|
|
Janssen, S. (2015). Inventory of data and data sharing mechanism for model linking and scaling exercises (Vol. 6).
Abstract: This deliverable lays out the work as done as part of MACSUR CropM on ‘Inventory of data and data sharing mechanism for model linking and scaling exercises’. In summary not much work was done, as it was found that there was not real demand for the activity in this task. The task in itself was servicing the other work as part of MACSUR, and as the service was not in demand, it was decided to take a low profile and wait for specific requests by partners for data in relation to model linking and upscaling. No Label
|
|
|
|
Janssen, S., Hansen, J. G., Jorgensen, J., & Jørgensen, M. S. (2015). Operational database for storing and extracting data (Vol. 6).
Abstract: This deliverable lays out the work as done as part of MACSUR CropM on data, with the focus on improving data management and have shared data curation for future use. The issue was tackled with help from the MACSUR central hub coordination in the form of Jason Jargenson from University of Reading. The data management as proposed and implemented in this deliverable is very much a bottom up process, in which partners in a meeting in Spring 2013 in Aarhus investigated the best way forward for data management across activities in CropM.As a follow up to this, the work was mainly divided in three parts: 1. The Open Data Journal for Agricultural Research, mainly focused on long term data archival and citation of data sets, as input and outputs to the modelling work, as part of MACSUR, lead by Wageningen UR 2. The Geonetwork data catalog hosted at Aarhus Universitet, that allows for operational access and storage of data sets as part of the ongoing work, also for restricted access of the consortium, and as a first step to visualization, lead by Aarhus Universitet. 3. The work on rating data sets, that provides a tool for improving data set access in an early phase for connecting them to models, lead by Reading University. At the end of the deliverable some next steps are giving for data activities in the context of AgMIP and beyond. No Label
|
|
|
|
Janssen, S., Houtkamp, J., De Groot, H., & Schils, R. (2015). Online web tool for data visualization (Vol. 6).
Abstract: This deliverable lays out the work as done as part of MACSUR CropM on data, with the focus on providing a web tool for visualization of model output. It was decided early on that not a specific MACSUR web tool would be developed as part of MACSUR for phase 1, and mostly results would be visualized in other available tools, such as the Global Yield Gap Atlas, which are recognised resources for visualizations. Only in relationship to the MACSUR Geonetwork data catalog hosted at Aarhus University some developments where started. Operationally speaking, most data was still being generated during phase 1, so there was not enough to visualize on specific websites and partners did not commit financial resources to their development, and only in kind was available. No Label
|
|
|
|
Jorgenson, J. (2013). Review of Cloud Computing Opportunities (Vol. 1).
Abstract: This paper will begin by defining some of the challenges that we face on the MACSUR project in terms of evaluating model uncertainty and carrying out model integration. I will briefly review what cloud technologies are available, followed with some suggestions about how those cloud technologies can be used in order to contribute to meeting the challenges set out in the first part of the paper.’Month 12’ deliverable for WP1 is a review of the opportunities for using cloud computing to develop the potential for model inter-comparison and interlinking in MACSUR. A challenging aspect of compiling this review is that before an ‘opportunity’ for any kind of model linking/comparison can be identified, a lot of information about the specifics of extant models and workflows must be gathered from each of the three themes (TradeM, CropM, and LiveM).This deliverable must, however, be more than just saying ‘these are the computing tools that we can use to.’. There are a number of different challenges at different levels; a hierarchy of challenges, if you like. For example, in order to get models ‘talking’ to one another, adequate protocols for the transference of data and scaleability will need to be established, and then things like uncertainty analysis for these integrated models will need to be addressed. Further issues exist relating to human behaviour and logistics (e.g. MACSUR is a large project with many members from all over Europe, with substantial distances between many of it’s members).The term “Cloud” is very ambiguous, and Cloud Computing covers a huge range of services, and a number of innovative tools exist which can make international collaborative research more effective. Two examples (already implemented on the MACSUR website) are: a discussion forum (where project members can create topics, make or reply to posts, and upload documents) and a complete surveying platform (to provide an un-restricted and fully featured survey platform for MACSUR members’ information gathering needs.) No Label
|
|
|
|
Kässi, P., Niskanen, O., & Känkänen, H. (2014). Farm level approach to manage grass yield variation in changing climate in Jokioinen and St. Petersburg (Vol. 3).
Abstract: Cattle’s feeding is based on grass silage in Northern Europe, but grass growth is highly dependent on weather conditions. In farms decision making, grass area is usually determined by the variation of yield. To be adequate in every situation, the lowest expected yield level determines the cultivated area. Other way to manage the grass yield risk is to increase silage storage capacity over annual consumption. Variation of grass yield in climate data from years 1961-1990 was compared with 15 different climate scenario models simulating years 2046-2065. A model was developed for evaluating the inadequacy risk in terms of cultivated area and storing capacity. The cost of risk is presented and discussed.In northern Europe a typical farm has storage for roughage consumption of almost one year. In addition, there can be a buffer storage. The extra storage is to be used before and during the harvest season. New harvest will be fed to animals only after the buffer empty. Shortage in the buffer storage is possible to be filled, when the yield exceeds the target level. For risk management, two alternative mechanisms are given: forage buffer and possibility to alter the field area.According to our results, there are no significant adverse effects in the cost of risk and implied farm profitability due to climate change. Selecting the risk management scenario of 30 % grass yield risk turned out to be the least cost solution. No Label
|
|
