Subproject 15

Project recipient: Leibniz-Centre for Agricultural Landscape Research (ZALF) e. V. Müncheberg, Institute for Landscape Systems Analysis, team: Dr. Karl-Otto Wenkel, Dr. Wilfried Mirschel, Dr. Ralf Wieland, Dr. Christian Kersebaum, MSc. Jans Bobert

Subproject title: Development and validation of a procedure for the interactive model-supported generation of yield expectation maps of heterogeneous fields

Subproject short title: Model supported generation of yield expectation maps

Subproject objectives<A[destinations|objectives]>: The development of procedures which facilitate the determination of realistic yield targets for management units within agricultural fields for the forthcoming vegetation period represents a scientific challenge which is not satisfactory solved up to now. In subproject 15 an applicable and scenario-apt<A[apt|suitable]> fuzzy expert system is to be developed, capable to take into account site- and crop production parameters for the derivation of the yield target as well as fuzzy variables like site experience and management considerations. The fuzzy expert system is to be tested scientifically and on farm under practical conditions.

Position of the subproject in the compound: The results of subproject 15 are directly transferred to the units production systems / production management of the compound (subprojects 9 – 11). The produced yield expectation maps will be made available to subproject 3 (efficiency of precision farming) to assist continuing economic analysis.

A close cooperation with subproject 18 (software for farm data management) safeguards the integration of the planning tool into a farm information and management software system.

For its part subproject 15 is dependent on data and information from other subprojects. In particular a cooperation with the subprojects 13 (remote sensing), subproject 14 (integrated site characterization) and subproject 16 (site specific rooting depth) is striven for. Validation and practical testing of the procedure will in addition be carried out with data of and in close cooperation with the project farms.

---

Objectives:

· Development of a new interactively applicable tool for the generation of yield expectation maps of heterogeneous fields for winter wheat and corn

· Experimental testing, quantification and interpretation of possible deviations of generated yield expectation maps from measured yield maps of the project farms

· Iterative validation of the procedure by means of actual data from the project farms

· Comparison of different potentially suitable methods for the generation of yield expectation maps

· Testing of the practical applicability of the procedure together with farmers

C<A[Job|Wonception:

· A procedure<A[methodics|methodology]> which meets the following requirements is to be developed:

· Best possible processing of the current knowledge about<A[about|over]> the interdependences between biomass- and yield formation and climate-, site-, management- and genetic factors

· Processing of incomplete and fuzzy data and information

· flexible, transparent and easily comprehensible

· interactive and spatially<A[spatially|three-dimensional]> applicable

· scenario-apt<A[apt|suitable]> and suitable for risk assessment

· data requirements similar to the generally available data of the project farms

· adjustable onto actual operating conditions on farm and possibilities for direct implementation of empirical knowledge of the farmers

· simple applicability

From contemporary<A[contemporary|present]> point of view a spatially<A[spatially|three-dimensional]> applicable fuzzy expert system is very suitable for these requirements.

The development of the planned ambitious procedure<A[intention|project]> occurs in following steps:

1. Comparative analysis and evaluation of potentially suitable methods for site specific yield potential and – target assessment<A[destination|objective]>

2. Development of a concept for a spatially<A[spatially|three-dimensional]> explicit, scenario-apt<A[apt|suitable]> fuzzy based expert system for interactive generation of yield expectation maps

3. Analysis and aim-oriented processing of current knowledge about<A[about|over]> interdependences between biomass- and yield formation and climate-, site-, management- and genetic factors

4. Build up of a relational knowledge data base

5. Development of algorithms and fuzzy sub models with help of the spatial<A[spatial|three-dimensional]> analysis and modeling tool SAMT, designed in the Institute for Landscape Systems Analysis of the ZALF Muencheberg

6. Programming of sub modules of a spatial<A[spatial|three-dimensional]> fuzzy based expert system for site specific yield expectation including functional testing and test of its stability

7. Scientific testing of the accuracy of the procedure by means of experimental databases from project farms, complemented with data of own experimental investigations

8. Validation of the procedure with data from the compound project preagro I and comparison with results of the procedure "Lisso” by means of data of the farm Reinstedt (Saxony- Anhalt) and/or further long term data sets

9. Documentation and scientific publication of the procedure

This subproject summary was updated on 04.04.2005