Field experiments 2012-2014
Field experiments 2012–2014
In the field experiments 2012–2014 two different substances were used, limestone and slaked lime, to treat the acid sulfate soil underneath the plough layer in a field in agricultural use. Limestone consists mostly of calcium carbonate (CaCO3) and the ultrafine-grained product used was Nordkalk’s FC2,5 which is limestone grinded to a median particle diameter of 2.5 µm. This product has later been renamed C2. The slaked lime consists of calcium hydroxide (Ca(OH)2) and the product used was Nordkalk’s SL 90 T where 96.3 % of the particles have a diameter smaller than 90 µm.
Close by the control well of the subfield to be treated, a mixing vessel with a volume of 3 m3 was placed (Fig. 2). The vessel was filled with river water and the treatment powder was added. The suspension was mixed continuously to avoid sedimentation using an electric motor and a generator. The particle sizes of the used powders were however small enough to make the suspensions relatively stable.
The soil in a subfield typically could receive a subirrigation flow of about 3–4 litres/second and the treatments were done with a flow of about 10 m3 per hour or about 2,8 litres/second. The outflow from the mixing vessel was adjusted so that the water level in the receiving control well was high and stable. Water head pressure from the water in the well helped in pushing the suspension out from the drainpipes into the soil through cracks and other macropores of the soil. The flow of water from the pump by the river to the mixing vessel was adjusted with a valve on the water pipe and the level in the mixing vessel was also kept high and stable.
Material for the suspensions was kept in big bags on a tractor trailer by the mixing vessel. Dosing was done manually (Fig. 3). The additions were made regularly, and the target concentration of the suspension was at a maximum 6–7 grams/litre or 6–7 kg/m3. The finished suspension was led directly to the control well (Fig. 4).
The first large-scale experiments were done in the summer of 2012. Two subfields, 3 and 5, were treated with 680–770 kg of FC2,5 ultrafine-grained limestone in about 110 m3 of river water. Two other subfields, 1 and 8, were treated with 110–150 kg of slaked lime in about 120 m3 of river water. The reference subfields 2, 4 and 7 received about 100 m3 river water each.
The field experiments continued in 2013. Two subfields, 6 and 9, were taken into use this year. The same chemicals as in 2012 were used in the experiments but the doses were different. Subfields 6 and 9 were treated with FC2,5 but the amounts were halved compared to 2012. Subfields 1 and 8 were treated with slaked lime again but the dose was now two to three times higher than in 2012.
In the experiments in 2014 subfields 6 and 9 were treated again with FC2,5 but the amount was now doubled, i.e., about 1600 kg of ultrafine-grained limestone in about 450 m3 of river water was used.
After each treatment, about 10 m3 of river water was pumped into the control well to rinse the subsurface drainage and to push any sedimented suspension out into the soil.
Most treatments could be done during a long workday. The flow was about 10 m3/hour and a treatment requiring 120 m3 therefore took about 12 hours. It was only the treatments in 2014 that took several days.
Table 1. Summary of the treatments done during the summers of 2012, 2013 and 2014.
Methodology used in following up treatment effects on drainage water
The procedures followed in sampling and analysis of drainage water developed during the years, and the final steps and methods used were:
1. Using a petrol pump, Honda WX15, stagnant water in the control well and subsurface drainage pipes was pumped out and discarded. The year after a treatment, some sedimented suspension was left in the pipes and the water was turbid. To avoid taking samples affected by sediments in the pipes, and to make sure the samples correctly represented groundwater in the soil, three drainage system volumes (i.e., three times the combined volume of the subsurface pipes and the control well) were pumped before sampling. This took about 20 minutes. However, the pumping time was prolonged if the water was still turbid after 20 minutes.
2. Using a battery-operated peristaltic pump, Eijkelkamp 12.25, and a handheld YSI Professional Plus meter together with a Quatro cable and a flow-through cell, pH, conductivity, and redox potential were measured during continuous pumping of drainage water through the flow-through cell. The intake end of the pump tube was kept as close as possible to the outlet of the collector pipe in the control well to avoid influence of atmospheric oxygen especially on the redox potential.
3. Samples for the determination of the anions sulfate, nitrate, and chloride, as well as the acidity titration, were taken with a handheld swing sampler directly by the outlet of the collector pipe in the control well and 250-ml bottles was filled completely. Contact with the atmosphere was also in this case minimized as, e.g., carbon dioxide losses to the atmosphere will affect the result of the acidity titration.
4. Samples for the determination of metals were taken with the swing sampler and transferred to a syringe and filtered through a Filtropur S 0.45 µm filter directly into a Falcon tube. The samples were acidified with suprapur HNO3. Early in the project, the target pH in the acidification was 2, but this was soon changed to pH 1.7.
5. The very first determinations of ferrous iron were made on unfiltered samples, but very soon this was changed, and the determinations were made on samples filtered as above. To stabilise the balance between ferric and ferrous iron in the sample, a method (described in detail here) where ferric iron is forming a complex with NTA and ferrous iron is forming a coloured complex with phenanthroline was used. Before sampling, a reagent mixture is pipetted into the sample bottles. In the field 40 ml of sample is added to the bottle. This way the ferric iron is forming the NTA-complex and ferrous iron is forming the coloured complex directly when sampling. Early on the sample was added by pipette in the field, but this was soon changed to the use of a battery-powered Ohaus YA102 field scale. The scale can be calibrated in the field while a pipette is usually calibrated for use in the laboratory at room temperature. Sampling took place outdoors usually at temperatures close to 0 °C. The samples were stored on ice if there was reason to believe that iron-reducing bacteria were present in the sample. This was especially important when sampling the river water.
The acidities were titrated, the anions were determined by ion chromatography, and the ferrous iron was determined by spectrometry in the laboratories of the Novia and Vaasa Universities of Applied Sciences. The metal analyses by ICP were subcontracted.
Visual examination of the drainage system in the autumn of 2014
After harvest in September 2014, trenches were dug in subfield 1 and 9. In both subfields, the trenches were dug perpendicular to a drainage pipe with one trench close to the control well and one close to the end of the drainage pipe. In subfield 1, treated with slaked lime, some of the suspended material was left in the drainage pipe as a sediment (Fig. 5). Least sedimentation in the drainage pipe was found close to the control well and more towards to the end of the drainage pipe. The phenomenon was clearly related to the flow rate of the suspension. There were no visible traces of the suspended material in the gravel surrounding the drainage pipe in in the nearby subsoil.
In subfield 9, which had been treated with ultrafine-grained limestone, only a limited amount of sedimentation was seen (Fig. 6). Sedimentation was evident towards the end of the drainage pipe, but hardly any was visible close to the control well. There was clearly less sedimentation compared to subfield 1, this most likely being a consequence of the smaller particle size used. The drainage gravel surrounding the pipe was loose and permeable to water.
The full length of the drainage pipes in subfields 1 and 9 were inspected by video on 12 September 2014 by the Eerola company. The results agreed with what had been seen in the dug trenches. It was once again concluded that the flow rate of the suspension and the particle size are of great importance for minimizing sedimentation and promoting the suspension’s further spread into the soil.
The drainage pipes in all treated subfields were flushed on 26–29 September 2014 by Nybacks Gräv. The water was pumped into the drainage pipes using the flush pipes and sedimented material was flushed to the control wells, from where it was pumped into the open ditch.
Summary of the results of the field experiments
Clear traces of the treatments were visible in the soil when trenches were dug about one month after the treatments in September 2014. Especially in subfield 9, where the treatment had consisted of the largest amount of ultrafine-grained limestone, it was possible to see visible limestone powder on the surfaces of cracks in the soil down to the reduced zone about 1.8 metres below ground level, and up to four metres from the drainage pipe close to the control well (Fig. 7). This was seen as a very encouraging result in the form of the extent to which the suspension spreads in the soil. A pH gradient could also be detected in the pore water using a handheld field pH meter. Close to the drainage pipe the pH was elevated to around pH 5 and about two metres from the pipe the pH of the pore water was below pH 4.
Directly visible traces of the treatment substance were not seen in subfield 1. This may very well be because the treatments were made in 2012 and 2013, and so when the trenches were dug in 2014 most of the treatment substance in the form of slaked lime had dissolved. Elevated pH values could be measured in the drainage gravel surrounding the pipe closer to the control well.
Treatment effects in treated subfields are also evident in the form of an improved quality of the drainage water. E.g., after treatments higher pH values, and lower values for titrated acidities, as well as lower metal concentrations were seen. In reference subfields, the pH is consistently below 4 and titrated acidity and metal concentrations are quite high.
A low dose of slaked lime (calcium hydroxide) did not produce a clear effect on the drainage water, while a higher dose raised the pH from 4 to 5 and lowered the titrated acidity from 4 mmol/litre to less than 1 mmol/litre. Of the metals, e.g., the aluminium concentration was lowered from about 25 mg/litre to less than 1 mg/litre. Even the following summer the aluminium concentration was still below 1 mg/litre. This clearly shows that the amount of treatment substance used is important when it comes to whether the quality of the drainage water will be improved or not.
Similar results were achieved with treatments with limestone suspensions. A low dose gave an effect in the form of an improved drainage water quality that lasted a shorter period of about 4 months, while a medium dose gave a noteworthy effect in the form of a pH value of about 6 and lowered values for titrated acidity (below 3 mmol/litre) and aluminium (1 mg/litre). However, the effect is maintained for about 1.5 years after which it slowly fades. With the large amount of limestone used in 2014, the effect on drainage water quality is more long-lasting.
In the article Subsurface hydrochemical precision treatment of a coastal acid sulfate soil by K. Dalhem et al. published in the journal Applied Geochemistry in 2019, an in-depth analysis of the results from the field experiments can be found. A popularized article in Finnish, Kemiallinen täsmäkäsittely happamilla sulfaattimailla, that describes the field experiments and results was published in the journal Vesitalous in 2019.
The microbiology at Risöfladan
During 2012, the first samples for an initial microbiological characterisation of the soil at Risöfladan, and the treatments effects on soil microbiota, were taken. The study was funded by K.H. Renlunds Stiftelse. The analyses were done at the Linnaeus University in Kalmar, Sweden, and published in the article Microbial community potentially responsible for acid and metal release from an Ostrobothnian acid sulfate soil published in the journal FEMS Microbiology Ecology in 2013.
