Newsletter from Danish Research Centre for Organic Farming • September 2003 • No. 3

Grass-clover and catch crops

How does grass-clover and catch crops affect crop performance and soil microbiology in unfertilised organic crop rotations?

By Finn P. Vinther, Elly M. Hansen and Jørgen E. Olesen, Department of Agroecology, Danish Institute of Agricultural Sciences

In most organic cropping systems the use of N2 fixing crops, such as grass-clover, is of crucial importance. Additionally, the use of catch crops will reduce the loss of nutrients from the soil and thereby maintain the nutritional status at a higher level leading to a sustainable crop production. The use of N2 fixing crops and catch crops may also affect the microbial communities in the soil with consequences for the microbial biomass and activity.

The objective of this study, which was a part of the DARCOF-project BIOMOD, was to investigate the influence of crop rotation, particularly the prehistory regarding plant residue amendment, on crop yields, soil respiration, N mineralisation, on emissions of the green house gas N2O and on the microbial biomass and activity. The study is described and discussed in more details in Vinther et al. (2003), and only an extended abstract is presented here.

Experimental field and crop rotations

The study was performed in an organic crop rotation field trial established in 1996 at Research Centre Foulum. The trial includes several crop rotations. However, the measurements were only carried out in four selected rotations, which were present in two replicates. Crop rotation 1 and 2 represent rotations where a grass-clover crop is included and where cereals and pulses are undersown with a catch crop of ryegrass or grass clover (‘high input’ rotations), whereas the crop rotation 3 and 4 only includes cereals and pulses without catch crops (‘low input’ rotations). None of the selected rotations have been fertilized since 1996.


Measurements of various parameters were carried out during the growing season of 2001:

  1. Crop parameters, such as growth rate, N uptake, grain yield and N in grains, were measured in the 2001-crops
  2. Soil microbiological parameters, such as substrate induced respiration (SIR), fluorescein diacetate hydrolysis (FDA) and arylsulfatase activity (ASA), were measured in bare-soil subplots
  3. In situ measurements of N mineralisation, N2O emission and soil respiration were measured in bare-soil subplots

Three bare-soil subplots (1 m2) where all plants were removed, were established in each of the main plots. This means that the soil microbial parameters and the in situ measurements were not affected by the 2001-crops, but rather expressed the accumulated effects of the crop rotations from 1997 to 2000.

Input or return of nitrogen

Based on measurements of dry matter yields and N concentration it was possible to calculate the amount of N-input from above-ground plant residues. The input of N was significantly higher in crop rotation 1 and 2 than in 3 and 4, due to differences in the use of catch crops and grass clover. The relatively high variation in crop rotations 1 and 2 was due to grass-clover crops. In crop rotation 1 the return of N in 1998 from this crop was 465 kg N ha-1 and the grass-clover gave 190 kg N ha-1 in 2000 in crop rotation 2. The amounts shown only include aboveground plant material, which was returned to soil after ploughing. The root biomass itself and the amount of N, which was deposited as root exudates during the growing season, may however constitute a considerable part of the total C and N input to the system. Roots are known to excrete several forms of organic materials, and between 40 and 90% of the C transferred to the roots are lost as rhizodeposition in the soil and thus represent a considerable pool of C and N. Hence, the return of N as shown should be considered as minimum values.

Effects on crop parameters

Because crops were different, a direct comparison between the crop rotations was not possible. However, based on linear regressions, daily production rates were calculated showing that the rates were significantly higher in high-input rotations with catch crops and grass-clover than in the low-input crop rotations without catch crops and grass-clover. The production rates (± SE) were 122 ± 22 and 116 ± 12 kg dry matter ha-1 d-1 in rotations 1 and 2, and 77 ± 1 and 86 ± 9 kg dry matter ha-1 d-1 in rotations 3 and 4, respectively.

The total uptake of N during the growing season 2001 was 89 ± 9, 104 ± 8, 62 ± 8 and 61 ± 1 kg N ha-1 in crop rotation 1, 2, 3 and 4, respectively. The amount of dry matter removed in 2001 from the cropping systems as grain yield was 4113 ± 538, 3771 ± 71, 2832 ± 17 and 3225 ± 29 kg ha-1 in crop rotation 1, 2, 3 and 4, respectively. Similarly, 54 ± 6, 57 ± 1, 36 ± 1 and 43 ± 1 kg N ha-1 were removed from the system with harvested grains.

The summarised results of measurements on the 2001-crops showed that the ‘high input’ rotations were significantly higher than the ‘low input’ rotations for all measured parameters.

Effects on soil microbiology

Results of the microbial parameters substrate induced respiration (SIR), hydrolysis of fluorescein diacetate (FDA) and arylsulfatase activity (ASA) indicated that incorporation of plant residues from the different crops in the four crop rotations did not affect SIR during the initial months after incorporation. Only at the late sampling, SIR tended to be higher in the high-input rotations than in the low-input rotations. Using a relationship between SIR and microbial biomass-C as described in the literature, an average of 305, 304, 305 and 230 µg biomass-C g-1 soil could be calculated for crop rotation 1 to 4, respectively.

Hydrolysis of FDA were in the range from 51 to 182 µg fluorescein g-1 soil h-1, and the average values over the entire season were 107, 120, 90 and 89 µg fluorescein g-1 soil h-1 for crop rotations 1 to 4, respectively. The FDA-values in the high-input rotations were significantly higher than those of the low-input rotations. Values of ASA were in the range from 13 to 48 µg p-nitrophenol g-1 soil h-1, and the average values over the entire season were 27, 31, 24 and 22 µg p-nitrophenol g-1 soil h-1 for crop rotations 1 to 4, respectively.

The effects of crop rotation on the soil microbial parameters generally showed as the crop parameters the similar trends with higher activity in ‘high input’ rotations than in ‘low input’ rotation.

Effects measured by in situ methods

The net N mineralisation rates increased from 0.1 – 0.4 kg N d-1 to 1.0 – 2.1 kg N d-1 in the period, probably as a result of increasing soil temperatures. During the initial periods the mineralisation rates seemed to be slightly higher in the plots where pea:barley/grass and grass-clover residues were incorporated than in plots where pea:barley or winter wheat residues were incorporated.

During the period from the middle of March to the beginning of July 2001 the accumulated net N mineralisation in the upper 20 cm of the bare-soil plots were 87, 105, 88 and 71 kg N ha-1 in crop rotation 1 to 4, respectively. In comparison, 89, 104, 62 and 61 kg N ha-1 was measured as crop N uptake. For crop rotations 1 and 2 there is a very good agreement between the measured N mineralisation and uptake of N, whereas the N uptake in the crops in rotation 3 and 4 seemed to be less than potentially available from mineralisation, indicating that other growth parameters than N supply may have been sub-optimal in these rotations, or that the efficiency of N uptake is higher in the 'high-input' rotations.

The soil respiration ranged between 0.20 and 0.85 g CO2 m-2 h-1 during the experimental period from early April to mid September. During this period of 150 days the cumulated soil respiration amounted to 5292, 5260, 5430 and 4337 kg CO2-C ha-1 in crop rotation 1 to 4, respectively.

The N2O emission rates were characterised by a considerable temporal variation ranging from less than 1 mg N2O-N m-2 d-1 to peak values of 4-5 mg N2O-N m-2 d-1. The peak values coincided with rain events. Cumulated N2O emission during a period of 140 days from April to August corresponded to 3.4, 3.1, 2.7 and 2.7 kg N ha-1 in crop rotation 1 to 4, respectively, but differences between rotations were not statistically significant.

Overall effects

A principal component analysis of the above described results shows that the crop rotations are fairly well separated from each other, when soil microbial parameters or crop related parameters are included in the calculation. Especially when all parameters are included there is a clear difference between ‘high input’ rotations and ‘low input’ rotations.

Relationships between parameters

Some relationships and interactions between parameters are described and discussed below. These relationships are derived from data, some of which are encumbered with large statistical variation, and can therefore only be considered as supplementary indicators for the crop rotations studied.

Microbial degradation of soil organic matter results in the formation of inorganic N and CO2 in ratios depending on soil properties and origin of the organic matter. Therefore, C/N-ratios estimated from measurements of soil respiration and N mineralisation may give additional information on the soil properties and fertility. The slightly higher C/N-ratios in crop rotations 3 and 4 contrary to 1 and 2 indicate that they are deficit in N or that the microbial communities in rotations 3 and 4 are less efficient in utilising the C sources. The efficiency of microbial communities can be evaluated from estimates of metabolic quotients qCO2, i.e. units of CO2 produced per unit of microbial biomass C. Usually, the qCO2 is estimated from laboratory measurements of basal respiration and content of microbial biomass C in a few grams of soil. However, here we estimated qCO2 by combining laboratory measurements of microbial biomass-C using the SIR-method with field measurements of soil respiration, which were measured in bare-soil plots and therefore was assumed to represent the basal microbial respiration. Although the differences are small and not statically different, the qCO2-values in high-input rotations tended to be lower than in low-input rotations. This indicates that the microbial communities in the latter rotations were less efficient in utilising the C sources in the soil, as also indicated by the differences in C/N-ratios mentioned above.

Traditionally, N2O emission factors, i.e., N2O-N as the proportion of N added, are related to input of fertiliser N, and an emission factor at 1.25% is presently recommended by the Intergovernmental Panel for Climate Change (IPCC) for N2O emission from fertilised agricultural soils. In the present study, however, no fertilisers were added, and traditional N2O emission factors could therefore not be calculated. As ‘input’ the nitrogen N-mineralisation was used, and N2O emission factors were calculated as the proportion of N mineralised. The emission factors were ranging from 1.8 to 2.9%, and were higher than the 1.25% recommended by IPCC. However, the emission factors found are in the same order of magnitude as the 2.5% recommended after application of organic fertilisers.


Overall, the results of this study showed the importance of grass-clover crops and catch crops for the maintenance of soil fertility in non-fertilised organic cropping systems. These crops had a positive impact on parameters related to the performance of the cereal crops as well as on parameters related to the microbial communities in the soil.


Vinther, F. P., Hansen, E. M. and Olesen, J. E. (2003) Effects of plant residues on crop performance, N mineralisation and microbial activity including field CO2 and N2O fluxes in unfertilised crop rotations. Nutrient Cycling in Agroecosystems (in press).