Tatjana Krimly

Integrated Modeling of Global Change Impacts on Agriculture and Groundwater Resources

The GLOWA-Danube research cooperation has developed the integrated simulation system DANUBIA to simulate water-related influences of global change in different spatial and temporal contexts. DANUBIA is a modular system comprised of 17 dynamically-coupled, process-based model components and a framework which controls the interaction of these components with respect to space and time. This article describes approaches and capabilities of DANUBIA with regard to the simulation of global change effects on agriculture and groundwater. To the agriculture-groundwater-relation, the direct effects that climate change has on the water balance are just as important as decisions made by land managers about land use and farming intensity. This article provides firstly a brief review of the research efforts which have been undertaken in the field of integrated modeling of agriculture and groundwater under conditions of global change. Then, the DANUBIA simulation framework and the associated DeepActor-framework for simulation of decision-making by agricultural actors are presented together with the model components which are most relevant to the interactions between agriculture and groundwater. The approach for developing combination climate and socio-economic scenarios is explained. Exemplary scenario results are shown for the Upper Danube Catchment in Southern Germany. Finally issues related to integrated simulation of global change effects on agriculture and groundwater are discussed.

DeepActor Models in DANUBIA

This chapter describes the representation of decision processes of socio-economic actors by means of actor models, which is one distinctive feature of DANUBIA. An actor model (also called an agent model) describes socio-economic processes as the sum of the individual actions taken by a range of different actors. The DeepActor framework provides a basis for modelling and implementing the socio-economic DANUBIA models. Decision makers, such as individuals, organisations or businesses, are modelled as “actors”. Each actor is localised within his physical and social environment (a proxel or a social network) and takes decisions as responses to his observations about which action to execute from a range of possible action options. Actors have various preferences and action options, represented by customised plans and decision procedures that are specific to each actor type. Further, actors have a “memory” (history) for recalling previous decisions. One special feature of DANUBIA is the coupling of physically based scientific models with socio-economic components. It is described how the transformation from quantitative states in nature to qualitative notifications for the actor model is realised using the flag concept.

Combining choice modeling estimates and stochastic simulations to assess the potential of new crops-The case of lignocellulosic perennials in Southwestern Germany

In the future, the lignocellulosic perennial crops short rotation coppice (SRC) and miscanthus are supposed to provide renewable raw materials for a bio‐based economy. To assess the potential regional supply of these crops, which are not yet widespread in Baden–Wuerttemberg (Southwest Germany), we used a two‐step approach. In a first step, we conducted a Discrete Choice Experiment (DCE) in regions of Baden–Wuerttemberg that—given their site conditions—are suitable for SRC or miscanthus. The respondents were characterized by significant preference heterogeneity for both (negatively valued) perennial crops and for all presented choice attributes. Thus, it was appropriate to estimate a random parameter logit model (RPL). The attributes average yearly contribution margin, long‐term purchase guarantee and cultivation by colleagues in the neighborhood had a significantly positive effect on the likelihood of cultivation, whereas the attributes contribution margin variability and initial investment need showed a significantly negative effect. In a second step, assuming realistic values for the levels of the attributes considered in the DCE, in stochastic simulations, we randomly draw part‐worth utilities from the multivariate normal distribution of these parameters according to the RPL results. This way, for alternative biomass prices, we derived shares of farmers’ willing to engage in perennial crop production and produced related regional supply functions. Under moderate yield and realistic input and farmland opportunity cost assumptions, the full regional miscanthus potential can only be achieved when farmers are offered either subsidies or price‐risk‐reducing long‐term contracts. Based on empirically determined heterogeneous farmer preferences, our two‐step approach is suitable to yield realistic estimations of any not yet implemented farming practices. We finally note caveats related to our analysis and discuss some policy implications of the major findings.

Effects of Future Climate Trends on Crop Management

To assess the impact of climate change and variability in the weather on the timing of crop management activities, dynamically coupled model runs of the DANUBIA components Biological, SNT, NaturalEnvironment and Farming were performed. Calculations are based on a GLOWA-Danube scenario including the climate trend REMO regional, the climate variant Baseline and the societal scenario Baseline. Results of the scenario calculation are compared with the reference period for four sample districts, which represent different site conditions within the drainage basin. The increase in air temperature over the scenario period leads to a shortening of the growing period of winter wheat and spring barley and, therefore, also for the cultivation periods. With very similar sowing dates, on average, the harvest of winter wheat is earlier by 21 days at the end of the scenario period compared to the reference period for all districts. Furthermore, the harvest dates for winter and spring cereals move closer to each other.

Agricultural Land Use and Drinking Water Demand

The assessment of agricultural water demand and agricultural land use is a common element in quantitative model-based policy analysis. Economic, political and pre-existing climatic and soil conditions shape the structure of agricultural land use which determines drinking water demand and nutrient input from agriculture. These factors vary strongly across regions. A nonlinear process analytical model based on agricultural statistics on district level was used to calculate parameters such as farm incomes, optimum cultivation plan, animal husbandry, quantities of fertiliser and the demand for drinking water at district level. To produce a more detailed picture with a higher spatial resolution, the regional optimization model was combined with an agricultural actor model that simulates farmers’ crop management decisions. An allocation tool was developed to allocate the different farm actors on a spatial unit of 1 × 1 km. For the analysis modelled yield changes from interdisciplinary global change scenarios were used.

Actor Model for Farmers’ Crop Management Decisions: The DeepFarming Model

Plant growth and farmers’ crop management decisions are strongly influenced by the climatic conditions. To simulate the interplay between crop management activities, crop growth and weather and its changes due to long-term development of climate change, the actor model DeepFarming was developed. DeepFarming has a high spatial and temporal resolution and is very closely linked with the agricultural sector model ACRE und the plant growth model Biological. ACRE delivers data on the yearly cultivation plan and Biological on the daily development stage of the plants to the actors. DeepFarming represents 28 different actor types, which were derived from statistical data on types of farm and their management practices at district level and are allocated to the drainage basin using specific rules. Key task of the actors is to make decisions on the timing of the crop management such as sowing, fertilising and harvesting for each crop. Relevant decision parameters are information on the daily weather conditions, soil saturation level and the development stage of the plants. As a result, the initialisation of the actors in the drainage basin is presented.

Land Use and Land Cover

One of the key challenges in Global Change Research is the modelling of future changes in land use and land cover as a result of socio-economic and global climate change. Such changes are calculated dynamically in the Global Change decision support system for the Upper Danube (DANUBIA), thus helping to predict the effects of Global Change in this region. For any such spatially explicit modelling, land use/land cover is the key information layer which integrates all involved natural scientific and socio-economic process models. Land cover characterises the condition of the earth’s surface and hence the properties that most directly influence water and energy fluxes, whereas land use describes the type of anthropogenic use, thus being a key parameter for modelling socio-economic processes. In order to meet all model requirements, a new tailored land use/land cover map was created with a representative overall area approximation and localisation of land use and land cover classes on basin scale, using a combination of CORINE Land Cover (CLC) data, official agricultural statistics and rule-based GIS operations. The resulting land use/land cover map can be used as initial state for modelling in DANUBIA. A total of 27 unique land use/land cover categories are distinguished in DANUBIA. The top hierarchical level is used consistently in all models, whereas the additional levels (e.g. detailed arable land classes) are individually used depending on the information needed by each model.

Effects of Future Climate Changes on Yields, Land Use and Agricultural Incomes

Dynamically coupled model runs of the DANUBIA components Biological, SNT, NaturalEnvironment and Farming were performed to estimate the effects of climate change on crop yields, agricultural land use and income. Calculations are based on a GLOWA-Danube scenario including the climate trend REMO regional, the climate variant Baseline and the societal scenario Baseline. Results of the scenario calculation are compared with the reference period for four sample districts, which represent different site conditions within the drainage basin. In general, the scenario results show an increase in yields for the considered groups of crops. However, changes of individual crops within these groups differ between the districts. All districts have an increase in income that at the beginning of the scenario period is mainly caused by the increase in premium payments of the CAP compared to the reference period. The further income increase at the end of the scenario period, which is significantly higher in districts with a higher proportion of arable land, can be attributed to the increase in yields. With respect to land use, all districts show a decrease in forage crops and an increase in the cultivation of cereals. Overall, the results indicate that no negative impacts on the productivity of the agricultural land and the income situation of the farms are to be expected up to the middle of the century.