Abstract: Communication is the key link between the generation of information by MACSUR about the uncertainty of climate change impacts on future food security and how information is used by decision makers. It is therefore important to make available the common tools for reporting uncertainty, with a discussion of the advantages or difficulties of each. That is the purpose of this report. No Label

Abstract: Climate variability (CV) normally influences production and farm management, and climate change (CC) has precisely the effect of changing this variability. Thus, models that estimate the economic impact of CC, integrating with climatic models, agronomic, and livestock, must represent the implications of this variability on farm management. This study describes an economic model based on Discrete Stochastic Programming (DSP) which assesses the impact of CC on milk production in the Grana Padano area. The model is based on 23 farm typologies from FADN that represent 856 farms in Piacenza and Cremona, two of the most important provinces for Grana Padano production. The results of the model were projected at the regional scale. The climate scenarios, current and future, are generated with a Regional Atmospheric Modeling System. The forage production under these scenarios is estimated with the EPIC agronomic model. Estimates on milk production and livestock mortality are based on studies conducted in the Po valley. The nutritional needs of the cattle are estimated with the CNCPS model. Probability distribution functions (PDF) express the relations between the CV and the productive variables under both climate scenarios. These PDFs represent the expectations of farmers on the productive-climate variability in the DSP model, which is PMP calibrated based on land distribution observed in a reference year. Comparing the model results in the two scenarios indicates the effects of CC, given the opportunity to adapt the use of resources and techniques of cultivation. The structure of the model, and its economic results are presented and discussed, along with the strengths and weaknesses of this approach. No Label

Abstract: Climate change suggests long periods without rainfall will occur in the future quite often. Previous approach on dependence crop-yields from size of rain confirms the existence of a statistically significant relation. We built a model describing the amount of precipitation and taking into account periods of drought, using a mixture of gamma distribution and one point-distribution. Parameter estimators were constructed from rainfall data using the method of maximum likelihood. Long series of days or decades of drought allow to determine the probabilities of adverse developments in agriculture as the basis for forecasting crop yields in the future (years 2030, 2050). Forecasted yields can be used for assessment of productivity and profitability of some selected crops in Kujavian-Pomeranian region. Assumptions and parameters of large-scale spatial economic models will be applied to build up relevant solutions. Calculated with this approach output could be useful to expect decrease in agricultural output in the region. It will enable to shape effective agricultural policy to know how to balance food supply and demand through appropriate managing with stored food raw material and/or import/export policies. Used precipitation-yields dependencies method let verify earlier used methodology through comparison of obtained solutions concerning forecasted yields and closed to it uncertainty analysis.This work was co-financed by NCBiR, Contract no. FACCE JPI/04/2012 – P100 PARTNER No Label