Turnover of organic matter in a Miscanthus field: effect of time in Miscanthus cultivation and inorganic nitrogen supply

Abstract

To accurately predict the potential environmental benefits of energy crops, the sequestration of carbon in soil needs to be quantified. The aim of this study was to investigate the mineralisation rate of the perennial C₄ grass Miscanthus giganteus and Miscanthus-derived soil organic matter under contrasting nitrogen supply. Soils were collected from sites where Miscanthus had been grown for 11 and 18 years, respectively, and where a C₃-grass (Lolium spp.) had been grown for 7 years. The soils were incubated for 4 months at two levels of soil inorganic nitrogen with or without dead root material of Miscanthus.

Addition of root material (residues) increased carbon mineralisation of indigenous organic matter when no nitrogen was added. Added inorganic nitrogen decreased carbon mineralisation in all soils. Nitrogen addition did not affect carbon mineralisation of the residues. Using the ¹³C fraction to calculate the proportion of respiratory CO₂ derived from Miscanthus showed that nitrogen addition decreased carbon mineralisation in soils, but it did not affect carbon mineralisation of the residues. Nitrogen mineralisation was highest in the C₃ grass soil without added residues. Nitrification decreased pH, especially in the treatments where nitrogen was added. The Miscanthus-derived organic matter is at least as stable as C₃ grassland-derived organic matter. Furthermore, the turnover time of the organic matter increases with time under Miscanthus cultivation.

The CENTURY soil organic matter sub-model was used to simulate the organic matter decomposition in the experiment. Carbon mineralisation was accurately simulated but there were unexplained discrepancies in the simulation of the δ¹³C in the respiration from the treatment with residues. The δ¹³C in respiration did not decrease with time as predicted, indicating that lignin accumulation did not influence the measurements.

Introduction

Growing energy crops for biofuel have been shown to be the best way to use surplus agricultural land in Europe for greenhouse gas mitigation (Smith et al., 2000). These calculations were based on both the direct substitution of fossil fuels and on carbon sequestration in the soils. However, there is uncertainty in the estimates of carbon sequestration in soils because the total increase or decrease in soil organic carbon (SOC) is determined by the amount of crop residues left in the field and their turnover-time.

Most studies of decomposition have focused on plant litter. When predicting carbon balance, both the build-up and the mineralisation of soil organic matter (SOM) must be taken into account. This has been investigated in long term field experiments where unplanted soils have also been included (Christensen, 1990). However, process studies under controlled conditions are lacking on this matter. Only the temperature response of soil carbon mineralisation has received some attention (Kirschbaum, 1995, Kättener et al., 1998, Lomander et al., 1998, Giardina and Ryan, 2000, Ågren and Bosatta, 2002). The study of SOM dynamics is complicated by the priming effect (Jenkinson et al., 1985, Kuzyakov et al., 2000). In addition, decomposition of native SOM may be influenced by organic matter amendments, living plants, tillage and, in some cases fertiliser additions.

Miscanthus giganteus is a perennial C₄-grass originating from South–East Asia, which has been proposed as a promising energy crop for Europe (Lewandowski et al., 2000, Vleeshouvers, 2001). It is also known to have a deep and extensive root system (Neukirchen et al., 1999) and, therefore, has a potential to increase carbon stores in the soil (Fisher et al., 1994), at least when established on arable land. The decomposability of Miscanthus roots is comparable to that of farmyard manure, while decomposability of harvest residues and rhizomes are comparable to cereal straw (Beuch et al., 2000). The need to know more about the subsequent decomposition of the SOM formed was emphasised in that study. As most of the aboveground parts of energy crops are removed, the roots will form the main soil carbon input.

About 1.11% of the carbon in CO₂ in the atmosphere is ¹³CO₂. As C₃-photosynthesis discriminates against ¹³C to a greater extent than C₄-photosynthesis (Farquhar et al., 1989) the difference in the ratio between ¹²C and ¹³C in C₃- and C₄-derived material can be used as a natural tracer. This has been used to trace recently formed carbon in fields that have been changed from C₃ to C₄ plant cover or vice versa, or to label plant material from different sources (Cheng, 1996, Balesdent and Balabane, 1996, Six et al., 1998, Collins et al., 1999, Collins et al., 2000, Bol et al., 2000, Molina et al., 2001). However, the accuracy of estimating carbon source using this natural labelling is somewhat lower than when using radiocarbon (¹⁴C) because the natural discrimination varies with growing conditions (Jenkinson et al., 1995). In northern temperate areas as Denmark, there are no wild C₄-plants and C₄-crops have not been grown until recently.

In the field, the addition of nitrogen usually results in an increase in SOM due to greater plant production (Christensen and Johnston, 1997). It is argued that this could be partially offset by an increased decomposition of plant litter since nitrogen limited decomposition has been observed in some cases (Henriksen and Breland, 1999). However, although nitrogen content in residues is positively correlated with decomposition rate (Melillo et al., 1982, de Neergaard et al., 2002, Tian et al., 1995, Mafongoya et al., 1998) the effect of inorganic nitrogen on decomposition is variable and sometimes nitrogen addition can even result in decreased decomposition rate (Magill and Aber, 1998). Inorganic nitrogen seems to speed up the decomposition of soluble carbohydrates but may well have the opposite effect on more recalcitrant materials such as lignin (Fog, 1988). The reason for this is still not fully understood but it may be related to (1) a change in the competition between different groups of decomposers; (2) ammonium inhibition of lignin-degrading enzymes or (3) promotion of the formation of recalcitrant compounds (Fog, 1988, Carreiro et al., 2000).

Models are useful when predicting carbon sequestration over long time-scales (Paustian et al., 1992, Parton and Rasmussen, 1994, Falloon et al., 2001, Foereid and Høgh-Jensen, 2004). Although testing the models over long time-scales may not be possible in all cases, the description of processes in the model can be validated. Most SOM models either include an assumption of no or positive effect of added nitrogen on decomposition. Consequences of violating this assumption are unknown.

Since Miscanthus is a C₄ plant, natural labelling may be used to track the carbon in the soil that had been stored during its cultivation in soils that have previously only been cropped C₃-plants. C₃-grass was used as a control because disturbance regime is similar to that of Miscanthus. The objectives of this study were: (1) to determine the stability of Miscanthus-derived carbon and plant residues to disturbance and the addition of inorganic nitrogen; (2) evaluate how carbon stability developed with time in Miscanthus cultivation; and (3) test if an established simulation model could adequately describe the development given the assumptions of the method.