Abstract: The effects of aggregating soil input data on modelling crop yields at regional scale have been explored within the MACSUR- Crop M – WP3 scaling exercise for an ensemble of crop models 1. The models were run for the North Rhine-Westphalia region in Germany with an average climate time-series (30 years) and soil data at resolution 1 km to 100 km. Aggregation effects showed substantial differences between the models 1. This could be linked to differences in model structure and concepts and to different procedures for the parameterization of soil properties. A further analysis of the sensitivity of the outputs to key soil properties, for each ‘model – method of parameterization’, could help in understanding differences observed within the model ensemble. In this study, we explored the relationship between winter wheat yields, water and N-stress indexes and simple key-soil properties, based on the COUP-model 2 simulations. Soils were grouped into classes according to selected parameters (i.e. soil depth, soil texture and soil organic content). Preliminary results show that some of those soil classes are clearly associated with high water and / or N-stress and lower yields or with high inter-annual variation of the yield. As such they represent key factors explaining the spatial pattern of the simulated yield at the different resolutions. In addition we identified differences in the fractional area of those soil classes between high and low spatial resolutions (‘inherent errors’ due to data aggregation). How this may influence soil data aggregation effects on simulated yields will be further analyzed. No Label

Abstract: Framing scenario selection (RCP/SSP)Ongoing scenario development in FP7 IMPRESSIONSSome examples of sources of data and scenarios No Label

Abstract: The latest Common Agricultural Policy (CAP) 2015 reforms bring a substantial change in the way farm support is paid in Scotland where previous direct CAP payments were largely based on historical entitlements. Under the new payment scheme, three rates of payment are designated based on land uses and capabilities. As a result, it is anticipated that, average large dairy farms will lose out up to 32% of their farm net margins, while small dairy farms will lose out between 7-20% of their farm net margins. Such reductions of payment support may force dairy farmers to cut costs of production on farms especially livestock variable costs including labour costs and costs of prevention, control, treatment and management of livestock diseases and welfare conditions. This will have direct and indirect consequences on health and welfare of dairy cattle. This study aims to assess the impact of new support payments under CAP 2015 reforms on financial capabilities of dairy herds in tackling three conditions namely: infertility, mastitis and lameness. A detailed inventory of 42 commercial dairy farms in Scotland that contains both physical (i.e. farm area, nutrition and labour supply, etc.) and health data collected in 2013 and was used to parameterise an optimisation model. The model is a linear programme (LP) model which optimises farm net margin under limiting farm resources. The model also consists of feed demand and supply components that are used to determine monthly feed requirements for each of the animals on a farm as well as grass yield for pasture area of the land. The model is run for both ‘healthy’ and ‘diseased’ herds under previous and future CAP support payments. Details of the model and the dataset used as well as some results will be presented at the conference. No Label