Abstract: Highlights • A framework to study climate and policy uncertainty in agriculture is presented. • Combining both sources of uncertainty has ambiguous effects on agriculture. • Uncertainty needs to be highlighted in modelling tools for policy analysis. Abstract Farmers are exposed to climate change and uncertainty about how that change will develop. As farm incomes, in Norway and elsewhere, greatly depend on government subsidies, the risk of a policy change constitutes an additional uncertainty source. Hence, climate and policy uncertainty could substantially impact agricultural production and farm income. However, these sources of uncertainty have, so far, rarely been combined in food production analyses. The aim of this study was to determine the effects of a combination of policy and climate uncertainty on agricultural production, land use, and social welfare in Norway. Output yield distributions of spring wheat and timothy, a major forage grass, from simulations with the weather-driven crop models, CSM-CERES-Wheat and, LINGRA, were processed in the a stochastic version Jordmod, a price-endogenous spatial economic sector model of the Norwegian agriculture. To account for potential effects of climate uncertainty within a given future greenhouse gas emission scenario on farm profitability, effects on conditions that represented the projected climate for 2050 under the emission scenario A1B from the 4th assessment report of the Intergovernmental Panel on Climate Change and four Global Climate Models (GCM) was investigated. The uncertainty about the level of payment rates at the time farmers make their management decisions was handled by varying the distribution of payment rates applied in the Jordmod model. These changes were based on the change in the overall level of agricultural support in the past. Three uncertainty scenarios were developed and tested: one with climate change uncertainty, another with payment rate uncertainty, and a third where both types of uncertainty were combined. The three scenarios were compared with results from a deterministic scenario where crop yields and payment rates were constant. Climate change resulted in on average 9% lower cereal production, unchanged grass production and more volatile crop yield as well as 4% higher farm incomes on average compared to the deterministic scenario. The scenario with a combination of climate change and policy uncertainty increased the mean farm income more than a scenario with only one source of uncertainty. On the other hand, land use and farm labour were negatively affected under these conditions compared to the deterministic case. Highlighting the potential influence of climate change and policy uncertainty on the performance of the farm sector our results underline the potential error in neglecting either of these two uncertainties in studies of agricultural production, land use and welfare.

Abstract: Highlights • Regularization techniques for regression outperformed the classical regression techniques in predicting crop yields. • Different regression techniques with similar prediction accuracy showed different responses of major climatic variables to crop yield. • The regression models showed some responses of crop yield to climatic conditions that is mostly absent in process based crop models. Abstract Statistical regression models represent alternatives to process-based dynamic models for predicting the response of crop yields to variation in climatic conditions. Regression models can be used to quantify the effect of change in temperature and precipitation on yields. However, it is difficult to identify the most relevant input variables that should be included in regression models due to the high number of candidate variables and to their correlations. This paper compares several regression techniques for modeling response of winter oilseed rape yield to a high number of correlated input variables. Several statistical regression methods were fitted to a dataset including 689 observations of winter oilseed rape yield from replicated field experiments conducted in 239 sites in Denmark, covering nearly all regions of the country from 1992 to 2013. Regression methods were compared by cross-validation. The regression methods leading to the most accurate yield predictions were Lasso and Elastic Net, and the least accurate methods were ordinary least squares and stepwise regression. Partial least squares and ridge regression methods gave intermediate results. The estimated relative yield change for a +1°C temperature increase during flowering was estimated to range between 0 and +6 %, depending on choice of regression method. Precipitation was found to have an adverse effect on yield during autumn and winter. It was estimated that an increase in precipitation of +1 mm/day would result in a relative yield change ranging from 0 to −4 %. Soil type was also important for crop yields with lower yields on sandy soils compared to loamy soils. Later sowing was found to result in increased crop yield. The estimated effect of climate on yield was highly sensitive to the chosen regression method. Regression models showing similar performance led in some cases to different conclusions with respect to effect of temperature and precipitation. Hence, it is recommended to apply an ensemble of regression models, in order to account for the sensitivity of the data driven models for projecting crop yield under climate change.