|dc.description.abstract||This report describes the establishment, experimental plan and initial soil characteristics of the field experiment linked to the project “Effects of anaerobically digested manure on soil fertility - establishment of a long-term study under Norwegian conditions” (SoilEffects, 2010-14). The aim of the SoilEffects project is to identify potential risks and benefits for soil fertility when animal manure is anaerobically digested for biogas production.
The field experiment was established on Tingvoll research farm in 2011. A biogas plant was built at this farm in 2010, to digest the manure from a herd of about 25 organically managed dairy cows. This report describes the initial characterization of the soil biology, chemistry and physics, along with the background of the project, the selection process of the research field and the project design. Effects of the manure treatment and application will be studied during 2011-14. The aim of this report is to function as a reference for later publications, and to inform other scientists establishing medium long-term field trials. Except from a few results on water-soluble C and soil microbiology from 2011, all results presented here are based on studies conducted in autumn 2010 or spring 2011, before any manure was applied. Effects of the experimental treatments were studied for different soil characteristics in 2011, 2012 and 2013 and will be presented in separate publications.
The field experiment has two cropping systems; grass (perennial grass-clover ley) with 20 experimental plots, and arable with 20 plots. Each experimental plot measures 3 m x 8 m. The ley was established in 2009 with cereals as a cover crop. The arable system was established in 2011, by ploughing a part of this ley. In the arable system, the soil is ploughed annually in spring, no legumes are grown, and aboveground plant material is removed at harvest. This practice is intended to stress the maintenance of soil organic matter in the arable system, to possibly reveal clearer effects of the experimental treatments. Within each cropping system, five experimental treatments are compared. They comprise two fertilization levels for each type of manure, plus a control treatment with no manure application. Each treatment has four replicates, randomly distributed within four blocks in each system.
There is a significant soil variation on the experimental field. However, for most of the studied characteristics, no statistically valid differences were found between average values across blocks within each cropping system.
The content of soil organic matter (SOM) is higher in the grass system than the arable system. In the upper soil layer (0-20 cm) the average SOM content measured by ignition loss was 11.3 % in the grass and 6.6 % in the arable system. Analyzed by total-C measurements, the corresponding SOM values were 11.03 % and 5.97 %. In Norwegian soil, SOM values between 3 and 6 % are regarded as high humus contents (“moldrik”), whereas values between 6 and 12 % are regarded as very high. The average values for total C (0-20 cm) were 6.41 in the grass and 3.47 % in the arable system, and for total-N 0.39 and 0.21 %.
On average for all treatments in the grass system (n= 20), the upper soil layer contained 0.32 mg organic C per g soil (air dried) by extraction in cold water (CWEC), increasing to 1.7 mg by hot water extraction (HWEC). In the arable system, the corresponding values were 0.23 and 1.1 mg.
The SOM content of the grass system was higher and more variable than that of the arable system, and differences between blocks were greater and more statistically significant in the grass than in the arable system. Differences in the initial SOM between the means of plots that have been assigned to different subsequent treatments of manure applications were on the whole much smaller than those between blocks within the same crop system. Nevertheless, significant differences were found in some cases, and thus the initial SOM status of the soil should be taken into account when interpreting differences that may arise after the treatments have been carried out for a number of years.
The soil texture, loamy sand (‘siltig mellomsand’) was similar in all replicate blocks and both depths in the grass system. It was slightly heavier and somewhat more variable in the arable system, with on average 16 % less sand, 11 % more silt and 4 % more clay. Somewhat heavier soil in the deeper parts of the terrain may be explained by washing out the soil layer during postglacial land elevation. The gravel contents were fairly low (< 10 %) in all cases.
Soil moisture retention and aeration properties of the upper soil layer were measured on each plot. Total porosity, aeration properties and moisture retention at low tension were all clearly greater in the grass system than in the arable system. Satisfactorily high levels of aeration and plant-available water-holding capacity were found in both systems. Close relationships were seen between the moisture retention and the soil organic matter content. This accounts for many of the differences in such properties that were found between blocks.
Soil aggregate size distribution was measured in the seedbed of the arable system plots. This confirmed that the predominant structure of the soil may be described as ‘single-grain’, with only 16 % aggregates of 2-6 mm and 16 % aggregates > 6 mm. There was little variation between blocks in the aggregate size distribution. The stability of soil aggregates (2-6 and 6-10 mm) to simulated rainfall was high (>85 %) in all cases, with little variation between blocks or treatment means.
The soil nutrient content was comparable in the two cropping systems. The nutrient concentrations in the upper soil layer (0-20 cm) were P-AL 2.87/2.31; K-AL 5.25/5.24; Mg-AL 4.34/3.53; Ca-AL 107.6/80.7; K-HNO3 122.4/175 mg of nutrient 100 g-1 dry soil in the grass/arable system. The pH value (H2O) was 5.82/5.87.
The accumulated soil respiration and the microbial community structure differed between the grass and the arable system. Soil respiration seemed to be influenced both by manure application and cropping system. In 2011, no significant change in the soil microbial community structure was found five days after manure application. This, however, may change with repeated manure applications over several years.
Five earthworm species were identified in the field experiment. Apporrectodea caliginosa was the most common, but also Lumbricus terrestris was abundant. Octolasion cyaneum was found mainly in the arable system. The average density was 133 earthworms m-2 in the grass system and 117 in the arable system. The average biomass was somewhat higher in the arable system (63.5 g m-2) than in the grass system (42.1 g m-2).
Collembolans were sampled from the grass system, in treatments with no or high manure application (but before manure application), from totally 12 plots. 17 species of collembolans were found, with an average density of and 7950 individuals m-2. The variation in species composition and density was high, and larger between treatments than between blocks. 11 species were found in all treatments. The most numerous collembolan species were the soil dwelling, white Mesaphorura macrochaeta and Protaphorura armata, and the litter dwelling greenish Isotomurus graminis.||nb_NO