Parameter Estimation in Sediment Submodels

In simple steady-state models with integration over time and space, net retention in the whole body including the sediment can be estimated by annual mass balances. If the sediment is considered as a separate pool in a model, both gross sedimentation rates, release rates from the sediment, and net sedimentation rates have to be estimated. A calibration by trial and error can be used and important information can be extracted on the sensitivity of the models to initial pool sizes and to rate constants. But information can also be hidden since the three process rates are interrelated.

If a higher resolution in time and space is needed, either laboratory experiments or field measurements, or a combination of these, have to be performed. Gross sedimentation rates can be measured in the field by exposure of sediment traps carefully designed with a length:diameter ratio >5-7 or by frequent sampling in meter-long settling tubes in the laboratory. In shallow water bodies, resuspension from the sediment may cause an error, but by deploying a series of traps in different depths, both gross sedimentation and resuspension rates can be estimated. Assuming that the settling is produced in the photic zone at the surface, it can be expected that the organic content of the trapped material will decline with increasing depth of deployment due to decay over time. Linear extrapolation of the decline observed in the upper traps to the traps just above the bottom will reveal the resuspension. By subtracting the extrapolated value from the observed value in the lower traps, the resuspension rate can be estimated. Short-term exposures of sediment traps are recommended since sedimentation rates may vary considerably over time. Corrections for resuspension can also be estimated by element concentration in suspended material in the water body and the concentration in the traps and in the very surface of the sediment. For modeling purposes, rate constants can be calculated by dividing average concentration in the water above the traps with the trap catch per unit area.

The size of the exchangeable or active pool of an element in the sediment can be estimated assuming a first-order decay and performing a non-transport-limited exhaustion experiment, eventually under optimum temperature conditions. Specific pools can also be estimated from differential extraction of sediment from different depths in the sediment. For sediment phosphorus, several schemes for extraction exist, but only a few really identify the relation to actual release rates. Loosely sorbed phosphate, reductable iron and manganese phosphate, and a fraction of the organically bound phosphorus is expected to be exchangeable. Phosphate bound to calcium is considered to be stable, since over time it is converted to hydroxyapatite.

Mineralization rate constants and sorption equilibrium constants can be experimentally separated by addition of poisons like antibiotics or mercury, suppressing biological activity. Such experiments can be performed at various temperatures, pH values, and concentration ranges. Carefully designed multifactorial experiments carried out with sediment slices from different sediment depths can provide the needed constants for complex models.

Transport processes in the sediment are determined from changes in porewater profiles over time. This is most easily done in the laboratory on undisturbed sediment cores, whereby the apparent diffusion can be estimated experimentally by using a conservative tracer. The tracer can be dissolved lithium ions applied to the well-mixed water column above an undisturbed sediment core. With time intervals, porewater from different sediment depths is retrieved, either by suction with a filter-mounted syringe from one core, or by the sectioning of a whole core, from an incubation of a series of parallel cores. The porewater from sediment slices can be retrieved by pressure, suction, or centrifugation. By correction for porosity and differences in ion size between the tracer and an element, an apparent diffusion coefficient can be estimated. The importance of bioturbation can be estimated by introducing known numbers of organisms in undisturbed sediment cores.

By comparing process rates in the water body and in the sediments, guidelines can be set up for the selection of complexity of sediment models. Diffusional fluxes across the sediment-water interface in lakes vary from 0 to 0.1 yr-1, sedimentation and growth of porewater volume vary from 0 to 0.05 yr- , water exchange through outflow varies from 0.1 to 1 yr- , and first-order decay rates vary from 0 to 35 yr- . In general, dilution and decay rates are faster than sediment dilution and diffusional fluxes. This means that decomposition at the sediment surface is more important than transport-limited decay in the deeper parts of the sediment. If the range of diffusion is increased by moving deeper in the sediment or if the gradients are becoming weaker, the influence of sediment pools is reduced. Consequently, in lakes with short hydraulic residence time the sediment fluxes are not important, if both water renewal and sediment fluxes are evenly distributed over the year. But if exchange rates both to and from the sediment are fast in the summer and water renewal is low at that time, the sediment can cause a significant delay in response to reduced loading. In such cases high resolutions in time and space are necessary, for the understanding of lake ecosystem's responses to reduced loading. In lakes with residence times of several years, the sediment will not be the reason for a delay in the response, since the hydraulic residence time and the sediment dilution effects will be the most important.

Project Earth Conservation

Project Earth Conservation

Get All The Support And Guidance You Need To Be A Success At Helping Save The Earth. This Book Is One Of The Most Valuable Resources In The World When It Comes To How To Recycle to Create a Better Future for Our Children.

Get My Free Ebook


Post a comment