modelling and analysis
In times of climate change and increasingly limited resources, Austrian agriculture is caught between economic efficiency and ecologically sustainable production. In addition to general questions on forms of agricultural land management, changing framework conditions and agricultural policy measures are also of importance. In Austria, a wide range of measures under the ÖPUL agri-environmental program promote environmentally friendly farming. There are numerous interactions between the different economic and ecological sub-areas of the agricultural, environmental and food systems, which can result in conflicts of objectives but also synergy effects.
In addition to theoretical, conceptual considerations, quantitative methods can also be helpful in demonstrating these complex interrelationships. For example, mathematical models for the simulation of plant growth (crop models) depict biophysical processes such as biomass production, nutrient cycles, soil water balance and erosion. With the help of model scenarios and depending on the site conditions, relationships between plant yields and management measures as well as the environmental effects of cultivation can be analyzed. In addition to the complex and data-intensive crop models, there are a number of other quantitative methods with varying degrees of detail, data requirements and explanatory potential.
Objective
The aim of the project is to develop an analysis tool for investigating the negative and positive environmental effects of agricultural land use and, conversely, the effects of current and expected future ecological conditions on agricultural land use. This tool will consist of a selection of quantitative models and the associated database. In future, the tool will serve as a valuable basis for decision-making on a wide range of issues in the agri-environmental sector. Selected findings will also be published in specialist journals.
Status of the project
Data on site conditions for agricultural land use, such as natural conditions like topography, soil properties and meteorological data, as well as on management practices such as crop rotation, irrigation, tillage and the use of fertilisers and plant protection products, were taken from a wide variety of data sources (including the BMLUK geodata catalogue, Geo-sphere Austria, soil map) and collated on a location-specific basis. In addition, remote sensing data from the JRC on soil moisture is included in the analyses. Yield data from accounting farms (FADN) is available for calibration and validation purposes.
As part of the ‘Biophysical Processes’ project, some of the work steps were taken over by the ICT staff unit. This mainly involved preparatory work in the areas of remote sensing, spatial analysis and climate data: a survey was conducted to determine which data portals provide raw data, processed data and indices from satellite data. A GIS analysis located areas of deforestation that are in close proximity to the measurement networks of climate stations, soil profiles and groundwater levels in order to analyse the associated influences on model accuracy. The third step involved comparing national and international sources of climate data. Ultimately, the INCA data from Geosphere Austria was determined to be the most suitable for the project purpose.
In the area of methodology, research was conducted on the various modelling methods. The large amount of data required for process-based crop models, which can be difficult to obtain in some cases, can theoretically lead to very detailed analyses, but in practice it is often a source of uncertainty. In contrast, the statistical and machine learning models used in this area are usually less data-intensive and more flexible in terms of data requirements, but are often less detailed and designed as a black box, i.e. without replicating causal relationships. Using the database, a process-based model was parameterised and calibrated, and several machine learning models (Elastic Net, MARS, Random Forest, Extreme Gradient Boosted Trees) were tuned and trained.
The first application examined the suitability of the various models for analysing the variance of quantity yields. Two papers on this topic were submitted and accepted at the ICROPM Symposium 2026. In the first paper, several machine learning models were trained at three different aggregation levels with three different sets of explanatory variables (management, biophysical variables, all variables). The results were supplemented and interpreted by explanatory models based on Shapley values (SHAP, dependence plots). The second paper used the work from the work package ‘Yield model for financial soil estimation’ (see following paragraph) and deals with the performance of crop, ML and hybrid yield models under different meteorological conditions.
Work package: Yield model for financial soil assessment (2025)
In financial soil assessment, the influence of site conditions on yield levels plays a significant role. For arable land assessment, so-called soil climate curves have previously been used to determine a reduction or increase in yield capacity depending on soil type, precipitation and heat conditions.
At the beginning of 2025, the recalculation of the surcharges and discounts was started in cooperation with the Federal Ministry of Finance. The creation of the database was very time-consuming: yield data and soil data at the test sites were processed and combined. Meteorological data from four different data sources were reviewed, processed and tested. The methodological implementation was also very extensive: on the one hand, the ideas behind the financial soil assessment were very clearly outlined; due to the tight time frame, it was also necessary to fall back on existing data. On the other hand, it was not possible to reproduce the previous so-called ‘soil climate curves’, for which no documentation on the methodology is available, on the basis of the existing data. Several models based on different methodological approaches (biophysical, statistical models, machine learning) were therefore developed, validated and compared. A final proposal from the BAB was presented to the BMF at the end of 2025.
Work in 2026
The contributions for the ICROPM Symposium 2026 will be worked out in detail. Findings from the symposium will be incorporated into further project development. The yield data set created by AGES as part of the ‘Yield Model for Financial Soil Assessment’ working pact will form the basis for a joint publication analysing yield fluctuations.
The analysis tool developed can be used to investigate a wide variety of relationships between agricultural land use and environmental effects, such as the impact of regionally increased drought on yields, growth periods and soil conditions. The possibility of using remote sensing to close data gaps and improve the accuracy of the models will also be investigated. In addition, synergies with other BAB projects at the interface between agriculture, the environment and modelling will be exploited. Regular information exchange to promote cooperation takes place within the framework of the BAB working group ‘Methods and Modelling’.
Timetable
Project start: 01/2020
Project end: 12/2026
This text has been automatically translated with www.DeepL.com/Translator.
