Field-scale evaluation of the oil remediation capacity of the legume Galega orientalis
Lijuan Yan1, Miiro Jääskeläinen1,Petri Penttinen1,Tuula Puhakainen2, Kristina Lindström1, Asko Simojoki2, Fred Stoddard2
Departments of 1 Food and Environmental Science; 2 Agricultural Sciences; University of Helsinki, Finland
In previous studies it has been shown that deep-rooted Galega orientalis has potential for re-vegetation of oil-contaminated sites. In greenhouse experiments, biological activity associated with Galega roots accelerated the degradation of oil contamination, and plant growth promoting bacteria (PGPB) improved nodulation and plant yield (Mikkonen et al., 2011a). This needed to be confirmed in the field. Grass-legume mixtures often out-yield their components grown in pure stands, and grasses have different root-associated biological activity. Hence a multi-year trial was prepared using the legume Galega orientalis and the grass Bromus inermis, singly and in combination, to remediate oil-contaminated soil. Plant responses were determined by observing dry matter production and photosynthesis, and changes in the bacterial community structure were followed. This offered also an opportunity to measure N2O release.
Plots were established in June 2009 in a Randomized Complete Block Design at the Viikki Experimental Farm, University of Helsinki (60°23’N 253’°E). Galega orientalis cv Gale, Bromus inermis cv Lahis were sown in pure stands and in a 75 grass: 25 legume ratio, with a bare soil (fallow) control. Each plot was divided into 4 subplots, approximately 2 m x 2 m, with factorial combinations of ±fuel oil (6 kg per plot, 7000 ppm), ± PGPB. In 2009, 60 kg/ha N fertilizer was applied to the grass plots, but none subsequently. Soil samples were taken 4 times: A: July 2009, B: May 2010, C: November 2010 and D: May, 2011. Top soil (0-20 cm) was passed through a 5 mm sieve and stored in -20 °C freezer before analysis. DNA was extracted using standard techniques. Soil bacterial community structures were studied by length heterogeneity PCR (LH-PCR), using domain-specific primers to amplify 16S rDNA (Mikkonen et al., 2011b). Gas traps were installed in May 2011. Gas samples were taken 6 times during the 2011 growing season. N2O content of the collected gas was determined by GC.
In 2010, plant dry matter production was greatest in the crop mixture plots and in the +Oil +PGPB treatment. In 2011, there was no significant difference in plant DM production among the main plot or subplot treatments. Galega proportion in the mixtures declined through each growing season but recovered in spring.
At the start of the experiment, LH-PCR fragment lengths differed greatly between oil and no-oil plots, but showed few differences between crops . Multi-dimensional scaling ordination (on the non-PGPB treatment only) separated the LH-PCR profiles of bare soil, Bromus, Galega, and mixture. Oil-treated plots at times A and B were clearly separated from non-oil plots. By time C the oil plots clustered with the non-oil plots. At time D, all treatments had shifted.
N2O release was greatest from the bare soil except at the start of the season. N2O release from the Galega plots exceeded that from the grass and mixture plots.
It is concluded that the soil bacterial community can recover from oil contamination. The levels of oil contamination, although high, were not sufficient to impede plant growth. N2O emission was higher from legume plots than unfertilized, legume-free grass plots, and next year the comparison will be made using fertilized grass plots. The harvested crop is suitable for digestion to biogas.
Mikkonen et al., 2011a. Geoderma 160: 336-346;
Mikkonen et al., 2011b FEMS Microbiology Ecology 78: 604-616.