«International 17 Workshop th Nitrogen The was jointly organised by Teagasc and AFBI Printed by Print Depot Suggested citation Authors, 2012. Title ...»
2. Materials & Methods CASIMOD’N aims at ensuring farm consistency (e.g., matching feeding needs and effluent production of livestock with crop plans and management practices), by modeling farmer strategy and subsequent practices and to model the nitrogen transfer and transformation at the catchment scale. It results mainly from the coupling of a farm model and an agro-hydrological model. The farm model was based on the existing TOURNESOL and FUMIGENE models (Chardon et al., 2008), and allowed simulating crop allocation and manure spreading in mixed farming systems. It allowed considering multiple farm structures and strategies. The strategy was implemented by setting the farmer preferences in the feeding ration and assigning a set of priorities to each crop, waste management and time of application for each field. The hypothesis underlying the crop allocation modeling was that in dairy farms, the feeding requirements of the herd are a major driver of farmer strategy to design cropping plans. Crop allocation and waste management were generated yearly for each farm during the simulation period. The agro-hydrological model was TNT2, topographic-based nitrogen transfer and transformations (Beaujouan et al., 2002). TNT2 is a detailed agro-hydrological model that simulated in a process-based, spatially distributed way the nitrogen transfers and transformations associated with management practices, crop growth and hydrological processes within the catchment.
Surveys were conducted on 54 farms and crop successions on each field were extracted from remote sensing analysis from 1997 to 2006. The validation of the farming systems were performed by testing ability of the CASIMOD’N model to reproduce the crop succession and allocation, the management of mineral and organic fertilization and the satisfaction of the herd alimentation needs.
Two prospective scenarios, based on indicators emerging from discussions among stakeholders and
Nitrogen Workshop 2012
decision makers were tested. These indicators have the ambition to be rather simple to compute, generic among the farming systems but still structuring the farming systems. The selected indicators were: (a) a maximum stocking rate of 1.4 Livestock Unit (LSU) per hectare of meadow, (b) a threshold of 100 kg N ha-1 input nitrogen at farm scale.
Figure 1 shows that in most cases the simulated scenario were compatible with maintaining the herd alimentation, but also highlighted some cases in which the farming systems could not comply with the proposed levels of the indicators. The simulation also suggested that this scenario would result in a significant descrease of N fluxes at the outlet of the catchment. Finally, the scenarios results gave valuable insights on the main components of N budget, with respect both to their temporal evolution and to their spatial distribution at the catchment scale.
4. Conclusion The new tool developed in this work, based on the coupling of a farm and a catchment models, proved its efficiency in reproducing past agricultural practices, and in generating prospective scenarios consistent with the farming system constrains. It will be used within the Acassya project (ANR-08-STRA-01) as a complementary tool to help farmers and stakeholders to build effective scenarios aiming at reaching the water quality objective while maintaining viable farming systems.
References Beaujouan, V., Durand, P., Ruiz, L., Aurousseau, P. and Cotteret, G. 2002. A hydrological model dedicated to topography-based simulation of nitrogen transfer and transformation: rationale and application to the geomorphologydenitrification relationship. Hydrological Processes 16, 493-507.
Chardon, X., Raison, C., Le Gall, A., Morvan, T. and Faverdin, P. 2008. Fumigene: a model to study the impact of management rules and constraints on agricultural waste allocation at the farm level. J. Agric. Sci. 146, 521-539.
Gouttenoire, L., Fiorelli, J.L., Trommenschlager, J.M., Coquil, X. and Cournut, S. 2010. Understanding the reproductive performance of a dairy cattle herd by using both analytical and systemic approaches: a case study based on a system experiment. Animal 4, 827-841.
Nitrogen Workshop 2012
Influencing factors on the nitrate residue levels in Flemish agricultural soils: a statistical analysis of 8 years of nitrate measurements Tits, M.a, Elsen, A.a,Vandervelpen, D.a, Bries J.a, Vandendriessche, H.a, Van Overtveld, K.b, Diels, J.b a Soil Service of Belgium, Heverlee, Belgium b Department of Earth and Environmental Science, K.U. Leuven, Belgium
1. Background & Objectives In Flanders, nitrate residues in the soil profile are used as an indicator for the risk of nitrate leaching from agricultural soils to surface and ground water. The nitrate residue is defined as the amount of nitrate-N present in the soil profile (0-90 cm) during the period October 1st to November 15th. They are annually measured, on one hand in a (more or less) directed selection of agricultural parcels (commissioned by the Manure Bank) and on the other hand in all the parcels having an agroenvironmental agreement “Water” (AEA Water). This produces datasets of 18 000 up to more than 30 000 nitrate residue measurements per year.
In the framework of the assignment of the Flemish Government to evaluate and differentiate the current nitrate residue standard, an extensive review of the historical nitrate residue measurements was made.
This descriptive analysis had to provide an answer to the following questions:
- which factors have a significant influence on the nitrate residue levels?
- what are the impacts of policy measures such as stricter fertilisation standards, agroenvironmental agreements, etc.?
2. Materials & Methods The available datasets for the analysis consisted of the nitrate residue measurements in parcels with AEA Water (173 022 usable measurements from 2001 to 2008) and the control measurements performed by the Manure Bank (35 916 usable measurements from 2004 to 2008). Despite the large size of the datasets, their representativeness for other agricultural parcels in Flanders is relative. The first dataset contains the complete population of all parcels with AEA Water in Flanders, but nitrate residues in these parcels are on average lower than in current agricultural parcels because of stricter fertilisation limits and a better implementation of optimal farming practices. The latter dataset consists each year of a directed sample of agricultural parcels towards derogation parcels, parcels in vulnerable zones, parcels in risk zones, parcels with a higher risk of excess fertilisation and a small amount of randomly selected parcels (5-10%). Therefore, it is assumed that an extrapolation of the results of this “non-random” dataset to the whole of Flanders would correspond to an overestimation of the nitrate residues. Despite these limitations, both datasets are considered as a unique and valuable resource to analyse the importance of influencing factors on nitrate residue levels.
For each measurement, additional information was available, such as sampling date, exact location of the parcel, main crop, catch crop (if present) and parcel surface. In addition, other datasets were linked containing climatic conditions, soil conditions (soil type, carbon content, pH), crop rotations, fertilisation limits and data on farm level concerning fertiliser use and manure production. Prior to the statistical analyses, a log-transformation had to be applied on the nitrate residue data in order to meet the statistical requirements of normality and homoscedasticity. The influence of the different parameters on the nitrate residues was then analysed through AN(C)OVA and regression techniques with the Statistica software (Statsoft Inc., 2007).
Nitrogen Workshop 2012
3. Results & Discussion Between both datasets (parcels with AEA Water and control measurements), significant differences in nitrate residue levels exist. On average nitrate residues in AEA Water parcels were lower (-22 kg N ha-1) than nitrate residues in the control measurements. This difference is mainly caused by the applied fertilisation practices. In parcels with AEA Water, fertilisation is generally better tuned to crop needs and as a consequence, effects of other parameters are smaller.
In both datasets, AEA Water parcels and control measurements, the nitrate residues show a significant decrease over the years. This decrease is attributed to a combined effect of stricter fertilisation limits and a gradual adoption of these limits in the farming practice, increased attention of the farmers to manuring practices and a better follow-up of fertilisation advices. Next to the generally decreasing trend in nitrate residues, the crop type is by far the most determining factor.
Grass and fruit trees show significantly the lowest nitrate residues (42-68 kg N ha-1), followed by sugar beets. Leguminosae, potatoes and vegetables give on average the highest nitrate residues (73kg N ha-1). Maize and cereals give intermediate values.
The effect of catch crops was considered per crop type. A catch crop after cereals is particularly important because cereals are harvested relatively early. From the results it appeared that yellow mustard sown as a catch crop after cereals reduced the nitrate residues significantly more than grassy catch crops. This is explained by the slower initial growth of grass. The effect of catch crops reflects in fact the effect of the presence of a crop and of the development (rooting) of this crop. A well developed (main or catch) crop at the moment of sampling absorbs the mineral nitrogen in the soil profile, leaving behind a smaller amount of nitrate. After crop types such as sugar beets and maize, the effect of catch crops on nitrate residues is less pronounced, because these crops have been harvested shortly before or even after the time of sampling and only limited nitrogen mineralisation has taken place since then. The carbon content of the ploughing layer (0-30 cm) has a significant and relatively important effect on the nitrate residues. Moreover, significant interaction effects were found between carbon content, soil pH and soil texture. Nitrate residues increase with increasing soil carbon contents. With higher pH-values and in heavier soil textures (loamy and clayey soils), this increase is more pronounced. Other parameters also had a significant effect on nitrate residues, such as agricultural region, soil type, climatic conditions, and farm type (with or without animals), but the relative importance of these parameters to explain the variation in the datasets was limited.
4. Conclusion The extensive analysis of the available nitrate residue measurements in Flanders since 2001 demonstrates the importance of fertilisation practices and policy measures. Parcels with an AEA Water have to meet stricter fertilisation standards and as a consequence show lower nitrate residue levels. Moreover, in all the measurements a significantly decreasing trend is observed over the years, parallel to the evolution of fertilisation standards becoming stricter and farmers adopting more the principles of good fertilisation practices.
References Statsoft Inc. 2007. Statistica Package 7.1. USA, Tulsa
Farm N balances in European landscapes and the effect of measures to reduce N-losses Dalgaard, T.a, Durand, P.b, Dragosits, U.c, Hutchings, N.J. a, Kedziora, A.d, Bienkowski, J. d, Frumau, A.e, Bleeker, A.e, Magliulo, E. f, Olesen, J.E. a, Hansen, B.g, Thorling, L.g, Theobald M.R. h, Drouet J.L. i, Cellier, P.i a Aarhus University, Dept. Agroecology. P.O. Box 50, DK-8830 Tjele, Denmark.
b INRA-AgroCampus, UMR SAS, Rennes, France.
c CEH, Penicuik, United Kingdom.
d RCAFE, Poznan, Poland.
e ECN, Petten, The Netherlands.
f CNR, Napoli, Italy.
g GEUS, Geological Survey of Denmark and Greenland, Aarhus, Denmark h UPM, Technical University of Madrid, Spain.
i INRA-AgroParisTech, UMR EGC, Thiverval-Grignon, France.
1. Background & Objectives The farm gate N balance is acknowledged as a valid and reliable indicator for potential nitrogen losses from agricultural systems, and the main driver for N pollution from intensive agricultural landscapes (Dalgaard et al., 2011a). Consequently, to investigate potentials for the reduction of Nlosses and greenhouse gas emissions, farm N balances from study landscapes in Denmark, France, Poland, Scotland, The Netherlands and Italy was collected during the www.NitroEurope.eu research project, and the remarkable differences between both the levels of-, and the within landscape heterogeneity were further investigated.
2. Materials & Methods Farm nitrogen balances were collected from study landscapes in Denmark, France, Poland, Scotland, The Netherlands and Italy in year 2007/2008. Based on the local farm data collections from 265 farms within these landscapes, the farm N-balances were calculated on an annual basis for each farm as the difference between N inputs and N outputs (equation 1).
Farm N-surplus = Noutputs - Ninputs= Nproducts - Nfeed - Nfertiliser - Nmanure - Nfixation - Ndeposition  Moreover for Denmark, additional data sets with farm N balances from farms in year 2002, 1996 and 1990 were included and compared to results from the NitroEurope year 2007/2008 campaign.
Nitrogen Workshop 2012 The Danish case study show significant reductions in farm N surpluses from year 1990 to 2008, and with significantly higher reductions at farms with a high livestock density, compared to farms with a low livestock density (Figure 1). This corresponds to the significant nitrate reductions found in groundwater samples, and reductions in the national farm N surplus in Denmark (Hansen et al., 2011), and the effects of measures to reduce N emissions listed by Kronvang et al. (2008) and Dalgaard et al. (2011b).
Nsurp2008 N-surplus (kg N/ha/yr)
Figure 1. Example on the development in farm N surpluses from Danish farms 1990-2008 (Nsurp), and the related exponential correlations, showing a significantly higher reduction at high compared to low livestock density farms.