Grasslands store large amounts of soil organic carbon (SOC) and this is an important element to be considered in land use systems that are part of the food production chain. The most central research tasks in this project were to find to what degree the ploughing frequency of a grassland together with fertilization practices affect SOC storage. To approach this, the existing long-term grassland experiments at southern west and northern Norway (maintained since 1968) and at western Norway (maintained since 1974) have been used to quantify SOC content down to approx. 60cm soil depth in grasslands managed with different ploughing frequency. We have collected soil samples from four depth increments (0-5, 5-20, 20-40 and 40-60 cm) at all sites. The results indicate that there is no difference between short- (ploughed every 3 years or 6 years) and long-term (not ploughed for almost 25 and 50 years) grasslands on SOC stocks at any of the sites. The experimental sites have different soil types and are therefore not directly comparable. However, former peatland soils in Svanhovd clearly contain the highest SOC stock as compared to the other sites. Soil organic carbon stocks decreased in the soil profile for both grassland types.
The project has also evaluated how ploughing frequency affect C accumulation and distribution in soil profile. Carefully excavated soil profiles were brought to the lab and analyzed with hyperspectral imaging with a method developed by NIBIO in collaboration with NMBU. One soil profile was selected for each study site and sampled cm by cm. Collected soil material were grinded at 200 mm and analyzed for carbon content with an isotopic carbon analyzer (Picarro). The goal was to find a predictive model between the carbon content and the hyperspectral signature of the corresponding scanned area using partial least square regression (PLSR). PLSR model can be built by combining different methods, such as: 1) variable selection (which wavelength to include), 2) signal pre-treatment, 3) cross validation strategy, and 4) number of predictive components to includes in the model. More than 35000 models were tested and 8 of them proved to be very successful at describing the C profiles used for calibration. However, when applied on the extra profiles the prediction of these model diverged substantially from one another indicating a weak predictive power. Electromagnetic soil mapping with EM38 at Særheim showed large soil variability within the experimental field. Probably large soil heterogeneity affected our results. As a consequence, assessing the impact of the ploughing frequency on the vertical carbon distribution in the profile via hyperspectral imaging does not seem to be possible in the present conditions.
Populations of earthworms were examined in 3 depths on selected plots at Særheim and Furuneset. Between 5-6 different earthworm species were identified at both fields, dominated with Apporectodea rosea and calignosa (Pink and gray earthworms). At Særheim, we found in 25 x 25 x 40 cm excavated earth columns, 17 earthworms on average (min 5, max 29, n = 4). At Furuneset, 36 earthworms on average (min 19, max 57, n = 4) per column, respectively. Rough calculations indicated that here were approximately 272 earthworms per m2 or 272,000 earthworms per dekar at Særheim and 576 earthworms per m2 or 576000 earthworms per dekar at Fureneset. It seems that earthworms can withstand pasture renewal at least every 6 years.
For the first time in Norway we have mapped the content of soil organic matter (SOM) content in soil based on soil samples taken at the farm level. We included over 100,000 soil samples that have been taken by farmers from 1990-2016 and sent to Eurofins and then subsequently stored in a database managed by NIBIO. Data was aggregated and displayed at the municipal level to protect the farmer's data security rights. We have compared SOM content affected by crop type cultivated (grass vs cereals) on a farm. There was a significant effect of soil type and interaction between type of crop and soil type on SOM content. There was a clear trend towards increased SOM with the cultivation of grass, especially in organic soils but also in sandy soils. Several maps of SOM across municipalities are made, which presented at our NIBIO homepage.
Studies, which combined historical with new data showed no or a weak decrease in SOC in 0-20cm depth in long-term grassland (between 1986 and 2019). In contrast, we found significant decrease in SOC in short-term grassland (ploughed every 6th year) over 30 years period.
A farm-economic analysis using data from the long-term experiments at Særheim and Fureneset recognised that ploughing every 3rd year was less profitable than ploughing every 6th year or unploughed.
På kort sikt 1. Forskere med ulik bakgrunn økte kunnskapen om eng som karbonlager og hvilke driftsmetoder bør brukes for å opprettholde karbon over tid. 2. Nyttig kunnskap om hvordan karbon fordeles i jorda i langvarig og kortvarige eng. 3. Tilgang på kunnskap om moldinnhold i norsk jord. 5. Mer kunnskap om sammenheng mellom engas avling og karbon innhold i jorda. 6. Økonomiske vurderinger om omløpstid i eng- nyttig kunnskap for bonden.
På lang sikt 1. Kunnskapen inkluderes i videre arbeid knyttet til klimaendringer. 2. Større interesse for eng som karbonlager blant produsenter.
There is a debate underway about how we can pursue simultaneously the goal of increasing food production while also reducing greenhouse gas emissions. The area under permanent grassland extends from Norway's southernmost agricultural territory along the mountainous west coast to arctic Northern Norway. Grasslands store carbon (C) and this can compensate emissions originated from food production. The most central research tasks in this project are to find to what degree the age of grassland together with management practices affect C storage and how to combine high areal productivity with high C sequestration.
To approach this, the existing long-term grassland experiments at southern west and northern Norway (maintained since 1968) and at western Norway (maintained since 1974) will be used to quantify C content down to 60-70 cm soil depth in grasslands older than 40, 20 and 6 years. The project will also evaluate how different grassland management systems including grassland age and productivity affect C accumulation and distribution in soil profile. Modelling studies will combine historical and new data to predict C sequestration in grasslands under changing climate. Economic value of C sequestration will be calculated and C content in forage production regions of Western and Northern Norway will be mapped. The project uses an integrated approach, combining experimental work, novel technologies and model-based methods. To accomplish this, researchers of Norwegian Institute of Bioeconomy and University of Life Sciences will collaborate. Extension service together with Norwegian Farmers association will ease the flow of information from research to farmers, local authorities and police makers. High-rating university in Sweden will also be involved. The project will contribute to the following research priorities of the FFL/JA call: sustainable food production, particularly, to increase knowledge on processes promoting better carbon balance in Norwegian agriculture.