) and diuron (3-(3,4-dichlorophenyl)-1, 1-dimethylurea) were studied in clayey soils from the Gharb area (Morocco). Soils A and B were planted with sun flower (Helianthus annuus) while soil C was planted with sugar cane (Saccharum offcinarum). Adsorption was studied for linuron in soils A and B, while mobility was studied only in soil B. Adsorption data were found to fit the Freundlich equation with correlation coefficients r² > 0.9. Freundlich coefficients (K_f, n_f) were in agreement with L and S isotherm types for soils A and B, respectively. Values of K_oc (195 and 102) indicate moderate adsorption. Desorption isotherms for linuron showed hysteresis for both soils. The pesticide would be more bound to soil A (H = 8.44) than to soil B (H = 4.01). The effect of alternating wet and dry conditions was tested for soils A and B. Results showed that retention would increase in soil subject to an additional wet and dry cycle. In the case of diuron isotherm was of type L in soil C. Desorption was noticeable at high concentrations and tended to decrease when concentrations diminished. Mobility of linuron was tested in polyvinyle chloride (PVC) columns, which received different treatments before their percolation. The pesticide was more mobile in a previously saturated column. In columns subject to a drying step after saturation with water, linuron mobility was greatly reduced.     

Climate change effects on aquatic biota, ecosystem structure and function

Climate change is projected to cause significant alterations to aquatic biogeochemical processes, (including carbon dynamics), aquatic food web structure, dynamics and biodiversity, primary and secondary production; and, affect the range, distribution and habitat quality/quantity of aquatic mammals and waterfowl. Projected enhanced permafrost thawing is very likely to increase nutrient, sediment, and carbon loadings to aquatic systems, resulting in both positive and negative effects on freshwater chemistry. Nutrient and carbon enrichment will enhance nutrient cycling and productivity, and alter the generation and consumption of carbon-based trace gases. Consequently, the status of aquatic ecosystems as carbon sinks or sources is very likely to change. Climate change will also very likely affect the biodiversity of freshwater ecosystems across most of the Arctic. The magnitude, extent, and duration of the impacts and responses will be system- and location-dependent. Projected effects on aquatic mammals and waterfowl include altered migration routes and timing; a possible increase in the incidence of mortality and decreased growth and productivity from disease and/or parasites; and, probable changes in habitat suitability and timing of availability.     

Spectrophotometric determination of uranium with arsenazo-III in perchloric acid

A short, sensitive and reliable spectrophotometric method, which has advantages over all known "wet chemistry" methods for uranium determination with regard to tolerance to common interferences, has been developed for the determination of uranium. Selectivity, molar absorptivity and the determination range of uranium have been enhanced by using 0.07% arsenazo-III as a chromogenic reagent. The use of 3 mol dm(-3) perchloric acid as a medium of determination was found to be excellent in terms of good solvent compatibility on dilution, destruction of organic contamination and simplicity of operation. The uranium-arsenazo-III complex formed instantly, and was found to be stable for more than 3 weeks with constant absorbance. Beer's law was obeyed up to a uranium concentration of 16 microg g(-1), with a molar absorptivity at 651 nm of 1.45x10(5) mol(-1) dm(3) cm(-1) at 24+/-2 degrees C. Only phosphate and citrate at 70-fold excess over uranium interfere seriously, whereas other anions studied could be tolerated up to a 70-fold excess over uranium. Of the cations studied, only Mn(II), Co(II), Ni(II), Cu(II) and Cr(III) decreased the normal absorbance of the complex. Iron(III), Ce(III) and Y(III) enhanced the absorbance. Other cations studied did not affect the absorbance up to a 50-fold excess. The accuracy was checked by determining uranium from standard solutions in the range 10-50 microg g(-1). It was found to be accurate with a 96.0-98.6% recovery rate. The method has been successfully applied to standard reference materials and ore samples at microg g(-1) levels.     

USE OF FIRST-ORDER UNCERTAINTY ANALYSIS TO OPTIMIZE SUCCESSFUL STREAM WATER QUALITY SIMULATION1

ABSTRACT: A first-order uncertainty technique is developed to quantify the relationship between field data collection and a modeling exercise involving both calibration and subsequent verification. A simple statistic (LTOTAL) is used to quantify the total likelihood (probability) of successfully calibrating and verifying the model. Results from the first-order technique are compared with those from a traditional Monte Carlo simulation approach using a simple Streeter-Phelps dissolved oxygen model. The largest single difference is caused by the filtering or removal of unrealistic outcomes within the Monte Carlo framework. The amount of bias inherent in the first-order approach is also a function of the magnitude of input variability and sampling location. The minimum bias of the first-order technique is approximately 20 percent for a case involving relatively large uncertainties. However the bias is well behaved (consistent) so as to allow for correct decision making regarding the relative efficacy of various sampling strategies. The utility of the first-order technique is demonstrated by linking data collection costs with modeling performance. For a simple and inexpensive project, a wise and informed selection resulted in an LTOTAL value of 86 percent, while an uninformed selection could result in an LTOTAL value of only 55 percent.     

HYDRODYNAMIC MODELING OF THE CARSON RIVER AND LAHONTAN RESERVOIR,NEVADA1

ABSTRACT: The RIVMOD hydrodynamic model was used to route upstream flows through a 115 km section of the Carson River and Lahontan Reservoir, Nevada. RIVMOD results will later be used to predict sediment movement and ultimately to determine mercury transport within the river/reservoir system. Significant modifications to the model computer code were necessary to represent the narrow, steeply sloping rectangular channel and relatively shallow sloping floodplain of the Carson River and its confluence with the Lahontan Reservoir. These changes include expansion of the continuity and momentum equations to account for rapidly changing channel widths along with the characterization of a complex cross-sectional shape. This modified version of the RIVMOD model can handle shallower side slopes and much more severe flood flow simulations than the original version. A 0.25 km spatial increment was required in the zone of confluence between the river and reservoir. Model predictions show excellent agreement with observed downstream flow and reservoir stage for the entire 1986 water year, which includes one of the most severe flood events of recent record. (KEY TERMS: hydraulics; modeling; simulation; surface water hydrology.)     

COMPUTING THE RISKS ASSOCIATED WITH WASTELOAD) ALLOCATION MODELING1

ABSTRACT: The risks associated with a traditional wasteload allocation (WLA) analysis were quantified with data from a recent study of the Upper Trinity River (Texas). Risk is define here as the probability of failing to meet an established in-stream water quality standard. The QUAL-TX dissolved oxygen (DO) water quality model was modified to a Monte Carlo framework. Flow augmentation coding was also modified to allow an exact match to be computed between the predicted and an established DO concentration standard, thereby providing an avenue for linking input parameter uncertainty to the assignment of a wasteload permit (allowable mass loading rate). Monte Carlo simulation techniques were employed to propagate input parameter uncertainties, typically encountered during WLA analysis, to the computed effluent five-day carbonaceous biochemical oxygen demand requirements for a single major wastewater treatment plant (WWTP). The risk of failing to meet an established in-stream DO criterion may be as high as 96 percent. The uncertainty associated with estimation of the future total Kjeldahl nitrogen concentration for a single tributary was found to have the greatest impact on the determination of allowable WWTP loadings.