As stated in the previous article in this series, the “Introduction to the Effects of Leachate Recirculation Series”, leachate data was compiled and analyzed using time series graphs and statistical trend testing.  The time series graphs allow for a visual determination of trends.  However, because there is significant variation in the data due to both natural variation and anomalous data points, visually assessing the trends can sometimes be difficult and subjective.  The statistical trend testing using two methods, Sen’s Slope/Mann-Kendall slope estimator and Linear Regression, allowed for a more objective approach to determining trends. 

            The trend analysis showed primarily decreasing trends or no observed trend for the majority of analytes, including iron and manganese.   However, increasing trends were noted for ammonia-N at all three sites, which will be discussed in a later article.  Some trends were only observed at one site, whereas others were observed at two or more sites.  For the purposes of this article series, articles will focus primarily on analytes where trends were observed in at least two of the three sites.  The following analytes showed statistically significant trends in only one or in none of the sites. Aluminum, antimony, barium, copper, cyanide, strontium, specific conductance, chloride, sodium, sulfate, boron, iron, nitrate/nitrite-N, total alkalinity, and total dissolved solids did not demonstrate consistent trends at each of the sites. 

            For those analytes with statistically significant trends observed at two or more sites, the observed trends and potential causes for the changes in concentrations with leachate recirculation will be discussed as part of this article series.  Constituents that typically have similar chemical behaviors will be discussed together. To start, manganese exhibited statistically decreasing trends in all three sites.  Iron and manganese typically behave similarly chemically and will, therefore, both be discussed in this article despite the weak correlation for iron.

            Iron and manganese are frequently detected at elevated concentrations in C&DD leachate.  Both of these elements are abundant metallic elements and are readily mobilized in the chemical environment created in a landfill.  Thus, these constituents are commonly present in the leachate.  In certain conditions (i.e. oxidizing), iron and manganese exist in solid mineral phases that are insoluble in water.  However, when there are chemically reducing conditions (i.e. anoxic), which are very common in the landfill environment where materials are decaying, iron will be reduced from the insoluble Iron III to the soluble Iron II, and manganese will be reduced from the Manganese IV to the soluble Manganese III.  In chemically reducing conditions, both iron and manganese are mobilized from the waste and the cover soils, resulting in elevated concentrations in leachate. 

            To determine how leachate recirculation affects the iron and manganese concentrations, the author utilized visual and statistical methods.  The iron and manganese data for the three C&DD landfill sites, A, B, and C, were plotted on time series graphs and statistically tested for trends using Sen’s Slope/Mann-Kendall and Linear regression methods.  See the first article in this series for additional background information.

           Due to the different sampling histories, the number of data points varies between the sites.  Site B has the longest sampling history, with leachate data available dating back to 2005, totaling 30 data points for iron and manganese.  In contrast, historical data for Site A only stretches back to 2011, and for Site B only to 2013.  The Site A time series consists of 18 data points and the Site B time series consists of 16 data points.  Prior to conducting the trend testing, the statistical software, Sanitas, was used to identify outliers in the data.  Outliers were flagged as such and were not included in the statistical trend testing.  For the time series, all data points were included with the exception of the first data point for Site B for both iron and manganese, which were exceedingly high and not representative for the site. 

           The Site A and B time-series graphs showed apparent decreasing trends for iron, see figures, above.  However, the trend in Site A appears to be based on two early time data points with the later time data being relatively stable.  Site C shows no apparent trend; only two data points differ significantly from the rest of the existing data, with the majority of the data showing little variation.  The later time data for Site B appears to level off, or even slightly increase for iron, but the concentrations remain significantly lower than the early time data. 

           The time series graphs for manganese show clear decreasing trends for all three sites.  See the figures below for the manganese time-series graphs.

           As discussed in Article 1 of this series, “Introduction to the Effects of Leachate Recirculation”, a visually apparent trend does not necessarily mean the trend is statistically significant. The data was analysed for statistically signficant trends using Sanitas, a statistical analysis software.  The trend analysis showed a statistically significant decreasing trend for iron only at Site A using linear regression.  No other statistically significant trends were detected for iron using either Sen’s Slope or linear regression analysis.  In contrast, both statistical trend analysis methods detected decreasing trends for manganese for each of the three study sites.   Therefore it appears that leachate recirculation is correlated with decreasing managanese concentrations, and no apparent effect on iron concentrations.  The table below summarizes the trend testing results for each site. 

            Why would the act of recirculating leachate reduce the concentrations of iron or manganese?  As discussed above, both iron and manganese exist in different forms depending on the chemical environment.  Changes in the chemical environment will influence whether these constituents exist in a soluble (reduced) or insoluble (oxidized) form.  The chemical environment in a landfill is typically reducing, therefore iron and manganese concentrations would be expected to be increasing over time. 

           Each of the sites studied recirculates leachate by spraying (as opposed to sub-surface application through trenches or pipe systems).  One possible explanation for the decreased or stable concentrations is that the act of spraying is introducing oxygen into the leachate, which will generate oxidizing conditions where the iron and manganese commonly precipitate out of solution.  Therefore, when the leachate reinfiltrates into the waste, some of the originally dissolved iron and manganese is left on the surface as precipitated solid mineral phase iron and manganese.  The act of infiltration then provides a second opportunity for iron and manganese to be sequestered out of the leachate.  As the infiltrating leachate comes into contact with soil,  iron and manganese can be taken out of solution by a processes such as cation-exchange and adsorption, leaving the leachate with lower concentrations of iron and manganese.  Essentially, the act of recirculation is treating the leachate by precipitating out some iron and manganese when exposed to air, then filtering the leachate through the soils.  Bacteria may also play a role in affecting iron concentrations in the leachate.  Iron-eating or iron-oxidizing bacteria derive energy from chemical reactions, with iron-oxidizing bacteria pushing the reaction to produce oxidized or solid iron. 

           A combination of one or more of these different factors (precipitation, adsorbtion, cation-exchange, and/or bacterial oxidation) is likely responsible for reducing dissolved concentrations of manganese.  These same factors also work on iron, however, iron does not show the statistically significant decreasing trends that manganese does in each of the three landfill sites studied. 

           One possible explanation for this is that iron is typically more common, and naturally occurs in soils and rocks at much higher levels than manganese.  According to the U.S. Geological Survey Water Supply Paper 2254, “Study and Interpretation of the Chemical Characteristics of Natural Water” by John D. Hem, there is only 1/50 as much manganese in the earth’s crust as there is iron, which is the second most abundant metallic element in the crust.  Both iron and manganese are mobilized in the reducing conditions of the landfill environment, but there is more iron, therefore the rate of iron dissolution in the landfill may be much closer or roughly equivalent to the rates of iron removal with leachate reciruculation resulting in no trend in concentrations.  Manganese naturally occurs in rocks and soils at lower quantities than iron, therefore, less manganese is dissolved into leachate overall and the same factors (precipitation, adsorption, etc.) have more of an effect. There is also a lot more iron in C&D waste in the form of rebar and other metal wastes than there is manganese.  The data available is insufficient to definitely identify the reason for the different trends observed between iron and manganese.

           The concentrations of iron and manganese in the leachate is still elevated in comparison to water quality standards.  However, with recirculation, the concentrations of these constituents are not increasing and for manganese continued reciruclation will result in lower levels over time.  As additional data is accumulated, this research will be updated to determine if the detected trends remain consistent.

           As for the original question posed in the first article: “Does leachate recirculation increase constituent concentrations?”  The answer for both iron and manganese is no. 

Previous Articles: Introduction to the Effects of Leachate Recirculation Series