Combining crop growth models with the Precision Agriculture concept of yield gap analysis to evaluate yield limiting and reducing factors

The agricultural sector is considered as one of the main elements of a “functional” society and its welfare for providing means to fulfil the most basic human need for food. In the past, an improved overall welfare of developed countries was directly related to early investment in conventional farming machinery parallel to the industrial development. With the expected future population increase and climate change, farm yield gaps are expected to increase and become more volatile. The awareness of negative externalities of agricultural production and yield maximisation led to many regulations e.g. in the EU aiming to control agricultural inputs to prevent ground water and environmental pollution. The problems arising from maintaining the balance between socio-economic and environmental aspects of agricultural production while maximising yield are complex and cannot be solved with one single solution. For the studies conducted in this dissertation a simplified yield definition was used with a major focus on yield limiting (nutrient) and yield reducing factors (leaf disease). There are still major knowledge gaps in fully understanding dynamics of potential yield (attainable under fully controlled conditions on e.g. research stations or in controlled environments) and actual yield attained by farmers due to spatial and temporal heterogeneity occurring in the field on large scale crop production. As more insight into soil- and weather-related dynamics was gained with various technological solutions available for agricultural practices, a more detailed investigation of the variabilities affecting yield came about. These technologies include various remote- and near-sensing technologies specialised for measuring various aspects of crop growth and field spatial and temporal variabilities. The measurement range of remote- and near-sensing technologies is able to capture soil×crop×weather dynamics by direct (e.g. N status measuring sensors) and indirect measurement (e.g. measuring plant chlorophyll, multi-spectral reflectance etc.) of soil parameters, plant biomass accumulation (for predicting yield) and weather parameters affecting in-season crop growth. Decision support tools such as process oriented crop growth models (DSSAT-Decision Support System for Agrotechnlogy Transfer) have the potential to capture in-season dynamics and rely on feedback of various sensor data for more accurate depiction of plant growth and yield prediction. Some of the major problems related to crop growth models are the underlying mathematical abstractions of plant phenological development and required calibration of plant genetics. In the future, sensor-based N status can be used for checking the model prediction of plant N uptake and for adjusting further in-season N prescriptions on a site-specific level. Various sensors have also potential to be used for evaluating leaf disease status in the field for predicating yield losses.