We were able to show, that it is possible to reduce selenium concentrations in an end-pit lake using ecosystem manipulation. We combined the knowledge of existing bioreactor principles with the in-situ production of a carbon source within the lake, rather than a direct addition of organic carbon such as whey or molasses.
Mining, Ecological Engineering, and Metals Extraction for the 21st Century
The mechanisms of microbial Se reduction processes were already well understood, but the limitation to successfully treat a whole lake ecosystem was hindered in large parts by the need for enough energy in form of a carbon source Attempts have been made before to introduce a carbon source and prime a lake with the required bacterial community 39 — 42 , but the amount of organic matter required makes application in larger lakes not feasible By adding nutrients, the yield of captured carbon through primary production is orders of magnitude higher and requires only a fraction of the volume to be brought to the lake, making it logistically much more feasible.
A careful control of the nutrient balance should accelerate the process of Se reduction. If added nitrate can be prevented from sinking into the hypolimnion before being assimilated, the Se reduction phase will be less delayed. We estimate that the onset of Se reduction would have occurred 1. A more precise stoichiometric approach to the nutrient addition would have required only one amendment, which would reduce the response time by another year. Thus, we predict, that the addition of precise volumes of fertilizer to the surface of an end-pit lake can show first results shortly after oxygen is depleted in the hypolimnion and a measurable response of Se reduction should be possible within the second growing season after the nutrient amendment.
To further improve water quality of mine discharge, rerouting of smaller streams through the monimolimnion of a fertilized pit-lake may allow the transformed pit lake to be used as a flow-through bioreactor itself. Additional monitoring of the monimolimnion and water inflow would be required to ensure that anoxia is maintained.
A continuous supply of nutrients would also be required to maintain the input of organic material to the hypolimnion. The treatment of flowing water in an end-pit lake would negate the need to build physical structures for bioreactors and reduce the costs of mitigation from millions to thousands of dollars.
This method would also address the problem of re-contamination of treated water, when water discharging from a pit-lake flows through subsurface rock drains. Treating Se contamination by ecosystem manipulation is a method best suited for large and deep lakes with small surface area to volume ratios and circum neutral pH. There are in excess of 40 coal-mining end-pit lakes larger than one hectare in Alberta, Canada alone. This combination of relative depth and size of the lake insures, as a rough estimate, that they are deep enough to stratify and have steep enough slopes to promote meromixis.
Hence more than half of the coal-mining pit lakes in Alberta are suitable for the herein described method of Se treatment. With large numbers of coal mines predicted to close in the near future, the amount of pit lakes in the landscape will increase making inexpensive and efficient treatments of critical importance. Since all of these lakes require continuous water quality monitoring mandated by the rules for mine closure operations, the additional cost of the fertilizer is essentially minimal. We believe, that by adapting the herein described method to other industries, the concept offers an inexpensive adaptation and large scale application for the mitigation of aquatic pollutants that require chemical reduction processes in anoxic environments.
The study lake used in this experiment is located in the eastern rocky mountain foothills north of Grande Cache, Alberta Canada UTM The experimental lake is 7. The experimental lake does not have a surface outflow. A smaller upstream lake, connected by a small surface connection, was initially considered as a reference lake. It was however deemed not suitable as a directly comparable lake due to its small surface area of 2.
While monitoring commenced on both lakes with equal effort, the data on the smaller lake revealed very high variability on all water quality metrics due to continuous mixing, a lack of stratification, and strong influences of weather, especially precipitation events. We decided that any effects of the manipulation in the manipulated lake would be clearly identifiable and possible to separate from natural variability.
Hence, the upstream lake is not included in the analysis and data is not shown in this document. Both lakes are abandoned surface mine pits that were filled with water from the surrounding landscape as a remediation tool and remain fishless.
- Publication Details!
- Practical Industrial Data Communications: Best Practice Techniques!
- Foundations of basic nursing!
- Acid Mine Drainage.
- We had Dreams and Songs to Sing;
The manipulated lake was fertilized in May and August No long term data was available prior to fertilization. As a highly oligotrophic headwater pit-lake, we assumed phosphorus to be limiting with negligible concentrations and nitrogen to be potentially higher due to input from blasting operations during mining 43 , We estimated the volume of the lake from its surface area and depth and chose a target concentration for total P as roughly 0.
Half of the fertilizer was added from shore in one dump, the other half was stored in tanks on shore and added over two weeks in the propeller-wash of a boat to better mix the fertilizer at the surface. A second fertilizer amendment was necessary, after it was evident, that the low concentration of N in the lake was insufficient to sustain a phytoplankton bloom for more than one month. Dispersion of the fertilizer within the lake happened over a few weeks and raised the lakes nitrate concentration from 0.
The high specific gravity of the fertilizer caused it to accumulate in higher concentrations in the deeper parts of the lake. Samples were collected three times within one month before the first fertilization, and monthly during the ice free season after that. Additionally, water samples were taken once during the winter from under the ice, if conditions allowed for it.
The lake was inaccessible during the winters of — and —, due to unstable ice and extremely bad road conditions on unmaintained abandoned mine roads. Water quality samples and depth profiles were taken over the deepest spot of the lake. Water quality samples were taken at two depths, over the deep spot of the lake, using a Van-Dorn water sampler. Surface samples were taken at 0. Samples were stored in acid washed bottles, preserved in the field with trace metal grade acids and kept refrigerated until analysis. Samples were stored in acid washed bottles and shipped over night in coolers to maintain a preserving environment.
Data were statistically analyzed using the programming language R Graphical representations of the data were created using the libraries ggplot2 49 and lattice Water quality data was graphed over time. The three phases of the reduction process were identified a-priory. Timing of each phase was set as the change in slope of the element at the beginning of the phase and the stabilization of a low point at the end of the phase. The change of concentrations of each element was expressed as electron equivalencies, i.
We used an electron transfer of four electrons for oxygen, five electrons for Nitrate and eight electrons for sulfate. In the case of Se reduction, we used the ratio of the Se-oxyanions measured in to calculate Se electron equivalencies more precisely. We used the total Se concentrations and calculated the ratio of electrons transferred from the ratio of Se-oxyanions measured in We assumed a first order relationship for the reduction of electron acceptors 33 , i. We would like to thank our summer field technicians for help with sample collection in the field.
Joseph Rasmussen. Electronic supplementary material. Supplementary information accompanies this paper at doi Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
National Center for Biotechnology Information , U. Sci Rep. Published online Sep 5.
N-P Fertilization Stimulates Anaerobic Selenium Reduction in an End-Pit Lake
Andreas Luek , 1 David J. Rowan , 2 and Joseph B. Rasmussen 1. David J. Joseph B. Author information Article notes Copyright and License information Disclaimer. Andreas Luek, Email: ac. Corresponding author.
Received May 11; Accepted Aug Abstract Selenium Se , an essential micro nutrient, is toxic to aquatic life at slightly higher water concentrations. Introduction Selenium Se contamination of aquatic environments is predominantly caused by anthropogenic activity, and is a contaminant of global concern 1. Open in a separate window. Figure 1. Picture of the manipulated lake before and after nutrient amendments. Figure 2. Figure 3. Table 1 Phases of the reduction process in the manipulated lake.
Lake fertilization The manipulated lake was fertilized in May and August Water quality analysis Samples were collected three times within one month before the first fertilization, and monthly during the ice free season after that. Data analysis Data were statistically analyzed using the programming language R Electronic supplementary material Supplementary Information K, pdf. Acknowledgements We would like to thank our summer field technicians for help with sample collection in the field.
Related Acidic Pit Lakes: The Legacy of Coal and Metal Surface Mines
Copyright 2019 - All Right Reserved