Abstract: Climate and socio-economic change will affect the land use and the economic viability of Dutch dairy farms. Explorations of future scenarios, which include different drivers and impacts, are needed to perform ex-ante policy assessment. This study uses a bio-economic farm model to assess impacts of climate and socio-economic change on dairy farms in a sandy area in the Netherlands. Farm data from the Farm Accountancy Data Network provided information on the current production levels and available farm resources. First, the farm plans of individual farms were optimized in the current situation to benchmark farms and assess the current scope for improvement. Secondly, simulations for two scenarios were included: a Global Economy with 2 °C global temperature rise (GE/W+) and a Regional Community with 1 °C global temperature rise (RC/G). The impacts of climate change, extreme events, juridical change (including abolishment of milk quota), technological change and price changes were evaluated in separate model runs within the predefined scenarios. We found that farms can increase profit both by intensification and land expansion; the latter especially for medium sized farms (less than 70 cows). Climate change including the effect of increased occurrence of extreme events may negatively affect farm gross margin in the GE/W+ scenario. Lower gross margins are compensated for by the effects of technology and price changes. In contrast with the GE/W+ scenario, climate change has positive impacts on farm profit in RC/G, but less favourable farm input-output price ratios have a negative effect. Technological change is needed to compensate for revenue losses due to higher input prices. In both GE/W+ and RC/G scenarios, dairy farms increase production and the use of artificial fertilizer. Medium sized farms have more options to increase profit than the large farms: they benefit more from the abolishment of the milk quota and better adapt to negative and positive impacts of climate change. While the exact impact of different drivers will always remain uncertain, this study showed that changes in prices, technology and markets have a relatively larger impact than climate change, even when extreme events are taken into account. By using whole farm plans as activities that can be selected, the model is grounded in observations, and it was shown that half of the farms are gross margin maximizers as assumed in the model. The model therefore indicates ‘what could happen if’, and gives insights in drivers and impacts of dairy farming in the region.

Abstract: As modeling becomes a more widespread practice in the agro-environmental sciences, scientists need reliable tools to calibrate models against ever more complex and detailed data. We present a generic Bayesian computation framework for grassland simulation, which enables parameter estimation in the Bayesian formalism by using Monte Carlo approaches. We outline the underlying rationale, discuss the computational issues, and provide results from an application of the Pasture Simulation model (PaSim) to three European grasslands. The framework was suited to investigate the challenging problem of calibrating complex biophysical models to data from altered scenarios generated by precipitation reduction (water stress conditions). It was used to infer the parameters of manipulated grassland systems and to assess the gain in uncertainty reduction by updating parameter distributions using measurements of the output variables.

Abstract: Eddy covariance data from four European grassland sites are used to probabilistically invert the CARAIB (CARbon Assimilation In the Biosphere) dynamic vegetation model (DVM) with 10 unknown parameters, using the DREAM((ZS)) (DiffeRential Evolution Adaptive Metropolis) Markov chain Monte Carlo (MCMC) sampler. We focus on comparing model inversions, considering both homoscedastic and heteroscedastic eddy covariance residual errors, with variances either fixed a priori or jointly inferred together with the model parameters. Agreements between measured and simulated data during calibration are comparable with previous studies, with root mean square errors (RMSEs) of simulated daily gross primary productivity (GPP), ecosystem respiration (RECO) and evapotranspiration (ET) ranging from 1.73 to 2.19, 1.04 to 1.56 g C m(-2) day(-1) and 0.50 to 1.28 mm day(-1), respectively. For the calibration period, using a homoscedastic eddy covariance residual error model resulted in a better agreement between measured and modelled data than using a heteroscedastic residual error model. However, a model validation experiment showed that CARAIB models calibrated considering heteroscedastic residual errors perform better. Posterior parameter distributions derived from using a heteroscedastic model of the residuals thus appear to be more robust. This is the case even though the classical linear heteroscedastic error model assumed herein did not fully remove heteroscedasticity of the GPP residuals. Despite the fact that the calibrated model is generally capable of fitting the data within measurement errors, systematic bias in the model simulations are observed. These are likely due to model inadequacies such as shortcomings in the photosynthesis modelling. Besides the residual error treatment, differences between model parameter posterior distributions among the four grassland sites are also investigated. It is shown that the marginal distributions of the specific leaf area and characteristic mortality time parameters can be explained by site-specific ecophysiological characteristics.