Johann Ilmberger

Why some Mediterranean sapropels survived burn-down (and others did not)

Abstract We have evaluated, using a numerical model, the burn-down depth (depth of post-depositional oxygenation) of organic matter-rich sapropels under aerobic conditions at the seafloor. The model suggests that sapropels may be erased by burn-down to depths of at least 5 cm after 10 kyr of bottom-water reoxygenation. A test of the sensitivity shows that the parameters exerting a major control over the burn-down process are the flux of freshly sedimenting organic carbon, the oxygen concentration of the bottom water, the amount of organic matter in the sapropel, and the sedimentation rate. In comparison, bioturbation is of minor importance. The result implies that burial of elevated concentrations of non-refractory organic carbon is generally improbable when the sedimentation rate is lower than 1–2 cm kyr−1. The record of sapropel deposition in the Mediterranean Sea and elsewhere may thus be biased towards depositional environments where the sedimentation rate is large enough to shield the organic-rich layer from oxygenation.

Tracer experiment with sulfur hexafluoride to quantify the vertical transport in a meromictic pit lake

Abstract: This paper describes a tracer experiment with sulfurhexafluoride (SF6) in the monimolimnion of the meromictic mining lake Merseburg-Ost 1b. In October 1998, 1.1 mmol (160 mg) of the conservative gas SF6 was released at the site of greatest depth 3.5 m above the sediment to observe its vertical spreading. An easy-to-use system to collect ∼0.5 L water samples using evacuated flasks was developed. The headspace technique, gas chromatographic separation and ECD-detection were used to determine SF6. The main objective is the evaluation of mean vertical diffusion coefficients for SF6 and heat from periodically measured SF6 and CTD profiles. During the study period of 11/2 years, the heat transport was estimated to be molecular in the stratified portion. A thermal flux from the sediments of 0.23 W/m2 was found necessary to balance the heat. In the region of high stability (N2∼10-2 s-2) diffusivities for SF6 were ∼ 10-8 m2/s, whereas in the lower part of the monimolimnion both tracers resulted in K∼ 7 · 10-6 m2/s. We found K to be approximately proportional to N-2.4 ± 0.2.

Vertical mixing in Überlinger See, western part of Lake Constance

Depth variable vertical eddy diffusion coefficients for heat (Kz) were calculated from continuously measured temperature profiles in Überlinger See (western part of Lake Constance). The temperatures were averaged over vertical intervals of 10 m yielding 14 discrete values (maximum depth of Überlinger See: 147 m). A linear fit from 10 June to 29 September 1987 was used to smooth the significant temperature fluctuations caused by internal seiches of Lake Constance.Assuming horizontal homogeneity for the smoothed data the Gradient-Flux-Method was applied to compute vertical diffusion coefficientsKz at different depths using the depth variable volumes and surfaces of the 14 layers. The resulting mean diffusion coefficients for the period from June to September are 0.04 cm2/s near the thermocline and up to 0.8 cm2/s in deeper strata (accuracy: ± 50%). It is shown that horizontal mixing between Überlinger See and Obersee (main lake) alters the computation ofKz by less than 50%.A relationship betweenKz and stability (Brunt-Väisälä) frequencyN is found which corresponds well to the theory of internal wave induced turbulence.Combining the diffusion coefficients with measured phosphorus profiles, a phosphorus flux from the hypolimnion to the epilimnion of (0.7 ± 0.4) mg P m−2 d−1 was calculated, corresponding to about 20% of the average external loading per area of Lake Constance in 1986.

Vertical mixing in Überlingersee (Lake Constance) traced by SF6 and heat

Vertical mixing in Überlingersee is studied by releasing sulfur hexafluoride (SF6) as a tracer at a central hypolimnic depth of 60 m and measuring its subsequent vertical dispersion over a period of three months. The experiment started with a streaky tracer injection of 1 liter gaseous SF6 (STP) in August 1990. At that time the lake showed a typical strong summer stratification which in a weakened form lasts until November. From the SF6 profiles of fifteen surveys at three sampling sites vertical diffusivitiesKz are calculated compensating internal seiche displacement and horizontal tracer loss. Except of the bottom region no sampling site or time period is marked by significant differences in the hypolimnicKz profile. So vertical mixing in the whole Überlingersee is described by mean diffusivities decreasing from 1.7 cm2/s at 120 m depth to 0.4 cm2/s in 30 m. The minimal value of 0.3 cm2/s in the thermocline region at 20 m depth is only based on observations in autumn. For a strong summer stratification it is certainly lower. The gradient-flux-method for heat was applied to compute a meanKz (T) profile from continuously measured temperature profiles. Significant differences resulting from the two tracers showed, that theKz (T) values are underestimated by up to a factor of 5 if cooling by lateral exchange is neglected. Particularly, internal seiche pumping of colder water from the adjacent Lake Obersee over the separating sill of Mainau into the deep Überlingersee basin is observed in 1990 from August onward, obviously controlling the heat budget below the sill level.

A SF6 tracer study of horizontal mixing in Lake Constance

Horizontal mixing processes in the hypolimnion of the western part of Lake Constance are studied by measuring the dispersion of a sulfur hexafluoride (SF6) tracer plume. Only 1 liter gaseous SF6 (STP) was released at a central hypolimnic depth of 60 m in August 1990. Over a period of 3 months the horizontal dispersion of the tracer plume was measured by 19 surveys using a new, vertically integrating sampling device. The observed horizontal dispersion is marked by strong storm-induced stirring events. Nevertheless mean turbulent diffusion coefficients for the whole period can be computed. They rise about linear from 0.7 ⋅ 105 cm2/s to 3.0 ⋅ 105 cm2/s with the distance from the western end of the lake. For the hypolimnion of Überlingersee, a sill-separated basin in the western part of Lake Constance, a simple budget model gives an exchange time of 67 ± 6 days with the main basin (Obersee).

Vertical structure of currents in western Lake Constance

Simultaneous measurements of current velocity and density profiles at the Sill of Mainau (western Lake Constance) enabled the observation of the internal response of the lake and the calculation of gradient Richardson number over the entire water column at the sampling site. Periodic changes of current direction corresponded with predicted periods of the two-layer oscillation and could be confirmed with thermocline oscillation data. In general, the current profiles confirmed a two-layer structure. However, an activating wind of speeds higher than 4 m/s produced three-layer profiles for a period of several hours at the sampling site, when the velocities due to the two-layer oscillation traversed zero. At times the implied current shear produced supercritical values of gradient Richardson number on a vertical scale of a few meters. This occurred during winds in the epilimnion and after strong (> 4 m/s) winds in the lower thermocline and the hypolimnion. In the 20 m above the lake bed, topographical conditions helped the gradient Richardson number to go supercritical sporadically. Across the thermocline, no supercritical gradient Richardson numbers could be confirmed at the measuring site. A lower boundary for vertical mixing in the measuring site due to current shear could be evaluated.

Coupling of groundwater and surface water at Lake Willersinnweiher: Groundwater modeling and tracer studies

Abstract.Coupling between groundwater and surface water at Lake Willersinnweiher, a gravel pit lake in the Upper Rhine Graben without any surface in- or outflow, was investigated using both a groundwater model and the tracers 18O and SF6. Based on groundwater modeling, recharge and discharge areas around the lake as well as the residence time of the lake water were determined in a regional context. The uncertainty of the simulated flow field was assessed by sensitivity analysis. The tracers 18O and SF6 were measured in several observation wells and piezometers around the lake as well as in the lake’s water column. They were used to verify groundwater flow directions found in the modeling. We found that the groundwater-lake interaction model had a large uncertainty even though relatively detailed information on input data was available. Independent information obtained from the environmental tracers allowed us to improve and verify the model.

Behavior of a Medium-Sized Basin Connected to a Large Lake

Regular measurements—extending over about one year—of water temperature, of electric conductivity, and of the concentration of the stable isotope helium-3 are analyzed with regard to the lateral exchange of water between the main part of Lake Constance—the Obersee—and its smaller neighboring basin, the Uberlingersee. In the absence of a significant density stratification (as occurs in late winter and early spring), different wind speeds over the two basins cause different vertical mixing leading to horizontal concentration gradients between the basins. Wind-induced or density-driven horizontal currents are not strong enough to completely reduce the gradients. In contrast, during periods of strong vertical stratification (summer and autumn), internal waves extending along the whole lake cause periodical horizontal currents and mixing and thus lead to homogeneous concentration distributions along the horizontal axis. The seasonal variation of horizontal gradients couples the nutrient balance of Uberlingersee to the chemistry of the main lake.

Dating problems with selected mining lakes and the adjacent groundwater body in Lusatia, Germany

This study presents selected results, applying environmental tracers to investigate lake water–groundwater interactions at two study sites located in Lusatia, Germany. The focus of the investigations were two meromictic pit lakes and their adjacent aquifers. In order to follow hydrodynamic processes between lake and groundwater, mixing patterns within the lakes as well as ages of lake and groundwater, water samples of ground- and lake water were collected at three occasions, representing summer and winter conditions in the aquatic systems. The water samples were analysed for stable isotopes (deuterium, oxygen-18) and tritium and sulphurhexafluoride (SF6 concentration). Lake water profiles of conductivity and 18O could validate the permanent stratification pattern of both the lakes. Groundwater data sets showed a heterogeneous local distribution in stable isotope values between rain and lake water. A two-component mixing model had been adopted only from 18O data to determine lake water proportions in the surrounding groundwater wells in order to correct measured tritium and SF6 concentrations in groundwater samples. This procedure had been hampered by upstream-located wells indicating strong 18O enrichment in groundwater samples. However, rough groundwater ages were estimated. For both study sites, Piston flow ages between 12.9 and 27.7 years were calculated. The investigations showed the good agreement between two different environmental dating tools, considering the marginal data sets.

Lake Dynamics: Observation and High-Resolution Numerical Simulation

We have demonstrated the rich set of hydrodynamic phenomena that operate in a natural lake. State-of-the-art experimental methods yeild a wealth of data that are highly resolved in time and in one space axis. The orthogonal space is sampled only very sparsely, however. Still, such data allow the estimation of crucial effective parameters like vertical diffusion coefficients. Numerical simulations, on the other hand, yeild high resoution results for theentire space. They are thus inherently more powerful than experimental observations and heuristic analysis whenever a complicated to pography is crucial. However, even with the current most advanced slovers, such high-resolution studies of most phenomena on realistic time scales is beyond reach. Confining experiemntal and numerical approaches hence appears as the optimal way to reach a deeper understanding of the lake hydrodynamics.